Search

To understand the mass behaviour of people, an effectual platform was provided by the online social network, which aids in developing techniques for the surveillance of Human Immunodeficiency Viruses/Acquired Immunodeficiency Syndrome (HIV/AIDS). With the rapid advancement of social sites, namely Facebook, Twitter, and blogs, the social networking approach is the most promising factor in HIV/AIDS investigation. Recently, most of the prevailing works implemented various frameworks to classify HIV/AIDS-related information using Social Media (SM) data. However, the traditional techniques were not generalized well enough to handle the complex structure of SM data. Also, the existing models were less effective due to the lack of annotation processes and pre-processing strategies. In this paper, to identify as well as classify HIV/AIDS-related SM data, a novel strategy has been proposed by utilizing the Levenshtein Distance-KMeans algorithm (LD-KMeans) and Deep Belief Networks—Long Short Term Memory (DBN-LSTM) models. The proposed work mainly focuses on the discussions on HIV and AIDS-related issues taking place on Twitter. For an efficient HIV/AIDS-related tweet classification, the proposed work undergoes the following steps. Initially, the tweets from Twitter are extracted by using Twitter API, and then, the preprocessing function is performed on the Twitter data. Then, the annotation extraction is performed. Next, the tweets are separated into organization tweets and person tweets based on the annotation. In the proposed work, organization tweets are highly considered. After that, the text normalization is performed, which provides the cleaned structured tweets. Then, the hashtags related to HIV and AIDS are identified and grouped together by using the LD-KMeans algorithm. Thereafter, the word embedding is performed by means of M-Word2Vec. Once the embedding process is completed, the most important features are selected by the LS-DFO algorithm. Finally, on the basis of selected features, the classification is performed, which efficiently classifies the HIV/AIDS-related tweets into different categories like symptoms, awareness, medicine, and reason. In this research work, Twitter data are utilized. Then, the outcomes obtained by the proposed methodology are analogized with the prevailing algorithms. Thus, the analysis results proved that the research methodology obtained a better accuracy, sensitivity, and specificity of 94.65%, 94.56%, and 94.25%, respectively. Likewise, the proposed work reached a tweet identification time of 72154 ms. Finally, the experiential outcomes demonstrated that regarding sensitivity, specificity, along with accuracy, the proposed model outperformed the prevailing systems in the process of classifying the HIV\AIDS-related tweets. [ABSTRACT FROM AUTHOR]

This study models the rainfall-runoff relationship in the Kebir-Rhumel River watershed in the Constantine Highlands, Algeria, using data from three concomitant rainfall and hydrometric stations. Statistical tests confirmed the absence of breaks in the series. We applied four conceptual models (GR4J, IHAC6, MORDOR, TOPMO8) and neural network models (RNN, NARX, LSTM) over three- and ten-year periods. Among the conceptual models, GR4J provided the best fit, highlighting the non-stationary nature of the relationship. The PMC neural network model performed well over three years but was less effective over ten years due to low flow influence. Notably, the NARX-RNN and RNN-LSTM models showed excellent predictive accuracy, with NARX-RNN perfectlycapturing flow dynamics and RNN-LSTM achieving minimal RMSE and high correlation coefficients. This study lies the comparative analysis of conceptual and neural network models, specifically the NARX-RNN and RNN-LSTM models, which have not been extensively applied in this context. This research fills the gap in understanding the effectiveness of neural network models in modelling non-stationary rainfall-runoff relationships in the region. [ABSTRACT FROM AUTHOR]

DNS tunneling is the attempt to create a hidden tunnel through a domain name service. Such a tunnel would jeopardize the targeted network and open the door for illegal access, control, and data exfiltration. The information security research community showed the variety of techniques that have been proposed to detect the tunnel. The majority of these efforts were relying on machine learning techniques where features of tunneling are considered such as length of DNS query, size, and entropy of the query. However, an additional analysis of the lexical information of the DNS query has been depicted recently and showed remarkable performance. This paper aims to examine the role of Long Short Term Memory (LSTM) model in terms of DNS lexical analysis. Two benchmark datasets related to DNS have been used. In addition, a character mapping mechanism has been used to replace every possible character with an integer number. Consequentially, the mapped representation has been fed into an LSTM model for DNS tunneling detection. Results showed that the proposed method was able to obtain a weighted average F1-score of 98% for both datasets respectively. Such results are competitive in the context of the state of the art and demonstrate the efficacy of the lexical analysis within the DNS tunneling detection task. [ABSTRACT FROM AUTHOR]

Extracting biomedical entities has caught many researchers' attention in which the recent technique of word embedding is employed for such a task. Yet, the traditional word embedding architectures of Word2vec or Glove are still suffering from the 'out-of-vocabulary' (OOV) problem. This problem occurs when an unseen term might be encountered during the testing which leads to absence of embedding vector. Hence, this study aims to propose a character-level embedding through FastText architecture. In fact, handling the character-level seems a promising solution for the OOV problem. To this end, the proposed FastText architecture has been used to generate embedding vectors for the possible N-gram combinations of each word. Consequentially, these vectors have been fed to a Long Short Term Memory (LSTM) architecture for classifying the words into its biomedical classes. Using two benchmark datasets of BioCreative-II and NCBI, the proposed method was able to produce an f-measure of 0.912 and 0.918 respectively. Comparing these results with the baseline studies demonstrates the superiority of the proposed character-level embedding of FastText in terms of Biomedical Named Entity Recognition (BNER) task. [ABSTRACT FROM AUTHOR]

