Search results for: LSTM

Authors: Selim M. Khan, Aaron A. Goodarzi, Joshua M. Taron, Tryggve Rönnqvist

Abstract:

Indoor air quality, a prime determinant of health, is strongly influenced by the presence of hazardous radon gas within the built environment. As a health issue, dangerously high indoor radon arose within the 20th century to become the 2nd leading cause of lung cancer. While the 21st century building metrics and human behaviors have captured, contained, and concentrated radon to yet higher and more hazardous levels, the issue is rapidly worsening in Canada. It is established that Canadians in the Prairies are the 2nd highest radon-exposed population in the world, with 1 in 6 residences experiencing 0.2-6.5 millisieverts (mSv) radiation per week, whereas the Canadian Nuclear Safety Commission sets maximum 5-year occupational limits for atomic workplace exposure at only 20 mSv. This situation is also deteriorating over time within newer housing stocks containing higher levels of radon. Deep machine learning (LSTM) algorithms were applied to analyze multiple quantitative and qualitative features, determine the most important contributory factors, and predicted radon levels in the known past (1990-2020) and projected future (2021-2050). The findings showed gradual downwards patterns in Sweden, whereas it would continue to go from high to higher levels in Canada over time. The contributory factors found to be the basement porosity, roof insulation depthness, R-factor, and air dynamics of the indoor environment related to human window opening behaviour. Building codes must consider including these factors to ensure adequate indoor ventilation and healthy living that can prevent lung cancer in non-smokers.

Keywords: radon, building metrics, deep learning, LSTM prediction model, lung cancer, canada, sweden

Procedia PDF Downloads 79

Authors: Mst Shapna Akter, Hossain Shahriar

Abstract:

One of the most important challenges in the field of software code audit is the presence of vulnerabilities in software source code. Every year, more and more software flaws are found, either internally in proprietary code or revealed publicly. These flaws are highly likely exploited and lead to system compromise, data leakage, or denial of service. C and C++ open-source code are now available in order to create a largescale, machine-learning system for function-level vulnerability identification. We assembled a sizable dataset of millions of opensource functions that point to potential exploits. We developed an efficient and scalable vulnerability detection method based on deep neural network models that learn features extracted from the source codes. The source code is first converted into a minimal intermediate representation to remove the pointless components and shorten the dependency. Moreover, we keep the semantic and syntactic information using state-of-the-art word embedding algorithms such as glove and fastText. The embedded vectors are subsequently fed into deep learning networks such as LSTM, BilSTM, LSTM-Autoencoder, word2vec, BERT, and GPT-2 to classify the possible vulnerabilities. Furthermore, we proposed a neural network model which can overcome issues associated with traditional neural networks. Evaluation metrics such as f1 score, precision, recall, accuracy, and total execution time have been used to measure the performance. We made a comparative analysis between results derived from features containing a minimal text representation and semantic and syntactic information. We found that all of the deep learning models provide comparatively higher accuracy when we use semantic and syntactic information as the features but require higher execution time as the word embedding the algorithm puts on a bit of complexity to the overall system.

Keywords: cyber security, vulnerability detection, neural networks, feature extraction

Procedia PDF Downloads 40

Authors: Rohan Bhasin

Abstract:

Agrammatism in non-fluent Aphasia Cases can be defined as a language disorder wherein a patient can only use content words ( nouns, verbs and adjectives ) for communication and their speech is devoid of functional word types like conjunctions and articles, generating speech of with extremely rudimentary grammar . Past approaches involve Speech Therapy of some order with conversation analysis used to analyse pre-therapy speech patterns and qualitative changes in conversational behaviour after therapy. We describe this approach as a novel method to generate functional words (prepositions, articles, ) around content words ( nouns, verbs and adjectives ) using a combination of Natural Language Processing and Deep Learning algorithms. The applications of this approach can be used to assist communication. The approach the paper investigates is : LSTMs or Seq2Seq: A sequence2sequence approach (seq2seq) or LSTM would take in a sequence of inputs and output sequence. This approach needs a significant amount of training data, with each training data containing pairs such as (content words, complete sentence). We generate such data by starting with complete sentences from a text source, removing functional words to get just the content words. However, this approach would require a lot of training data to get a coherent input. The assumptions of this approach is that the content words received in the inputs of both text models are to be preserved, i.e, won't alter after the functional grammar is slotted in. This is a potential limit to cases of severe Agrammatism where such order might not be inherently correct. The applications of this approach can be used to assist communication mild Agrammatism in non-fluent Aphasia Cases. Thus by generating these function words around the content words, we can provide meaningful sentence options to the patient for articulate conversations. Thus our project translates the use case of generating sentences from content-specific words into an assistive technology for non-Fluent Aphasia Patients.

