Attention-Based Deep Recurrent Model for Survival Prediction

2021 ◽  
Vol 2 (4) ◽  
pp. 1-18
Author(s):  
Zhaohong Sun ◽  
Wei Dong ◽  
Jinlong Shi ◽  
Kunlun He ◽  
Zhengxing Huang
Keyword(s):  
Risk Factors ◽  
Survival Analysis ◽  
Deep Learning ◽  
Service Management ◽  
Short Term Memory ◽  
Survival Prediction ◽  
Learning Techniques ◽  
Proposed Model ◽  
Traditional Approaches

Survival analysis exhibits profound effects on health service management. Traditional approaches for survival analysis have a pre-assumption on the time-to-event probability distribution and seldom consider sequential visits of patients on medical facilities. Although recent studies leverage the merits of deep learning techniques to capture non-linear features and long-term dependencies within multiple visits for survival analysis, the lack of interpretability prevents deep learning models from being applied to clinical practice. To address this challenge, this article proposes a novel attention-based deep recurrent model, named AttenSurv , for clinical survival analysis. Specifically, a global attention mechanism is proposed to extract essential/critical risk factors for interpretability improvement. Thereafter, Bi-directional Long Short-Term Memory is employed to capture the long-term dependency on data from a series of visits of patients. To further improve both the prediction performance and the interpretability of the proposed model, we propose another model, named GNNAttenSurv , by incorporating a graph neural network into AttenSurv, to extract the latent correlations between risk factors. We validated our solution on three public follow-up datasets and two electronic health record datasets. The results demonstrated that our proposed models yielded consistent improvement compared to the state-of-the-art baselines on survival analysis.

scholarly journals Stock Pattern Classification from Charts using Deep Learning Algorithms

2020 ◽  
Vol 3 (1) ◽  
pp. 445-454
Author(s):  
Celal Buğra Kaya ◽  
Alperen Yılmaz ◽  
Gizem Nur Uzun ◽  
Zeynep Hilal Kilimci
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Convolutional Neural Networks ◽  
Pattern Classification ◽  
Short Term Memory ◽  
Stock Exchange ◽  
Learning Techniques ◽  
Proposed Model ◽  
Istanbul Stock Exchange

Pattern classification is related with the automatic finding of regularities in dataset through the utilization of various learning techniques. Thus, the classification of the objects into a set of categories or classes is provided. This study is undertaken to evaluate deep learning methodologies to the classification of stock patterns. In order to classify patterns that are obtained from stock charts, convolutional neural networks (CNNs), recurrent neural networks (RNNs), and long-short term memory networks (LSTMs) are employed. To demonstrate the efficiency of proposed model in categorizing patterns, hand-crafted image dataset is constructed from stock charts in Istanbul Stock Exchange and NASDAQ Stock Exchange. Experimental results show that the usage of convolutional neural networks exhibits superior classification success in recognizing patterns compared to the other deep learning methodologies.

Data Transcription for India's Supreme Court Documents Using Deep Learning Algorithms

2020 ◽  
Vol 16 (4) ◽  
pp. 21-41
Author(s):  
Vaissnave V. ◽  
P. Deepalakshmi
Keyword(s):  
Deep Learning ◽  
Short Term Memory ◽  
Performance Comparison ◽  
Legal Cases ◽  
Learning Techniques ◽  
Proposed Model ◽  
Comparison Results ◽  
Complex Forms ◽  
Indian Law ◽  
Day By Day

The Indian legal system is one of the largest judiciary systems in the world and handles a huge number of legal cases which is increasing rapidly day by day. The computerized documentation of Indian law is highly voluminous and complex forms. This article proposes a model using deep learning techniques to split the judgment text into the issue, facts, arguments, reasoning, and decision. To evaluate the proposed model, the authors conducted experiments that revealed that the convolutional neural network and long short-term memory transcription technique could achieve better accuracy and obtain superior transcription performance. Comparison results indicate that the proposed algorithm gives the highest classification accuracy rate of 95.6%. The adaptation of splitting the judgment text into the issue, facts, arguments, reasoning, and decision helps to find specific portions of the judgment within a second, making the job of analyzing the case more effective, efficient, and faster.

scholarly journals Through-Wall Radar Classification of Human Posture Using Convolutional Neural Networks

