Search

Many complex physical systems exhibit a rich variety of discrete behavioural modes. Often, the system complexity limits the applicability of standard modelling tools. Hence, understanding the underlying physics of different behaviours and distinguishing between them is challenging. Although traditional machine learning techniques could predict and classify behaviour well, typically they do not provide any meaningful insight into the underlying physics of the system. In this paper we present a novel method for extracting physically meaningful clusters of discrete behaviour from limited experimental observations. This method obtains a set of physically plausible functions that both facilitate behavioural clustering and aid in system understanding. We demonstrate the approach on the V-shaped falling paper system, a new falling paper type system that exhibits four distinct behavioural modes depending on a few morphological parameters. Using just 49 experimental observations, the method discovered a set of candidate functions that distinguish behaviours with an error of 2.04%, while also aiding insight into the physical phenomena driving each behaviour. © 2019 Howison et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

It is important to implement detection and assessment of plant diseases based on remotely sensed data for disease monitoring and control. Hyperspectral data of healthy leaves, leaves in incubation period and leaves in diseased period of wheat stripe rust and wheat leaf rust were collected under in-field conditions using a black-paper-based measuring method developed in this study. After data preprocessing, the models to identify the diseases were built using distinguished partial least squares (DPLS) and support vector machine (SVM), and the disease severity inversion models of stripe rust and the disease severity inversion models of leaf rust were built using quantitative partial least squares (QPLS) and support vector regression (SVR). All the models were validated by using leave-one-out cross validation and external validation. The diseases could be discriminated using both distinguished partial least squares and support vector machine with the accuracies of more than 99%. For each wheat rust, disease severity levels were accurately retrieved using both the optimal QPLS models and the optimal SVR models with the coefficients of determination (R2) of more than 0.90 and the root mean square errors (RMSE) of less than 0.15. The results demonstrated that identification and severity evaluation of stripe rust and leaf rust at the leaf level could be implemented based on the hyperspectral data acquired using the developed method. A scientific basis was provided for implementing disease monitoring by using aerial and space remote sensing technologies.

The soybean cyst nematode (SCN), Heterodera glycines, is the most damaging pathogen of soybeans in the United States. To assess the severity of nematode infestations in the field, SCN egg population densities are determined. Cysts (dead females) of the nematode must be extracted from soil samples and then ground to extract the eggs within. Sucrose centrifugation commonly is used to separate debris from suspensions of extracted nematode eggs. We present a method using OptiPrep as a density gradient medium with improved separation and recovery of extracted eggs compared to the sucrose centrifugation technique. Also, computerized methods were developed to automate the identification and counting of nematode eggs from the processed samples. In one approach, a high-resolution scanner was used to take static images of extracted eggs and debris on filter papers, and a deep learning network was trained to identify and count the eggs among the debris. In the second approach, a lensless imaging setup was developed using off-the-shelf components, and the processed egg samples were passed through a microfluidic flow chip made from double-sided adhesive tape. Holographic videos were recorded of the passing eggs and debris, and the videos were reconstructed and processed by custom software program to obtain egg counts. The performance of the software programs for egg counting was characterized with SCN-infested soil collected from two farms, and the results using these methods were compared with those obtained through manual counting.

Background Neglected tropical diseases (NTDs) primarily affect the poorest populations, often living in remote, rural areas, urban slums or conflict zones. Arboviruses are a significant NTD category spread by mosquitoes. Dengue, Chikungunya, and Zika are three arboviruses that affect a large proportion of the population in Latin and South America. The clinical diagnosis of these arboviral diseases is a difficult task due to the concurrent circulation of several arboviruses which present similar symptoms, inaccurate serologic tests resulting from cross-reaction and co-infection with other arboviruses. Objective The goal of this paper is to present evidence on the state of the art of studies investigating the automatic classification of arboviral diseases to support clinical diagnosis based on Machine Learning (ML) and Deep Learning (DL) models. Method We carried out a Systematic Literature Review (SLR) in which Google Scholar was searched to identify key papers on the topic. From an initial 963 records (956 from string-based search and seven from a single backward snowballing procedure), only 15 relevant papers were identified. Results Results show that current research is focused on the binary classification of Dengue, primarily using tree-based ML algorithms. Only one paper was identified using DL. Five papers presented solutions for multi-class problems, covering Dengue (and its variants) and Chikungunya. No papers were identified that investigated models to differentiate between Dengue, Chikungunya, and Zika. Conclusions The use of an efficient clinical decision support system for arboviral diseases can improve the quality of the entire clinical process, thus increasing the accuracy of the diagnosis and the associated treatment. It should help physicians in their decision-making process and, consequently, improve the use of resources and the patient’s quality of life., Author summary Neglected tropical diseases (NTDs) primarily affect the poorest populations, often living in remote, rural areas, urban slums or conflict zones. Arboviruses are a significant NTD category spread by mosquitoes. Dengue, Chikungunya, and Zika are three arboviruses that affect a large proportion of the population in Latin and South America. The clinical diagnosis of these arboviral diseases is a difficult task due to the concurrent circulation of several arboviruses which present similar symptoms and, sometimes, inaccurate test results. In this paper, we present the state of the art of studies investigating the automatic classification of arboviral diseases based on Machine Learning (ML) and Deep Learning (DL) models. Results show that current research is focused on the classification of Dengue, primarily using tree-based ML algorithms. The use of an efficient clinical decision support system for arboviral diseases can improve the quality of the entire clinical process, thus increasing the accuracy of the diagnosis and the associated treatment. It should help physicians in their decision-making process and, consequently, improve the use of resources and the patient’s quality of life.

Recurrent neural networks with memory and attention mechanisms are widely used in natural language processing because they can capture short and long term sequential information for diverse tasks. We propose an integrated deep learning model for microbial DNA sequence data, which exploits convolutional neural networks, recurrent neural networks, and attention mechanisms to predict taxonomic classifications and sample-associated attributes, such as the relationship between the microbiome and host phenotype, on the read/sequence level. In this paper, we develop this novel deep learning approach and evaluate its application to amplicon sequences. We apply our approach to short DNA reads and full sequences of 16S ribosomal RNA (rRNA) marker genes, which identify the heterogeneity of a microbial community sample. We demonstrate that our implementation of a novel attention-based deep network architecture, Read2Pheno, achieves read-level phenotypic prediction. Training Read2Pheno models will encode sequences (reads) into dense, meaningful representations: learned embedded vectors output from the intermediate layer of the network model, which can provide biological insight when visualized. The attention layer of Read2Pheno models can also automatically identify nucleotide regions in reads/sequences which are particularly informative for classification. As such, this novel approach can avoid pre/post-processing and manual interpretation required with conventional approaches to microbiome sequence classification. We further show, as proof-of-concept, that aggregating read-level information can robustly predict microbial community properties, host phenotype, and taxonomic classification, with performance at least comparable to conventional approaches. An implementation of the attention-based deep learning network is available at https://github.com/EESI/sequence_attention (a python package) and https://github.com/EESI/seq2att (a command line tool)., Author summary Microbiomes are communities of microscopic organisms found everywhere, including on and in the human body. For example, the gut microbiome plays an important role in digestion, and changes in composition are associated with changes in health or disease, e.g., inflammatory bowel disease (IBD). Today, microbiome composition is often obtained from high-throughput sequencing, which generates many short DNA reads from multiple organisms in a sample. In this paper, we present a novel deep learning framework, Read2Pheno, to predict phenotype from all the reads in a set of biological samples. An attention mechanism allows visualization of specific subregions (sets of bases) which are important in classifying the reads according to phenotype or taxon labels. We evaluate the framework on sequencing data for 16S rRNA genes, genetic markers used to identify microbial taxonomy. We show that Read2Pheno performs comparably as conventional methods on three distinct data sets from the American Gut Project, IBD patients and controls, and a comprehensive taxonomic database. Moreover, Read2Pheno results can be readily interpreted—e.g., to identify regions of the 16S rRNA gene to target for PCR diagnostics—without additional pre/post-processing steps that can introduce complexity and error.