Keywords: aphasia, expressive aphasia, assistive algorithms, neurology, machine learning, natural language processing, language disorder, behaviour disorder, sequence to sequence, LSTM

Procedia PDF Downloads 134

Authors: Despoina Mouratidis, Katia Lida Kermanidis

Abstract:

Machine translation (MT) is a very important task in Natural Language Processing (NLP). MT evaluation is crucial in MT development, as it constitutes the means to assess the success of an MT system, and also helps improve its performance. Several methods have been proposed for the evaluation of (MT) systems. Some of the most popular ones in automatic MT evaluation are score-based, such as the BLEU score, and others are based on lexical similarity or syntactic similarity between the MT outputs and the reference involving higher-level information like part of speech tagging (POS). This paper presents a language-independent machine learning framework for classifying pairwise translations. This framework uses vector representations of two machine-produced translations, one from a statistical machine translation model (SMT) and one from a neural machine translation model (NMT). The vector representations consist of automatically extracted word embeddings and string-like language-independent features. These vector representations used as an input to a multi-layer neural network (NN) that models the similarity between each MT output and the reference, as well as between the two MT outputs. To evaluate the proposed approach, a professional translation and a "ground-truth" annotation are used. The parallel corpora used are English-Greek (EN-GR) and English-Italian (EN-IT), in the educational domain and of informal genres (video lecture subtitles, course forum text, etc.) that are difficult to be reliably translated. They have tested three basic deep learning (DL) architectures to this schema: (i) fully-connected dense, (ii) Convolutional Neural Network (CNN), and (iii) Long Short-Term Memory (LSTM). Experiments show that all tested architectures achieved better results when compared against those of some of the well-known basic approaches, such as Random Forest (RF) and Support Vector Machine (SVM). Better accuracy results are obtained when LSTM layers are used in our schema. In terms of a balance between the results, better accuracy results are obtained when dense layers are used. The reason for this is that the model correctly classifies more sentences of the minority class (SMT). For a more integrated analysis of the accuracy results, a qualitative linguistic analysis is carried out. In this context, problems have been identified about some figures of speech, as the metaphors, or about certain linguistic phenomena, such as per etymology: paronyms. It is quite interesting to find out why all the classifiers led to worse accuracy results in Italian as compared to Greek, taking into account that the linguistic features employed are language independent.

Keywords: machine learning, machine translation evaluation, neural network architecture, pairwise classification

Procedia PDF Downloads 98

Authors: Sumit Sen, Sohan Khedekar, Umang Shinde, Shivam Bhargava

Abstract:

This study has been undertaken to investigate whether the deep learning models are able to predict the future stock prices by training the model with the historical stock price data. Since this work required time series analysis, various models are present today to perform time series analysis such as Recurrent Neural Network LSTM, ARIMA and Facebook Prophet. Applying these models the movement of stock price of stocks are predicted and also tried to provide the future prediction of the stock price of a stock. Final product will be a stock price prediction web application that is developed for providing the user the ease of analysis of the stocks and will also provide the predicted stock price for the next seven days.

Keywords: Autoregressive Integrated Moving Average, Deep Learning, Long Short Term Memory, Time-series

Procedia PDF Downloads 105

Authors: Supria Sarkar, Vasit Sagan, Sourav Bhadra, Meghnath Pokharel, Felix B.Fritschi

Abstract:

Soybean and their derivatives are very important agricultural commodities around the world because of their wide applicability in human food, animal feed, biofuel, and industries. However, the significance of soybean production depends on the quality of the soybean seeds rather than the yield alone. Seed composition is widely dependent on plant physiological properties, aerobic and anaerobic environmental conditions, nutrient content, and plant phenological characteristics, which can be captured by high temporal resolution remote sensing datasets. Planet scope (PS) satellite images have high potential in sequential information of crop growth due to their frequent revisit throughout the world. In this study, we estimate soybean seed composition while the plants are in the field by utilizing PlanetScope (PS) satellite images and different machine learning algorithms. Several experimental fields were established with varying genotypes and different seed compositions were measured from the samples as ground truth data. The PS images were processed to extract 462 hand-crafted vegetative and textural features. Four machine learning algorithms, i.e., partial least squares (PLSR), random forest (RFR), gradient boosting machine (GBM), support vector machine (SVM), and two recurrent neural network architectures, i.e., long short-term memory (LSTM) and gated recurrent unit (GRU) were used in this study to predict oil, protein, sucrose, ash, starch, and fiber of soybean seed samples. The GRU and LSTM architectures had two separate branches, one for vegetative features and the other for textures features, which were later concatenated together to predict seed composition. The results show that sucrose, ash, protein, and oil yielded comparable prediction results. Machine learning algorithms that best predicted the six seed composition traits differed. GRU worked well for oil (R-Squared: of 0.53) and protein (R-Squared: 0.36), whereas SVR and PLSR showed the best result for sucrose (R-Squared: 0.74) and ash (R-Squared: 0.60), respectively. Although, the RFR and GBM provided comparable performance, the models tended to extremely overfit. Among the features, vegetative features were found as the most important variables compared to texture features. It is suggested to utilize many vegetation indices for machine learning training and select the best ones by using feature selection methods. Overall, the study reveals the feasibility and efficiency of PS images and machine learning for plot-level seed composition estimation. However, special care should be given while designing the plot size in the experiments to avoid mixed pixel issues.

