Deep BiLSTM Research Articles

A hybrid CNN-LSTM approach for intelligent cyber intrusion detection system

As the technology is advancing more and more in the era of increasing digitalization, safeguarding networks from cyber threats is crucial. As cyber-attacks on critical infrastructure are becoming more and more sophisticated, enhancing cyber intrusion detection systems (IDS) is imperative. This paper proposes and evaluates a deep learning-based IDS using the NSL-KDD dataset, a benchmark for intrusion detection. The system pre-processes data with Recursive Feature Elimination (RFE) and a Decision Tree classifier to identify the most significant features, optimizing model performance. Various deep learning models, including ANN, LSTM, BiLSTM, CNN-LSTM, GRU, and BiGRU, have been evaluated. The CNN-LSTM model outperformed the others, with 95 % accuracy, 0.89 recall, and 0.94 f1-score. These results prove the effectiveness of the proposed IDS in accurately distinguishing between malicious and benign network traffic. Future research can explore ensemble techniques like boosting or bagging to further enhance IDS performance.

An Automated Intrusion Detection System in IoT system using Attention based Deep Bidirectional Sparse Auto Encoder model

Nowadays, the Internet of Things (IoT) is a smart network connected to the Internet for transmitting gathered data with verified protocols. Attackers frequently use communication protocol defects as the basis for their attacks. Better protection measures are required since attacks affect the reputations of service providers. Both machine learning (ML) and deep learning (DL) methods have been developed in a number of research works to detect network intrusions. However, the system's security is limited by the rising number of new threats. Critical problems in IoT platforms, cyber-physical systems, wireless networks, and fog computing are caused by such attacks. The development of various cyber-security attacks reinforces the need for a strong intrusion detection system (IDS) in the IoT platform. The proposed study introduced a robust deep-feature learning mechanism for automatically detecting network intruders in the IoT platform. Initially, input data are gathered from the given dataset. Pre-processing helps reduce any noise in the data and improves the data quality using cleaning, outlier removal, and min-max normalization. The proposed Attention-based Deep Bidirectional Sparse Auto Encoder (AD-BiSA) model is the most important feature retrieved using the attention-based deep Bi-LSTM model. The different IoT network threats are categorized using a sparse Autoencoder approach. The chaotic Seagull Optimization (CSGO) algorithm decreases the loss and enhances the weight in the proposed DL technique. The UNSW NB15_IDS and NSL-KDD datasets achieve accuracy rates of 99.71% and 98.97%, respectively, for the proposed technique. The proposed method achieves better performance than existing approaches.

Enhancing Agricultural Commodity Price Forecasting Using Generative Models: A Deep Learning Approach

Predicting stock market prices is a critical yet challenging task in finance. Technical analysis, a widely used methodology in investment theory, involves forecasting price movements by analyzing historical market data. Recently, deep learning has gained prominence for its exceptional ability to process complex data, making it a popular tool for financial applications such as stock prediction, portfolio optimization, financial information analysis, and trade execution strategies. In this study, we propose a novel deep learning architecture that integrates a Generative Adversarial Network (GAN) with a Convolutional Neural Network (CNN) as the discriminator and Gated Recurrent Units (GRU) as the generator. This framework generates distributions of daily stock prices through adversarial learning to predict stock closing prices. Using daily trading data from Ruchi Soya Industries Limited, an empirical analysis was performed across a broad time frame. Results show that the proposed GAN-based architecture significantly outperforms traditional deep learning models, including GRU, LSTM, and Bi-LSTM, in predicting stock closing prices. These findings demonstrate the potential of this novel approach for improving stock price forecasting accuracy.

Novel hawk swarm‐optimized deep learning classification with K‐nearest neighbor based decision making for autonomous vehicle movement controller

SummaryNowadays, intelligent transportation systems pay a lot of attention to autonomous vehicles it is believed that an autonomous vehicle improves mobility, comfort, safety, and energy efficiency. Making decisions is essential for the development of autonomous vehicles since these algorithms must be able to manage dynamic and complex urban crossings. In this research an optimal deep BiLSTM‐GAN classifier to detect the movement of smart vehicles, initially the preprocessing stage is performed to decrease noise in the received data after that the essential regions are next be extracted in the region of interest (ROI) to make the right decision. The extracted data are forwarded to the GAN for road segmentation as well as the optimized deep BiLSTM classifier, which recognizes the traffic sign, simultaneously making it possible to do a modified Hough line‐based maneuver prediction using the segmented information from the roads. Finally, the GAN determines the lane, and the BiLSTM predicts the traffic sign. The K‐nearest neighbor (KNN)‐based autonomous vehicle movement controllers are used to make the decision based on the predicted traffic sign and information about the lane. The proposed HSO algorithm was developed as the outcome of the common fusion of hawk and swarm optimization. Based on lane detecting achievements, at training percentage (TP) 90, the accuracy is 91.75%, Peak signal‐to‐noise ratio (PSNR) is 64.84%, mean square error (MSE) is 28.78, and mean absolute error (MAE) is 20.20, respectively, similarly based on the traffic sign prediction achievements at TP 90, the accuracy is 93.71%, sensitivity is 95.15%, specificity is 93.91%, and MSE is 28.78%, respectively.

