Search

With the development of hand-held photographic devices, document images in unconstrained environments can be captured in high-speed and high-resolution. It will be more efficient to process the text information of multiple documents simultaneously. In this paper, we propose a multi-document detection approach. It can estimate the amount of documents and also detect their accurate locations via iteratively searching the four corners and their direction maps from individual document in the image. Even for slightly occluded documents, the proposed method can infer the hidden corner positions. The model is designed to jointly learn the corner categories, locations and their directions in attentional regions via two branches of the same sequential prediction process. The association score is calculated based on the them between two corner connections. The graph theory, considering corners as nodes and association scores as edges, is applied to get the quadrangle for each document in image. For evaluation, we collect a Multi-Doc data set which contains 2,200 document images in various natural scenes. We show that the baseline model trained on this collected data and bench-mark SmartDoc 2015 can detect both single and multiple documents accurately and effectively.

In this paper we present C o R o NN a a deep sequential framework for epidemic prediction that leverages a flexible combination of sequential and convolutional components to analyse the transmission of COVID-19 and, perhaps, other undiscovered viruses. Importantly, our methodology is generic and may be tailored to specific analysis goals. We exploit C o R o NN a to analyse the impact of various mobility containment policies on the pandemic using cumulative viral dissemination statistics with local demographic and movement data from several nations. Our experiments show that data on mobility has a significant, but delayed, impact on viral propagation. When compared to alternative frameworks that combine multivariate lagged predictors and basic LSTM models, C o R o NN a outperforms them. On the contrary, no technique based solely on lagged viral dissemination statistics can forecast daily cases.

The accurate prediction of human movement trajectory has a variety of benefits for many applications such as optimizing nurse’s trajectory in a hospital, optimizing movements of old or disabled people to minimize their routine efforts, etc. To perform human movement prediction, large amount of historical positioning data from sensors has to be collected and mined. We analyzed different human sequential movement prediction approaches and their limitations. In this work, we propose a new classifier named Apriori based Probability Tree Classifier (APTC) which predicts the human movement sequence patterns in indoor environment. The APTC classifier is integrated into Bagged J48 Machine learning algorithm which results in an ensemble model to predict the future human movement patterns. The patterns are mined based on spatial, temporal and social data which add more accuracy to our prediction. Our model also performs clustering mechanism to detect the abnormal patterns.

In this study, the authors use deep learning sequence prediction models for the continuous monitoring of the epidemic while considering the potential impacts of Bacille Calmette-Guérin (BCG) vaccination and tuberculosis (TB) infection rates in populations. Three models were built based on the epidemic data evolution in several countries between the date of their first case and April 1, 2020. The data was based on 14 variables for cases prediction, 15 variables for recoveries prediction, and 16 variables for deaths prediction. Prevision results were very promising, and the suspicions on the BCG vaccination and TB infections rates' implications turned out to be warranted. The model can evolve by continuously updating and enriching data, adding the experiences of all affected countries.

This article presents a novel methodology to detect insurance claim related frauds in the healthcare system using concepts of sequence mining and sequence prediction. Fraud detection in healthcare is a non-trivial task due to the heterogeneous nature of healthcare records. Fraudsters behave as normal patients and with the passage of time keep on changing their way of planting frauds; hence, there is a need to develop fraud detection models. The sequence generation is not the part of previous researches which mostly focus on amount based analysis or medication versus diseases sequential analysis. The proposed methodology is able to generate sequences of services availed or prescribed by each specialty and analyse via two cascaded checks for the detection of insurance claim related frauds. The methodology addresses these challenges and self learns from historical medical records. It is based on two modules namely “Sequence rule engine and Prediction based engine”. The sequence rule engine generates frequent sequences and probabilities of rare sequences for each specialty of the hospital. The comparison of such sequences with the actual patient sequences leads to the identification of anomalies as both sequences are not compliant to the sequences of the rule engine. The system performs further in detail analysis on all non-compliant sequences in the prediction based engine. The proposed methodology is validated by generating patient sequences from last five years transactional data of a local hospital and identifies patterns of service procedures administered to patients using Prefixspan algorithm and Compact prediction tree. Various experiments have been performed to validate the applicability of the developed methodology and the results demonstrate that the methodology is pertinent to detect healthcare frauds and provides on average 85% of accuracy. Thus can help in preventing fraudulent claims and provides better insight into how to improve patient management and treatment procedures.