Overfitting is one of the critical problems in developing models by machine learning. With machine learning becoming an essential technology in computational biology, we must include training about overfitting in all courses that introduce this technology to students and practitioners. We here propose a hands-on training for overfitting that is suitable for introductory level courses and can be carried out on its own or embedded within any data science course. We use workflow-based design of machine learning pipelines, experimentation-based teaching, and hands-on approach that focuses on concepts rather than underlying mathematics. We here detail the data analysis workflows we use in training and motivate them from the viewpoint of teaching goals. Our proposed approach relies on Orange, an open-source data science toolbox that combines data visualization and machine learning, and that is tailored for education in machine learning and explorative data analysis., Author summary Every teacher strives for an a-ha moment, a sudden revelation by the student who gained a fundamental insight she will always remember. In the past years, authors of this paper have been tailoring their courses in machine learning to include material that could lead students to such discoveries. We aim to expose machine learning to practitioners–not only computer scientists but also molecular biologists and students of biomedicine, that is, the end-users of bioinformatics’ computational approaches. In this article, we lay out a course that aims to teach about overfitting, one of the key concepts in machine learning that needs to be understood, mastered, and avoided in data science applications. We propose a hands-on approach that uses an open-source workflow-based data science toolbox that combines data visualization and machine learning. In the proposed training about overfitting, we first deceive the students, then expose the problem, and finally challenge them to find the solution. In the paper, we present three lessons in overfitting and associated data analysis workflows and motivate the use of introduced computation methods by relating them to concepts conveyed by instructors.

Breast cancer is the cancer with the highest incidence of malignant tumors in women, which seriously endangers women’s health. With the help of computer vision technology, it has important application value to automatically classify pathological tissue images to assist doctors in rapid and accurate diagnosis. Breast pathological tissue images have complex and diverse characteristics, and the medical data set of breast pathological tissue images is small, which makes it difficult to automatically classify breast pathological tissues. In recent years, most of the researches have focused on the simple binary classification of benign and malignant, which cannot meet the actual needs for classification of pathological tissues. Therefore, based on deep convolutional neural network, model ensembleing, transfer learning, feature fusion technology, this paper designs an eight-class classification breast pathology diagnosis model BCDnet. A user inputs the patient’s breast pathological tissue image, and the model can automatically determine what the disease is (Adenosis, Fibroadenoma, Tubular Adenoma, Phyllodes Tumor, Ductal Carcinoma, Lobular Carcinoma, Mucinous Carcinoma or Papillary Carcinoma). The model uses the VGG16 convolution base and Resnet50 convolution base as the parallel convolution base of the model. Two convolutional bases (VGG16 convolutional base and Resnet50 convolutional base) obtain breast tissue image features from different fields of view. After the information output by the fully connected layer of the two convolutional bases is fused, it is classified and output by the SoftMax function. The model experiment uses the publicly available BreaKHis data set. The number of samples of each class in the data set is extremely unevenly distributed. Compared with the binary classification, the number of samples in each class of the eight-class classification is also smaller. Therefore, the image segmentation method is used to expand the data set and the non-repeated random cropping method is used to balance the data set. Based on the balanced data set and the unbalanced data set, the BCDnet model, the pre-trained model Resnet50+ fine-tuning, and the pre-trained model VGG16+ fine-tuning are used for multiple comparison experiments. In the comparison experiment, the BCDnet model performed outstandingly, and the correct recognition rate of the eight-class classification model is higher than 98%. The results show that the model proposed in this paper and the method of improving the data set are reasonable and effective.

Fin-tech is an emerging field, inspiring revolutionary innovations in the financial field. It may initiate the evolutionary episode of the financial research, where volatility forecasting is a crucial topic in finance. For forecasting volatility, GARCH model is a prevailing model, however, further improvement of the GARCH model is still challenging. In this paper, we demonstrate how Fintech can play a part in volatility forecasting by employing a metaheuristic procedure called Genetic Programming. On the basis, we are able to develop a new volatility forecasting model, which can beat GARCH family models (including GARCH, IGARCH and TGARCH models) in a significant way. Since genetic programming is an evolutionary algorithm based on the principles of natural selection, this innovative work will be a breakthrough point in the financial area. The innovation of this paper demonstrates how GP technology can be applied in the financial field, attempting to explore the volatility forecasting area from the combination of new technology and finance, known as fintech. More importantly, when the formula of volatility forecasting is unknown as we introduce a new factor, namely, the liquidity factor, we unveil that how GP method can be helpful in determining the specific volatility forecasting model format. We thereby exhibit the liquidity effects on volatility forecasting filed from the fintech perspective.

Phoneme pronunciations are usually considered as basic skills for learning a foreign language. Practicing the pronunciations in a computer-assisted way is helpful in a self-directed or long-distance learning environment. Recent researches indicate that machine learning is a promising method to build high-performance computer-assisted pronunciation training modalities. Many data-driven classifying models, such as support vector machines, back-propagation networks, deep neural networks and convolutional neural networks, are increasingly widely used for it. Yet, the acoustic waveforms of phoneme are essentially modulated from the base vibrations of vocal cords, and this fact somehow makes the predictors collinear, distorting the classifying models. A commonly-used solution to address this issue is to suppressing the collinearity of predictors via partial least square regressing algorithm. It allows to obtain high-quality predictor weighting results via predictor relationship analysis. However, as a linear regressor, the classifiers of this type possess very simple topology structures, constraining the universality of the regressors. For this issue, this paper presents an heterogeneous phoneme recognition framework which can further benefit the phoneme pronunciation diagnostic tasks by combining the partial least square with support vector machines. A French phoneme data set containing 4830 samples is established for the evaluation experiments. The experiments of this paper demonstrates that the new method improves the accuracy performance of the phoneme classifiers by 0.21 - 8.47% comparing to state-of-the-arts with different data training data density.

Artificial intelligence (AI) is emerging as a technology at the center of many political, economic, and societal debates. This paper formulates a new AI patent search strategy and applies this to provide a landscape analysis of AI innovation dynamics and technology evolution. The paper uses patent analyses, network analyses, and source path link count algorithms to examine AI spatial and temporal trends, cooperation features, cross-organization knowledge flow and technological routes. Results indicate a growing yet concentrated, non-collaborative and multi-path development and protection profile for AI patenting, with cross-organization knowledge flows based mainly on interorganizational knowledge citation links.

In this paper, we introduce the concept of learning to sense, which aims to emulate a complex characteristic of human reasoning: the ability to monitor and understand a set of interdependent events for decision-making processes. Event datasets are composed of textual data and spatio-temporal features that determine where and when a given phenomenon occurred. In learning to sense, related events are mapped closely to each other in a semantic vector space, thereby identifying that they contain similar contextual meaning. However, learning a semantic vector space that satisfies both textual similarities and spatio-temporal constraints is a crucial challenge for event analysis and sensing. This paper investigates a Semantic Variational Autoencoder (SVAE) to fine-tune pre-trained embeddings according to both textual and spatio-temporal events of the class of interest. Experiments involving more than one hundred sensors show that our SVAE outperforms a competitive one-class classification baseline. Moreover, our proposal provides desirable learning requirements to sense scenarios, such as visualization of the sensor decision function and heat maps with the sensor’s geographic impact.

