Search

These days, the integration of big data and information systems enables industries such as sports, health, and medical sciences in resource optimization by enhancing their operational efficiency. Personalized health management, driven by advancements in big data analysis, feature extraction, and internet of things (IoT) has emerged as a prominent trend in the sports industry. Based on these advancements, this paper focuses on leveraging machine vision, big data, and IoT to conduct a comprehensive study for extracting football players' offense features by aiming to improve accuracy and ensure the players' safety in sports industry. First, this paper discusses a framework for sports health, which is based on big data analysis. Next, we discuss the use of big data and feature extraction for IoT and sports health informatics. The action image of an offense for football players is collected by an intelligent camera, and the gathered imaging code is read and analyzed by radio frequency technology. Using the recognition method of offense action using deep learning and motion information, the offense action's image features and the three-color features of video RGB are fused into deep learning features to complete the offense action recognition. Based on the offense action recognition, the Harris 3D operator is applied to construct the sequence potential function of the offense action, and the sequence possible function of the offense action is used as the operational technique for this purpose. In addition, the AdaBoost algorithm is used to comprehensively screen the characterized data of athletes' offense actions, and the filtered data are used as training samples to complete the feature extraction of athletes' offense actions. The experimental results show that the proposed method can effectively reduce the recognition error of offense action and improve the efficiency of action's feature extraction and the accuracy of feature extraction. [ABSTRACT FROM AUTHOR]

The goal of the paper is to present a Stacked autoencoder approach for enhancing Intrusion Detection Systems (IDSs) in Mobile Ad-Hoc Networks (MANETs). The paper proposes a Stacked autoencoder-based approach for MANET (Stacked AE-IDS) to reduce correlation and model relevant features with high-level representation. This method reproduces the input with a reduced correlation, and the output of the autoencoder is used as the input of the Deep Neural Network (DNN) classifier (DNN-IDS). The proposed Deep Learning-based IDS focuses on Denial of Services (DoS) attacks within labeled datasets, which are available for intrusion detection, and employs the most potential attacks that impact routing services in Mobile Networks. The proposed Stacked AE-IDS method enhances the effectiveness of IDSs in detecting attacks in MANETs by reducing the correlation and modeling high-level representations of relevant features. The focus on DoS attacks and their impact on routing services in Mobile Networks makes the proposed approach particularly relevant for MANET security. The proposed Stacked AE-IDS approach has potential applications in enhancing the security of MANETs by improving the effectiveness of IDSs. This approach can be used to detect different types of attacks, particularly DoS attacks, and their impact on routing services in Mobile Networks. [ABSTRACT FROM AUTHOR]

With the actual demand of international communication, intelligent translation of English has become a key direction of artificial intelligence development in the current English field. For the existing English intelligent translation, how to deal with the massive data effectively is always a big problem. Therefore, it is necessary to use the principle of machine learning to optimize the English intelligent translation model. The purpose of this paper is to optimize the existing English intelligent translation model through spectral clustering to remove outliers, so as to make it more suitable for the use of massive data. Moreover, this paper uses deep learning methods to improve on the basis of the PoseNet network structure and adds regularization to the convolutional layer, which reduces the problem of gradient disappearance and reduces the computational complexity. In addition, this paper uses adaptive weighting to remove invalid model assumptions. In the conceptual space of the similarity matrix, the interior point is farther from the origin than the outlier point. The algorithm in this paper can detect and eliminate the outliers in a paragraph of English. At the same time, the normal English content will be classified into data categories, and the final translation result will be obtained. Through the experimental test, it can be seen that the model proposed in this paper has good performance and can cope with massive data, so it has certain superiority. [ABSTRACT FROM AUTHOR]

Pedestrian detection is crucial for crowd surveillance applications and cyber-physical systems that can deliver timely and sophisticated solutions, especially with applications like person identification, person count, and tracking as the number of people rises. Even though the number of cutting-edge neural network-based frameworks for object detection models and pedestrian detection in images has steadily increased, object detection and image classification have made progress in terms of accuracy levels greater than 99% and level of granularity. A powerful object detector, specifically designed for high-end surveillance applications, is needed to position the bounding box, label it and return its relative positions. The size of these bounding boxes can vary depending on the object and its interaction with the physical world. To overcome these limitations and requirements, an extensive evaluation of the state-of-the-art algorithms has been presented in this paper. The work presented in this paper performs object detections on the MOT20 dataset using various algorithms and testing on a custom dataset recorded in our organization premises using an unmanned aerial vehicle (UAV). The main contribution of the paper is to tune the hyperparameters of the Yolov5 object detection algorithm to develop a new efficient, simple, and reliable object detector in Surveillance. The modified Yolov5 model outperforms the existing Faster-RCNN, SSD, and YOLO algorithms, according to real-world data experiments, and ensures better object detection with a 61% precision and 44% F measure value. [ABSTRACT FROM AUTHOR]

Calculus equation is an important tool for mathematical research and plays an important role in most natural science research. Since the beginning of the eighteenth century, people have gradually used differential and integral equations to solve physical problems. In general, several different aspects of differential equations in the field of mathematics are concerned and studied by most scholars. However, this paper studies and establishes the optimal model for numerical solution of differential equations through deep learning and genetic algorithm. In this paper, the solution of ordinary differential equations is solved through the use of polynomial function space, while the linear combination of simple function x and its product nx can obtain multinomial function space. The space function form of polynomial is very simple, and the operation ability is very strong. Almost all functions can be approximated, and the function space can be transformed by a simple function. Through data simulation test results, it can be found that the oscillation of neural network output is stronger and stronger with the increase in depth, that is to say, the deeper depth endows the neural network with stronger oscillation properties, so for the oscillation function, the depth neural network fitting effect is better than the shallow neural network. Therefore, by combining deep learning and genetic algorithm, this paper studies and establishes the optimal model for numerical solution of differential equations, and finds that the deep neural network can largely complete data simulation. [ABSTRACT FROM AUTHOR]

With the rapid development of machine intelligence, an increasing number of websites and servers have been amicably visited or sometimes attached by intelligent machines intensively. Therefore, how to empower a host machine to intelligently distinguish intelligent machines from humans is a challenging work. In this paper, the Mirror Turing Test (MTT) is conceived and implemented. Unlike the standard Turing Test, the tester in the MTT is replaced by a machine instead of a human. Current advancements on deep learning enable machines to recognize subtle differences between genuine and counterfeit works. Sometimes, the ability of machines is even superior to that of humans. Will machines transcend humans in an irreversible trend? Not completely right. The detection of soul in an artwork remains far beyond the capacity of machines. The two sets of MTT based on poetry generated by a machine and a novel imitated by a human were conducted in this paper and neither of them passed the MTT. Poetry is one of the art forms in which authors reveal their souls. Thus, we chose poetry in the MTT experiments on the basis of our soul computing model, thus clearly discriminating machine from human. [ABSTRACT FROM AUTHOR]