Hard interaction learning between source sequences and their next targets is challenging, which exists in a myriad of sequential prediction tasks. During the training process, most existing methods focus on explicitly hard interactions caused by wrong responses. However, a model might conduct correct responses by capturing a subset of learnable patterns, which results in implicitly hard interactions with some unlearned patterns. As such, its generalization performance is weakened. The problem gets more serious in sequential prediction due to the interference of substantial similar candidate targets. To this end, we propose a Hardness Aware Interaction Learning framework (HAIL) that mainly consists of two base sequential learning networks and mutual exclusivity distillation (MED). The base networks are initialized differently to learn distinctive view patterns, thus gaining different training experiences. The experiences in the form of the unlikelihood of correct responses are drawn from each other by MED, which provides mutual exclusivity knowledge to figure out implicitly hard interactions. Moreover, we deduce that the unlikelihood essentially introduces additional gradients to push the pattern learning of correct responses. Our framework can be easily extended to more peer base networks. Evaluation is conducted on four datasets covering cyber and physical spaces. The experimental results demonstrate that our framework outperforms several state-of-the-art methods in terms of top-k based metrics., 10 pages; 4 figures; Proceedings of the 30th ACM International Conference on Information and Knowledge Management (CIKM '21), November 1--5, 2021, Virtual Event, QLD, Australia

Laboratory performance testing of the phase angle of asphalt concrete (AC) mixtures is often expensive and requires enormous human effort and time. To circumvent this problem, several regre...

keypoint detection is mainly a single-type target detection. In this paper, we present an approach for multi-type target keypoints detection. To allow for a learning-based approach, we developed large scale-dataset of multiple-type of UAVs and customized the OpenPose network architecture. The architecture is designed to jointly learn part locations and their association via two branches of the same sequential prediction process. The proposed approach learns from three different (Skylark fixed-wing, the Phantom 4 multi-rotor, and aggregated) UAVs datasets. Furthermore, to evaluate our approach, we conduct an extensive experiment on all three datasets. The results show the effectiveness of the proposed approach with respect to accuracy and dramatically improves the speed compared to other state-of-the-art methods.

Dementia is a combination of systematic symptoms of a long-term decline in human memory and thinking capabilities, typically caused by the aging process. The primary aim of this research is to employ a human activity prediction algorithm to distinguish between healthy subjects and dementia-affected patients, in order to provide diagnosis at the early stages of dementia. A new algorithm, viz. Sequence Prediction via All Discoverable Episodes (SPADE), is introduced in this research to find out distinct parameters that can be used to deliver successful diagnosis. Since dementia patients do not tend to have a recognisable activity pattern, this would make it difficult for the algorithm to function well. The experiment results establish a noticeable difference of 11% in the peak accuracy of sequence prediction performed between healthy adults and dementia-affected residents. SPADE has achieved an average accuracy of 80%, i.e. 12% improvement over M-SPEED in predicting future events. This is thus evidenced that the activity prediction algorithms possess the potentials to detect the early symptoms of dementia without using any expensive clinical procedures.

The generalization of predictive models that follow the standard risk minimization paradigm of machine learning can be hindered by the presence of spurious correlations in the data. Identifying invariant predictors while training on data from multiple environments can influence models to focus on features that have an invariant causal relationship with the target, while reducing the effect of spurious features. Such invariant risk minimization approaches heavily rely on clearly defined environments and data being perfectly segmented into these environments for training. However, in real-world settings, perfect segmentation is challenging to achieve and these environment-aware approaches prove to be sensitive to segmentation errors. In this work, we present an environment-agnostic approach to develop generalizable models for classification tasks in sequential datasets without needing prior knowledge of environments. We show that our approach results in models that can generalize to out-of-distribution data and are not influenced by spurious correlations. We evaluate our approach on real-world sequential datasets from various domains.

Sequential prediction is an essential component of Artificial Intelligence (AI) and Machine Learning (ML) in a number of domains. In this paper, we analyze the performance of sequential prediction algorithms that are used to predict the next event in smart environments. We first employ three sequential prediction algorithms, namely LZ78, Active LeZi (ALZ) and the Sequence Prediction via Enhanced Episode Discovery (SPEED) to build sequential trees, and then an efficient prediction algorithm with different window lengths is applied to these trees to predict the next sensor event for a sequence of activities. The sequential algorithms are tested with different window lengths for a dataset consists of 25 sensors. The results show that the best prediction accuracy is achieved by SPEED with a peak accuracy of 92.86%, but with higher storage requirement compared to LZ78 and ALZ.

In this article, we propose a hybrid-Bayesian frequentist approach using a Bayesian sequential prediction of the index of satisfaction. For interim analysis that addresses prediction hypoth...