Recent relation extraction models’ architecture are evolved from the shallow neural networks to natural language model, such as convolutional neural networks or recurrent neural networks to Bert. However, these methods did not consider the semantic information in the sequence or the distance dependence problem, the internal semantic information may contain the useful knowledge which can help relation classification. Focus on these problems, this paper proposed a BERT-based relation classification method. Compare with the existing Bert-based architecture, the proposed model can obtain the internal semantic information between entity pair and solve the distance semantic dependence better. The pre-trained BERT model after fine tuning is used in this paper to abstract the semantic representation of sequence, then adopt the piecewise convolution to obtain semantic information which influence the extraction results. Compare with the existing methods, the proposed method can achieve a better accuracy on relational extraction task because of the internal semantic information extracted in the sequence. While, the generalization ability is still a problem that cannot be ignored, and the numbers of the relationships are difference between different categories. In this paper, the focal loss function is adopted to solve this problem by assigning a heavy weight to less number or hard classify categories. Finally, comparing with the existing methods, the F1 metric of the proposed method can reach a superior result 89.95% on the SemEval-2010 Task 8 dataset.

Keeping an eye on pedestrians as they navigate through a scene, surveillance cameras are everywhere. With this context, our paper addresses the problem of pedestrian attribute recognition (PAR). This problem entails recognizing attributes such as age-group, clothing style, accessories, footwear style etc. This multi-label problem is extremely challenging even for human observers and has rightly garnered attention from the computer vision community. Towards a solution to this problem, in this paper, we adopt trainable Gabor wavelets (TGW) layers and cascade them with a convolution neural network (CNN). Whereas other researchers are using fixed Gabor filters with the CNN, the proposed layers are learnable and adapt to the dataset for a better recognition. We propose a two-branch neural network where mixed layers, a combination of the TGW and convolutional layers, make up the building block of our deep neural network. We test our method on twoo challenging publicly available datasets and compare our results with state of the art.

The vibration generated by tractor field operations will seriously affect the comfort and health of the driver. The low frequency vibration generated by the engine and ground excitation is similar to the natural frequency of human organs. Long term operation in this environment will resonate with the organs and affect drivers’ health. To investigate this possibility, in this paper we carried out a collection experiment of human physiological indicators relevant to vibration fatigue. Four physiological signals of surface electromyography, skin electricity, skin temperature, and photoplethysmography signal were collected while the subjects experienced vibration. Several features of physiological signals as well as the law of signal features changing with fatigue are studied. The test results show that with the increase of human fatigue, the overall physiological parameters show the following trends: The median frequency of the human body surface electromyography and the slope of skin surface temperature decreases, the value of skin conductivity and the mean value of the photoplethysmography signal increases. Furthermore, this paper proposes a vibration comfort evaluation method based on multiple physiological parameters of the human body. An artificial neural network model is trained with test samples, and the prediction accuracy rate reaches 88.9%. Finally, the vibration conditions are changed by the shock-absorbing suspension of a tractor, verifying the effectiveness of the physiological signal changing with the vibration of the human body. The established prediction model can also be used to objectively reflect the discomfort of the human body under different working conditions and provide a basis for structural design optimization.

With the rapid development of the mobile internet, people are becoming more dependent on the internet to express their comments on products or stores; meanwhile, text sentiment classification of these comments has become a research hotspot. In existing methods, it is fairly popular to apply a deep learning method to the text classification task. Aiming at solving information loss, weak context and other problems, this paper makes an improvement based on the transformer model to reduce the difficulty of model training and training time cost and achieve higher overall model recall and accuracy in text sentiment classification. The transformer model replaces the traditional convolutional neural network (CNN) and the recurrent neural network (RNN) and is fully based on the attention mechanism; therefore, the transformer model effectively improves the training speed and reduces training difficulty. This paper selects e-commerce reviews as research objects and applies deep learning theory. First, the text is preprocessed by word vectorization. Then the IN standardized method and the GELUs activation function are applied based on the original model to analyze the emotional tendencies of online users towards stores or products. The experimental results show that our method improves by 9.71%, 6.05%, 5.58% and 5.12% in terms of recall and approaches the peak level of the F1 value in the test model by comparing BiLSTM, Naive Bayesian Model, the serial BiLSTM_CNN model and BiLSTM with an attention mechanism model. Therefore, this finding proves that our method can be used to improve the text sentiment classification accuracy and effectively apply the method to text classification.

With the increase of patients with retinopathy, retinopathy recognition has become a research hotspot. In this article, we describe the etiology and symptoms of three kinds of retinal diseases, including drusen(DRUSEN), choroidal neovascularization(CNV) and diabetic macular edema(DME). In addition, we also propose a hybrid attention mechanism to classify and recognize different types of retinopathy images. In particular, the hybrid attention mechanism proposed in this paper includes parallel spatial attention mechanism and channel attention mechanism. It can extract the key features in the channel dimension and spatial dimension of retinopathy images, and reduce the negative impact of background information on classification results. The experimental results show that the hybrid attention mechanism proposed in this paper can better assist the network to focus on extracting thr fetures of the retinopathy area and enhance the adaptability to the differences of different data sets. Finally, the hybrid attention mechanism achieved 96.5% and 99.76% classification accuracy on two public OCT data sets of retinopathy, respectively.

Stochastic models of biomolecular reaction networks are commonly employed in systems and synthetic biology to study the effects of stochastic fluctuations emanating from reactions involving species with low copy-numbers. For such models, the Kolmogorov’s forward equation is called the chemical master equation (CME), and it is a fundamental system of linear ordinary differential equations (ODEs) that describes the evolution of the probability distribution of the random state-vector representing the copy-numbers of all the reacting species. The size of this system is given by the number of states that are accessible by the chemical system, and for most examples of interest this number is either very large or infinite. Moreover, approximations that reduce the size of the system by retaining only a finite number of important chemical states (e.g. those with non-negligible probability) result in high-dimensional ODE systems, even when the number of reacting species is small. Consequently, accurate numerical solution of the CME is very challenging, despite the linear nature of the underlying ODEs. One often resorts to estimating the solutions via computationally intensive stochastic simulations. The goal of the present paper is to develop a novel deep-learning approach for computing solution statistics of high-dimensional CMEs by reformulating the stochastic dynamics using Kolmogorov’s backward equation. The proposed method leverages superior approximation properties of Deep Neural Networks (DNNs) to reliably estimate expectations under the CME solution for several user-defined functions of the state-vector. This method is algorithmically based on reinforcement learning and it only requires a moderate number of stochastic simulations (in comparison to typical simulation-based approaches) to train the “policy function”. This allows not just the numerical approximation of various expectations for the CME solution but also of its sensitivities with respect to all the reaction network parameters (e.g. rate constants). We provide four examples to illustrate our methodology and provide several directions for future research., Author summary We develop a deep learning framework for estimating solutions of the chemical master equation (CME) that is fundamental to stochastic analysis of reaction networks. The CME is a system of ordinary differential equations that describes the time-evolution of the probability density of the random state-vector, and owing to an inherent curse of dimensionality, directly solving the CME is generally impractical with existing approaches. Moreover, the commonly employed simulation-based approaches for estimating CME solutions often require an exorbitant amount of computational time, even for moderately-sized networks. To counter these issues, we develop a deep reinforcement learning based method, called DeepCME, in this paper. DeepCME not only estimates function expectations based on the CME solution, but it also solves the more challenging problem of estimating their sensitivities with respect to all the model parameters. We illustrate our approach with four carefully chosen reaction network examples with varying sizes. Our results demonstrate that DeepCME reliably estimates the expectations of interest, along with all the parametric sensitivities, at a fraction of the computational cost of simulation-based estimators. We present many directions for future research and suggest further improvements to DeepCME that can greatly enhance its accuracy and applicability.