Every second, a huge volume of multi-dimensional data is generated in fields such as Social Networking, Industrial Internet of Things, Stock market and E-commerce applications. Knowledge and pattern extraction are a challenging task in the evolving nature of data stream. Major issues are (i) ‘concept drift’ occurs as a result of pattern changes in the data distribution and (ii) ‘concept evolution’ occurs when a new class evolves in the data stream. These issues degrade the performance of learning models. In this paper, we focus on detection of concept evolution and enhance the performance of classifiers. For this, we propose a new model to identify novel classes, namely, Detection of Novel Classes (DNC). The proposed method adopts long short term memory to continuously observe the streaming data in order to detect emerging classes. The continuous monitoring allows the model to distinguish between existing classes and the novel classes which save time and memory. Also, the proposed method is demonstrated for identifying more than one novel class. The experiments are performed over seven different datasets. The results confirm the efficiency is increased ranging from 6% to 34% by the proposed method in identifying new concepts in the evolving data stream than the existing methods available in the literature. [ABSTRACT FROM AUTHOR]

Purpose: Accurate detection of Atrial Fibrillation (AF) has great significance in the field of medical science which can reduce the rate of mortality and morbidity. The present study focuses on Electrocardiography (ECG) signal classification using dimensionality reduction techniques combined with R wave to R wave interval (RR interval) features. Materials and Methods: In the first approach, Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), Independent Component Analysis (ICA), and Probabilistic Principal Component Analysis (PPCA) are performed independently on denoised ECG signal using Discrete Wavelet Transform (DWT) for the classification of ECG signal. In the second approach, the dimensionality reduction techniques combined with RR interval features are used for the classification of ECG signal. Results: Machine Learning (ML) algorithms such as Decision Tree (DT), Support Vector Machine (SVM), and Deep Learning (DL) algorithms such as Long Short Term Memory (LSTM) and Bi-Directional LSTM (BiLSTM) are used for classification purposes. Conclusion: The proposed methodology provided an overall accuracy of 93.65% with PCA and LSTM classifier and an overall accuracy of 99.45% with PCA combined with RR interval features and LSTM classifier. The developed technology has potential applications in many practical solutions. [ABSTRACT FROM AUTHOR]

Pneumonia is a serious inflammatory disease that causes lung ulcers, and it is one of the leading reasons for pediatric death in the world. Chest X-rays are perhaps the most commonly utilized modalities to recognize pneumonia. Generally, the illness could be analyzed by a specialist radiologist. But for some reason, the diagnosis may be subjective. Thus, the physicians must be guided by computer-aided diagnosis frameworks in this challenging task. In this study, we propose a combined deep learning architecture to identify pneumonia in chest radiography images. We first, use Adaptive Median Filter for images enhancement, then we employ a regularized Convolutional Neural Network for features extraction, and then we use Long Short Term Memory as a classifier. Finally, the attention mechanism is used to direct the network attention to relevant features. The suggested approach was tested on two publicly available pneumonia X-ray datasets provided by Kermany and the Radiological Society of North America. On the Kermany and RSNA datasets, the suggested technique attained accuracy rates of 99.91% and 88.86%, respectively. In the last stage of our experiments, we employed a Grad-CAM-based color visualization technique to precisely interpret the detection of pneumonia in radiological images. The results outperformed those of state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Traditional music emotion recognition (MER) faces problems such as lack of contextual information, inaccurate recognition of music emotions, and difficulty in handling nonlinear relationships. This article first used long short-term memory (LSTM) networks to capture global information and contextual relationships of music. Subsequently, the DCNN was chosen to process sequence data and capture global dependencies to improve the accuracy of MER. Finally, a MER model was constructed based on DCNN to recognize and classify music emotions. This article obtained the impact of different parameter values on model training iterations by adjusting hyperparameters related to training. The optimal values for learning rate μ , momentum coefficient α , weight attenuation coefficient γ , and Dropout coefficient were 0.01, 0.7, 0.0003, and 0.5, respectively. The DCNN used in this article was iteratively trained with recurrent neural networks, convolutional recurrent neural networks, and transform domain neural networks for audio spectrograms, and the results were compared. The experimental findings indicated that the spectral recognition accuracy of DCNN was stable at 95.68%, far higher than the other three different networks. The results showed that the DCNN method used in this article could more accurately distinguish different negative emotions and positive emotions. [ABSTRACT FROM AUTHOR]