As a great Chinese thinker and writer in the twentieth century, Lu Xun and his literary works are widely known. However, as a successful cultural communication activist, editor and publisher, we still need to conduct in-depth research on Lu Xun in many aspects. Therefore, based on deep learning, this paper constructs an online multimedia learning system of Lu Xun literature. This system takes the relationship between classical Lu Xun literature and modern multimedia technology as the research object, and compare the calculation effect of other different types of algorithms and this dhraa algorithm. Through the availability of data, the dhraa algorithm is significantly better than other algorithms in the recommendation accuracy, thus proving its effectiveness. This system is managed by two servers and one system. The two servers are database server and web server, respectively. After testing, the system has good bearing capacity, can make up for the limited processing capacity of the server, and ensure the system has high performance. Its performance characteristics also show that the system achieved the expected performance. This paper systematically combines Lu Xun's literature with modern multimedia, which can provide online learning services for Lu Xun's literature lovers, thus helping scholars to expand Lu Xun's research field and academic vision. This paper designs an effective online learning system of Lu Xun's literature by combining deep learning, multimedia technology and Lu Xun's literature. [ABSTRACT FROM AUTHOR]

At present, wearable devices have some problems, such as poor adaptability to human motion behavior, and the recognition accuracy required for different wearers cannot be achieved. Based on the principle of deep learning algorithm, this paper realizes the development of intelligent rope skipping movement data monitoring system. Through the universal human body analysis model, the attention mechanism is introduced and embedded into the decoding network. The data set of rope skipping is classified by multiple labels, and the convolution of spatial graph is constructed, which is extended to the time series dynamics of moving human skeleton data. Aiming at the problem of complex information data in the process of moving human body recognition, we use pose estimation to calculate the key points of moving human body, extract the dynamic structure information of human skeleton sequence. Due to the problems of line of sight occlusion in the process of moving human target tracking, a target tracking algorithm based on multi domain convolution neural network is adopted to improve the feature extraction ability of the algorithm by segmenting the target to be tracked and identifying the area around the target. The data set of rope skipping is collected by wearable sensors, and the difference in the numerical range may be large, so the data is normalized. Finally, through the loss function, the fitting effect of neural network can be evaluated, and the gradient optimization model parameters can be calculated, and coping with different data changes. Through the final system performance test, it is verified that the accuracy rate of the system designed in this paper is above 90%, which can effectively monitor the data of skipping rope and be used in the actual operation of skipping rope. [ABSTRACT FROM AUTHOR]

Focusing on non-iterative approaches in training feed-forward neural networks, this special issue includes 12 papers to share the latest progress, current challenges, and potential applications of this topic. This editorial presents a background of the special issue and a brief introduction to the 12 contributions. [ABSTRACT FROM AUTHOR]

The physical movement of the human hand produces gestures, and hand gesture recognition leads to the advancement in automated vehicle movement system. In this paper, the human hand gestures are detected and recognized using convolutional neural networks (CNN) classification approach. This process flow consists of hand region of interest segmentation using mask image, fingers segmentation, normalization of segmented finger image and finger recognition using CNN classifier. The hand region of the image is segmented from the whole image using mask images. The adaptive histogram equalization method is used as enhancement method for improving the contrast of each pixel in an image. In this paper, connected component analysis algorithm is used in order to segment the finger tips from hand image. The segmented finger regions from hand image are given to the CNN classification algorithm which classifies the image into various classes. The proposed hand gesture detection and recognition methodology using CNN classification approach with enhancement technique stated in this paper achieves high performance with state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The identification of germ integrity is very important for the study of germ rice grains. However, the traditional algorithms lack to achieve better recognition results. This article proposes a specific germ integrity detection model that improves the inception-resnet v1 model and adopts topological structure branch of inception-v4. The proposed model dilates the convolution based on the original model that gives an explicit description of image features for germ rice. The proposed scheme achieves the better approximation of abstract representations for potential space. Simultaneously, this paper is based on color image features for germ rice to identify its image location. Afterward, the rice germ gathered in the training will be unified its location by image rotation in accordance with the location information of germ. The proposed model is increasing the identification precision of rice germ integrity. It further strengthens the training of germ detail features. The experiment result shows that comprehensive recognition accuracy in this paper is up to 90.43%, compared with other traditional recognition methods, the algorithm accuracy has highly improved. [ABSTRACT FROM AUTHOR]

The need for security in healthcare environments has become increasingly important due to the rise of cyber-attacks and data breaches. In order to address this issue, this paper proposes an Internet of Things (IoT)-based intrusion detection system for healthcare using a deep learning strategy with custom features. This paper proposes an IoT-based intrusion detection system for healthcare using a deep learning strategy with custom features. The proposed system utilizes the IoT technology to gather real-time data from various medical devices and sensors deployed in a healthcare environment. The system incorporates a recurrent neural network (RNN) and a bidirectional long short-term memory (BiLSTM) algorithm to detect and classify intrusion attempts. The custom features are extracted from the incoming data streams and used to train the deep learning models. The proposed system is evaluated on a dataset comprising different types of intrusion scenarios, and it achieves an accuracy of 99.16%, error rate of 0.008371%, sensitivity ratio of 99.89% and specificity ratio of 98.203% for IoTID20 with custom features using RNNBiLSTM. The results demonstrate the effectiveness of the proposed system in detecting and mitigating security threats in a healthcare environment. The system has the potential to improve patient privacy and security, ultimately leading to better healthcare outcomes. [ABSTRACT FROM AUTHOR]

In today's generation, many healthcare organizations moving towards patient-centred quality assessment, transparency strategies, and practices. Therefore, online reviews are the part of everyday life and millions of online reviews being generated daily; unfortunately, all online reviews are not genuine do happen negative reviews written by malicious users are everywhere on the social media platform such as Google, Yelp, Facebook, TripAdvisor, Amazon, Healthgrades, etc. Every fake, spam, and negative online review can directly impact a potential customer's purchasing decision, company's reputation, profitability and misleading the consumers, and damage the reputation of review websites. Hence, detecting fake online reviews has become a big challenge in the rapid development of e-commerce but it can be more dangerous when it comes to medical products, e-healthcare platforms to classify online reviews correctly due to the high false-positive rate (FPR) and the false-negative rate (FNR). In the medical environment, trust is the best reason to pursue transparency and deserve their confidence then patients have made online reviews an integral part of their decision-making process. Patient online reviews and ratings can be a good indicator of clinical effectiveness, efficiently control their health and patient safety. Researchers performed various investigations to predict fake online reviews, and their models could have provided low prediction accuracy due to the selection of lesser features and lack of essential prediction labels. To handle this issue, this paper presents an efficient approach to specifically predict fake online reviews with multiple feature analyses in an automated way with different perspectives of patient safety using qualitative methods. The primary aim of this paper is to provide a solid and complete comparative study of present research on detecting fake online reviews designed a deep learning index system to examine multiple features such as sentiment, test, and behaviour features of the online reviewers. The input samples are trained with the novel dense network model to extract essential features. And the labels are provided tentatively, and the model adopts some initial samples to train the proposed classifier model as the model intends to give better accuracy. The model's novelty relies on the feature selection and label classification of reviews. The experimental results indicated that fake review detection yields better accuracy of 99.5% on the anticipated model than other baseline approaches across the Amazon and medical datasets, using deep learning techniques to reduce the FPR, and FNR then improves fake review detection capabilities. The proposed model significantly improved the classifier's performance compared to existing state-of-the-art techniques in the Amazon and medical domains, demonstrating the effectiveness and relevance of the method. [ABSTRACT FROM AUTHOR]