2019 ◽  
Vol 2019 ◽  
pp. 1-10
Author(s):  
Alper Kılıç ◽  
İsmail Babaoğlu ◽  
Ahmet Babalık ◽  
Ahmet Arslan
Keyword(s):  
Deep Learning ◽  
Continuous Wave ◽  
Human Posture ◽  
Learning Techniques ◽  
Military Applications ◽  
Radar Signals ◽  
Traditional Approaches ◽  
Term Data

Through-wall detection and classification are highly desirable for surveillance, security, and military applications in areas that cannot be sensed using conventional measures. In the domain of these applications, a key challenge is an ability not only to sense the presence of individuals behind the wall but also to classify their actions and postures. Researchers have applied ultrawideband (UWB) radars to penetrate wall materials and make intelligent decisions about the contents of rooms and buildings. As a form of UWB radar, stepped frequency continuous wave (SFCW) radars have been preferred due to their advantages. On the other hand, the success of classification with deep learning methods in different problems is remarkable. Since the radar signals contain valuable information about the objects behind the wall, the use of deep learning techniques for classification purposes will give a different direction to the research. This paper focuses on the classification of the human posture behind the wall using through-wall radar signals and a convolutional neural network (CNN). The SFCW radar is used to collect radar signals reflected from the human target behind the wall. These signals are employed to classify the presence of the human and the human posture whether he/she is standing or sitting by using CNN. The proposed approach achieves remarkable and successful results without the need for detailed preprocessing operations and long-term data used in the traditional approaches.

scholarly journals A New Hybrid CNN-LSTM Model with Non-SoftMax Functions for Face Spoof Detection

2021 ◽  
Author(s):  
S Lokesh Kumar ◽  
Yamani Sai Asish ◽  
Sannasi Ganapathy
Keyword(s):  
Deep Learning ◽  
Short Term Memory ◽  
Hybrid Approach ◽  
Mobile Payments ◽  
Hybrid Techniques ◽  
Video Frames ◽  
Learning Techniques ◽  
Proposed Model ◽  
Full Picture ◽  
The Impact

Abstract Recently, the emerging applications such as banking, mobile payments, face recognition technology are gradually booming and also increases the users count around the world. The extensive deployment of facial recognition systems has drawn close attention to the dependability of facial biometrics in the fight against spoof attacks, in which a picture, video or 3D mask of a real user's face may be used to access facilities or services illegitimately. While a number of anti-spoofing or liveness detection approaches (which identify whether a face is live or spoof when captured) were suggested, the problem is still unresolved because of the difficulty in discovering discriminatory and computer-cost characteristics and techniques for spoof assaults. Existing methods also utilise a full picture or video to determine luminosity. Often though, some facial areas (video frames) are redundant or relate to the confusion of the picture (video). In this paper, we propose a new hybrid deep learning technique called Hybrid Convolutional Neural Network (CNN) based architecture with Long Short-Term Memory (LSTM) units to study the impact in classification. In this technique is applied a non-softmax function for making effective decision on classification. The hybrid approach is implemented followed by a comparative analysis with existing conventional and hybrid techniques used for spoof detection. The proposed model is proved as better than the existing deep learning techniques and other hybrid models in terms of precision, recall, f-measure and accuracy.

Construction of an Ensemble Scheme for Stock Price Prediction Using Deep Learning Techniques

2021 ◽  
Vol 17 (2) ◽  
pp. 72-95
Author(s):  
Justice Kwame Appati ◽  
Ismail Wafaa Denwar ◽  
Ebenezer Owusu ◽  
Michael Agbo Tettey Soli
Keyword(s):  
Deep Learning ◽  
Stock Price ◽  
Short Term Memory ◽  
Absolute Error ◽  
Stock Price Prediction ◽  
Price Prediction ◽  
Learning Techniques ◽  
Proposed Model ◽  
Bidirectional Lstm ◽  
Gated Recurrent Unit

This study proposes a deep learning approach for stock price prediction by bridging the long short-term memory with gated recurrent unit. In its evaluation, the mean absolute error and mean square error were used. The model proposed is an extension of the study of Hossain et al. established in 2018 with an MSE of 0.00098 as its lowest error. The current proposed model is a mix of the bidirectional LSTM and bidirectional GRU resulting in 0.00000008 MSE as the lowest error recorded. The LSTM model recorded 0.00000025 MSE, the GRU model recorded 0.00000077 MSE, and the LSTM + GRU model recorded 0.00000023 MSE. Other combinations of the existing models such as the bi-directional LSTM model recorded 0.00000019 MSE, bi-directional GRU recorded 0.00000011 MSE, bidirectional LSTM + GRU recorded 0.00000027 MSE, LSTM and bi-directional GRU recorded 0.00000020 MSE.