• Deep auto-encoder model is used to handle the non-linear and redundant information in dam key parameters. • A hybrid system based on intelligent optimization algorithms and deep learning for predicting tailings dam parameters. • The uncertainty analysis provides more comprehensive and precise interval predictions. • The prediction accuracy of monitoring parameters of tailings dam is significantly improved. Tailings dams, as critical infrastructure, play a vital role in ensuring the safety and reliability of tailings pond systems. Predicting the trend of tailings dam monitoring parameters helps monitor its operational status and support safety management and decision-making. The traditional time series prediction model focuses on point prediction while neglecting the nonlinear feature of data and the uncertainty of prediction results, which is far from meeting the requirements for risk monitoring and safety management of tailings dam systems. To address this, Deep Auto-Encoder is used to extract and optimize features of multi-dimensional time series; Whale Optimization Algorithm, Long Short Term Memory, and Kernel Density Estimation are used to build a monitoring parameters trend interval prediction model. It can effectively predict and analyze the uncertainty of tailings dam monitoring parameters. Time series data of monitoring parameters are collected from a tailings dam in Anhui, China and modeled to verify the effectiveness of the model. The results show that the hybrid prediction model performs the best in monitoring parameter prediction accuracy and trend. The Root Mean Square Error values of the Whale Optimization Algorithm and Long Short Term Memory prediction for reservoir water level, surface displacement, internal displacement, phreatic line, and beach width are 0.0160, 1.0405, 0.0389, 0.0176, and 0.7525, with errors reduced by 0.37 % to 17.92 % compared to other models. The uncertainty analysis confirms the high reliability of the model. The Kernel Density Estimation effectively forecasts the fluctuation range of predicted values, significantly improving the prediction effectiveness of the monitoring parameters. [ABSTRACT FROM AUTHOR]

Security is considered the primary challenge in the healthcare industry in the aspect of the ethical and legal perspective of patient medical data. In existing approaches, the accessibility, reliability, and confidentiality of medical data are needed for more security in the healthcare industry. The main obstacles involved in the communication between the smartphone and monitoring devices allow a data theft attack. Hence, the secured model needs to be designed for providing data security in healthcare applications. By avoiding these challenges, this task provides secured intrusion detection with blockchain-based healthcare data transmission with the help of novel intelligence techniques. Initially, in the data acquisition phase, the gathered data from online sources consist of brain, skin, and retinal medical images. Then, the Weight Optimized Deep Belief Network (WO-DBN) checks the nodes to confirm whether the gathered data are in a malicious state or not at the time of transmission. The medical data are secured with the support of encryption and blockchain technology, where the medical images are enciphered by chaotic-map-aided image encryption through optimal key generation. These encrypted images are uploaded to the blockchain for secure data communication to the cloud server. Here, the authorized person can replace the data using the same optimal key and also it can be done after the decryption in the proposed chaotic map. The decrypted data are given to the final disease diagnosis phase for classifying the images without any information loss with the support of a Residual Network (Resnet101), where the final layer is restored by the "Deep Neural Network (DNN) and Long Short Term Memory (LSTM)" to enhance the classification accuracy and it is named Res-LSTM+DNN. The weight optimization in DBN, optimal key generation, and hyper-parameter tuning in classification are done by the Improved Dingo Optimizer (IDOX). From the overall result validation, the accuracy rate of the recommended approach scores 98.6%. [ABSTRACT FROM AUTHOR]

Scene classification in remote sensing is challenging due to high inter-class similarity and low intra-class similarity. Numerous techniques have been introduced, but accurately classifying scenes remains arduous. To address this challenge, To address this, we propose a hybrid framework, DF-DNLSTM, integrating DenseNet-121 for feature extraction and BiLSTM for sequential modeling, enhancing accuracy and contextual understanding. Second, a Conditional Generative Adversarial Network (CGAN) is employed for data augmentation, improving training data quantity and quality. Finally, the study introduces SwarmHawk, a hybrid optimization algorithm that combines particle swarm optimization (PSO) and Harris hawk optimization (HHO). SwarmHawk ensures the selection of informative features while concurrently eliminating duplicates and redundancies. It also reduces computational time to 4863 s. The proposed DF-DNLSTM model is rigorously assessed on three public datasets-UCM, AID, and NWPU. Results demonstrate its superior efficacy, achieving 99.87% accuracy on UCM, equivalent accuracy on NWPU, and sustaining 98.57% accuracy on AID. This study establishes DF-DNLSTM's effectiveness, highlighting its potential contributions to advancing remote sensing scene classification. [ABSTRACT FROM AUTHOR]

It is generally recognized that accurate and timely prediction of future traffic flow information is one of the important conditions for improving the utilization rate of road networks and relieving traffic congestion. Since long short-term memory (LSTM) networks have obvious advantages in dealing with time series data, LSTM-based prediction models, as well as the improved and hybrid models based on LSTM, have been widely used in the field of traffic flow prediction and have achieved a series of important results. However, although some existing reviews have provided a comprehensive summary of and prospects for traffic flow prediction methods, there is no recent review specifically summarizing the traffic flow prediction methods based on LSTM. Therefore, this article presents a comprehensive review of LSTM-based methods for traffic flow prediction. First, we briefly introduce the development history of traffic flow prediction, the advantages of using LSTM for traffic flow prediction, and the basic principles of LSTM. Then, the representative model structures, training methods, public datasets, attractive results, the advantages and disadvantages of different LSTM-based methods are presented and analyzed in detail. Furthermore, to illustrate the performance of various refinements and hybrid models based on LSTM for traffic flow prediction, we meticulously select several representative approaches to conduct comparative experiments and quantitative analysis. Finally, we discuss the current limitations and challenges of LSTM and traffic flow prediction and present potential areas of future research interest to the readers. [ABSTRACT FROM AUTHOR]