The identification of apple leaf diseases is crucial to reduce yield reduction and timely take disease control measures. Employing deep learning for apple leaf disease identification is challenging because of the limited availability of samples for supervised training and the serious class imbalance. Hence, this paper proposes an accurate deep learning-based pipeline to solve the problem of limited data sets on farms and reduce the partiality due to serious class imbalance. Firstly, an improved cycle-consistent adversarial networks (CycleGAN) is used to generate synthetic samples to improve the learning of data distribution and solve the problems of small data sets and class imbalance. Secondly, ResNet is trained as a baseline convolutional neural network classifier to classify apple leaf diseases. The experimental results show that ResNet has the highest recognition accuracy on the test set, reaching 97.78%, and the classification accuracy is significantly improved by the generated synthetic samples (+ 14.7%). In addition, the experiment results of t-distributed stochastic neighbor embedding (t-SNE) and visual Turing test visually confirmed that the images generated by improved CycleGAN have much better quality and are more convincing. [ABSTRACT FROM AUTHOR]

A traditional approach for panoramic image generation consists of a random sample consensus (RANSAC) algorithm on a set of scale-invariant feature transform (SIFT) correspondences to generate a homography matrix between two images. Although producing adequate results for some type of images, hand-crafted SIFT features are not robust enough for highly varying natural images and the iterative RANSAC algorithm with its randomness does not always find the desired homography matrix. Recently, deep neural networks have been producing significant results in many challenging computer vision problems by learning features from large amounts of data. However, only very few recent works have been applied deep learning to panoramic image generation with the objective of finding feature correspondences and estimating homography matrix. Moreover, the absence of a proper dataset for the image stitching task hinders the standardization of models and comparison of their results. This paper attempts to generate panoramic images by extensively experimenting with various approaches using deep neural networks. The best proposed deep learning model achieved 7.31 and 1.07 pixels of the average absolute value loss for corner difference in X and Y directions, respectively. At the same time, qualitative results demonstrate superiority in comparison with the state-of-the-art SIFT+RANSAC algorithm. Specifically, in 72% of time the proposed framework either produced better results than SIFT+RANSAC or results of the proposed approach and SIFT+RANSAC were indistinguishable. Although SIFT + RANSAC produces better results in 28% of the time with respect to the loss function, our results are still visually comparable in many of these cases. Finally, a novel panoramic image generation dataset is introduced in this paper. [ABSTRACT FROM AUTHOR]

Hurricanes are tropical storms that cause immense damage to human life and property. Rapid assessment of damage caused by hurricanes is extremely important for the first responders. But this process is usually slow, expensive, labor intensive and prone to errors. The advancements in remote sensing and computer vision help in observing Earth at a different scale. In this paper, a new Convolutional Neural Network model has been designed with the help of satellite images captured from the areas affected by hurricanes. The model will be able to assess the damage by detecting damaged and undamaged buildings based upon which the relief aid can be provided to the affected people on an immediate basis. The model is composed of five convolutional layers, five pooling layers, one flattening layer, one dropout layer and two dense layers. Hurricane Harvey dataset consisting of 23,000 images of size 128 × 128 pixels has been used in this paper. The proposed model is simulated on 5750 test images at a learning rate of 0.00001 and 30 epochs with the Adam optimizer obtaining an accuracy of 0.95 and precision of 0.97. The proposed model will help the emergency responders to determine whether there has been damage or not due to the hurricane and also help those to provide relief aid to the affected people. [ABSTRACT FROM AUTHOR]

Deep learning is a rapidly growing research area having state of art achievement in various applications including but not limited to speech recognition, object recognition, machine translation, and image segmentation. In the current modern industrial manufacturing system, Machine Health Surveillance System (MHSS) is achieving increasing popularity because of the widespread availability of low cost sensors internet connectivity. Deep learning architecture gives useful tools to analyze and process these vast amounts of machinery data. In this paper, we review the latest deep learning techniques and their variant used for MHSS. We used Gearbox Fault Diagnosis dataset in this paper that contains the sets of vibration attributes recorded by SpectraQuest's Gearbox Fault Diagnostics Simulator. In addition, we used the variant of auto encoders for feature extraction to achieve higher accuracy in machine health surveillance. The results showed that the bagging ensemble classifier based on voting techniques achieved 99% accuracy. [ABSTRACT FROM AUTHOR]

In order to improve the effect of intelligent language translation, this paper analyzes the problems of the MSE cost function used by most of the current DNN-based speech enhancement algorithms and proposes a deep learning speech enhancement algorithm based on perception-related cost functions. Moreover, this paper embeds the suppression gain parameter estimation into the architecture of the traditional speech enhancement algorithm and converts the relationship between the noisy speech spectrum and the enhanced speech spectrum into a simple multiplication relationship based on suppression gain combined with deep learning algorithms to construct an intelligent language translation system. Moreover, this paper evaluates the translation effect of the system, analyzes the actual results, and uses simulation tests to verify the performance of the intelligent language translation model constructed in this paper. From the experimental results, it can be seen that the intelligent language translation system based on deep learning algorithms has good results. [ABSTRACT FROM AUTHOR]

This paper focuses on the problem of motorcyclist helmet detection in single images. Although some previous works have been developed to deal with this problem, yet most of them are designed for videos and not suitable for single-image helmet detection. In view of this, in this paper, we propose a novel dual-detection framework for single-image motorcyclist helmet detection with multi-head self-attention. Particularly, two types of detectors are first trained, namely the rider detector and the head-shoulder detector, which are jointly leveraged in the dual-detection scheme. To take advantage of the contextual relevance information, the multi-head self-attention mechanism is incorporated, where multiple self-attention layers are integrated to capture the complex relationships among the input features so as to further enhance the detection accuracy. A new benchmark dataset, termed the SCAU helmet detection on motorcyclists (SCAU-HDM) dataset, is presented, which consists of 8000 training images and 2000 test images. Extensive experiments on the benchmark dataset demonstrate the effectiveness of the proposed framework. The code is available at https://github.com/LiChunHong2020/SCAUHDM. [ABSTRACT FROM AUTHOR]