Computational drug repositioning and drug-target prediction have become essential tasks in the early stage of drug discovery. In previous studies, these two tasks have often been considered separately. However, the entities studied in these two tasks (i.e., drugs, targets, and diseases) are inherently related. On one hand, drugs interact with targets in cells to modulate target activities, which in turn alter biological pathways to promote healthy functions and to treat diseases. On the other hand, both drug repositioning and drug-target prediction involve the same drug feature space, which naturally connects these two problems and the two domains (diseases and targets). By using the wisdom of the crowds, it is possible to transfer knowledge from one of the domains to the other. The existence of relationships among drug-target-disease motivates us to jointly consider drug repositioning and drug-target prediction in drug discovery. In this paper, we present a novel approach called iDrug, which seamlessly integrates drug repositioning and drug-target prediction into one coherent model via cross-network embedding. In particular, we provide a principled way to transfer knowledge from these two domains and to enhance prediction performance for both tasks. Using real-world datasets, we demonstrate that iDrug achieves superior performance on both learning tasks compared to several state-of-the-art approaches. Our code and datasets are available at: https://github.com/Case-esaC/iDrug., Author summary Traditional high-throughput techniques for drug discovery are often expensive, time-consuming, and with high failure rates. Computational drug repositioning and drug-target prediction have thus become essential tasks in the early stage drug discovery. The emergence of large-scale heterogeneous biological networks has offered unprecedented opportunities for developing machine learning approaches to identify novel drug-disease or drug-target interactions. However, most existing works focused either on the drug-disease network or on the drug-target network, thus failed to capture the inherent dependencies between these two networks. These two biological networks are naturally connected since they involve the same drug feature space. In our opinion, ignoring this rich source of information is a major shortcoming of some existing works. In this paper, we present a novel approach called iDrug, which seamlessly integrates the drug-disease network and the drug-target network into one coherent model via cross-network embedding. As a result, iDrug is able to take full usage of the knowledge within these two biological networks to better exploit new biomedical insights of drug-target-disease. Therefore, iDrug has broad applications in drug discovery.

As a core driving force of the most recent round of industrial transformation, artificial intelligence has triggered significant changes in the world economic structure, profoundly changed our life and way of thinking, and achieved an overall leap in social productivity. This paper aims to examine the effect of knowledge transfer performance on the artificial intelligence industry innovation network and the path artificial intelligence enterprises can take to promote sustainable development through knowledge transfer in the above context. First, we construct a theoretical hypothesis and conceptual model of the innovation network knowledge transfer mechanism within the artificial intelligence industry. Then, we collect data from questionnaires distributed to Chinese artificial intelligence enterprises that participate in the innovation network. Moreover, we empirically analyze the impact of innovation network characteristics, organizational distance, knowledge transfer characteristics, and knowledge receiver characteristics on knowledge transfer performance and verify the hypotheses proposed in the conceptual model. The results indicate that innovation network centrality and organizational culture distance have a significant effect on knowledge transfer performance, with influencing factors including network scale, implicit knowledge transfer, receiver’s willingness to receive, and receiver’s capacity to absorb knowledge. For sustainable knowledge transfer performance on promoting Chinese artificial intelligence enterprises innovation, this paper finally delivers valuable insights and suggestions.

Accurate visual tracking is a challenging issue in computer vision. Correlation filter (CF) based methods are sought in visual tracking based on their efficiency and high performance. Nonetheless, CF-based trackers are sensitive to partial occlusion, which may reduce their overall performance and even lead to failure in tracking challenge. In this paper, we presented a very powerful tracker based on the kernelized correlation filter tracker (KCF). Firstly, we employ an intelligent multi-part tracking algorithm to improve the overall capability of correlation filter based tracker, especially in partial-occlusion challenges. Secondly, to cope with the problem of scale variation, we employ an effective scale adaptive scheme, which divided the target into four patches and computed the scale factor by finding the maximum response position of each patch via kernelized correlation filter. With this method, the scale computation was transformed into locating the centers of the patches. Thirdly, because the small deviation of the central function value will bring the problem of location ambiguity. To solve this problem, the new Gaussian kernel functions are introduced in this paper. Experiments on the default 51 video sequences in Visual Tracker Benchmark demonstrate that our proposed tracker provides significant improvement compared with the state-of-art trackers.

This paper studies the inverse kinematics of two non-spherical wrist configurations of painting robot. The simplest analytical solution of orthogonal wrist configuration is deduced in this paper for the first time. For the oblique wrist configuration, there is no analytical solution for the configuration. So it is necessary to solve by general method, which cannot achieve high precision and high speed as analytic solution. Two general methods are optimized in this paper. Firstly, the elimination method is optimized to reduce the solving speed to 20% of the original one, and the completeness of the method is supplemented. Based on the Gauss damped least squares method, a new optimization method is proposed to improve the solving speed. The enhanced step length coefficient is introduced to conduct studies with the machine learning correlation method. It has been proved that, on the basis of ensuring the stability of motion, the number of iterations can be effectively reduced and the average number of iterations can be less than 5 times, which can effectively improve the speed of solution. In the simulation and experimental environment, it is verified.

Understanding the digital jump of bank customers is key to design strategies to bring on board and keep online users, as well as to explain the increasing competition from new providers of financial services (such as BigTech and FinTech). This paper employs a machine learning approach to examine the digitalization process of bank customers using a comprehensive consumer finance survey. By employing a set of algorithms (random forests, conditional inference trees and causal forests) this paper identities the features predicting bank customers’ digitalization process, illustrates the sequence of consumers’ decision-making actions and explores the existence of causal relationships in the digitalization process. Random forests are found to provide the highest performance–they accurately predict 88.41% of bank customers’ online banking adoption and usage decisions. We find that the adoption of digital banking services begins with information-based services (e.g., checking account balance), conditional on the awareness of the range of online services by customers, and then is followed by transactional services (e.g., online/mobile money transfer). The diversification of the use of online channels is explained by the consciousness about the range of services available and the safety perception. A certain degree of complementarity between bank and non-bank digital channels is also found. The treatment effect estimations of the causal forest algorithms confirm causality of the identified explanatory factors. These results suggest that banks should address the digital transformation of their customers by segmenting them according to their revealed preferences and offering them personalized digital services. Additionally, policymakers should promote financial digitalization, designing policies oriented towards making consumers aware of the range of online services available.

Multiscale geometric analysis (MGA) is not only characterized by multi-resolution, time-frequency localization, multidirectionality and anisotropy, but also outdoes the limitations of wavelet transform in representing high-dimensional singular data such as edges and contours. Therefore, researchers have been exploring new MGA-based image compression standards rather than the JPEG2000 standard. However, due to the difference in terms of the data structure, redundancy and decorrelation between wavelet and MGA, as well as the complexity of the coding scheme, so far, no definitive researches have been reported on the MGA-based image coding schemes. In addressing this problem, this paper proposes an image data compression approach using the hidden Markov model (HMM)/pulse-coupled neural network (PCNN) model in the contourlet domain. First, a sparse decomposition of an image was performed using a contourlet transform to obtain the coefficients that show the multiscale and multidirectional characteristics. An HMM was then adopted to establish links between coefficients in neighboring subbands of different levels and directions. An Expectation-Maximization (EM) algorithm was also adopted in training the HMM in order to estimate the state probability matrix, which maintains the same structure of the contourlet decomposition coefficients. In addition, each state probability can be classified by the PCNN based on the state probability distribution. Experimental results show that the HMM/PCNN -contourlet model proposed in this paper leads to better compression performance and offer a more flexible encoding scheme.