Keywords: agriculture, computer vision, data science, geospatial technology

Procedia PDF Downloads 95

Authors: Ying Ying Liu, Austin Cooke, Parimala Thulasiraman

Abstract:

Internet of Things (IoT) has attracted a recent spark in research on Internet of Vehicles (IoV). In this paper, we focus on one research area in IoV: preserving location privacy of vehicle data. We discuss existing location privacy preserving techniques and provide a scheme for evaluating these techniques under IoV traffic condition. We propose a different strategy in applying Differential Privacy using k-d tree data structure to preserve location privacy and experiment on real world Gowalla data set. We show that our strategy produces differentially private data, good preservation of utility by achieving similar regression accuracy to the original dataset on an LSTM (Long Term Short Term Memory) neural network traffic predictor.

Keywords: differential privacy, internet of things, internet of vehicles, location privacy, privacy preservation scheme

Procedia PDF Downloads 126

Authors: Davide Fuscà, Kanan Rahimli, Roberto Leuzzi

Abstract:

Preventing abuse and illegal activities in a given area of the sea is a very difficult and expensive task. Artificial intelligence offers the possibility to implement new methods to identify the vessel class type from the kinematic features of the vessel itself. The task strictly depends on the quality of the data. This paper explores the application of a deep, long short-term memory model by using AIS flow only with a relatively low quality. The proposed model reaches high accuracy on detecting nine vessel classes representing the most common vessel types in the Ionian-Adriatic Sea. The model has been applied during the Adriatic-Ionian trial period of the international EU ANDROMEDA H2020 project to identify vessels performing behaviors far from the expected one depending on the declared type.

Keywords: maritime surveillance, artificial intelligence, behavior analysis, LSTM

Procedia PDF Downloads 195

Authors: Vinayaka Gude Divya Sampath

Abstract:

The research introduces an integrated prediction model to assess the susceptibility of roads in a future flooding event. The model consists of deep learning algorithm for forecasting gauge height data and Flood Inundation Mapper (FIM) for spatial flooding. An optimal architecture for Long short-term memory network (LSTM) was identified for the gauge located on Tangipahoa River at Robert, LA. Dropout was applied to the model to evaluate the uncertainty associated with the predictions. The estimates are then used along with FIM to identify the spatial flooding. Further geoprocessing in ArcGIS provides the susceptibility values for different roads. The model was validated based on the devastating flood of August 2016. The paper discusses the challenges for generalization the methodology for other locations and also for various types of flooding. The developed model can be used by the transportation department and other emergency response organizations for effective disaster management.

Keywords: deep learning, disaster management, flood prediction, urban flooding

Procedia PDF Downloads 107

Authors: Amir Mabhout, Toktam Ghafarian, Amirhossein Farzin, Zahra Makki, Sajjad Alizadeh, Amirhossein Ghavi

Abstract:

The high number of Netflix subscribers makes it attractive for data scientists to extract valuable knowledge from the viewers' behavioural analyses. This paper presents a set of statistical insights into viewers' viewing history. After that, a deep learning model is used to predict the future watching behaviour of the users based on previous watching history within the Netflixlatte data pool. Netflixlatte in an aggregated and anonymized data pool of 320 Netflix viewers with a length 250 000 data points recorded between 2008-2022. We observe insightful correlations between the distribution of viewing time and the COVID-19 pandemic outbreak. The presented deep learning model predicts future movie and TV series viewing habits with an average loss of 0.175.

Keywords: data analysis, deep learning, LSTM neural network, netflix

Procedia PDF Downloads 181

Authors: Rodrigo S. Fonseca, Antônio C. P. Veiga

Abstract:

Aiming at helping people with some movement limitation that makes typing and communication difficult, there is a need to customize an assistive tool with a learning environment that helps the user in order to optimize text input, identifying the error and providing the correction and possibilities of choice in the Portuguese language. The work presents an Orthographic and Grammatical System that can be incorporated into writing environments, improving and facilitating the use of an alphanumeric keyboard, using a prototype built using a genetic algorithm in addition to carrying out the prediction, which can occur based on the quantity and position of the inserted letters and even placement in the sentence, ensuring the sequence of ideas using a Long Short Term Memory (LSTM) neural network. The prototype optimizes data entry, being a component of assistive technology for the textual formulation, detecting errors, seeking solutions and informing the user of accurate predictions quickly and effectively through machine learning.