In energy-harvesting wireless body area networks (EH-WBAN), the self-sustainability of body sensor nodes without compromising the service quality criteria is very important. In remote vital signs monitoring applications, the sampling rate of body nodes in each period should be determined to achieve this goal. There are two fundamental challenges in determining the sampling rate: 1) In EH-WBAN, the rate of the harvestable energy is time-dependent and unpredictable. 2) The vital signs of the patient have different change rates relevant to the time. Therefore, a technique is required that can learn the changes of the harvestable energy and the change rates of vital signs simultaneously and specify the proper sampling rate for BNs. The reinforcement learning algorithms are among the efficient solutions for this problem because they are capable of learning environmental uncertainties. Previous RL-based methods for determining the BN's sampling rate have three fundamental problems: 1) focus on the optimization of the transmission energy and ignore the sensing energy, 2) only affect one of the two aspects of energy or sensed data in determining the sampling rate and 3) discretization of the problem space does not ensure the determination of the optimal sampling rate. Therefore, this paper proposes a method named “deep reinforcement learning-based dynamic sampling” (DRDS) which first formulates the sampling rate determination problem as a Markov decision process (MDP) and proposes a deep deterministic policy gradient algorithm (DDPG) to solve it. The proposed method considers both energy and data variability aspects in determining the sampling rate. Two parameters, the super-capacitor’s voltage level, and ambient light intensity, are considered for the energy aspect; for the data variability aspect, the long short-term memory (LSTM) algorithm is developed to predict the change rate of data in the next round. Simulations indicate that by preserving the data integrity, the proposed method can decrease the sampling rate and unnecessary data transmission by about 49.95 % and 89.7 %, respectively, compared with state-of-the-art methods, and allow the sensor to achieve self-sustainability.

This paper has aimed to develop the Breast Cancer Detection (BCD) technique along with the mammography reports of the patients. For this proposed framework, the required images are collected from publicly available datasets. Further, the collected images are given into the phase of feature extraction, where the deep features based on texture features, ResNet, morphological features and density features are extracted and fused together. In addition, the weighted fused feature selection is performed with the fused features, in which the weight is optimised with the Modified Hybrid based Leader Optimization (MHLBO) obtained from Hybrid Leader-Based Optimization(HLBO) to elevate the feature performance on classification. Then, breast cancer can be classified with the developed Adaptive Cascaded Deep Network (ACDN) by combing the classification results of, Recurrent Neural Network (RNN), Gate Recurrent Unit (GRU), and Long Short-Term Memory (LSTM). Here, the parameter tuning has occurred in all three classifiers using the same enhanced HLBO. Throughout the experimentation, the designed model shows 97% and 96% regarding accuracy and F1-score, respectively. The tentative outcomes show the efficacy of the designed BCD method is compared with other baseline approaches. [ABSTRACT FROM AUTHOR]

Abstract Quantitative measurement of RNA expression levels through RNA-Seq is an ideal replacement for conventional cancer diagnosis via microscope examination. Currently, cancer-related RNA-Seq studies focus on two aspects: classifying the status and tissue of origin of a sample and discovering marker genes. Existing studies typically identify marker genes by statistically comparing healthy and cancer samples. However, this approach overlooks marker genes with low expression level differences and may be influenced by experimental results. This paper introduces “GENESO,” a novel framework for pan-cancer classification and marker gene discovery using the occlusion method in conjunction with deep learning. we first trained a baseline deep LSTM neural network capable of distinguishing the origins and statuses of samples utilizing RNA-Seq data. Then, we propose a novel marker gene discovery method called “Symmetrical Occlusion (SO)”. It collaborates with the baseline LSTM network, mimicking the “gain of function” and “loss of function” of genes to evaluate their importance in pan-cancer classification quantitatively. By identifying the genes of utmost importance, we then isolate them to train new neural networks, resulting in higher-performance LSTM models that utilize only a reduced set of highly relevant genes. The baseline neural network achieves an impressive validation accuracy of 96.59% in pan-cancer classification. With the help of SO, the accuracy of the second network reaches 98.30%, while using 67% fewer genes. Notably, our method excels in identifying marker genes that are not differentially expressed. Moreover, we assessed the feasibility of our method using single-cell RNA-Seq data, employing known marker genes as a validation test.

Abstract Physical fitness is one of the most important traits a person could have for health longevity. Conducting regular exercise is fundamental to maintaining physical fitness, but with the caveat of occurring injury if not done properly. Several algorithms exists to automatically monitor and evaluate exercise using the user’s pose. However, it is not an easy task to accurately monitor and evaluate exercise poses automatically. Moreover, there are limited number of datasets exists in this area. In our work, we attempt to construct a neural network model that could be used to evaluate exercise poses based on key points extracted from exercise video frames. First, we collected several images consists of different exercise poses. We utilize the the OpenPose library to extract key points from exercise video datasets and LSTM neural network to learn exercise patterns. The result of our experiment has shown that the methods used are quite effective for exercise types of push-up, sit-up, squat, and plank. The neural-network model achieved more than 90% accuracy for the four exercise types.