The number of papers on machine learning and deep neural networks applied to intrusion detection systems (IDS) is ever-increasing. Differently from existing work on the topic, this paper explores the effect of training-time randomness of deep neural networks, which is overlooked by the related literature. Training-time randomness is regulated by the seed of the pseudorandom number generator, and affects the performance of IDS models. The seed selection is studied in conjunction with other critical learning parameters: to the best of our knowledge, there are no similar studies in IDS. The experiments are done with a recent and widely consolidated intrusion detection benchmark, which is used to train and test a neural network under different combinations of seeds and parameters both in supervised and semi-supervised learning modes. The results are inferred by a mixture of explorative analysis, design of experiments, and analysis of variance. According to the results, the choice of the seed yields either excellent or scarce detection metrics; more importantly, the seed selection might be as relevant as the other major learning parameters assessed. [ABSTRACT FROM AUTHOR]

Automated software transformations and the process of automated program repair and improvement are an important aspect of modern software engineering practices. In this paper, we describe a system which uses a graph-based deep learning model that can be trained to automatically transform and improve computer programs. By operating on language-agnostic, universal graph-like structures easily extractable from source code files (abstract syntax trees), the deep learning agent learns which transformations should be effectively applied to various structures recognized in the source code in order to improve it. By defining a metric which we want to improve and introducing an optimization task—a reinforcement learning setting, an agent learns to automatically apply a chain of transformations to the program, drastically improving it. While similar program improvement processes exist, they exclusively use exhaustive search algorithms to try all the possible code transformations which is a long process susceptible to local optimum issues. Our solution aims to model and embed structural knowledge about the programs being transformed which greatly helps the agent to choose best possible code transformations to apply. Elements of the approach we present in this paper are further applicable not just to automatic software improvement tasks, but also to other code-related tasks. [ABSTRACT FROM AUTHOR]

In recent years, the rapid advancement of artificial intelligence (AI) technology has made remarkable strides in the music industry, profoundly impacting various aspects of our lives. This progress has extended to revolutionizing industries, streamlining processes, and enhancing personalization in music. AI's expanding capabilities have driven computer vision, natural language processing, and machine learning breakthroughs. As music enthusiasts increasingly turn to online platforms, the need for effective recommendation systems has become more pressing. The exponential growth of musical content poses a challenge in discovering music tailored to individual tastes. This paper proposed a personalized music recommendation system powered by deep learning, explicitly utilizing convolutional neural networks (CNNs) to address the challenges posed by expanding music libraries and diverse song resources. In today's era of ever-growing music databases and increasing content diversity, more than traditional retrieval methods are needed to efficiently connect users with their preferred track. Traditional music platforms rely solely on retrieval mechanisms, necessitating manual inputs such as artist names and titles to locate specific songs. However, this approach needs in-depth music data analysis and delivers personalized recommendations generated by various deep learning-based algorithms. This paper bridges this gap by harnessing deep learning algorithms for comprehensive music data analysis and personalized song suggestions. The proposed CNN-based recommendation system explores the domains of deep learning and content-based approaches, empowering users to discover their favorite tunes swiftly. This innovative mechanism leverages CNNs and deep learning algorithms to offer efficient, personalized, and data-driven music recommendations, transforming the user experience with their music collections. In experimental results, the proposed CNN model is compared with various deep learning algorithms, including DNN, RNN, and LSTM, as well as traditional algorithms like SVM, KNN, RF, and DT. Impressively, the CNN model achieved the highest accuracy at 95%, surpassing all other algorithms. [ABSTRACT FROM AUTHOR]

The emergence of data-driven analytics and intelligent monitoring systems is radically transforming the world of competitive sports. However, robust real-time activity recognition remains elusive for rapid-paced sports like table tennis due to the complexity of strokes and quick maneuvering. This paper tries to overcome this challenge through an innovative cloud-supported system integrating the Internet of Things (IoT), machine learning, and wearable sensors for automated analysis of table tennis gameplay. We strategically set up a multi-camera IoT system around a table tennis court to compile a rich labeled image dataset encapsulating over 15,000 frames. To improve model generalization, the images captured various playing styles, lighting conditions, and camera angles. We developed tailored SVM and CNN architectures optimized for table tennis activity classification. The models were trained on GPU-accelerated platforms using the curated dataset. After hyperparameter tuning and cross-validation, the CNN model achieved 96.2% accuracy, a precision value of 96.3%, a recall value of 95.2%, and an F1-score of 96.3% in classifying standard strokes and serves. Additionally, the models exhibited impressive processing speeds of 22–34 fps, enabling real-time utilization. Our proposed model outperformed the latest models in this field with higher accuracy, recall values, F1-score, and precision. While expanding the dataset diversity and testing variations of deep network architectures could further enhance performance, this paper demonstrates a crucial leap toward helping real-time analytics in table tennis using an AI-powered computer vision approach. The hybrid system combining IoT, wearables, and machine learning establishes a framework to transform data into actionable and timely insights for table tennis players and coaches. [ABSTRACT FROM AUTHOR]

Since frequent traffic accidents bring great losses to people's safety and social property, this paper takes the results of traffic accident risk prediction as the basis so that it can help city managers to reasonably deploy police force, relieve traffic pressure, avoid traffic accidents and provide safe guidance to pedestrians. Based on this paper, a deep learning framework including spatiotemporal attention mechanism is proposed to solve the problem of traffic accident prediction in urban areas, and the experimental simulation shows that the accuracy of traffic accident risk prediction proposed in this paper reaches 94%. [ABSTRACT FROM AUTHOR]

Advances in artificial intelligence and data acquisition technology are growing, the research on deep learning algorithm has gone deep into various fields. At this stage, the demand supply matching model under the comprehensive passenger transport hub travel system is obtained by analyzing the travel mode data. This paper takes traffic time as the research direction, uses machine learning and complex network theory to conduct in-depth learning algorithm research, respectively, discusses, explores and forecasts the traffic supply and demand mode data, and explores the traffic mode supply and demand model under the comprehensive passenger transport hub travel service system. It is shown in traffic data that using the prediction form of deep learning to predict traffic conditions and travel pressure can enable traffic managers to master traffic dynamics and lead the direction for future traffic development. Finally, from the perspective of MaaS system, the paper uses big data and information processing methods to explore the supply and demand matching model in terms of transportation modes. The research shows that MaaS system can make overall planning in terms of traffic resources, provide reasonable travel modes for traffic travelers and match corresponding travel services. It not only makes overall planning and guarantee for travel services, but also promotes the sustainable development of the transportation supply and demand system. [ABSTRACT FROM AUTHOR]