Keywords: genetic algorithm, neural networks, word prediction, machine learning

Procedia PDF Downloads 156

Authors: Wanchalerm Patanacharoenwong, Panaya Sudta, Prachya Bumrungkun

Abstract:

The paper focuses on the development of a system that combines Internet of Things (IoT) technologies and deep learning algorithms for anomaly detection in residential Electric Vehicle (EV) charging in smart cities. With the increasing number of EVs, ensuring efficient and reliable charging systems has become crucial. The aim of this research is to develop an integrated IoT and deep learning system for detecting anomalies in residential EV charging and enhancing EV load profiling and event detection in smart cities. This approach utilizes IoT devices equipped with infrared cameras to collect thermal images and household EV charging profiles from the database of Thailand utility, subsequently transmitting this data to a cloud database for comprehensive analysis. The methodology includes the use of advanced deep learning techniques such as Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM) algorithms. IoT devices equipped with infrared cameras are used to collect thermal images and EV charging profiles. The data is transmitted to a cloud database for comprehensive analysis. The researchers also utilize feature-based Gaussian mixture models for EV load profiling and event detection. Moreover, the research findings demonstrate the effectiveness of the developed system in detecting anomalies and critical profiles in EV charging behavior. The system provides timely alarms to users regarding potential issues and categorizes the severity of detected problems based on a health index for each charging device. The system also outperforms existing models in event detection accuracy. This research contributes to the field by showcasing the potential of integrating IoT and deep learning techniques in managing residential EV charging in smart cities. The system ensures operational safety and efficiency while also promoting sustainable energy management. The data is collected using IoT devices equipped with infrared cameras and is stored in a cloud database for analysis. The collected data is then analyzed using RNN, LSTM, and feature-based Gaussian mixture models. The approach includes both EV load profiling and event detection, utilizing a feature-based Gaussian mixture model. This comprehensive method aids in identifying unique power consumption patterns among EV owners and outperforms existing models in event detection accuracy. In summary, the research concludes that integrating IoT and deep learning techniques can effectively detect anomalies in residential EV charging and enhance EV load profiling and event detection accuracy. The developed system ensures operational safety and efficiency, contributing to sustainable energy management in smart cities.

Keywords: cloud computing framework, recurrent neural networks, long short-term memory, Iot, EV charging, smart grids

Procedia PDF Downloads 22

Authors: Leonardo Alfredo Mendoza, Cristian Munoz, Marco Aurelio Pacheco, Manoela Kohler, Evelyn Batista, Rodrigo Moura

Abstract:

Natural Language Processing (NLP) techniques are increasingly more powerful to be able to interpret the feelings and reactions of a person to a product or service. Sentiment analysis has become a fundamental tool for this interpretation but has few applications in languages other than English. This paper presents a classification of sentiment analysis in Portuguese with a base of comments from social networks in Portuguese. A word embedding's representation was used with a 50-Dimension GloVe pre-trained model, generated through a corpus completely in Portuguese. To generate this classification, the bidirectional long short-term memory and bidirectional Gated Recurrent Unit (GRU) models are used, reaching results of 99.1%.

Keywords: natural processing language, sentiment analysis, bidirectional long short-term memory, BI-LSTM, gated recurrent unit, GRU

Procedia PDF Downloads 120

Authors: Yagya Raj Pandeya, Joonwhoan Lee

Abstract:

The dynamic behavior of music and video makes it difficult to evaluate musical instrument playing in a video by computer system. Any television or film video clip with music information are rich sources for analyzing musical instruments using modern machine learning technologies. In this research, we integrate the audio and video information sources using convolutional neural network (CNN) and pass network learned features through recurrent neural network (RNN) to preserve the dynamic behaviors of audio and video. We use different pre-trained CNN for music and video feature extraction and then fine tune each model. The music network use 2D convolutional network and video network use 3D convolution (C3D). Finally, we concatenate each music and video feature by preserving the time varying features. The long short term memory (LSTM) network is used for long-term dynamic feature characterization and then use late fusion with generalized mean. The proposed network performs better performance to recognize the musical instrument using audio-video multimodal neural network.