Predicting a click sequence in mobile applications improves the user experience in various ways. By predicting which button will be clicked next, one can predict how the application will work and how the device will operate. However, predicting the click sequence is difficult because of the problems involved in collecting click sequences in real application usage. More importantly, accurate predictions are extremely challenging. In this paper, we address these issues. We propose PathFinder, a scheme for collecting click events and based on them predicting the next click in the application. The clicks are collected with the Android Accessibility Service and the next click is predicted via long short-term memory (LSTM). For the prediction, the base click sequence model is first generated from all users’ data; then, a personalized model is trained with an individual click sequence. As training data considerably influences the performance of LSTM, several techniques are developed to enhance the quality of the training data. The experimental results for 100 popular applications showed that the coverage and accuracy of click sequence tracing were 95% and 96%, respectively. Furthermore, PathFinder predicted the top three buttons that would be clicked next with a 0.76 F -measure for 1 775 043 real click data.

An automated smart home system is a key contributor to user assistance technology in modern civilization. Crucial merit of such a system is its ability to train itself through recorded data and recognize patterns in resident behaviors. Lack of sufficient prediction accuracy, exponential memory consumption, and extensive runtime prevent many of the current activity prediction approaches from being seamlessly integrated into consumer residences. This research introduces a sequence prediction algorithm which uses a prefix tree-based data model in order to learn and predict user actions. The algorithm applies episode discovery to detect correlated sensor events and learns the activities using a lossless data compression technique. This process assigns a probability of occurrence to sensor events and uses these probabilities to detect patterns in resident behavior. A complexity analysis of the algorithm is done to prove its efficiency in terms of memory usage and runtime. Using the presented technique, predictions are performed on popular datasets and contrasted with existing algorithms. The proposed algorithm achieves an 8.22% improvement in prediction accuracy over its predecessors, along with 66.69% better memory efficiency and 37% faster runtime.

As an important index of aero-engine, Remaining Useful Life (RUL) is the key content of prediction. Due to the good generation characteristics of Variational Auto-encoder (VAE) and Generation Adversarial Network (GAN) networks, this paper proposes a Health Index (HI) curve generation method based on VAE-GAN. After that, sensor sequence prediction is carried out through Bidirectional Long Short-Term Memory Network (BLSTM). The two networks are parallel, and then RUL prediction is carried out by synthesizing the data of the two networks. As far as the author knows, this is the first use of VAE-GAN in Prognostics Health Management (PHM). It is verified on the Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) dataset. Finally, the results show that the VAE-GAN network is effective and superior in RUL prediction. At the same time, the proposed parallel network is superior to other RUL prediction methods by generating HI curves.

In the third paper in a series of papers on autism savants, detection of giftedness and the use of mental arithmetic as an intervention in autism and a practice of metal wellness, we describe the use of python scripts towards primality detection exercises, of both small primes and arbitrary sized numbers and several other exercises including sequence prediction, inspired by branch prediction architectures. Sequence prediction as a mental exercise, is used as infotainment and as a wellness exercise, and a possible intervention in ASD. Several prediction mechanisms inspired by data and prediction algorithms are described. Keywords: ASD, Autism Savants, Education For The Gifted, primality detection, data mining, basket of associations, sequences, branch prediction, plotting graphs.

Sepsis is a leading cause of morbidity and mortality worldwide. Early identification of sepsis is important as it allows timely administration of potentially life-saving resuscitation and antimicrobial therapy. We present COMPOSER (COnformal Multidimensional Prediction Of SEpsis Risk), a deep learning model for the early prediction of sepsis, specifically designed to reduce false alarms by detecting unfamiliar patients/situations arising from erroneous data, missingness, distributional shift and data drifts. COMPOSER flags these unfamiliar cases as indeterminate rather than making spurious predictions. Six patient cohorts (515,720 patients) curated from two healthcare systems in the United States across intensive care units (ICU) and emergency departments (ED) were used to train and externally and temporally validate this model. In a sequential prediction setting, COMPOSER achieved a consistently high area under the curve (AUC) (ICU: 0.925–0.953; ED: 0.938–0.945). Out of over 6 million prediction windows roughly 20% and 8% were identified as indeterminate amongst non-septic and septic patients, respectively. COMPOSER provided early warning within a clinically actionable timeframe (ICU: 12.2 [3.2 22.8] and ED: 2.1 [0.8 4.5] hours prior to first antibiotics order) across all six cohorts, thus allowing for identification and prioritization of patients at high risk for sepsis.

Sequence prediction problems have been a major problem for a long time. Recurrent Neural Network (RNN) has been a good solution for sequential prediction problems. This work aims to create a generative model for text. Even though, RNN has its own limitations such as vanishing and exploding gradient descent problems, and inefficiency to keep track of long-term dependencies. To overcome these drawbacks, Long Short Term Memory (LSTM) has been a path-breaking solution to deal with sequential data and text data in particular. This paper delineates the design and working of text generation using word-level LSTM-RNN.