Over exploitation of groundwater in Changping District of Beijing city has caused serious land subsidence in the past decades. In recent years, the operation of the South-to-North Water Transfer Project has reduced the land subsidence rate. In this paper, Experimental tests are performed using the GDS Consolidation Testing System to characterize the compression and rebound of soils at depths of less than 100 m caused by groundwater withdrawal and recharge in Changping District. The results indicate that the compressible layers are the main contributors to land subsidence. The first compressible layer experiences greater deformation and more considerable hysteresis than the other compressible layers with the same decrease in the pore water pressure. Therefore, the exploitation of the adjacent aquifer should be controlled in the future. The deformation in the second and third compressible layers is a gradual and long-term process with little rebound; therefore, the subsidence should be seriously addressed when the groundwater in the two compressible layers is exploited on a large scale. In the same compressible layer, silty clay is more compressible and hysteretic than silt. For the same soil sample, the deformation rate decreases gradually as the pore water pressure decreases, whereas the creep deformation shows an overall increasing trend. A parameter named the subsidence index Cw is proposed in this paper to describe the soil compressibility during groundwater withdrawal. All the soil samples are characterized by elastic-plastic deformation, and the shallow soil samples with less pore water pressure decrease are more likely to rebound.

A well-designed self-reported scale is highly applicable to current clinical and research practices. However, the problems with the scale method, such as quantitative analysis of content validity and test-retest reliability analysis of state-like variables are yet to be resolved. The main purpose of this paper is to propose an operational method for solving these problems. Additionally, it aims to enhance understanding of the research paradigm for the scale method (excluding criterion-related validity). This paper used a study that involved screening of high-risk groups for OCD (Obsessive-Compulsive Disorder), conducted 5 rounds of tests, and developed scales, reliability, and validity analysis (using sample sizes of 496, 610, 600, 600 and 990). The operational method we propose is practical, feasible, and can be used to develop and validate a scale.

High-throughput data such as metabolomics, genomics, transcriptomics, and proteomics have become familiar data types within the “-omics” family. For this work, we focus on subsets that interact with one another and represent these “pathways” as graphs. Observed pathways often have disjoint components, i.e., nodes or sets of nodes (metabolites, etc.) not connected to any other within the pathway, which notably lessens testing power. In this paper we propose the Pathway Integrated Regression-based Kernel Association Test (PaIRKAT), a new kernel machine regression method for incorporating known pathway information into the semi-parametric kernel regression framework. This work extends previous kernel machine approaches. This paper also contributes an application of a graph kernel regularization method for overcoming disconnected pathways. By incorporating a regularized or “smoothed” graph into a score test, PaIRKAT can provide more powerful tests for associations between biological pathways and phenotypes of interest and will be helpful in identifying novel pathways for targeted clinical research. We evaluate this method through several simulation studies and an application to real metabolomics data from the COPDGene study. Our simulation studies illustrate the robustness of this method to incorrect and incomplete pathway knowledge, and the real data analysis shows meaningful improvements of testing power in pathways. PaIRKAT was developed for application to metabolomic pathway data, but the techniques are easily generalizable to other data sources with a graph-like structure., Author summary PaIRKAT is a tool for improving testing power on high dimensional data by including graph topography in the kernel machine regression setting. Studies on high-dimensional data can struggle to include the complex relationships between variables. The semi-parametric kernel machine regression model is a powerful tool for capturing these types of relationships. They provide a framework for testing for relationships between outcomes of interest and high dimensional data such as metabolomic, genomic, or proteomic pathways. Our paper proposes a kernel machine method for including known biological connections between high dimensional variables by representing them as edges of ‘graphs’ or ‘networks.’ It is common for nodes (e.g., metabolites) to be disconnected from all others within the graph, which leads to meaningful decreases in testing power when graph information is ignored. We include a graph regularization or ‘smoothing’ approach for managing this issue. We demonstrate the benefits of this approach through simulation studies and an application to the metabolomic data from the COPDGene study.

A query optimizer attempts to predict a performance metric based on the amount of time elapsed. Theoretically, this would necessitate the creation of a significant overhead on the core engine to provide the necessary query optimizing statistics. Machine learning is increasingly being used to improve query performance by incorporating regression models. To predict the response time for a query, most query performance approaches rely on DBMS optimizing statistics and the cost estimation of each operator in the query execution plan, which also focuses on resource utilization (CPU, I/O). Modeling query features is thus a critical step in developing a robust query performance prediction model. In this paper, we propose a new framework based on query feature modeling and ensemble learning to predict query performance and use this framework as a query performance predictor simulator to optimize the query features that influence query performance. In query feature modeling, we propose five dimensions used to model query features. The query features dimensions are syntax, hardware, software, data architecture, and historical performance logs. These features will be based on developing training datasets for the performance prediction model that employs the ensemble learning model. As a result, ensemble learning leverages the query performance prediction problem to deal with missing values. Handling overfitting via regularization. The section on experimental work will go over how to use the proposed framework in experimental work. The training dataset in this paper is made up of performance data logs from various real-world environments. The outcomes were compared to show the difference between the actual and expected performance of the proposed prediction model. Empirical work shows the effectiveness of the proposed approach compared to related work.

Read-depths (RDs) are frequently used in identifying structural variants (SVs) from sequencing data. For existing RD-based SV callers, it is difficult for them to determine breakpoints in single-nucleotide resolution due to the noisiness of RD data and the bin-based calculation. In this paper, we propose to use the deep segmentation model UNet to learn base-wise RD patterns surrounding breakpoints of known SVs. We integrate model predictions with an RD-based SV caller to enhance breakpoints in single-nucleotide resolution. We show that UNet can be trained with a small amount of data and can be applied both in-sample and cross-sample. An enhancement pipeline named RDBKE significantly increases the number of SVs with more precise breakpoints on simulated and real data. The source code of RDBKE is freely available at https://github.com/yaozhong/deepIntraSV., Author summary In this paper, we used the deep segmentation model UNet to alleviate the bin size limitation of RD-based SV callers. UNet was initially proposed for image data. Here, we demonstrated that the UNet model could also be applied for one-dimensional genomic data. We formalized the breakpoint prediction as a segmentation task and inferred breakpoints in single-nucleotide resolution from predicted label marks. Through a set of experiments on both simulated and real WGS data, we demonstrated that the UNet model could be trained with a small amount of data, and an enhancement pipeline called RDBKE significantly increased the number of SVs with more precise breakpoints.

Fake news is a complex problem that leads to different approaches used to identify them. In our paper, we focus on identifying fake news using its content. The used dataset containing fake and real news was pre-processed using syntactic analysis. Dependency grammar methods were used for the sentences of the dataset and based on them the importance of each word within the sentence was determined. This information about the importance of words in sentences was utilized to create the input vectors for classifications. The paper aims to find out whether it is possible to use the dependency grammar to improve the classification of fake news. We compared these methods with the TfIdf method. The results show that it is possible to use the dependency grammar information with acceptable accuracy for the classification of fake news. An important finding is that the dependency grammar can improve existing techniques. We have improved the traditional TfIdf technique in our experiment.

In this paper, we propose a hierarchical statistical model for a single repairable system subject to several failure modes (competing risks). The paper describes how complex engineered systems may be modelled hierarchically by use of Bayesian methods. It is also assumed that repairs are minimal and each failure mode has a power-law intensity. Our proposed model generalizes another one already presented in the literature and continues the study initiated by us in another published paper. Some properties of the new model are discussed. We conduct statistical inference under an objective Bayesian framework. A simulation study is carried out to investigate the efficiency of the proposed methods. Finally, our methodology is illustrated by two practical situations currently addressed in a project under development arising from a partnership between Petrobras and six research institutes.