Keywords: multimodal, 3D convolution, music-video feature extraction, generalized mean

Procedia PDF Downloads 183

Authors: Haozhe Xiang

Abstract:

With the rise of the Internet of Things era, intelligent products are gradually integrating into people's lives. Pedestrian trajectory prediction has become a key issue, which is crucial for the motion path planning of intelligent agents such as autonomous vehicles, robots, and drones. In the current technological context, deep learning technology is becoming increasingly sophisticated and gradually replacing traditional models. The pedestrian trajectory prediction algorithm combining neural networks and attention mechanisms has significantly improved prediction accuracy. Based on in-depth research on deep learning and pedestrian trajectory prediction algorithms, this article focuses on physical environment modeling and learning of historical trajectory time dependence. At the same time, social interaction between pedestrians and scene interaction between pedestrians and the environment were handled. An improved pedestrian trajectory prediction algorithm is proposed by analyzing the existing model architecture. With the help of these improvements, acceptable predicted trajectories were successfully obtained. Experiments on public datasets have demonstrated the algorithm's effectiveness and achieved acceptable results.

Keywords: deep learning, graph convolutional network, attention mechanism, LSTM

Procedia PDF Downloads 14

Authors: Sneha Thakur, Sanjeev Karmakar

Abstract:

This paper proposed hybrid Particle Swarm Optimization (PSO) – Long-Short Term Memory (LSTM) model for groundwater level prediction. The evaluation of the performance is realized using the parameters: root mean square error (RMSE) and mean absolute error (MAE). Ground water level forecasting will be very effective for planning water harvesting. Proper calculation of water level forecasting can overcome the problem of drought and flood to some extent. The objective of this work is to develop a ground water level forecasting model using deep learning technique integrated with optimization technique PSO by applying 29 years data of Chhattisgarh state, In-dia. It is important to find the precise forecasting in case of ground water level so that various water resource planning and water harvesting can be managed effectively.

Keywords: long short-term memory, particle swarm optimization, prediction, deep learning, groundwater level

Procedia PDF Downloads 34

Authors: Jiajun Wang, Xiaoge Li

Abstract:

The purpose of the aspect-level sentiment analysis task is to identify the sentiment polarity of aspects in a sentence. Currently, most methods mainly focus on using neural networks and attention mechanisms to model the relationship between aspects and context, but they ignore the dependence of words in different ranges in the sentence, resulting in deviation when assigning relationship weight to other words other than aspect words. To solve these problems, we propose a new aspect-level sentiment analysis model that combines a multi-channel convolutional network and graph convolutional network (GCN). Firstly, the context and the degree of association between words are characterized by Long Short-Term Memory (LSTM) and self-attention mechanism. Besides, a multi-channel convolutional network is used to extract the features of words in different ranges. Finally, a convolutional graph network is used to associate the node information of the dependency tree structure. We conduct experiments on four benchmark datasets. The experimental results are compared with those of other models, which shows that our model is better and more effective.

Keywords: aspect-level sentiment analysis, attention, multi-channel convolution network, graph convolution network, dependency tree

Procedia PDF Downloads 161

Authors: Chaudhary Muhammad Aqdus Ilyas, Matthias Rehm, Kamal Nasrollahi, Thomas B. Moeslund

Abstract:

This paper presents the investigation of data collection procedures, associated with robots when placed with traumatic brain injured (TBI) patients for rehabilitation purposes through facial expression and mood analysis. Rehabilitation after TBI is very crucial due to nature of injury and variation in recovery time. It is advantageous to analyze these emotional signals in a contactless manner, due to the non-supportive behavior of patients, limited muscle movements and increase in negative emotional expressions. This work aims at the development of framework where robots can recognize TBI emotions through facial expressions to perform rehabilitation tasks by physical, cognitive or interactive activities. The result of these studies shows that with customized data collection strategies, proposed framework identify facial and emotional expressions more accurately that can be utilized in enhancing recovery treatment and social interaction in robotic context.

Keywords: computer vision, convolution neural network- long short term memory network (CNN-LSTM), facial expression and mood recognition, multimodal (RGB-thermal) analysis, rehabilitation, robots, traumatic brain injured patients

Procedia PDF Downloads 110

Authors: Muhammad Zubair Khan, Yugyung Lee

Abstract:

Deep learning has recently achieved enormous response in semantic image segmentation. The previously developed U-Net inspired architectures operate with continuous stride and pooling operations, leading to spatial data loss. Also, the methods lack establishing long-term pixels connection to preserve context knowledge and reduce spatial loss in prediction. This article developed encoder-decoder architecture with bi-directional LSTM embedded in long skip-connections and densely connected convolution blocks. The network non-linearly combines the feature maps across encoder-decoder paths for finding dependency and correlation between image pixels. Additionally, the densely connected convolutional blocks are kept in the final encoding layer to reuse features and prevent redundant data sharing. The method applied batch-normalization for reducing internal covariate shift in data distributions. The empirical evidence shows a promising response to our method compared with other semantic segmentation techniques.