Abstract With technological innovations, enterprises in the real world are managing every iota of data as it can be mined to derive business intelligence (BI). However, when data comes from multiple sources, it may result in duplicate records. As data is given paramount importance, it is also significant to eliminate duplicate entities towards data integration, performance and resource optimization. To realize reliable systems for record deduplication, late, deep learning could offer exciting provisions with a learning-based approach. Deep ER is one of the deep learning-based methods used recently for dealing with the elimination of duplicates in structured data. Using it as a reference model, in this paper, we propose a framework known as Enhanced Deep Learning-based Record Deduplication (EDL-RD) for improving performance further. Towards this end, we exploited a variant of Long Short Term Memory (LSTM) along with various attribute compositions, similarity metrics, and numerical and null value resolution. We proposed an algorithm known as Efficient Learning based Record Deduplication (ELbRD). The algorithm extends the reference model with the aforementioned enhancements. An empirical study has revealed that the proposed framework with extensions outperforms existing methods.

We explore the use of caching both at the network edge and within User Equipment (UE) to alleviate traffic load of wireless networks. We develop a joint cache placement and delivery policy that maximizes the Quality of Service (QoS) while simultaneously minimizing backhaul load and UE power consumption, in the presence of an unknown time-variant file popularity. With file requests in a time slot being affected by download success in the previous slot, the caching system becomes a non-stationary Partial Observable Markov Decision Process (POMDP). We solve the problem in a deep reinforcement learning framework based on the Advantageous Actor-Critic (A2C) algorithm, comparing Feed Forward Neural Networks (FFNN) with a Long Short-Term Memory (LSTM) approach specifically designed to exploit the correlation of file popularity distribution across time slots. Simulation results show that using LSTM-based A2C outperforms FFNN-based A2C in terms of sample efficiency and optimality, demonstrating superior performance for the non-stationary POMDP problem. For caching at the UEs, we provide a distributed algorithm that reaches the objectives dictated by the agent controlling the network, with minimum energy consumption at the UEs, and minimum communication overhead.

Objective To construct a time series analysis fusion tool using multisource internet data and then accurately predict the incidence trend of hepatitis in Chongqing. Methods The incidence rate of hepatitis were obtained from the database of the Centre for Health and Disease Control. Air pollutant data were obtained from the official website of the China Environmental Monitoring Station, climate data were obtained from the National Meteorological Galaxy Center, and network index data were obtained through Baidu search engine. The time duration was from November 2013 to May 2023. Based on existing time series analysis methods, multisource data were used to correct the residual part of the decomposition model. A delayed input neural network (DINN) was constructed based on the respective advantages of non autoregressive (NAR) and long short-term memory (LSTM) recurrent neural networks. Afterwards, optimization modules such as the Nutcracker Optimization Algorithm (NOA) and Joint Quantile Huber Loss (JQHL) were added to the foundation, and then DINN+ was constructed. Results Compared to common single-input models and synchronous multi-input models, DINN achieved the best prediction performance. After adding hyperparameters and loss function optimization, the predictive performance of DINN+ was further improved, with a mean-square error (MSE) of 0.170 9, a mean absolute error (MAE) of 0.461 2, a root-mean-square error (RMSE) of 0.582 1, a mean absolute percentage error (MAPE) of 0.062 6, and a R-square (R2) of 0.884 0 in a testing set. Conclusion Based on the ideas of diverse methods and multidimensional data fusion, we propose a DINN+ optimization model with good accuracy and generalization ability on the basis of previous time series analysis. This model enriches and supplements the methodological research content of using multisource data to calibrate infectious disease time series prediction analysis and can serve as a new benchmark for future analysis of influencing factors and trend prediction of infectious disease public health.

In the profession of education, predicting students' academic success is an essential responsibility. This study introduces a novel methodology for predicting students' pass or fail outcome in certain courses. The system utilises academic, demographic, emotional, and VLE sequence information of students. Traditional prediction methods often struggle to capture the temporal dynamics inherent in student data, such as learning trajectories, study habits, and evolving performance patterns. In response, this research leverages Recurrent Neural Network (RNNs) and Long Short Term Memory (LSTM) network (LSTMs), which are specifically designed to model sequences and long-term dependencies from OULAD and self-generated Emotional dataset. By incorporating these architectures, the proposed methodology excels in capturing the intricate relationships between various factors over time. Further, various ML models such as Random Forest (RF), Support Vector Machine (SVM), Naive Bayes (NB) and Decision Tree (DT) are integrated with RNN + LSTM to enhance the predictive power of model. The proposed system with RNN + LSTM + RF techniques gained approximately 97% accuracy that is comparatively higher than RNN + LSTM + SVM, RNN + LSTM + NB and RNN + LSTM + DT i.e., 90.67%, 86.45% & 84.42% respectively. [ABSTRACT FROM AUTHOR]