Peer-to-Peer (P2P) lending provides convenient and efficient financing channels for small and medium-sized enterprises and individuals, and therefore it has developed rapidly since entering the market. However, due to the imperfection of the credit system and the influence of cyberspace restrictions, P2P network lending faces frequent borrower credit risk crises during the transaction process, with a high proportion of borrowers default. This paper first analyzes the basic development of China’s P2P online lending and the credit risks of borrowers in the industry. Then according to the characteristics of P2P network lending and previous studies, a credit risk assessment indicators system for borrowers in P2P lending is formulated with 29 indicators. Finally, on the basis of the credit risk assessment indicators system constructed in this paper, BP neural network is built based on the BP algorithm, which is trained by the LM algorithm (Levenberg-Marquardt), Scaled Conjugate Gradient, and Bayesian Regularization respectively, to complete the credit risk assessment model. By comparing the results of three mentioned training methodologies, the BP neural network trained by the LM algorithm is finally adopted to construct the credit risk assessment model of borrowers in P2P lending, in which the input layer node is 9, the hidden layer node is 11 and output layer node is 1. The model can provide practical guidance for China and other countries’ P2P lending platforms, and therefore to establish and improve an accurate and effective borrower credit risk management system.

This paper establishes a fully automatic real-time image segmentation and recognition system for breast ultrasound intervention robots. It adopts the basic architecture of a U-shaped convolutional network (U-Net), analyses the actual application scenarios of semantic segmentation of breast ultrasound images, and adds dropout layers to the U-Net architecture to reduce the redundancy in texture details and prevent overfitting. The main innovation of this paper is proposing an expanded training approach to obtain an expanded of U-Net. The output map of the expanded U-Net can retain texture details and edge features of breast tumours. Using the grey-level probability labels to train the U-Net is faster than using ordinary labels. The average Dice coefficient (standard deviation) and the average IOU coefficient (standard deviation) are 90.5% (±0.02) and 82.7% (±0.02), respectively, when using the expanded training approach. The Dice coefficient of the expanded U-Net is 7.6 larger than that of a general U-Net, and the IOU coefficient of the expanded U-Net is 11 larger than that of the general U-Net. The context of breast ultrasound images can be extracted, and texture details and edge features of tumours can be retained by the expanded U-Net. Using an expanded U-Net can quickly and automatically achieve precise segmentation and multi-class recognition of breast ultrasound images.

Objective Some researchers have studied about early prediction and diagnosis of major adverse cardiovascular events (MACE), but their accuracies were not high. Therefore, this paper proposes a soft voting ensemble classifier (SVE) using machine learning (ML) algorithms. Methods We used the Korea Acute Myocardial Infarction Registry dataset and selected 11,189 subjects among 13,104 with the 2-year follow-up. It was subdivided into two groups (ST-segment elevation myocardial infarction (STEMI), non ST-segment elevation myocardial infarction NSTEMI), and then subdivided into training (70%) and test dataset (30%). Third, we selected the ranges of hyper-parameters to find the best prediction model from random forest (RF), extra tree (ET), gradient boosting machine (GBM), and SVE. We generated each ML-based model with the best hyper-parameters, evaluated by 5-fold stratified cross-validation, and then verified by test dataset. Lastly, we compared the performance in the area under the ROC curve (AUC), accuracy, precision, recall, and F-score. Results The accuracies for RF, ET, GBM, and SVE were (88.85%, 88.94%, 87.84%, 90.93%) for complete dataset, (84.81%, 85.00%, 83.70%, 89.07%) STEMI, (88.81%, 88.05%, 91.23%, 91.38%) NSTEMI. The AUC values in RF were (98.96%, 98.15%, 98.81%), ET (99.54%, 99.02%, 99.00%), GBM (98.92%, 99.33%, 99.41%), and SVE (99.61%, 99.49%, 99.42%) for complete dataset, STEMI, and NSTEMI, respectively. Consequently, the accuracy and AUC in SVE outperformed other ML models. Conclusions The performance of our SVE was significantly higher than other machine learning models (RF, ET, GBM) and its major prognostic factors were different. This paper will lead to the development of early risk prediction and diagnosis tool of MACE in ACS patients.

This paper contributes to the literature on topology identification (TI) in distribution networks and, in particular, on change detection in switching devices’ status. The lack of measurements in distribution networks compared to transmission networks is a notable challenge. In this paper, we propose an approach to topology identification (TI) of distribution systems based on supervised machine learning (SML) algorithms. This methodology is capable of analyzing the feeder’s voltage profile without requiring the utilization of sensors or any other extraneous measurement device. We show that machine learning algorithms can track the voltage profile’s behavior in each feeder, detect the status of switching devices, identify the distribution system’s typologies, reveal the kind of loads connected or disconnected in the system, and estimate their values. Results are demonstrated under the implementation of the ANSI case study.

Unmanned ground vehicles (UGVs) are an important research application of artificial intelligence. In particular, the deep learning-based object detection method is widely used in UGV-based environmental perception. Good experimental results are achieved by the deep learning-based object detection method Faster region-based convolutional neural network (Faster R-CNN). However, the exploration space of the region proposal network (RPN) is restricted by its expression. In our paper, a boosted RPN (BRPN) with three improvements is developed to solve this problem. First, a novel enhanced pooling network is designed in this paper. Therefore, the BRPN can adapt to objects with different shapes. Second, the expression of BRPN loss function is improved to learn the negative samples. Furthermore, the grey wolf optimizer (GWO) is used to optimize the parameters of the improved BRPN loss function. Thereafter, the performance of the BRPN loss function is promoted. Third, a novel GA-SVM classifier is applied to strengthen the classification capacity. The PASCAL VOC 2007, VOC 2012 and KITTI datasets are used to test the BRPN. Consequently, excellent experimental results are obtained by our deep learning-based object detection method.

An extreme learning machine (ELM) is a novel training method for single-hidden layer feedforward neural networks (SLFNs) in which the hidden nodes are randomly assigned and fixed without iterative tuning. ELMs have earned widespread global interest due to their fast learning speed, satisfactory generalization ability and ease of implementation. In this paper, we extend this theory to hypercomplex space and attempt to simultaneously consider multisource information using a hypercomplex representation. To illustrate the performance of the proposed hypercomplex extreme learning machine (HELM), we have applied this scheme to the task of multispectral palmprint recognition. Images from different spectral bands are utilized to construct the hypercomplex space. Extensive experiments conducted on the PolyU and CASIA multispectral databases demonstrate that the HELM scheme can achieve competitive results. The source code together with datasets involved in this paper can be available for free download at https://figshare.com/s/01aef7d48840afab9d6d.

We measure how accurately replication of experimental results can be predicted by black-box statistical models. With data from four large-scale replication projects in experimental psychology and economics, and techniques from machine learning, we train predictive models and study which variables drive predictable replication. The models predicts binary replication with a cross-validated accuracy rate of 70% (AUC of 0.77) and estimates of relative effect sizes with a Spearman ρ of 0.38. The accuracy level is similar to market-aggregated beliefs of peer scientists [1, 2]. The predictive power is validated in a pre-registered out of sample test of the outcome of [3], where 71% (AUC of 0.73) of replications are predicted correctly and effect size correlations amount to ρ = 0.25. Basic features such as the sample and effect sizes in original papers, and whether reported effects are single-variable main effects or two-variable interactions, are predictive of successful replication. The models presented in this paper are simple tools to produce cheap, prognostic replicability metrics. These models could be useful in institutionalizing the process of evaluation of new findings and guiding resources to those direct replications that are likely to be most informative.

Several lifetime distributions have played an important role to fit survival data. However, for some of these models, the computation of maximum likelihood estimators is quite difficult due to presence of flat regions in the search space, among other factors. Several well-known derivative-based optimization tools are unsuitable for obtaining such estimates. To circumvent this problem, we introduce the AdequacyModel computational library version 2.0.0 for the R statistical environment with two major contributions: a general optimization technique based on the Particle Swarm Optimization (PSO) method (with a minor modification of the original algorithm) and a set of statistical measures for assessment of the adequacy of the fitted model. This library is very useful for researchers in probability and statistics and has been cited in various papers in these areas. It serves as the basis for the Newdistns library (version 2.1) published in an impact journal in the area of computational statistics, see https://CRAN.R-project.org/package=Newdistns. It is also the basis of the Wrapped library (version 2.0), see https://CRAN.R-project.org/package=Wrapped. A third package making use of the AdequacyModel library can be found in https://CRAN.Rproject.org/package=sglg. In addition, the proposed library has proved to be very useful for maximizing log-likelihood functions with complex search regions. The library provides a greater control of the optimization process by introducing a stop criterion based on a minimum number of iterations and the variance of a given proportion of optimal values. We emphasize that the new library can be used not only in statistics but in physics and mathematics as proved in several examples throughout the paper.