Keywords: deep learning, semantic segmentation, image analysis, pixels connection, convolution neural network

Procedia PDF Downloads 70

Authors: Parisa Golbayani, Dan Wang, Ionut¸ Florescu

Abstract:

In this work we implement machine learning techniques to financial statement reports in order to asses company’s credit rating. Specifically, the work analyzes the performance of four neural network architectures (MLP, CNN, CNN2D, LSTM) in predicting corporate credit rating as issued by Standard and Poor’s. The paper focuses on companies from the energy, financial, and healthcare sectors in the US. The goal of this analysis is to improve application of machine learning algorithms to credit assessment. To accomplish this, the study investigates three questions. First, we investigate if the algorithms perform better when using a selected subset of important features or whether better performance is obtained by allowing the algorithms to select features themselves. Second, we address the temporal aspect inherent in financial data and study whether it is important for the results obtained by a machine learning algorithm. Third, we aim to answer if one of the four particular neural network architectures considered consistently outperforms the others, and if so under which conditions. This work frames the problem as several case studies to answer these questions and analyze the results using ANOVA and multiple comparison testing procedures.

Keywords: convolutional neural network, long short term memory, multilayer perceptron, credit rating

Procedia PDF Downloads 201

Authors: Jacqueline Rose T. Alipo-on, Francesca Isabelle F. Escobar, Myles Joshua T. Tan, Hezerul Abdul Karim, Nouar Al Dahoul

Abstract:

Electrocardiogram (ECG) signal analysis and processing are crucial in the diagnosis of cardiovascular diseases, which are considered one of the leading causes of mortality worldwide. However, the traditional rule-based analysis of large volumes of ECG data is time-consuming, labor-intensive, and prone to human errors. With the advancement of the programming paradigm, algorithms such as machine learning have been increasingly used to perform an analysis of ECG signals. In this paper, various deep learning algorithms were adapted to classify five classes of heartbeat types. The dataset used in this work is the synthetic MIT-BIH Arrhythmia dataset produced from generative adversarial networks (GANs). Various deep learning models such as ResNet-50 convolutional neural network (CNN), 1-D CNN, and long short-term memory (LSTM) were evaluated and compared. ResNet-50 was found to outperform other models in terms of recall and F1 score using a five-fold average score of 98.88% and 98.87%, respectively. 1-D CNN, on the other hand, was found to have the highest average precision of 98.93%.

Keywords: heartbeat classification, convolutional neural network, electrocardiogram signals, generative adversarial networks, long short-term memory, ResNet-50

Procedia PDF Downloads 76

Authors: Christina Adly, Meena Abdelmeseeh, Tamer Basha

Abstract:

This paper proposes a system for hand movement recognition using multichannel surface EMG(sEMG) signals obtained from 40 subjects using 40 different exercises, which are available on the Ninapro(Non-Invasive Adaptive Prosthetics) database. First, we applied processing methods to the raw sEMG signals to convert them to their amplitudes. Second, we used deep learning methods to solve our problem by passing the preprocessed signals to Fully connected neural networks(FCNN) and recurrent neural networks(RNN) with Long Short Term Memory(LSTM). Using intrasubject evaluation, The accuracy using the FCNN is 72%, with a processing time for training around 76 minutes, and for RNN's accuracy is 79.9%, with 8 minutes and 22 seconds processing time. Third, we applied some postprocessing methods to improve the accuracy, like majority voting(MV) and Movement Error Rate(MER). The accuracy after applying MV is 75% and 86% for FCNN and RNN, respectively. The MER value has an inverse relationship with the prediction delay while varying the window length for measuring the MV. The different part uses the RNN with the intersubject evaluation. The experimental results showed that to get a good accuracy for testing with reasonable processing time, we should use around 20 subjects.

Keywords: hand movement recognition, recurrent neural network, movement error rate, intrasubject evaluation, intersubject evaluation

Procedia PDF Downloads 99

Authors: Lan Zhang, Cailing Zhang

Abstract:

In recent years, many face anti-spoofing algorithms have high detection accuracy when detecting 2D face anti-spoofing or 3D mask face anti-spoofing alone in the field of face anti-spoofing, but their detection performance is greatly reduced in multidimensional and cross-datasets tests. The rPPG method used for face anti-spoofing uses the unique vital information of real face to judge real faces and face anti-spoofing, so rPPG method has strong stability compared with other methods, but its detection rate of 2D face anti-spoofing needs to be improved. Therefore, in this paper, we improve an rPPG(Remote Photoplethysmography) method(MrPPG) for face anti-spoofing which through color space fusion, using the correlation of pulse signals between real face regions and background regions, and introducing the cyclic neural network (LSTM) method to improve accuracy in 2D face anti-spoofing. Meanwhile, the MrPPG also has high accuracy and good stability in face anti-spoofing of multi-dimensional and cross-data datasets. The improved method was validated on Replay-Attack, CASIA-FASD, Siw and HKBU_MARs_V2 datasets, the experimental results show that the performance and stability of the improved algorithm proposed in this paper is superior to many advanced algorithms.