Quantitative measurement of RNA expression levels through RNA-Seq is an ideal replacement for conventional cancer diagnosis via microscope examination. Currently, cancer-related RNA-Seq studies focus on two aspects: classifying the status and tissue of origin of a sample and discovering marker genes. Existing studies typically identify marker genes by statistically comparing healthy and cancer samples. However, this approach overlooks marker genes with low expression level differences and may be influenced by experimental results. This paper introduces "GENESO," a novel framework for pan-cancer classification and marker gene discovery using the occlusion method in conjunction with deep learning. we first trained a baseline deep LSTM neural network capable of distinguishing the origins and statuses of samples utilizing RNA-Seq data. Then, we propose a novel marker gene discovery method called "Symmetrical Occlusion (SO)". It collaborates with the baseline LSTM network, mimicking the "gain of function" and "loss of function" of genes to evaluate their importance in pan-cancer classification quantitatively. By identifying the genes of utmost importance, we then isolate them to train new neural networks, resulting in higher-performance LSTM models that utilize only a reduced set of highly relevant genes. The baseline neural network achieves an impressive validation accuracy of 96.59% in pan-cancer classification. With the help of SO, the accuracy of the second network reaches 98.30%, while using 67% fewer genes. Notably, our method excels in identifying marker genes that are not differentially expressed. Moreover, we assessed the feasibility of our method using single-cell RNA-Seq data, employing known marker genes as a validation test. [ABSTRACT FROM AUTHOR]

In light of the complexity of failure mechanisms in rotating machinery and the significant differences in feature information, traditional diagnostic models often rely on prior knowledge, leading to low accuracy and poor adaptability. We propose a rotating machinery fault diagnosis method named "Randomized Quantization-Randomized LSTM Block Mapping Method-Stochastic Configuration Network" (RQ-RLBM-SCN). To address the challenges posed by limited feature information in failing machinery and insufficient training samples, we employ random quantization data augmentation to expand the original data samples from multiple sensors. This enhances the model's adaptability, accuracy and mitigating overfitting. Next, we employ a random LSTM block mapping method to extract features, solving the difficulty of extracting temporal data features in stochastic configuration network (SCN). Sub-sequently, we utilize a stochastic configuration network (SCN) for classification. SCN dynamically configures parameters without the need for backpropagation to update parameters, ensuring a proper learning rate while effectively avoiding issues such as gradient explosion or vanishing. Research results indicate that the RQ-RLBM-SCN method can accurately identify bearing and gear faults. In 10 repeated experiments, the average accuracy on the multi-sensor dataset for bearings and gears reaches 99.80% and 98.75% respectively which are higher than the original SCN, TSC-SCN, VMD-SCN, SVM and KNN fault diagnosis methods. This approach provides a dynamic method and diagnostic insight for establishing a health monitoring model for rotating machinery. [ABSTRACT FROM AUTHOR]

In recent years, cloud computing applications have facilitated the distribution of heterogeneous, unstructured digitized data among users' social networks of varying opinions. Text processing at a large scale requires high-precision computational techniques, which increases the computational burden. The advent of big data analytics along with Natural Language Processing is a powerful factor in improving the efficiency of processing large-scale text data, using the kernel of the MapReduce big data analytics frameworks that allow parallelization of large computational operations. In this paper, we propose an intelligent sentiment prediction approach based on deep learning, batch, and streaming big data analytics. In fact, our main objective is to take advantages of the powerful tools provided by the distributed platforms, such as, Hadoop and Spark to preprocess streaming data. This involves various tasks such as cleaning the data, reducing its size, minimizing access time, and decreasing storage volume. This step prepares the big streaming data to be fully exploited by Deep Learning models. This work includes a research study on processing big data related to short volume scripts based on batch and streaming distributed frameworks as well as deep learning approaches in Natural Language Processing. We detail our idea for analyzing short texts to determine their semantic context and categorize them into pros and cons poles. There were different stages in building this model, the first involving data reduction and refinement using selected features and big data analysis tools. In the second stage, words are embedded by global vector to be computed in layers of convolutional and recurrent neural networks. The experimental study and the analysis of the results confirm the usefulness of our proposed model and its superiority over the main approaches studied in the literature. Our model achieved a performance of 96% in terms of accuracy. [ABSTRACT FROM AUTHOR]

Physical fitness is one of the most important traits a person could have for health longevity. Conducting regular exercise is fundamental to maintaining physical fitness, but with the caveat of occurring injury if not done properly. Several algorithms exists to automatically monitor and evaluate exercise using the user's pose. However, it is not an easy task to accurately monitor and evaluate exercise poses automatically. Moreover, there are limited number of datasets exists in this area. In our work, we attempt to construct a neural network model that could be used to evaluate exercise poses based on key points extracted from exercise video frames. First, we collected several images consists of different exercise poses. We utilize the the OpenPose library to extract key points from exercise video datasets and LSTM neural network to learn exercise patterns. The result of our experiment has shown that the methods used are quite effective for exercise types of push-up, sit-up, squat, and plank. The neural-network model achieved more than 90% accuracy for the four exercise types. [ABSTRACT FROM AUTHOR]