Machine intelligence methods originated as effective tools for generating learning representations of features directly from the data and have indicated usefulness in the area of deception detection. The success of machine intelligence-based methods covers resolving multiple complex tasks that combine multiple low-level image features with high-level contexts, from feature extraction to classification. The goal of this paper, given this period of rapid evolution, is to provide a detailed overview of the recent developments in the domain of automated deception detection mainly brought about by machine intelligence-based techniques. This study examines about 100 research papers that explores diverse areas of common deception detection through text, speech, and video data analysis. We performed a critical analysis of the existing techniques, tools and available datasets which have been used within the existing works, followed by possible directions for the future developments in this domain. [ABSTRACT FROM AUTHOR]

Image classification has received extensive attention as an important technical means of acquiring image information. It has been widely used in various engineering fields. Although the existing traditional image classification methods have been widely applied in practical problems, there are some problems in the application process, such as unsatisfactory effects, low classification accuracy and weak adaptive ability. This is because this type of method relies on the designer's prior knowledge and cognitive understanding of the classification task. At the same time, this method separates image feature extraction and classification into two steps for classification operation. However, the deep learning model has a powerful learning ability, which integrates the feature extraction and classification process into a whole to complete the image classification test, which can effectively improve the image classification accuracy. At the same time, the image classification method based on deep learning also has the following problems in the application process: First, it is impossible to effectively approximate the complex functions in the deep learning model. Second, the deep learning model comes with a low classifier with low accuracy. To this end, this paper introduces the idea of sparse representation into the architecture of deep learning network, comprehensively utilizes the sparse representation of good multidimensional data linear decomposition ability and the deep structural advantages of multi-layer nonlinear mapping to complete the complex function approximation in deep learning model. It constructs a deep learning model with adaptive approximation ability, which solves the function approximation problem of deep learning models. At the same time, in order to further improve the classification effect of the deep learning classifier, a sparse representation classification method based on the optimized kernel function is proposed to replace the classifier in the deep learning model, thereby improving the image classification effect. Based on the above explanation, this paper proposes an image classification algorithm based on the stacked sparse coding depth learning model-optimized kernel function nonnegative sparse representation. The experimental results show that the proposed method not only has a higher average accuracy than other mainstream methods, but also can be well adapted to various image databases. This is because the proposed method can extract more image feature information than the traditional image classification method and can better adaptively match the image information. Compared with other deep learning methods, it can better solve the problems of complex function approximation and poor classifier effect, thus further improving image classification accuracy. [ABSTRACT FROM AUTHOR]

Poultry farms across the world house animals such as cows, sheep's, pigs and hen These farms are the places from where the market gets meat, eggs, wool and other animal products which are used in our daily lives. But sometimes the business of the farms is severely hit primarily due to poultry animals catching diseases or these animals succumbing to injuries caused due to fighting each other. The identification of such animals showing unusual characteristics or behavior is necessary. We through our paper use deep learning concepts to help the poultry owner to identify these unusual characteristics in a sheep. Some of the characteristics that our paper identifies are skinny, redness, non-aggression, aggression, beefy and foraging. Our paper makes use of a video, shot from a camera and the implementation of a sequential model as well as make use of the SSD algorithm to identify and characterize the sheep's depicted in the video. The sequential model is trained using the training dataset which contains static images of sheep's and the attribute/characteristics depicted by those sheep's that are stored in a CSV file. The testing dataset contains the images that are extracted from the input video as frames. The testing dataset is passed through the sequential model to get the characteristics/attributes depicted by the sheep in each frame and store those characteristics/attributes in a CSV file. The SSD algorithm is trained on identifying the various animals and not only does it display the name of the animal detected, it displays the confidence percentage of the animal as well, in our paper it is used for identifying sheep. The above stated algorithm also creates a border around the identified sheep such that it can be used for tracking purposes during the entirety of the video. The SSD algorithm also takes in the attributes depicted by these frames and displays them along the border which identifies the sheep. The accuracy is compared with YOLO algorithms and shows 3 to 6% improvement on prediction rate. [ABSTRACT FROM AUTHOR]

Diabetic retinopathy (DR) is a complication of diabetes mellitus that leads to vision loss if not treated timely. Microaneurysms (MA) is the first clinical sign of DR. An automatic MA detection method through retinal fundus images has been proposed in this paper. The MA detection method consists of three steps: preprocessing, MA candidate detection, and pixel-wise classification. A novel convolutional neural network (CNN) architecture has been proposed in this paper to train MA and non-MA patches, and the majority voting technique has been used to detect the MA patches. The presented method has been evaluated using an openly available dataset, namely Retinopathy Online Challenge (ROC). The proposed method produces 92% of area under the receiver operating characteristics (AUC) curve, which is better than other state-of-the-art methods. The developed CNN architecture produced the highest accuracy even for fewer images, which helps effective DR screening. [ABSTRACT FROM AUTHOR]

Healthcare is the most important need of today's era. Healthcare refers to the improvement of the human health by preventing, curing, diagnosing, recovering from a health hazard caused. Thus, to improve the health condition of a human system technology, such as machine learning, deep learning and artificial intelligence has come into play. The combination of artificial technology with the health sector has made a huge impact and success on the world. Curing millions of diseases, analysis of various infections, providing accurate test results and high-level maintenance check are now all possible with the evolution of technology. Every part of human body can now be diagnosed and analyze to study all kinds of tissues, blood vessels, organs, cells for improvement of health and curing of diseases. Research sector has been working with a continuous pace to accomplish various studies to identify different body organs and have a descriptive study for the identification of proper working mechanism of the human body. One such study has also shown a huge progress in the recent times, the identification of glomeruli in human kidney tissue. The tiny ball like structured which is composed of blood vessels that has an actively participation in the filtration of the blood to form urine. Thus, the paper presents a deep learning-based model formed for the identification of these glomeruli present in the human kidney. After implementing, the proposed model obtained an accuracy of 99.68% with a dice coefficient of 0.9060. [ABSTRACT FROM AUTHOR]

Automatic target detection plays a significant role during war operations. The concept behind automatic target detection in war is military object recognition from the captured images. For object recognition in the given image, convolutional neural network (CNN) architectures are used efficiently. However, CNN suffers from a problem of location invariant and their performance depends mainly on the size of the training set. Generally, the available training data will be in less proportion for military objects due to its operational and security issues. Due to these two issues of CNN, the performance of CNN may degrade abruptly. To address military object recognition, a relatively new neural network architecture based on capsule network (CapsNet) is introduced in this paper. A variant of CapsNet called the multi-level CapsNet framework is projecting in this paper for efficient military object recognition under the case of a small training dataset. The validation of the introduced framework is done on a dataset of military objects collected from the Internet. The dataset contains particularly five different military objects and similar civil ones. Experiments demonstrate that the proposed framework can accomplish a high recognition precision, superior to many other algorithms such as conventional support vector machines and transfer learning-based CNNs. [ABSTRACT FROM AUTHOR]