Keywords: face anti-spoofing, face presentation attack detection, remote photoplethysmography, MrPPG

Procedia PDF Downloads 137

Authors: Surajit Chakrabarty, Piyush Chauhan, Subhasis Panda, Sujoy Bhattacharya

Abstract:

A large proportion of roads in India are not well maintained as per the laid down public safety guidelines leading to loss of direction control and fatal accidents. We propose a technique to detect speed breakers and potholes using mobile sensor data captured from multiple vehicles and provide a profile of the road. This would, in turn, help in monitoring roads and revolutionize digital maps. Incorporating randomness in the model formulation for detection of speed breakers and potholes is crucial due to substantial heterogeneity observed in data obtained using a mobile application from multiple vehicles driven by different drivers. This is accomplished with Hidden Markov Models, whose hidden state sequence is found for each time step given the observables sequence, and are then fed as input to LSTM network with peephole connections. A precision score of 0.96 and 0.63 is obtained for classifying bumps and potholes, respectively, a significant improvement from the machine learning based models. Further visualization of bumps/potholes is done by converting time series to images using Markov Transition Fields where a significant demarcation among bump/potholes is observed.

Keywords: deep learning, hidden Markov model, pothole, speed breaker

Procedia PDF Downloads 94

Authors: Victor Breux, Jérôme Boutet, Alain Goret, Viviane Cattin

Abstract:

Early detection of anomalies in data centers is important to reduce downtimes and the costs of periodic maintenance. However, there is little research on this topic and even fewer on the fusion of sensor data for the detection of abnormal events. The goal of this paper is to propose a method for anomaly detection in data centers by combining sensor data (temperature, humidity, power) and deep learning models. The model described in the paper uses one autoencoder per sensor to reconstruct the inputs. The auto-encoders contain Long-Short Term Memory (LSTM) layers and are trained using the normal samples of the relevant sensors selected by correlation analysis. The difference signal between the input and its reconstruction is then used to classify the samples using feature extraction and a random forest classifier. The data measured by the sensors of a data center between January 2019 and May 2020 are used to train the model, while the data between June 2020 and May 2021 are used to assess it. Performances of the model are assessed a posteriori through F1-score by comparing detected anomalies with the data center’s history. The proposed model outperforms the state-of-the-art reconstruction method, which uses only one autoencoder taking multivariate sequences and detects an anomaly with a threshold on the reconstruction error, with an F1-score of 83.60% compared to 24.16%.

Keywords: anomaly detection, autoencoder, data centers, deep learning

Procedia PDF Downloads 153

Authors: YingWei Tan, XueFeng Ding

Abstract:

Attention-based voice activity detection models have gained significant attention in recent years due to their fast training speed and ability to capture a wide contextual range. The inclusion of multi-head style and position embedding in the attention architecture are crucial. Having multiple attention heads allows for differential focus on different parts of the sequence, while position embedding provides guidance for modeling dependencies between elements at various positions in the input sequence. In this work, we propose an approach by considering each head as a node, enabling the application of graph neural networks (GNN) to identify correlations among the different nodes. In addition, we adopt an implementation named rotary position embedding (RoPE), which encodes absolute positional information into the input sequence by a rotation matrix, and naturally incorporates explicit relative position information into a self-attention module. We evaluate the effectiveness of our method on a synthetic dataset, and the results demonstrate its superiority over the baseline CRNN in scenarios with low signal-to-noise ratio and noise, while also exhibiting robustness across different noise types. In summary, our proposed framework effectively combines the strengths of CNN and RNN (LSTM), and further enhances detection performance through the integration of graph neural networks and rotary position embedding.