In this paper we propose a novel method for location recognition and prediction in smart homes based on semi-supervised learning. We use data collected from low-resolution depth video cameras installed in four apartments with older adults over 70 years of age, and collected during a period of one to seven weeks. The location of the person in the depth images is detected by a person detection algorithm adapted from YOLO (You Only Look Once). The locations extracted from the videos are then clustered using K-means clustering. Sequence prediction algorithms are used to predict the next cluster (location) based on the previous clusters (locations). The accuracy of predicting the next location is up to 91%, a significant improvement compared to the case where binary sensors are placed in the apartment based on human intuition. The paper presents an analysis on the effect of the memory length (i.e. the number of previous clusters used to predict the next one), and on the amount of recorded data required to converge.

Clinical event sequences consist of thousands of clinical events that represent records of patient care in time. Developing accurate prediction models for such sequences is of a great importance for defining representations of a patient state and for improving patient care. One important challenge of learning a good predictive model of clinical sequences is patient-specific variability. Based on underlying clinical complications, each patient’s sequence may consist of different sets of clinical events. However, population-based models learned from such sequences may not accurately predict patient-specific dynamics of event sequences. To address the problem, we develop a new adaptive event sequence prediction framework that learns to adjust its prediction for individual patients through an online model update.

Predicting the future paths of an agent's neighbors accurately and in a timely manner is central to the autonomous applications for collision avoidance. Conventional approaches, e.g., LSTM-based models, take considerable computational costs in the prediction, especially for the long sequence prediction. To support more efficient and accurate trajectory predictions, we propose a novel CNN-based spatial-temporal graph framework GraphTCN, which models the spatial interactions as social graphs and captures the spatio-temporal interactions with a modified temporal convolutional network. In contrast to conventional models, both the spatial and temporal modeling of our model are computed within each local time window. Therefore, it can be executed in parallel for much higher efficiency, and meanwhile with accuracy comparable to best-performing approaches. Experimental results confirm that our model achieves better performance in terms of both efficiency and accuracy as compared with state-of-the-art models on various trajectory prediction benchmark datasets., 10 pages, 7 figures, 3 tables

This paper presents an analysis of informative presentations using sequential multimodal modeling for automatic assessment of presentation performance. For this purpose, we transform a single video into multiple time-series segments that are provided as inputs to sequential models, such as Long Short-Term Memory (LSTM). This sequence modeling approach enables us to capture the time-series change of multimodal behaviors during the presentation. We proposed variants of sequential models that improve the accuracy of performance prediction over non-sequential models. Moreover, we performed segment analysis on the sequential models to analyze how relevant information from various segments can lead to better performance in sequential prediction models.

Human beings nowadays spend more than 90% of the lifetime indoors, leading to the dramatic increase of energy consumption in various buildings. Therefore, research regarding the environment friendly building becomes much more popular recently in which the prediction of energy consumption is a promised method. Nevertheless, the accuracy of prediction is not sound due to insufficient samples. A novel data generation model, termed HMSP, based on the generative adversarial networks, is proposed in this paper to generate much more data robustly, depending on a small number of samples available. The prediction CV-RMSE results, adopting data from the hybrid model, reach 3.03% at best and 7.99% at worst respectively compared to the samples recorded.

Automatic text summarization is one of the mentionable research areas of natural language processing. The amount of data is increasing rapidly, and the necessity of understanding the gist of any text is just a mandatory tool, nowadays. The area of text summarization has been developing since many years. Mentionable research has been already done through extractive summarization approach; in other side, abstractive summarization approach is the way to summarize any text as like human. Machine will be able to provide a new type of summarization, where the understanding of given summary may found as like as human-generated summary. Several research developments have already been done for abstractive summarization in English language. This paper shows a necessary method—“text generation” in context of Bengali abstractive text summarization development. Text generation helps the machine to understand the pattern of human-written text and then produce the output as is human-written text. A basis recurrent neural network (RNN) has been applied for this text generation approach. The most applicable and successful RNN—long short-term memory (LSTM)—has been applied. Contextual tokens have been used for the better sequence prediction. The proposed method has been developed in the context of making it useable for further development of abstractive text summarization.