This paper proposes to predict traffic accidents based on IoTs and deep learning to address the current problem of inaccurate traffic accident prediction. Since traditional traffic accident prediction often applies classical prediction algorithms to a small portion of data, the obtained models can only predict a small range of traffic accidents. Most accident prediction models are limited by the lack of data features, do not consider the problems of practical application scenarios, and do not incorporate regional heterogeneity, so the prediction accuracy of accident prediction models is poor. This paper analyzes and summarizes the relationship between traffic accidents and influencing factors from five aspects, such as people, vehicles, roads and environment, and proves the influence of regional heterogeneity on accidents, which paves the way for traffic accident prediction. The data and heterogeneous spatial data are preprocessed and feature selected, respectively. Logistic regression and random forest algorithm are used to train the corresponding prediction models. The results show that the prediction model combined with regional heterogeneity has better comprehensive performance than the original data. [ABSTRACT FROM AUTHOR]

This research paper proposes sentiment analysis of pets using deep learning technologies in the artificial intelligence of things system. Mask R-CNN is used to detect image objects and generate contour mask maps, pose analysis algorithms are used to obtain object posture information, and at the same time object sound signals are converted into spectrograms as features, using deep learning image recognition technology to obtain object emotion information. By using the fusion of object posture and emotional characteristics as the basis for pet emotion identification and analysis, the detected specific pet behaviour states will be actively notified to the owner for processing. Compared with traditional speech recognition, which uses mel-frequency cepstral coefficients for feature extraction, coupled with a Gaussian mixture model-hidden Markov model for voice recognition, the experimental method of this research paper effectively improves the accuracy by 70%. Prior work on the implementation of smart pet surveillance systems has used the pet's tail and mouth as features, and has combined these with sound features to analyse the pet's emotions. This research paper proposes a new method of sentiment analysis in pets, and our method is compared with previous related work. Experimental results show that our approach increases the accuracy rate by 70%. [ABSTRACT FROM AUTHOR]

The emergence of Artificial Intelligence (AI) provides brand-new technical support and a broader resource platform for college oral English teaching and evaluation. The advancement of science, technology, and networks has resulted in significant educational changes. Because of these, the demand for English instruction is rising quickly, along with the level of cultural integration. However, due to the limits of conventional teaching techniques and the slow expansion of information technology, oral English classes have been discarded and discontinued for a long time. Based on the preceding, this paper first addresses the shortcomings of traditional college-level spoken English instruction by exposing content and timing deficiencies. It then describes how spoken English data are collected, preprocessed, and framed before being standardized for consistency. After that, this paper uses mapping recognition results using a DBN speech model and correlation coefficients in the evaluation. In addition, it uses the Pairwise Variability Index in rhythm evaluation to assess stress distribution differences between test and standard pronunciation. Besides the above, the proposed approach uses Support Vector Machine (SVM) optimization inside statistical learning to improve phoneme recognition and reliability, especially when dealing with difficult-to-distinguish phoneme sets. Further analysis is carried out using neural networks, which include excitation functions and error computations. Finally, the complete design emphasizes user-centric functional characteristics such as practicability, user needs, and robust system management subsystems. Experiments suggest that the natural language processing-based oral English teaching mode can increase students' overall oral English skills. These results reveal that students' excitement increased by 33.3%, their verbal fluency increased by 86%, and their vocabulary learning increased by 16.1%. The results show that this method can effectively assist students in learning oral English. [ABSTRACT FROM AUTHOR]

Music expresses thoughts and emotions in artistic form and is made up of different components including harmony, rhythm, and melody. Several musical elements are tied together during the songwriting process in order to generate melodies that are harmonic. A music melody is the essential element of any music to generate strong feelings from listeners and capture their interest. In the process of music appreciation, melody controls the emotional changes of music. It is an efficiently perceived part and the tone of a song. In recent years, Sichuan unvoiced music has developed rapidly and attracted much attention. This paper selects Sichuan unvoiced music as the main research theme and constructs a melody generation algorithm by utilizing the state-of-the-art techniques of deep learning (DL) and evolutionary algorithms (EAs) such as recurrent neural network-long short-term memory (RNN-LSTM) and genetic algorithm (GA). Firstly, this paper briefly describes the concept of DL algorithms, the deep generation model, and sequence to sequence model, as they constitute the technological foundation for this research. Secondly, this paper proposes a melody generation algorithm that utilizes RNN-LSTM for melody generation and GA for melody optimization. More specifically, the melody is generated by preprocessing data, creating, and training the RNN-LSTM model. A GA was used to determine the melodic fitness function for eight songs as the fitness function directly affects the selected termination condition. The fitness function can be thought of as either a person or an evolutionary rule. Finally, the average score of these songs, before and after evolution, is calculated, which demonstrates that the analysis and rotation creation methods are more precise and that the song's average melody score is higher. The method proposed in this study has been thoroughly compared to the existing approaches proposed in earlier studies, and it was found that the approach we propose is more effective in terms of accuracy and the average melody score. [ABSTRACT FROM AUTHOR]

In the analysis of agricultural product price time series, the detection of abnormal fluctuations is the primary task. Accurately judging the abnormal fluctuations of agricultural product prices can help farmers avoid potential economic losses in a timely manner and improve the circulation efficiency of agricultural products. This paper introduces a convolutional neural network (CNN) classification model based on time series images (TSI), and identifies abnormal fluctuations in agricultural product prices using improved standard deviation–slope judgment. First, a standard deviation–slope (SDS) time series abnormal fluctuation criterion is proposed. Second, the Markov Transfer Field (MTF) method is used to convert the sparse one-dimensional time series of agricultural product prices into a two-dimensional dense image. Third, the CNN model is used to automatically extract features from time series images containing abnormal fluctuations, and they are divided into two categories: "normal" and "abnormal"; Finally, the performance of the proposed model was evaluated using China's corn and wheat price dataset. Comparing with other abnormal fluctuation judgment methods, the accuracy of the proposed algorithm is about 20% higher on average. This confirms the applicability of the standard deviation–slope time series Image-Resnet-34 (SDS-TSI-Resnet34) model in practical scenarios. At the end of the paper, some feasible suggestions for the efficient development of agricultural economy are proposed based on the abnormal fluctuation judgment method. [ABSTRACT FROM AUTHOR]

The world population continues to increase which imposes rising demand in agriculture production. How to improve crop breeding to feed the growing population is a significant challenge. The traditional crop breeding is resource-intensive and time-consuming. Predictive modeling on crop yield can speed up the breeding process and make it resource-efficient. In this paper, a genetic algorithm (GA)-assisted deep learning solution method is proposed for the crop yield prediction. The proposed method consists of two phases, i.e., the global search phase and the local search phase. In the global search phase, GA is used to search for the best initial weights of the neural network. In the local search phase, random perturbation is added to avoid the local optimum and vanishing gradient problems. A case study of crop yield prediction is conducted to compare the proposed method and other gradient-based methods. The results show that the proposed method outperforms the gradient-based methods in terms of convergence speed and prediction accuracy. [ABSTRACT FROM AUTHOR]