Keywords: voice activity detection, CRNN, graph neural networks, rotary position embedding

Procedia PDF Downloads 25

Authors: Khalid Ali, Manar Jammal

Abstract:

In order to meet the stringent latency and reliability requirements of the upcoming 5G networks, Open Radio Access Networks (O-RAN) have been proposed. The virtualization of O-RAN has allowed it to be treated as a Network Function Virtualization (NFV) architecture, while its components are considered Virtualized Network Functions (VNFs). Hence, intelligent Machine Learning (ML) based solutions can be utilized to apply different resource management and allocation techniques on O-RAN. However, intelligently allocating resources for O-RAN VNFs can prove challenging due to the dynamicity of traffic in mobile networks. Network providers need to dynamically scale the allocated resources in response to the incoming traffic. Elastically allocating resources can provide a higher level of flexibility in the network in addition to reducing the OPerational EXpenditure (OPEX) and increasing the resources utilization. Most of the existing elastic solutions are reactive in nature, despite the fact that proactive approaches are more agile since they scale instances ahead of time by predicting the incoming traffic. In this work, we propose and evaluate traffic forecasting models based on the ML algorithm. The algorithms aim at predicting future O-RAN traffic by using previous traffic data. Detailed analysis of the traffic data was carried out to validate the quality and applicability of the traffic dataset. Hence, two ML models were proposed and evaluated based on their prediction capabilities.

Keywords: O-RAN, traffic forecasting, NFV, ARIMA, LSTM, elasticity

Procedia PDF Downloads 172

Authors: Leah Lee

Abstract:

This paper examines the correlation between deep learning models and emotions with noises to see whether or not noises mask emotions. The deep learning models used are plain convolutional neural networks (CNN), auto-encoder, long short-term memory (LSTM), and Visual Geometry Group-16 (VGG-16). Emotion datasets used are Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS), Crowd-sourced Emotional Multimodal Actors Dataset (CREMA-D), Toronto Emotional Speech Set (TESS), and Surrey Audio-Visual Expressed Emotion (SAVEE). To make it four times bigger, audio set files, stretch, and pitch augmentations are utilized. From the augmented datasets, five different features are extracted for inputs of the models. There are eight different emotions to be classified. Noise variations are white noise, dog barking, and cough sounds. The variation in the signal-to-noise ratio (SNR) is 0, 20, and 40. In summation, per a deep learning model, nine different sets with noise and SNR variations and just augmented audio files without any noises will be used in the experiment. To compare the results of the deep learning models, the accuracy and receiver operating characteristic (ROC) are checked.

Keywords: auto-encoder, convolutional neural networks, long short-term memory, speech emotion recognition, visual geometry group-16

Procedia PDF Downloads 35

Authors: Jimeng Shi, Mahek Jain, Giri Narasimhan

Abstract:

Time Series Forecasting (TSF) is used to predict the target variables at a future time point based on the learning from previous time points. To keep the problem tractable, learning methods use data from a fixed-length window in the past as an explicit input. In this paper, we study how the performance of predictive models changes as a function of different look-back window sizes and different amounts of time to predict the future. We also consider the performance of the recent attention-based Transformer models, which have had good success in the image processing and natural language processing domains. In all, we compare four different deep learning methods (RNN, LSTM, GRU, and Transformer) along with a baseline method. The dataset (hourly) we used is the Beijing Air Quality Dataset from the UCI website, which includes a multivariate time series of many factors measured on an hourly basis for a period of 5 years (2010-14). For each model, we also report on the relationship between the performance and the look-back window sizes and the number of predicted time points into the future. Our experiments suggest that Transformer models have the best performance with the lowest Mean Average Errors (MAE = 14.599, 23.273) and Root Mean Square Errors (RSME = 23.573, 38.131) for most of our single-step and multi-steps predictions. The best size for the look-back window to predict 1 hour into the future appears to be one day, while 2 or 4 days perform the best to predict 3 hours into the future.

Keywords: air quality prediction, deep learning algorithms, time series forecasting, look-back window

Procedia PDF Downloads 116

Authors: Niklas Panten, Eberhard Abele

Abstract:

This paper presents a novel approach for real-time and near-optimal control of industrial smart grids by deep reinforcement learning (DRL). To achieve highly energy-efficient factory systems, the energetic linkage of machines, technical building equipment and the building itself is desirable. However, the increased complexity of the interacting sub-systems, multiple time-variant target values and stochastic influences by the production environment, weather and energy markets make it difficult to efficiently control the energy production, storage and consumption in the hybrid industrial smart grids. The studied deep reinforcement learning approach allows to explore the solution space for proper control policies which minimize a cost function. The deep neural network of the DRL agent is based on a multilayer perceptron (MLP), Long Short-Term Memory (LSTM) and convolutional layers. The agent is trained within multiple Modelica-based factory simulation environments by the Advantage Actor Critic algorithm (A2C). The DRL controller is evaluated by means of the simulation and then compared to a conventional, rule-based approach. Finally, the results indicate that the DRL approach is able to improve the control performance and significantly reduce energy respectively operating costs of industrial smart grids.

Keywords: industrial smart grids, energy efficiency, deep reinforcement learning, optimal control

Procedia PDF Downloads 157