With the rise of Big Data, there has been a significant effort in increasing compute power through GPUs, TPUs, and heterogeneous architectures. As a result, many applications are memory bound, i.e., they are bottlenecked by the movement of data from main memory to compute units. One way to address this issue is through data prefetching, which relies on accurate prediction of memory accesses. While recent deep learning models have performed well on sequence prediction problems, they are far too heavy in terms of model size and inference latency to be practical for data prefetching. Here, we propose extremely compact LSTM models that can predict the next memory access with high accuracy. Prior LSTM based work on access prediction has used orders of magnitude more parameters and developed one model for each application (trace). While one (specialized) model per application can result in more accuracy, it is not a scalable approach. In contrast, our models can predict for a class of applications by trading off specialization at the cost of few retraining steps at runtime, for a more generalizable compact meta-model. Our experiments on 13 benchmark applications demonstrate that three compact meta-models can obtain accuracy close to specialized models using few batches of retraining for majority of the applications.

Speech recognition is a sequence prediction problem. Besides employing various deep learning approaches for framelevel classification, sequence-level discriminative training has been proved to be indispensable to achieve the state-of-the-art performance in large vocabulary continuous speech recognition (LVCSR). However, keyword spotting (KWS), as one of the most common speech recognition tasks, almost only benefits from frame-level deep learning due to the difficulty of getting competing sequence hypotheses. The few studies on sequence discriminative training for KWS are limited for fixed vocabulary or LVCSR based methods and have not been compared to the state-of-the-art deep learning based KWS approaches. In this paper, a sequence discriminative training framework is proposed for both fixed vocabulary and unrestricted acoustic KWS. Sequence discriminative training for both sequence-level generative and discriminative models are systematically investigated. By introducing word-independent phone lattices or non-keyword blank symbols to construct competing hypotheses, feasible and efficient sequence discriminative training approaches are proposed for acoustic KWS. Experiments showed that the proposed approaches obtained consistent and significant improvement in both fixed vocabulary and unrestricted KWS tasks, compared to previous frame-level deep learning based acoustic KWS methods., accepted by Speech Communication, 08/02/2018

A multiway classification model based on CNN is proposed in this paper for Breast Cancer Histopathalogy images of varied magnification level (40X, 100X, 200X, 400X) consisting of total 8 sub-classes (4 Benign and 4 malignant). The study describes the performance metrics when the Dynamic Selection of Classifiers (DSC) methodology was used in the Multiple Classifier Framework System (MCFS) instead of the traditional way of approaching the problem which treats all the sub-classes individually. The obtained metrics suggest that the DSC approach is safer as it firstly classifies the histopathology image as Benign or Malignant which significantly reduces the risk of misclassification. The next classifier then classifies the image into one of subsequent sub-classes in benign and malignant category to provide a secondary opinion to the pathologist such that relevant pathology tests can be performed. Furthermore, the data augmentation strategies applied (such as -Horizontal flip, Vertical Flip, Random Zoom, Random Rotations) to the existing dataset for class balancing seems to show no improvement when the models were trained using the augmented dataset. For evaluating their performances various metrics such as precision, sensitivity, F-score etc. are used.

A contemporary approach for acquiring the computational gains of depth in recurrent neural networks (RNNs) is to hierarchically stack multiple recurrent layers. However, such performance gains come with the cost of challenging optimization of hierarchal RNNs (HRNNs) which are deep both hierarchically and temporally. The researchers have exclusively highlighted the significance of using shortcuts for learning deep hierarchical representations and deep temporal dependencies. However, no significant efforts are made to unify these finding into a single framework for learning deep HRNNs. We propose residual recurrent highway network (R2HN) that contains highways within temporal structure of the network for unimpeded information propagation, thus alleviating gradient vanishing problem. The hierarchical structure learning is posed as residual learning framework to prevent performance degradation problem. The proposed R2HN contain significantly reduced data-dependent parameters as compared to related methods. The experiments on language modeling (LM) tasks have demonstrated that the proposed architecture leads to design effective models. On LM experiments with Penn TreeBank, the model achieved 60.3 perplexity and outperformed baseline and related models that we tested.

The ability to know in advance the trend of running process instances, with respect to different features, such as the expected completion time, would allow business managers to timely counteract to undesired situations, in order to prevent losses. Therefore, the ability to accurately predict future features of running business process instances would be a very helpful aid when managing processes, especially under service level agreement constraints. However, making such accurate forecasts is not easy: many factors may influence the predicted features. Many approaches have been proposed to cope with this problem but, generally, they assume that the underlying process is stationary. However, in real cases this assumption is not always true. In this work we present new methods for predicting the remaining time of running cases. In particular we propose a method, assuming process stationarity, which achieves state-of-the-art performances and two other methods which are able to make predictions even with non-stationary processes. We also describe an approach able to predict the full sequence of activities that a running case is going to take. All these methods are extensively evaluated on different real case studies.