Skin cancer is a terrifying disorder that affects all individuals. Due to the significant increase in the rate of melanoma skin cancer, early detection of skin cancer is now more critical than ever before. Malignant melanoma is one of the most serious forms of skin cancer, and it is caused by abnormal melanocyte cell growth. In recent years, skin cancer predictive categorization has become more accurate and predictive due to multiple deep learning algorithms. Malignant melanoma is diagnosed using the Recurrent Convolution Neural Network‐Long Short‐Term Memory (RCNN‐LSTM), which is one of the deep learning classification approaches. Using the International Skin Image Collection and the RCNN‐LSTM, the data are categorized and analyzed to gain a better understanding of skin cancer. The method begins with data preprocessing, which prepares the dataset for classification. Additionally, the RCNN is employed to extract the features that are vital to the prediction process. The LSTM is accountable for the final step, classification. There are further factors to examine, such as the precision of 94.60%, the sensitivity of 95.67%, and the F1‐score of 95.13%. Other benefits of the suggested study include shorter prediction durations of 95.314, 122.530, and 131.205 s and lower model loss of 0.25%, 0.19%, and 0.15% for input sizes 10, 15, and 20, respectively. Three datasets had a reduced categorization error of 5.11% and an accuracy of 95.42%. In comparison to previous approaches, the work discussed here produces superior outcomes. Research Highlights: Recurrent convolutional neural network (RCNN) deep learning approach for optimizing time prediction and error classification in early melanoma detection. It extracts a high number of specific features from the skin disease image, making the classification process easier and more accurate.To reduce classification errors in accurately detecting melanoma, context dependency is considered in this work. By accounting for context dependency, the deprivation state is avoided, preventing performance degradation in the model.To minimize melanoma detection model loss, a skin disease image augmentation or regularization process is performed in this work. This strategy improves the accuracy of the model when applied to fresh, previously unobserved data. [ABSTRACT FROM AUTHOR]

As more and more cars are connected to the internet, the threat of cyber-attacks and illegal access to one's automobile increases expeditiously. Illegal access to automobiles via. the car's integrated network system is quite a common phenomenon nowadays. Thus, it is essential to identify drivers on the basis of their driving patterns. Henceforth, this paper presents a driver profiling and identification method based on data acquired from car sensors. In-vehicle sensors generate dozens of operational data streams, and identifying the right representative features for driver profiling is a challenging task. Therefore, in our work to capture human driving behavior dynamics, we have designed a framework based on Long Short Term Memory. Moreover, for extracting relevant and independent features from the Controller Area Network (CAN) dataset, we suggested using feature selection algorithms. The proposed framework is evaluated on the publicly available vehicle CAN OBD-II dataset. While we demonstrate the effectiveness of the proposed architecture, an essential objective of this study is to verify that inter-driver heterogeneity and intra-driver homogeneity can be modeled using time series dependency. [ABSTRACT FROM AUTHOR]

A very crucial branch of Natural Language Processing is Sentiment Analysis, which seeks to elicit feelings in the public from feedback provided by users. This study proposes a Hybridized Deep Neural Network-based framework for the sentiment analysis, where we have modified Dispersive Flies Optimization by adjusting its neighbor counterpart and applied that Neighbour Adjusted Dispersive Flies optimization for optimizing feature space with the aid of sentiment information extracted using our specially developed SentiWordNet lexicon-linked fitness function, after preliminary processing of data. This modification helps to avoid the local optimal solution and supports the optimization process to approach the global optimal solution in more effective way. Next, to handle the textual features efficiently through Deep Learning approaches, we use pre-trained embedding technique to represent them mathematically. The Hybridized Deep Neural Network, which is made up of a Convolutional Neural Network and Long Short Term Memory, is then given the embedded features. In order to store locally implanted information, Convolutional Neural Networks construct hierarchical representations, while Long Short Term Memory attempts to recollect pertinent prior data for opinion categorization. This hybridization helps to take advantage of both the component networks. The deep neural network system ultimately delivers the desired sentiment category. To demonstrate its effectiveness, the suggested hybrid methodology is reckoned and contrasted with numerous cutting-edge methodologies utilizing a variety of performance indicators. Our proposed framework gives the best performance compared to the baselines with an accuracy of 89.0%, 81.9%, 67.9%, 64.6%, 83.2%, 79.8% and 91.3% for Amazon, ETSY, Big Basket, Facebook, Finance, Twitter and Wine dataset respectively. [ABSTRACT FROM AUTHOR]

With technological innovations, enterprises in the real world are managing every iota of data as it can be mined to derive business intelligence (BI). However, when data comes from multiple sources, it may result in duplicate records. As data is given paramount importance, it is also significant to eliminate duplicate entities towards data integration, performance and resource optimization. To realize reliable systems for record deduplication, late, deep learning could offer exciting provisions with a learning-based approach. Deep ER is one of the deep learning-based methods used recently for dealing with the elimination of duplicates in structured data. Using it as a reference model, in this paper, we propose a framework known as Enhanced Deep Learning-based Record Deduplication (EDL-RD) for improving performance further. Towards this end, we exploited a variant of Long Short Term Memory (LSTM) along with various attribute compositions, similarity metrics, and numerical and null value resolution. We proposed an algorithm known as Efficient Learning based Record Deduplication (ELbRD). The algorithm extends the reference model with the aforementioned enhancements. An empirical study has revealed that the proposed framework with extensions outperforms existing methods. [ABSTRACT FROM AUTHOR]