Ransomware Classification with Deep Neural Network and Bi-LSTM

Malicious attacks, malware, and ransomware families present essential risks to cybersecurity and may result in significant harm to computer systems, data clusters, networks, and mobile apps across a range of industries. Recently, there has been interest in ransomware classification using DNN and Bi-LSTM. DNN, a subset of machine learning techniques, has been found to improve ransomware detection and classification precision and efficacy. Ransomware has been affecting commercial, public, and governmental organizations' networks and computer systems for more than a decade, enabling new dynamic detection techniques to help DNNs detect ransomware. However, deep neural network-based architectures and DL classifiers (such as DNN, and Bi-LSTM classifiers) will be employed to detect ransomware. These networks may learn to correctly identify and categorize new ransomware incidents by integrating various datasets, including known and unknown ransomware samples. The classification of ransomware detection has been thoroughly investigated, and a model incorporating classic DL techniques with DNN and Bi-LSTM-based architecture will be proposed. A model execution experiment will be carried out to facilitate comparative testing of various approaches. This study focuses on the detection and classification of ransomware using DNN and Bi-LSTM. This study provides the groundwork for future investigations into the issues with ransomware detection. To protect against several ransomware attack types, deep neural networks have become an effective tool for ransomware detection. These networks combine machine learning and deep learning techniques.

A Computational Predictor for Accurate Identification of Tumor Homing Peptides by Integrating Sequential and Deep BiLSTM Features.

Cancer remains a severe illness, and current research indicates that tumor homing peptides (THPs) play an important part in cancer therapy. The identification of THPs can provide crucial insights for drug-discovery and pharmaceutical industries as they allow for tailored medication delivery towards cancer cells. These peptides have a high affinity enabling particular receptors present upon tumor surfaces, allowing for the creation of precision medications that reduce off-target consequences and enhance cancer patient treatment results. Wet-lab techniques are considered essential tools for studying THPs; however, they're labor-extensive and time-consuming, therefore making prediction of THPs a challenging task for the researchers. Computational-techniques, on the other hand, are considered significant tools in identifying THPs according to the sequence data. Despite many strategies have been presented to predict new THP, there is still a need to develop a robust method with higher rates of success. In this paper, we developed a novel framework, THP-DF, for accurately identifying THPs on a large-scale. Firstly, the peptide sequences are encoded through various sequential features. Secondly, each feature is passed to BiLSTM and attention layers to extract simplified deep features. Finally, an ensemble-framework is formed via integrating sequential- and deep features which are fed to a support vector machine which with 10-fold cross-validation to carry to validate the efficiency. The experimental results showed that THP-DF worked better on both [Formula: see text] and [Formula: see text] datasets by achieving accuracy of > 95% which are higher than existing predictors both datasets. This indicates that the proposed predictor could be a beneficial tool to precisely and rapidly identify THPs and will contribute to the cutting-edge cancer treatment strategies and pharmaceuticals.

Diagnostic biomarker discovery from brain EEG data using LSTM, reservoir-SNN, and NeuCube methods in a pilot study comparing epilepsy and migraine

The study introduces a new online spike encoding algorithm for spiking neural networks (SNN) and suggests new methods for learning and identifying diagnostic biomarkers using three prominent deep learning neural network models: deep BiLSTM, reservoir SNN, and NeuCube. EEG data from datasets related to epilepsy, migraine, and healthy subjects are employed. Results reveal that BiLSTM hidden neurons capture biological significance, while reservoir SNN activities and NeuCube spiking dynamics identify EEG channels as diagnostic biomarkers. BiLSTM and reservoir SNN achieve 90 and 85% classification accuracy, while NeuCube achieves 97%, all methods pinpointing potential biomarkers like T6, F7, C4, and F8. The research bears implications for refining online EEG classification, analysis, and early brain state diagnosis, enhancing AI models with interpretability and discovery. The proposed techniques hold promise for streamlined brain-computer interfaces and clinical applications, representing a significant advancement in pattern discovery across the three most popular neural network methods for addressing a crucial problem. Further research is planned to study how early can these diagnostic biomarkers predict an onset of brain states.

Optimized deep autoencoder and BiLSTM for intrusion detection in IoTs-Fog computing

Optimized deep autoencoder and BiLSTM for intrusion detection in IoTs-Fog computing

A novel deep dual self-attention and Bi-LSTM fusion framework for Parkinson’s disease prediction using freezing of gait: a biometric application

A novel deep dual self-attention and Bi-LSTM fusion framework for Parkinson’s disease prediction using freezing of gait: a biometric application

An adaptive search optimizer-based deep Bi-LSTM for emotion recognition using electroencephalogram signal