This article reviews the state of the art of prediction and optimization for sequence-driven scheduling in job shop flexible manufacturing systems (JS-FMSs). The objectives of the article are to (1) analyze the literature related to algorithms for sequencing and scheduling, considering domain, method, objective, sequence type, and uncertainty; and to (2) examine current challenges and future directions to promote the feasibility and usability of the relevant research. Current challenges are summarized as follows: less consideration of uncertainty factors causes a gap between the reality and the derived schedules; the use of stationary dispatching rules is limited to reflect the dynamics and flexibility; production-level scheduling is restricted to increase responsiveness owing to product-level uncertainty; and optimization is more focused, while prediction is used mostly for verification and validation, although prediction-then-optimization is the standard stream in data analytics. In future research, the degree of uncertainty should be quantified and modeled explicitly; both holistic and granular algorithms should be considered; product sequences should be incorporated; and sequence learning should be applied to implement the prediction-then-optimization stream. This would enable us to derive data-learned prediction and optimization models that output accurate and precise schedules; foresee individual product locations; and respond rapidly to dynamic and frequent changes in JS-FMSs.

While activity recognition has been shown to be valuable for pervasive computing applications, less work has focused on techniques for forecasting the future occurrence of activities. We present an activity forecasting method to predict the time that will elapse until a target activity occurs. This method generates an activity forecast using a regression tree classifier and offers an advantage over sequence prediction methods in that it can predict expected time until an activity occurs. We evaluate this algorithm on real-world smart home datasets and provide evidence that our proposed approach is most effective at predicting activity timings.

The information on unknown parameters of an oil formation can be express through a multivariate distribution of probabilities in space of parameters. A new quantitative method for carrying out of the discriminant analysis of models of an oil formation on the basis of hydrodynamic slits by sequential prediction of the probabilities of pressure modification is offered in this paper. Investigation of look-ahead probability for several models of the slit formation was conducted. The developed method can be used as a universal tool for carrying out of the discriminant analysis of models irrespective of the number of parameters.

Predicting users daily location visits - when and where they will go, and how long they will stay - is key for making effective location-based recommendations. Knowledge of an upcoming day allows the suggestion of relevant alternatives (e.g., a new coffee shop on the way to work) in advance, prior to a visit. This helps users make informed decisions and plan accordingly. People's visit routines, or just routines, can vary significantly from day to day, and visits from earlier in the day, week, or month may affect subsequent choices. Traditionally, routine prediction has been modeled with sequence methods, such as HMMs or more recently with RNN-based architectures. However, the problem with such architectures is that their predictive performance degrades when increasing the number of historical observations in the routine sequence. In this paper, we propose Masked-TCN (MTCN), a novel method based on time-dilated convolutional networks. The method implements custom dilations and masking which can process effectively long routine sequences, identifying recurring patterns at different resolution - hourly, daily, weekly, monthly. We demonstrate that MTCN achieves 8% improvement in accuracy over current state-of-the-art solutions on a large data set of visit routines.

Spatiotemporal sequence prediction is an important problem in deep learning. We study next-frame video prediction using a deep-learning-based predictive coding framework that uses convolutional, long short-term memory (convLSTM) modules. We introduce a novel reduced-gate convolutional LSTM architecture that achieves better next-frame prediction accuracy than the original convolutional LSTM while using a smaller parameter budget, thereby reducing training time and memory requirements. We tested our reduced gate modules within a predictive coding architecture on the gray-scale and RGB video datasets. We found that our reduced-gate model has a significant reduction of approximately 40 percent of the total number of training parameters and training time in comparison with the standard LSTM model which makes it attractive for hardware implementation especially on small devices.

Data sparsity is one of the most challenging problems in recommendation systems. In this paper, we tackle this problem by proposing a novel multi-modal behavioral information-aware recommendation method named MIAR which is based on recurrent neural networks and matrix factorization. First, an interaction context-aware sequential prediction model is designed to capture user-item interaction contextual information and behavioral sequence information. Second, an attributed context-aware rating prediction model is proposed to capture attribution contextual information and rating information. Finally, three fusion methods are developed to combine two sub-models. As a result, the MIAR method has several distinguished advantages in terms of mitigating the data sparsity problem. The method can well perceive diverse influences of interaction and attribution contextual information. Meanwhile, a large number of behavioral sequence and rating information can be utilized by the MIAR approach. The proposed algorithm is evaluated on real-world datasets and the experimental results show that MIAR can significantly improve recommendation performance compared to the existing state-of-art recommendation algorithms.