Human Activity Recognition using Fourier Transform Inspired Deep Learning Combination Model

2019 ◽  
Vol 9 (1) ◽  
pp. 16-31
Author(s):  
Kyungkoo Jun
Keyword(s):  
Fourier Transform ◽  
Deep Learning ◽  
Short Term Memory ◽  
Window Size ◽  
Sensor Data ◽  
Data Sets ◽  
Data Set ◽  
Proposed Model ◽  
Testing Data ◽  
Labeling Scheme

Background & Objective: This paper proposes a Fourier transform inspired method to classify human activities from time series sensor data. Methods: Our method begins by decomposing 1D input signal into 2D patterns, which is motivated by the Fourier conversion. The decomposition is helped by Long Short-Term Memory (LSTM) which captures the temporal dependency from the signal and then produces encoded sequences. The sequences, once arranged into the 2D array, can represent the fingerprints of the signals. The benefit of such transformation is that we can exploit the recent advances of the deep learning models for the image classification such as Convolutional Neural Network (CNN). Results: The proposed model, as a result, is the combination of LSTM and CNN. We evaluate the model over two data sets. For the first data set, which is more standardized than the other, our model outperforms previous works or at least equal. In the case of the second data set, we devise the schemes to generate training and testing data by changing the parameters of the window size, the sliding size, and the labeling scheme. Conclusion: The evaluation results show that the accuracy is over 95% for some cases. We also analyze the effect of the parameters on the performance.

Detection of Pavement Maintenance Treatments using Deep-Learning Network

2021 ◽  
pp. 036119812110078
Author(s):  
Lu Gao ◽  
Yao Yu ◽  
Yi Hao Ren ◽  
Pan Lu
Keyword(s):  
Deep Learning ◽  
Measurement Errors ◽  
Short Term Memory ◽  
Learning Networks ◽  
Pavement Maintenance ◽  
Time Period ◽  
Learning Techniques ◽  
Maintenance And Rehabilitation ◽  
History Of ◽  
Deep Learning Network

Pavement maintenance and rehabilitation (M&R) records are important as they provide documentation that M&R treatment is being performed and completed appropriately. Moreover, the development of pavement performance models relies heavily on the quality of the condition data collected and on the M&R records. However, the history of pavement M&R activities is often missing or unavailable to highway agencies for many reasons. Without accurate M&R records, it is difficult to determine if a condition change between two consecutive inspections is the result of M&R intervention, deterioration, or measurement errors. In this paper, we employed deep-learning networks of a convolutional neural network (CNN) model, a long short-term memory (LSTM) model, and a CNN-LSTM combination model to automatically detect if an M&R treatment was applied to a pavement section during a given time period. Unlike conventional analysis methods so far followed, deep-learning techniques do not require any feature extraction. The maximum accuracy obtained for test data is 87.5% using CNN-LSTM.

scholarly journals A Novel LSTM Model with Interaction Dual Attention for Radar Echo Extrapolation

2021 ◽  
Vol 13 (2) ◽  
pp. 164
Author(s):  
Chuyao Luo ◽  
Xutao Li ◽  
Yongliang Wen ◽  
Yunming Ye ◽  
Xiaofeng Zhang
Keyword(s):  
Short Term Memory ◽  
Weather Forecast ◽  
Vital Role ◽  
Data Sets ◽  
Short Term ◽  
Learning Techniques ◽  
Radar Echo ◽  
Hidden States ◽  
Better Than

The task of precipitation nowcasting is significant in the operational weather forecast. The radar echo map extrapolation plays a vital role in this task. Recently, deep learning techniques such as Convolutional Recurrent Neural Network (ConvRNN) models have been designed to solve the task. These models, albeit performing much better than conventional optical flow based approaches, suffer from a common problem of underestimating the high echo value parts. The drawback is fatal to precipitation nowcasting, as the parts often lead to heavy rains that may cause natural disasters. In this paper, we propose a novel interaction dual attention long short-term memory (IDA-LSTM) model to address the drawback. In the method, an interaction framework is developed for the ConvRNN unit to fully exploit the short-term context information by constructing a serial of coupled convolutions on the input and hidden states. Moreover, a dual attention mechanism on channels and positions is developed to recall the forgotten information in the long term. Comprehensive experiments have been conducted on CIKM AnalytiCup 2017 data sets, and the results show the effectiveness of the IDA-LSTM in addressing the underestimation drawback. The extrapolation performance of IDA-LSTM is superior to that of the state-of-the-art methods.