As the root cause of illegal cyber activities, botnets are evolving continuously over the last two decades. Current researches on botnet command and control mechanism based on blockchain network suffer from high economic cost, single point of failure, and limited scalability. In this paper, we present DUSTBot, a novel P2P botnet model based on Bitcoin transactions to prepare for new cyber threats. Specifically, a covert, duplex, and low-cost command and control (C&C) channel in the Bitcoin network is presented in our work. DUSTBot uses the Bitcoin main network as the downstream channel while using the Bitcoin testnet as the upstream channel. Furthermore, the peer list exchange algorithm based on the Ethereum block hash proposed in this paper is effective against routing table poisoning attack and P2P botnet crawling. The robustness of DUSTBot against node removal is studied through constructing the botnet with a P2P simulator. We deploy the implementation of DUSTBot on cloud platforms to test its feasibility and performance. Moreover, the stealthiness of DUSTBot and the effectiveness of the proposed peer list exchange algorithm are evaluated. The results demonstrate the feasibility, performance, stealthiness, and robustness of DUSTBot. In the end, possible countermeasures are discussed to mitigate similar threats in the future.

As the industry gradually enters the stage of unmanned and intelligent, factories in the future need to realize intelligent monitoring and diagnosis and maintenance of parts and components. In order to achieve this goal, it is first necessary to accurately identify and classify the parts in the factory. However, the existing literature rarely studies the classification and identification of parts of the entire factory. Due to the lack of existing data samples, this paper studies the identification and classification of small samples of industrial machine parts. In order to solve this problem, this paper establishes a convolutional neural network model based on the InceptionNet-V3 pretrained model through migration learning. Through experimental design, the influence of data expansion, learning rate and optimizer algorithm on the model effectiveness is studied, and the optimal model was finally determined, and the test accuracy rate reaches 99.74%. By comparing with the accuracy of other classifiers, the experimental results prove that the convolutional neural network model based on transfer learning can effectively solve the problem of recognition and classification of industrial machine parts with small samples and the idea of transfer learning can also be further promoted.

Precision point positioning (PPP) is widely used in maritime navigation and other scenarios because it does not require a reference station. In PPP, the satellite clock bias (SCB) cannot be eliminated by differential, thus leading to an increase in positioning error. The prediction accuracy of SCB has become one of the key factors restricting positioning accuracy. Although International GNSS Service (IGS) provides the ultra-rapid ephemeris prediction part (IGU-P), its quality and real-time performance can not meet the practical application. In order to improve the accuracy of PPP, this paper proposes to use the Prophet model to predict SCB. Specifically, SCB sequence is read from the observation part in the ultra-rapid ephemeris (IGU-O) released by IGS. Next, the SCB sequence between adjacent epochs are subtracted to obtain the corresponding SCB single difference sequence. Then using the Prophet model to predict SCB single difference sequence. Finally, the prediction result is substituted into the PPP positioning observation equation to obtain the positioning result. This paper uses the final ephemeris (IGF) published by IGS as a benchmark and compares the experimental results with IGU-P. For the selected four satellites, compared with the results of the IGU-P, the accuracy of SCB prediction of the model in this paper is improved by about 50.3%, 61.7%, 60.4%, and 48.8%. In terms of PPP positioning results, we use Real-time kinematic (RTK) measurements as a benchmark in this paper. Positioning accuracy has increased by 26%, 35%, and 19% in the N, E, and U directions, respectively. The results show that the Prophet model can improve the performance of PPP.

In recent years, China's e-commerce industry has developed at a high speed, and the scale of various industries has continued to expand. Service-oriented enterprises such as e-commerce transactions and information technology came into being. This paper analyzes the shortcomings and challenges of traditional online shopping behavior prediction methods, and proposes an online shopping behavior analysis and prediction system. The paper chooses linear model logistic regression and decision tree based XGBoost model. After optimizing the model, it is found that the nonlinear model can make better use of these features and get better prediction results. In this paper, we first combine the single model, and then use the model fusion algorithm to fuse the prediction results of the single model. The purpose is to avoid the accuracy of the linear model easy to fit and the decision tree model over-fitting. The results show that the model constructed by the article has further improvement than the single model. Finally, through two sets of contrast experiments, it is proved that the algorithm selected in this paper can effectively filter the features, which simplifies the complexity of the model to a certain extent and improves the classification accuracy of machine learning. The XGBoost hybrid model based on p/n samples is simpler than a single model. Machine learning models are not easily over-fitting and therefore more robust.

Color-based image segmentation classifies pixels of digital images in numerous groups for further analysis in computer vision, pattern recognition, image understanding, and image processing applications. Various algorithms have been developed for image segmentation, but clustering algorithms play an important role in the segmentation of digital images. This paper presents a novel and adaptive initialization approach to determine the number of clusters and find the initial central points of clusters for the standard K-means algorithm to solve the segmentation problem of color images. The presented scheme uses a scanning procedure of the paired Red, Green, and Blue (RGB) color-channel histograms for determining the most salient modes in every histogram. Next, the histogram thresholding is applied and a search in every histogram mode is performed to accomplish RGB pairs. These RGB pairs are used as the initial cluster centers and cluster numbers that clustered each pixel into the appropriate region for generating the homogeneous regions. The proposed technique determines the best initialization parameters for the conventional K-means clustering technique. In this paper, the proposed approach was compared with various unsupervised image segmentation techniques on various image segmentation benchmarks. Furthermore, we made use of a ranking approach inspired by the Evaluation Based on Distance from Average Solution (EDAS) method to account for segmentation integrity. The experimental results show that the proposed technique outperforms the other existing clustering techniques by optimizing the segmentation quality and possibly reducing the classification error.

BackgroundThe development of deep learning (DL) algorithms is a three-step process-training, tuning, and testing. Studies are inconsistent in the use of the term "validation", with some using it to refer to tuning and others testing, which hinders accurate delivery of information and may inadvertently exaggerate the performance of DL algorithms. We investigated the extent of inconsistency in usage of the term "validation" in studies on the accuracy of DL algorithms in providing diagnosis from medical imaging.Methods and findingsWe analyzed the full texts of research papers cited in two recent systematic reviews. The papers were categorized according to whether the term "validation" was used to refer to tuning alone, both tuning and testing, or testing alone. We analyzed whether paper characteristics (i.e., journal category, field of study, year of print publication, journal impact factor [JIF], and nature of test data) were associated with the usage of the terminology using multivariable logistic regression analysis with generalized estimating equations. Of 201 papers published in 125 journals, 118 (58.7%), 9 (4.5%), and 74 (36.8%) used the term to refer to tuning alone, both tuning and testing, and testing alone, respectively. A weak association was noted between higher JIF and using the term to refer to testing (i.e., testing alone or both tuning and testing) instead of tuning alone (vs. JIF 10: adjusted odds ratio 2.41, P = 0.089). Journal category, field of study, year of print publication, and nature of test data were not significantly associated with the terminology usage.ConclusionsExisting literature has a significant degree of inconsistency in using the term "validation" when referring to the steps in DL algorithm development. Efforts are needed to improve the accuracy and clarity in the terminology usage.