Transient muscle movements influence the temporal structure of myoelectric signal patterns, often leading to unstable prediction behavior from movement-pattern classification methods. We show that temporal convolutional network sequential models leverage the myoelectric signal's history to discover contextual temporal features that aid in correctly predicting movement intentions, especially during interclass transitions. We demonstrate myoelectric classification using temporal convolutional networks to effect 3 simultaneous hand and wrist degrees-of-freedom in an experiment involving nine human-subjects. Temporal convolutional networks yield significant $(p, 4 pages, 5 figures, accepted for Neural Engineering (NER) 2019 Conference

The natural language processing (NLP) denotes a technique used to process data such as text and speech. Some of the fundamental research in NLP includes the named entity recognition, which recognizes the named entities (i.e., persons and companies) from texts, the semantic parsing, which converts a natural language utterance to a logical form, and the co-reference resolution, which extracts the nouns (including pronouns and noun phrases) pointing to the same reference body. In this paper, we focus on the mention extraction and classification, proposing a neural-encoded mention-hypergraph model named the BILU-NEMH to extract the mention entities from a content. The proposed BILU-NEMH model combines a mention hypergraph model with the encoding schema and neural network. The proposed model can effectively capture the overlapping mention entities of an unbounded length. The proposed model was verified by the experiments, and the obtained experimental results showed that the proposed model achieved better performance and greater effectiveness than the existing related models on most standard datasets.

A latent variable conditional random fields (CRF) model is proposed to improve sequence labeling, which utilizes the BIO encoding schema as latent variable to capture the latent structure of hidden variables and observation data. The proposed model automatically selects the best encoding schema for each given input sequence. Through experimentation, it is demonstrated that the proposed model unveils the latent variable while performing robustly on sequence-labeling prediction tasks.

Data quality is essential in banking industry for the compliance with the standards of banking regulation, BCBS 239. But the quality is yet to be forecasted by many financial institutions. Machine learning has been recommended by the regulator in 2018 to resolve this. To assist on this, we develop a machine learning model to train several Long Short-Term Memory (“LSTM”) Recurrent Neural Networks (“RNNs”) for the prediction including forward LSTM RNN, backward LSTM RNN and bi-directional LSTM RNN (“BiLSTM”). With the prediction, financial institutions will understand what data quality is going to be. The networks make sequence predictions with optimizations followed by an evaluation with heterogeneous methodologies, validation techniques and algorithms.

Supertagging is a sequence prediction task where each word is assigned a piece of complex syntactic structure called a supertag. We provide a novel approach to multi-task learning for Tree Adjoining Grammar (TAG) supertagging by deconstructing these complex supertags in order to define a set of related but auxiliary sequence prediction tasks. Our multi-task prediction framework is trained over the exactly same training data used to train the original supertagger where each auxiliary task provides an alternative view on the original prediction task. Our experimental results show that our multi-task approach significantly improves TAG supertagging with a new state-of-the-art accuracy score of 91.39% on the Penn Treebank supertagging dataset.

In this paper, we investigate the performance of several sequence prediction techniques on the prediction of future events of human behavior in a smart home, as well as the timestamps of those next events. Prediction techniques in smart home environments have several use cases, such as the real-time identification of abnormal behavior, identifying coachable moments for e-coaching, and a plethora of applications in the area of home automation. We give an overview of several sequence prediction techniques, including techniques that originate from the areas of data mining, process mining, and data compression, and we evaluate the predictive accuracy of those techniques on a collection of publicly available real-life datasets from the smart home environments domain. This contrast our work with existing work on prediction in smart homes, which often evaluate their techniques on a single smart home instead of a larger collection of logs. We found that LSTM neural networks outperform the other prediction methods on the task of predicting the next activity as well as on the task of predicting the timestamp of the next event. However, surprisingly, we found that it is very dependent on the dataset which technique works best for the task of predicting a window of multiple next activities.

Deep recurrent neural networks (DRNNs) have recently demonstrated strong performance in sequential data analysis, such as natural language processing. These capabilities make them a promising tool for inferential analysis of sequentially structured bioinformatics data as well. Here, we assessed the ability of Long Short-Term Memory (LSTM) networks, a class of DRNNs, to predict properties of proteins based on their primary structures. The proposed architecture is trained and tested on two different datasets to predict whether a given sequence falls into a certain class or not. The first dataset, directly imported from Uniprot, was used to train the network on whether a given protein contained or did not contain a conserved sequence (homeodomain), and the second dataset, derived by literature mining, was used to train a network on whether a given protein binds or doesn't bind to Artemisinin, a drug typically used to treat malaria. In each case, the model was able to differentiate between the two different classes of sequences it was given with high accuracy, illustrating successful learning and generalization. Upon completion of training, an ROC curve was created using the homeodomain and artemisinin validation datasets. The AUC of these datasets was 0.80 and 0.87 respectively, further indicating the models' effectiveness. Furthermore, using these trained models, it was possible to derive a protocol for sequence detection of homeodomain and binding motif, which are well-documented in literature, and a known Artemisinin binding site, respectively [1-3]. Along with these contributions, we developed a python API to directly connect to Uniprot data sourcing, train deep neural networks on this primary sequence data using TensorFlow, and uniquely visualize the results of this analysis. Such an approach has the potential to drastically increase accuracy and reduce computational time and, current major limitations in informatics, from inquiry to discovery in protein function research