scholarly journals A Hybrid Optimized LSTM Models for Human Activity Recognition with IOT Devices

2021 ◽  
pp. 182-189
Author(s):  
S. Arokiaraj ◽  
Dr. N. Viswanathan
Keyword(s):  
Deep Learning ◽  
Internet Of Things ◽  
Short Term Memory ◽  
The Other ◽  
Learning Models ◽  
Computational Overhead ◽  
Temporal Features ◽  
Human Movements ◽  
Proposed Model ◽  
Iot Devices

With the advent of Internet of things(IoT),HA (HA) recognition has contributed the more application in health care in terms of diagnosis and Clinical process. These devices must be aware of human movements to provide better aid in the clinical applications as well as user’s daily activity.Also , In addition to machine and deep learning algorithms, HA recognition systems has significantly improved in terms of high accurate recognition. However, the most of the existing models designed needs improvisation in terms of accuracy and computational overhead. In this research paper, we proposed a BAT optimized Long Short term Memory (BAT-LSTM) for an effective recognition of human activities using real time IoT systems. The data are collected by implanting the Internet of things) devices invasively. Then, proposed BAT-LSTM is deployed to extract the temporal features which are then used for classification to HA. Nearly 10,0000 dataset were collected and used for evaluating the proposed model. For the validation of proposed framework, accuracy, precision, recall, specificity and F1-score parameters are chosen and comparison is done with the other state-of-art deep learning models. The finding shows the proposed model outperforms the other learning models and finds its suitability for the HA recognition.

scholarly journals BO-LSTM: Classifying relations via long short-term memory networks along biomedical ontologies

2018 ◽  
Author(s):  
Andre Lamurias ◽  
Luka A. Clarke ◽  
Francisco M. Couto
Keyword(s):  
Deep Learning ◽  
Text Mining ◽  
Drug Interactions ◽  
Short Term Memory ◽  
Biomedical Ontologies ◽  
Short Term ◽  
Term Memory ◽  
Domain Specific ◽  
Learning Techniques ◽  
Long Short Term Memory

AbstractRecent studies have proposed deep learning techniques, namely recurrent neural networks, to improve biomedical text mining tasks. However, these techniques rarely take advantage of existing domain-specific resources, such as ontologies. In Life and Health Sciences there is a vast and valuable set of such resources publicly available, which are continuously being updated. Biomedical ontologies are nowadays a mainstream approach to formalize existing knowledge about entities, such as genes, chemicals, phenotypes, and disorders. These resources contain supplementary information that may not be yet encoded in training data, particularly in domains with limited labeled data.We propose a new model, BO-LSTM, that takes advantage of domain-specific ontologies, by representing each entity as the sequence of its ancestors in the ontology. We implemented BO-LSTM as a recurrent neural network with long short-term memory units and using an open biomedical ontology, which in our case-study was Chemical Entities of Biological Interest (ChEBI). We assessed the performance of BO-LSTM on detecting and classifying drug-drug interactions in a publicly available corpus from an international challenge, composed of 792 drug descriptions and 233 scientific abstracts. By using the domain-specific ontology in addition to word embeddings and WordNet, BO-LSTM improved both the F1-score of the detection and classification of drug-drug interactions, particularly in a document set with a limited number of annotations. Our findings demonstrate that besides the high performance of current deep learning techniques, domain-specific ontologies can still be useful to mitigate the lack of labeled data.Author summaryA high quantity of biomedical information is only available in documents such as scientific articles and patents. Due to the rate at which new documents are produced, we need automatic methods to extract useful information from them. Text mining is a subfield of information retrieval which aims at extracting relevant information from text. Scientific literature is a challenge to text mining because of the complexity and specificity of the topics approached. In recent years, deep learning has obtained promising results in various text mining tasks by exploring large datasets. On the other hand, ontologies provide a detailed and sound representation of a domain and have been developed to diverse biomedical domains. We propose a model that combines deep learning algorithms with biomedical ontologies to identify relations between concepts in text. We demonstrate the potential of this model to extract drug-drug interactions from abstracts and drug descriptions. This model can be applied to other biomedical domains using an annotated corpus of documents and an ontology related to that domain to train a new classifier.

Sign in / Sign up

Export Citation Format

Share Document