Piano is a complex instrument, which humans learn to play after many years of practice. This paper investigates the complex dynamics of the embodied interactions between a human and piano, in order to gain insights into the nature of humans’ physical dexterity and adaptability. In this context, the dynamic interactions become particularly crucial for delicate expressions, often present in advanced music pieces, which is the main focus of this paper. This paper hypothesises that the relationship between motor control for key-pressing and the generated sound is a manifold problem, with high-degrees of non-linearity in nature. We employ a minimalistic experimental platform based on a robotic arm equipped with a single elastic finger in order to systematically investigate the motor control and resulting outcome of piano sounds. The robot was programmed to run 3125 key-presses on a physical digital piano with varied control parameters. The obtained data was applied to a Gaussian Process (GP) inference modelling method, to train a network in terms of 10 playing styles, corresponding to different expressions generated by a Musical Instrument Digital Interface (MIDI). By analysing the robot control parameters and the output sounds, the relationship was confirmed to be highly nonlinear, especially when the rich expressions (such as a broad range of sound dynamics) were necessary. Furthermore this relationship was difficult and time consuming to learn with linear regression models, compared to the developed GP-based approach. The performance of the robot controller was also compared to that of an experienced human player. The analysis shows that the robot is able to generate sounds closer to humans’ in some expressions, but requires additional investigations for others.

Neural networks are widely used in automatic credit scoring systems with high accuracy and outstanding efficiency. However, in the absence of prior knowledge, it is difficult to determine the set of hyper-parameters, which makes its application limited in practice. This paper presents a novel framework of credit-scoring model based on neural networks trained by the optimal swarm intelligence (SI) algorithm. This framework incorporates three procedures. Step 1, pre-processing, including imputation, normalization, and re-ordering of the samples. Step 2, training, where SI algorithms optimize hyper-parameters of back-propagation artificial neural networks (BP-ANN) with the area under curve (AUC) as the evaluation function. Step 3, test, applying the optimized model in Step 2 to predict new samples. The results show that the framework proposed in this paper searches the hyper-parameter space efficiently and finds the optimal set of hyper parameters with appropriate time complexity, which enhances the fitting and generalization ability of BP-ANN. Compared with existing credit-scoring models, the model in this paper predicts with a higher accuracy. Additionally, the model enjoys a greater robustness, for the difference of performance between training and testing phases.

Humans are able to track multiple objects at any given time in their daily activities—for example, we can drive a car while monitoring obstacles, pedestrians, and other vehicles. Several past studies have examined how humans track targets simultaneously and what underlying behavioral and neural mechanisms they use. At the same time, computer-vision researchers have proposed different algorithms to track multiple targets automatically. These algorithms are useful for video surveillance, team-sport analysis, video analysis, video summarization, and human–computer interaction. Although there are several efficient biologically inspired algorithms in artificial intelligence, the human multiple-target tracking (MTT) ability is rarely imitated in computer-vision algorithms. In this paper, we review MTT studies in neuroscience and biologically inspired MTT methods in computer vision and discuss the ways in which they can be seen as complementary., Author summary Multiple-target tracking (MTT) is a challenging task vital for both a human’s daily life and for many artificial intelligent systems, such as those used for urban traffic control. Neuroscientists are interested in discovering the underlying neural mechanisms that successfully exploit cognitive resources, e.g., spatial attention or memory, during MTT. Computer-vision specialists aim to develop powerful MTT algorithms based on advanced models or data-driven computational methods. In this paper, we review MTT studies from both communities and discuss how findings from cognitive studies can inspire developers to construct higher performing MTT algorithms. Moreover, some directions have been proposed through which MTT algorithms could raise new questions in the cognitive science domain, and answering them can shed light on neural processes underlying MTT.

Analyzing the rhythm of animals’ acoustic signals is of interest to a growing number of researchers: evolutionary biologists want to disentangle how these structures evolved and what patterns can be found, and ecologists and conservation biologists aim to discriminate cryptic species on the basis of parameters of acoustic signals such as temporal structures. Temporal structures are also relevant for research on vocal production learning, a part of which is for the animal to learn a temporal structure. These structures, in other words, these rhythms, are the topic of this paper. How can they be investigated in a meaningful, comparable and universal way? Several approaches exist. Here we used five methods to compare their suitability and interpretability for different questions and datasets and test how they support the reproducibility of results and bypass biases. Three very different datasets with regards to recording situation, length and context were analyzed: two social vocalizations of Neotropical bats (multisyllabic, medium long isolation calls of Saccopteryx bilineata, and monosyllabic, very short isolation calls of Carollia perspicillata) and click trains of sperm whales, Physeter macrocephalus. Techniques to be compared included Fourier analysis with a newly developed goodness-of-fit value, a generate-and-test approach where data was overlaid with varying artificial beats, and the analysis of inter-onset-intervals and calculations of a normalized Pairwise Variability Index (nPVI). We discuss the advantages and disadvantages of the methods and we also show suggestions on how to best visualize rhythm analysis results. Furthermore, we developed a decision tree that will enable researchers to select a suitable and comparable method on the basis of their data., Author summary In the analysis of animal communication more and more interest is shown in rhythm of animal communication and what information this might convey. In this paper, we establish a workflow to analyze the temporal structure–namely the rhythm–of any particular animals’acoustic signal with methods that are applicable for a wide range of signals and results that are easily comparable and interpretable. This workflow will enhance the understanding of rhythmicality in animals’ acoustic signals as well as facilitate comparison between species. Methods we conducted ranged from simple distributional and visual analysis to higher mathematics such as Fourier analysis. All analyses rely on Inter-Onset-Intervals, the duration between the beginning of one element and the next. We used different datasets from two neotropical bat species as well as from the sperm whale. With this selection, we cover very short sequences with only few elements up to sequences of around 200 elements, multisyllabic and monosyllabic sequences and social communication as well as sounds used for orientation and foraging.

In imaging systems, image blurs are a major source of degradation. This paper proposes a parameter estimation technique for linear motion blur, defocus blur, and atmospheric turbulence blur, and a nonlinear deconvolution algorithm based on sparse representation. Most blur removal techniques use image priors to estimate the point spread function (PSF); however, many common forms of image priors are unable to exploit local image information fully. In this paper, the proposed method does not require models of image priors. Further, it is capable of estimating the PSF accurately from a single input image. First, a blur feature in the image gradient domain is introduced, which has a positive correlation with the degree of blur. Next, the parameters for each blur type are estimated by a learning-based method using a general regression neural network. Finally, image restoration is performed using a half-quadratic optimization algorithm. Evaluation tests confirmed that the proposed method outperforms other similar methods and is suitable for dealing with motion blur in real-life applications.

In this paper, a novel 3D roaming algorithm considering collision detection and interaction is proposed that adopts a triangle mesh to organize and manage massive spatial data and uses a customized bounding box intersector to rapidly obtain the potential collided triangles. The proposed algorithm can satisfy the requirements of timeliness and practicability during complicated large 3D scene collision detection. Moreover, we designed a method to calculate the collision point coordinates according to the spatial position relation and distance change between the virtual collision detection sphere and triangles, with the triangle edges and three vertices being considered. Compared to the methods that use the native intersector of OpenSceneGraph (OSG) to obtain the collision point coordinates, the calculation efficiency of the proposed method is greatly improved. Usually, when there is a big split/pit in the scene, the viewpoints will fly off the scene due to the fall of the collision detection sphere, or the region interior cannot be accessed when the entrance of some local region (e.g., internal grotto) of the scene is too small. These problems are solved in this paper through 3D scene-path training and by self-adaptively adjusting the radius of the virtual collision detection sphere. The proposed 3D roaming and collision detection method applicable for massive spatial data overcomes the limitation that the existing roaming and collision detection methods are only applicable to 3D scenes with a small amount of data and simple models. It provides technical supports for freewill browsing and roaming of indoor/outdoor and overground/underground of the 3D scene in cases of massive spatial data.