In recent years deep learning models improve the diagnosis performance of many diseases especially respiratory diseases. This paper will propose an evaluation for the performance of different deep learning models associated with the raw lung auscultation sounds in detecting respiratory pathologies to help in providing diagnostic of respiratory pathologies in digital recorded respiratory sounds. Also, we will find out the best deep learning model for this task. In this paper, three different deep learning models have been evaluated on non-augmented and augmented datasets, where two different datasets have been utilized to generate four different sub-datasets. The results show that all the proposed deep learning methods were successful and achieved high performance in classifying the raw lung sounds, the methods were applied on different datasets and used either augmentation or non-augmentation. Among all proposed deep learning models, the CNN–LSTM model was the best model in all datasets for both augmentation and non-augmentation cases. The accuracy of CNN–LSTM model using non-augmentation was 99.6%, 99.8%, 82.4%, and 99.4% for datasets 1, 2, 3, and 4, respectively, and using augmentation was 100%, 99.8%, 98.0%, and 99.5% for datasets 1, 2, 3, and 4, respectively. While the augmentation process successfully helps the deep learning models in enhancing their performance on the testing datasets with a notable value. Moreover, the hybrid model that combines both CNN and LSTM techniques performed better than models that are based only on one of these techniques, this mainly refers to the use of CNN for automatic deep features extraction from lung sound while LSTM is used for classification. [ABSTRACT FROM AUTHOR]

Sentiment analysis (SA) is widely applied in practical applications and known as emotion AI or opinion mining. In fact, SA tasks are really hard even for human; there are many factors that can affect the overall performance such as the length of the text, idiom, metaphor, or slang. In this paper, mainly proposed multi-channel word embeddings for SA consists of three parts: improving word representation, applying attention mechanism, and applying the state-of-the-art deep learning models. Particularly, a better representation, first, allows our method to achieve accurate and higher performance. Second, the attention mechanism enhances the model ability in focusing on the place from where the useful and important information can be extracted. In final, in order to benchmark our proposed method, we implement deep learning models such as CNN, RNN, CNN variants. The experimental results show that our method achieved a higher performance compared to baseline methods. The experiment results highlighted the main contributions based on better word representation of multi-channel pre-trained word embeddings were shown. In addition, this model focuses on the words that contain useful and important information, and a higher accuracy was found compared to the previous models. [ABSTRACT FROM AUTHOR]

The performance of target detection algorithms for manual features is gradually saturated, and the development of target detection has stagnated. Computer vision is a discipline that studies how to use computers to replace the human eye and use visual information and visual algorithms to automatically detect, recognize, and track target objects. Target detection has also been fully developed as a basis for solving advanced vision tasks such as target tracking, instance segmentation, image understanding, and behavior recognition. The theories of artificial intelligence and deep learning has gradually developed, making convolutional neural networks a hot spot of attention in the field of computer vision and image processing. Based on the work in the field of target detection, this paper proposes a target detection algorithm based on the prior knowledge of the impact area of the tennis ball. Our experiments and evaluations show that the algorithm can achieve higher detection accuracy and faster detection speed. Moreover, the proposed algorithm accurately tracks and judge the impact area of tennis sports, and improve the accuracy and reliability of the tennis impact. [ABSTRACT FROM AUTHOR]

The existing research on data analysis and personalized recommendation algorithms for music mainly focuses on improving traditional recommendation systems such as classical collaborative filtering recommendation, content recommendation, and model recommendation. However, there has been no efficient solution to the cold start of recommendation systems and the sparse nature of scoring matrices. There is a dire need for an intelligent and efficient algorithm that can support the students in teaching music in an effective way. Deep belief neural network (DBN-DNN) is a deep learning model nested by many restricted Boltzmann machines (RBMs) to deal with sparse matrices and immediacy. At the same time, it can generate better initialization parameters and significantly improve the training speed of the model. This paper utilizes a decision support system that decides the matrix decomposition of user ratings by using a hidden semantic model to extract user preference features for hidden factors and the weights of music pieces on these k hidden factors. The proposed approach extracts the audio feature values from music clips by considering that music itself has classifiable information. This method is intelligent in terms of efficiency for effective music teaching and helps students learn. The experiments show that the personalized recommendation algorithm combined with a deep belief neural network has good recommendation performance. The recommendation coefficient predicted by the model is lower than the real rating for the samples with ratings. The recommendation is no longer limited to the original user group and song library and has better recommendation accuracy and scalability. [ABSTRACT FROM AUTHOR]

In high-throughput phenotyping, recognizing individual plant categories is a vital support process for plant breeding. However, different plant categories have different fine-grained characteristics, i.e., intra-class variation and inter-class similarity, making the process challenging. Existing deep learning-based recognition methods fail to effectively address this recognition task under challenging requirements, leading to technical difficulties such as low accuracy and lack of generalization robustness. To address these requirements, this paper proposes PlantNet, a fine-grained network for plant recognition based on transfer learning and a bilinear convolutional neural network, which achieves high recognition accuracy in high-throughput phenotyping requirements. The network operates as follows. First, two deep feature extractors are constructed using transfer learning. The outer product of the different spatial locations corresponding to the two features is then calculated, and the bilinear convergence is computed for the different spatial locations. Finally, the fused bilinear vectors are normalized via maximum expectation to generate the network output. Experiments on a publicly available Arabidopsis dataset show that the proposed bilinear model performed better than related state-of-the-art methods. The interclass recognition accuracy of the four different species of Arabidopsis Sf-2, Cvi, Landsberg and Columbia are found to be 98.48%, 96.53%, 96.79% and 97.33%, respectively, with an average accuracy of 97.25%. Thus, the network has good generalization ability and robust performance, satisfying the needs of fine-grained plant recognition in agricultural production. [ABSTRACT FROM AUTHOR]

In recent years, with the construction and development of smart cities, text recognition in building images can not only achieve geolocation but also provide guiding significance for GIS mapping and automatic updating. Since buildings have different orientations, angles and shapes, it is difficult to recognize textual features in images. With the wide application of convolutional neural networks and recurrent neural networks in image processing, this paper proposes a BFPN-RCNN algorithm for detecting and recognizing curved text in architectural images. A comparison with other image detection algorithms on different datasets proves that the algorithm can effectively identify curved text at different angles in natural scene images. [ABSTRACT FROM AUTHOR]