For multi-equipment maintenance of modern production equipment, the economic correlation and degradation uncertainty may lead to insufficient or excessive maintenance, increasing maintenance costs. This paper proposes a dynamic grouping maintenance method based on probabilistic remaining useful life (RUL) prediction for multiple equipment. Long short term memory (LSTM) is developed to predict the equipment probability RUL by the Variational Auto-Encoder (VAE) resampling. Then, the dynamic grouping maintenance model is constructed to minimize the maintenance cost rate under the known probabilistic RUL information. The gazelle optimization algorithm (GOA) is used to determine the optimal maintenance time for each equipment. To better verify the effectiveness of the proposed method, a numerical case with six wind turbines is introduced to analyse the performance of GOA. Moreover, the advantages of dynamic grouping maintenance is verified by comparing with independent maintenance, whose maintenance cost rate is reduced by 10.01%. [ABSTRACT FROM AUTHOR]

Pashtu is one of the most widely spoken languages in south-east Asia. Pashtu Numerics recognition poses challenges due to its cursive nature. Despite this, employing a machine learning-based optical character recognition (OCR) model can be an effective way to tackle this issue. The main aim of the study is to propose an optimized machine learning model which can efficiently identify Pashtu numerics from 0–9. The methodology includes data organizing into different directories each representing labels. After that, the data is preprocessed i.e., images are resized to 32 × 32 images, then they are normalized by dividing their pixel value by 255, and the data is reshaped for model input. The dataset was split in the ratio of 80:20. After this, optimized hyperparameters were selected for LSTM and CNN models with the help of trial-and-error technique. Models were evaluated by accuracy and loss graphs, classification report, and confusion matrix. The results indicate that the proposed LSTM model slightly outperforms the proposed CNN model with a macro-average of precision: 0.9877, recall: 0.9876, F1 score: 0.9876. Both models demonstrate remarkable performance in accurately recognizing Pashtu numerics, achieving an accuracy level of nearly 98%. Notably, the LSTM model exhibits a marginal advantage over the CNN model in this regard. [ABSTRACT FROM AUTHOR]

Stroke is a disease that is caused due to the blockage and burst in the blood vessels of the brain, thus resulting in abrupt brain dysfunction, like sensory or motor disorders, unconsciousness, limb paralysis, and pronunciation disorders. The existing stroke prediction algorithms have some limitations because of the lengthy testing procedures and hefty testing expenses. The main goal of this study is to develop and implement the proposed fusion-based, optimized deep learning model for stroke disease prediction using multimodalities. For that, this research considers the Computed Tomography (CT) and electroencephalogram (EEG) signals as input, and all of these inputs are processed separately to predict the stroke disease. While predicting the stroke disease with a CT image, the bilateral filter performs the pre-processing and the disease prediction is done with the DenseNet model, which is tuned by the proposed Jaya Fractional Reptile Search Algorithm (Jaya FRSA). Similar to how the proposed FRSA does CNN-LSTM training, the Convolutional Neural Network-Long Short Term Memory (CNN-LSTM) predicts the stroke disease using the EEG data as an input after the Gaussian filter removes signal noise. Additionally, the CT image and the EEG signal are processed independently from the image and signal properties. Additionally, the CNN-LSTM model and DenseNet model results are combined using the overlap coefficient to get the final disease prediction. According to the experimental study, the suggested method achieved the maximum image accuracy, sensitivity, and specificity of 0.924, 0.930, and 0.935. [ABSTRACT FROM AUTHOR]

The Industrial Internet of Things is becoming the novel driving force in the automotive industry, assembly travel more suitable for individuals. Despite this, there are still a number of obstacles to overcome, such as detecting illegal drivers, identifying legitimate drivers, and evaluating driving behavior. The use of deep learning networks has been attempted by many academics to address the issues of detecting illicit drivers and identifying legal drivers, however the gathering and analysis of data on driving behavior are still subject to several restrictions. Furthermore, insufficient focus has been placed on examining a driver's behavior. To deal with the aforementioned concerns, we carried out a thorough investigation on driving behavior patterns and constructed an Multi-Task Learning (MTL) based network. In the first place, we gather real-world data from an automobile and analyze it for traits related to driving. After that, a novel MTL network is built utilizing a Long Short Term Memory network to detect unlawful drivers, identify legal drivers, and evaluate driving behavior. To strengthen their argument, the authors could conduct experiments or case studies comparing the performance and efficiency of MTL with single-task learning or other methods. By quantifying these factors, they could provide more concrete evidence to support their claim that MTL offers time and cost savings. [ABSTRACT FROM AUTHOR]

Purpose: Accurate detection of Atrial Fibrillation (AF) has great significance in the field of medical science which can reduce the rate of mortality and morbidity. The present study focuses on Electrocardiography (ECG) signal classification using dimensionality reduction techniques combined with R wave to R wave interval (RR interval) features. Materials and Methods: In the first approach, Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), Independent Component Analysis (ICA), and Probabilistic Principal Component Analysis (PPCA) are performed independently on denoised ECG signal using Discrete Wavelet Transform (DWT) for the classification of ECG signal. In the second approach, the dimensionality reduction techniques combined with RR interval features are used for the classification of ECG signal. Results: Machine Learning (ML) algorithms such as Decision Tree (DT), Support Vector Machine (SVM), and Deep Learning (DL) algorithms such as Long Short Term Memory (LSTM) and Bi-Directional LSTM (BiLSTM) are used for classification purposes. Conclusion: The proposed methodology provided an overall accuracy of 93.65% with PCA and LSTM classifier and an overall accuracy of 99.45% with PCA combined with RR interval features and LSTM classifier. The developed technology has potential applications in many practical solutions.