With the rising demand for smart devices and smart home systems, automation and activity prediction has become a vital aspect of people's everyday lives. Researchers have focused on developing approaches that detect user activity patterns and used them to predict future actions. One such system is Modified Sequence Prediction via Enhanced Episode Discovery (M-SPEED), which uses spatiotemporal daily life activities to analyze user behaviors. However, the low accuracy of this algorithm can be a limiting factor inefficient activity prediction. Also, the computational overhead of run time and memory causes this algorithm to show poor performance in large datasets. This research focuses on modifying the M-SPEED algorithm to improve its capability to run on a larger dataset while at the same time improving run time. The accuracy is also improved to make it more effective in real-world applications. Proof of algorithm efficiency is provided to ensure system validity, and simulation is carried out on real-life data. The results demonstrate a 66.69% improvement in cumulative memory efficiency, 37% faster run time, and 8.22% better accuracy confirming the proposal's effectiveness

Long-term load forecast (LTLF) for area distribution feeders is one of the most critical tasks frequently performed in electric distribution utility companies. For a specific planning area, cost-effective system upgrades can only be planned out based on accurate feeder LTLF results. In our previous research, we established a unique sequence prediction method which has the tremendous advantage of combining area top-down, feeder bottom-up and multi-year historical data all together for forecast and achieved a superior performance over various traditional methods by real-world tests. However, the previous method only focused on the forecast of the next one-year. In our current work, we significantly improved this method: the forecast can now be extended to a multi-year forecast window in the future; unsupervised learning techniques are used to group feeders by their load composition features to improve accuracy; we also propose a novel selective sequence learning mechanism which uses Gated Recurrent Unit network to not only learn how to predict sequence values but also learn to select the best-performing sequential configuration for each individual feeder. The proposed method was tested on an actual urban distribution system in West Canada. It was compared with traditional methods and our previous sequence prediction method. It demonstrates the best forecasting performance as well as the possibility of using sequence prediction models for multi-year component-level load forecast., Comment: 22 pages, 9 figures

In order to obtain stable returns, this paper aims to establish a quantitative model with higher prediction accuracy. According to the time series characteristics of financial data, the prediction model with financial time series data is constructed by using the time-memory sequence model LSTM, which is applied to the representative SSE 50 series stocks. And based on this, the LSTM model with Encoder-Decoder mode and a hybridized framework of LSTM with CNN are built to improve the original model. Feature extraction is performed on the input data by using CNN, and then as an input to the LSTM, the extracted features are used for sequence prediction with the LSTM model.

Skeleton-based action recognition distinguishes human actions using the trajectories of skeleton joints, which can be a good representation of human behaviors. Conventional methods usually construct classifiers with hand-crafted or the learned features to recognize human actions. Different from constructing a direct action classifier for action recognition task, this paper attempts to identify human actions based on the development trends of behavior sequences. Specifically, we first utilize the memory neural network to construct action predictors for each kind of activity. These action predictors can then output the action trends at the next time step. According to the predictions of these action predictors at each time step and the removal rule, the poor predictors can be eliminated step by step, and the IDentity(ID) number of the last predictor left is considered as the label of the action sequence to be categorized. We compare the proposed action recognition algorithm using sequence prediction learning with other methods on two publicly available datasets. Our experimental results consistently demonstrate the feasibility and effectiveness of the suggested method. It also proves the importance of prediction learning for action recognition.

In order to survive in a competitive environment, getting placed in a reputed company after graduation is not the end of education. Within a company also a person needs to keep re-skilling and up-skilling oneself to stay relevant and successful. A Recommender system would help new employees to identify the best choice (course) of their interest. In this paper, we have proposed a recommender system that would assist new employees by predicting for them the choices of certifications and skills that they should acquire to advance in their careers, given their personal histories (education and already acquired skills and certifications). We have employed a relatively new technique, Compact Prediction Trees, for this task. CPTs have shown convincing results in sequence prediction, performing better and faster than many traditional algorithms.