Invasive ductal carcinoma (IDC) is a common form of breast cancer that affects women. In traditional medical practice, physicians have to manually test and classify areas which are suspected to be cancerous. However, the literature strongly indicates that the manual segmentation process performed by medical practitioners is neither time efficient nor accurate, as it relies on their subjective judgment. This paper introduces a model called residual attention neural network breast cancer classification (RANN-BCC) to help medical practitioners in the cancer diagnostic process. RANN-BCC utilizes residual neural network (ResNet) as an expert-supportive method to aid medical practitioners in cancer diagnosis. The implementation of RANN-BCC can support the classification of whole slide imaging (WSI) into non-IDC and IDC without prior information about the presence of a cancerous lesion. The classification results demonstrate that the RANN-BCC model has achieved 92.45% accuracy, 0.98 recall, 0.91 precision, and 0.94 F-score which has outperformed other models such as CNN, AlexNet, Residual Neural Network 34 (ResNet34), and Feed-Forward Neural Network. The developed RANN-BCC model aims to help medical experts to classify IDC and non-IDC of breast cancer by learning the feature content of medical images. [ABSTRACT FROM AUTHOR]

This paper discussed an efficient technique to detect near-duplicate images with higher accuracy. The proposed method improved the existing accuracy of near-duplicate image detection and classification using Haar wavelet feature extraction and classification algorithm. Due to advancements in digital image acquisition and image editing software, near-duplicate image counterfeiting has become predictable. This expanding problem needs the creation of sophisticated detection algorithms capable of effectively preventing the spread of near-duplicate image deception. Support vector machine (SVM) and K-near neighbor (KNN) classifiers are used in the classification stage of a modified deep learning model with a convolutional neural network (CNN). The classification model finds the better result on the Imperial College London dataset, with 99.12% using CNN with KNN classifier and 98.63% for CNN with SVM classifier. The classification model also gives better results on the Nanyang Technological University dataset, with 98.36% using CNN with the KNN classifier and 97.85% for CNN with the SVM classifier. The model is tested using publicly available images to find the results. According to the comparison results, this model performed significantly better result than the previous model. As a result, digital image forensics has become more critical, particularly in identifying image duplication. [ABSTRACT FROM AUTHOR]

Fractional uncertain differential equations are used for forecast in this paper. It is crucial to provide an accurate estimation method since fractional uncertain differential equations hold memory effects and parameter estimation strongly depends on the accuracy of numerical schemes. A generalized moment estimation method is developed by the famous L 1 numerical scheme. Adam optimization algorithm is adopted in the minimum optimization problem and the obtained optimal parameters can pass the uncertain hypothesis test. Compared with the uncertain differential equations, the fractional one holds more parameter freedom degrees and captures long-term interactions of observed data. Finally, the applications to aftershock frequency and pharmacokinetics are provided as examples. It can be concluded that fractional uncertain differential equations lead to the higher forecast accuracy which also provides some potential works on "fractional deep learning" under uncertain environments. [ABSTRACT FROM AUTHOR]

Advertising Click-Through Rate (CTR) prediction is a relatively successful application in the field of recommendation system. Improving the accuracy of advertising click-through rate can not only make the user experience better, but also give more benefits to advertising platforms and advertisers. It can be seen from the research status that the interaction between learning features has become a very important part of the advertising CTR prediction model. Although the existing CTR prediction models based on deep learning have achieved good results, some models only consider a single interaction mode and lack the diversity of feature interaction. To resolve this problem, this paper proposes a CTR prediction model based on multi-feature interaction, called EDIF, which aims to enhance the diversity of feature interaction. Firstly, the model learns multiple different embedding vectors for each feature in the embedding layer, which reflects the correlation between features; secondly, in the interaction of high-order features, the embedding vectors of each feature are added and pooled to form the aggregation vector of the whole feature as the input, which reflects the integrity of the feature; finally, after the feature embedding operation, the model introduces two layers in parallel: compressed excitation network layer and explicit high-order interaction layer, which improves the ability of feature interaction. We have done a lot of experiments on two public data sets, Avazu and Criteo. The results show that the model effect of this paper has great advantages over the latest model. [ABSTRACT FROM AUTHOR]

Accurate sensing and understanding of gestures can improve the quality of human–computer interaction and show great theoretical significance and application potentials in the fields of smart home, assisted medical care and virtual reality. WiFi channel state information (CSI)-based device-free wireless gesture recognition requires no sensors and has a series of advantages such as permission for non-line-of-sight scenario, low cost, preserving for personal privacy and working in the dark night. Although most of the current WiFi CSI-based gesture recognition approaches can achieve good performance, they are difficult to adapt to the new domains. Therefore, this paper proposes ML-WiGR, a novel approach for device-free gesture recognition in cross-domain applications. ML-WiGR applies convolutional neural networks (CNN) and long short-term memory (LSTM) neural networks as the basic model for gesture recognition to extract spatial and temporal features. Combined with the meta-learning training mechanism, ML-WiGR can dynamically adjust the learning rate and meta-learning rate in training process adaptively and optimize the initial parameters of a basic model for gesture recognition, only using a few samples and several iterations to adapt to the new domain. In the experiments, the approach is tested under a variety of scenarios. The results show that ML-WiGR can achieve comparable performance against existing approaches with only a small number of samples for training in cross-domains. [ABSTRACT FROM AUTHOR]

Recently, a deep learning model, the deep forest (DF), was designed as an alternative to deep neural networks. Each cascade layer of the DF contains a set of random forests (RFs) with a large number of decision trees, some of which are of high redundancy and poor performance. To avoid the negative impacts of such decision trees, this paper proposes to optimize RFs in each cascade layer of the DF so as to realize a pruned deep forest (PDF) with higher performance and smaller ensemble size. In this paper, a new ordering-based ensemble pruning method is proposed based on feature vectorization and quantum walks. This method simultaneously considers the accuracy and the diversity of base classifiers, and it provides an integrated evaluation criterion for ordering base classifiers in the ensemble system. The effectiveness of the proposed method is verified by experiments and discussions. [ABSTRACT FROM AUTHOR]

Retinal diseases are most widely affected by various types of people without any proper reason. Nowadays, many retinal diseases are identified by experts. Detecting retinal diseases in the early stages is very important with better accuracy. Deep learning (DL) techniques are commonly used in the early prediction of retinal disorders. In DL, multiple layers accurately detect abnormalities in the retinal images. Various datasets are also present for this research work. This paper uses a hybrid multilayered classification (HMLC/CNN-VGG19). This system is developed to categorize four kinds of retinal disorders (age-related macular degeneration, choroidal neovascularization, Drusen, diabetic retinopathy, as well as typical cases). The proposed H.M.L.C. is applied to OCT to gather pictures from different data sources, like the U.C.I. repository, Kaggle, etc. The CNN and VGG-19 models used in the H.M.L.C. are implemented in Python over the datasets. The experimental results in terms of classification accuracy are verified. The classification accuracy is high since the H.M.L.C. used the advanced features from CNN and VGG-19 models. The performance is calculated using sensitivity, specificity, F1 score, and accuracy. [ABSTRACT FROM AUTHOR]