Emotion is a main element of human beings, and consists of an essential impact on their daily behaviors, comprising communication, learning, etc. Emotion recognition is the main challenging task in the arena of Human-machine communication depending on physical signals, which has developed and expanded a portion of deliberation with the technologies of the brain as well as a computer interface, in which recognizing the emotions utilizing the Electroencephalogram (EEG) signal is developed seriously in improving the interaction of human and computer. In this research, deep learning (DL) enabled emotion recognition with EEG data is presented to progress the precise identification with input datasets such as DEAP and SEED IV. The frequency bands involving the alpha, gamma, delta, and so on are created for EEG signals depending on their activities. Initially, the artifacts are eradicated from the EEG signal, and then the features are extracted from the generated frequency band at the final stage. The foremost purpose of the established Adaptive search optimizer -based deep Bidirectional Long Short-term Memory (ASO-based deep Bi-LSTM) model is to forecast the patient’s emotions from the EEG signals in an accurate manner, which discloses the efficiency in terms of metrics involving the specificity, accuracy, and sensitivity than the prior approaches are evaluated. Thus, the ASO-based deep Bi-LSTM model attains the maximum performance which is evaluated with an accuracy of 97.511%, sensitivity of 99.141%, and a specificity of 99.298 % for the DEAP dataset and SEED IV database concerning the Training Percentage (TP) that demonstrates the efficiency of the developed model compared with other existing techniques.

Chinese Cyberbullying Detection Using XLNet and Deep Bi-LSTM Hybrid Model

The popularization of the internet and the widespread use of smartphones have led to a rapid growth in the number of social media users. While information technology has brought convenience to people, it has also given rise to cyberbullying, which has a serious negative impact. The identity of online users is hidden, and due to the lack of supervision and the imperfections of relevant laws and policies, cyberbullying occurs from time to time, bringing serious mental harm and psychological trauma to the victims. The pre-trained language model BERT (Bidirectional Encoder Representations from Transformers) has achieved good results in the field of natural language processing, which can be used for cyberbullying detection. In this research, we construct a variety of traditional machine learning, deep learning and Chinese pre-trained language models as a baseline, and propose a hybrid model based on a variant of BERT: XLNet, and deep Bi-LSTM for Chinese cyberbullying detection. In addition, real cyber bullying remarks are collected to expand the Chinese offensive language dataset COLDATASET. The performance of the proposed model outperforms all baseline models on this dataset, improving 4.29% compared to SVM—the best performing method in traditional machine learning, 1.49% compared to GRU—the best performing method in deep learning, and 1.13% compared to BERT.

An efficient and optimized cooperative indoor localization for B5G communication

SummaryUltra‐accurate indoor localization/positioning technology is rapidly gaining interest in 5G and beyond (B5G) communications. Achieving 0.1 m level of precise indoor localization is a significant challenge that can be resolved using channel state information (CSI) of wireless signals from nearby access points. The complex indoor propagation issues like shadowing, blockage effects, and multi‐path fading diminish CSI, affecting precise position/location accuracy. Thus, to achieve ultra‐accurate localization, a novel MOIL framework is proposed using a deep learning BiLSTM model and CSI features in an efficient and optimized manner. The proposed MOIL integrates two functional blocks, SecLOP and CordNet. SecLOP conceals multi‐path effects and noise to optimize the input information using sequential responses of signals. CordNet minimizes precise position/location errors by taking all combinations of optimized signal features from the SecLOP block. Extensive simulation experiments were conducted to validate MOIL in different indoor scenarios, such as line‐of‐sight (LOS) and non‐line‐of‐sight (NLOS). Moreover, through a substantial ablation study, the efficiency of the proposed framework is validated/verified. The results outperformed existing works by providing cm‐level accuracy of 1 cm in LOS and 1.5 cm in NLOS scenarios. The proposed work outperformed the literature by 31.3% in LOS and 25.30% in NLOS scenarios.

A Deep Bidirectional LSTM Model Enhanced by Transfer-Learning-Based Feature Extraction for Dynamic Human Activity Recognition

Dynamic human activity recognition (HAR) is a domain of study that is currently receiving considerable attention within the fields of computer vision and pattern recognition. The growing need for artificial-intelligence (AI)-driven systems to evaluate human behaviour and bolster security underscores the timeliness of this research. Despite the strides made by numerous researchers in developing dynamic HAR frameworks utilizing diverse pre-trained architectures for feature extraction and classification, persisting challenges include suboptimal performance accuracy and the computational intricacies inherent in existing systems. These challenges arise due to the vast video-based datasets and the inherent similarity in the data. To address these challenges, we propose an innovative, dynamic HAR technique employing a deep-learning-based, deep bidirectional long short-term memory (Deep BiLSTM) model facilitated by a pre-trained transfer-learning-based feature-extraction approach. Our approach begins with the utilization of Convolutional Neural Network (CNN) models, specifically MobileNetV2, for extracting deep-level features from video frames. Subsequently, these features are fed into an optimized deep bidirectional long short-term memory (Deep BiLSTM) network to discern dependencies and process data, enabling optimal predictions. During the testing phase, an iterative fine-tuning procedure is introduced to update the high parameters of the trained model, ensuring adaptability to varying scenarios. The proposed model’s efficacy was rigorously evaluated using three benchmark datasets, namely UCF11, UCF Sport, and JHMDB, achieving notable accuracies of 99.20%, 93.3%, and 76.30%, respectively. This high-performance accuracy substantiates the superiority of our proposed model, signaling a promising advancement in the domain of activity recognition.

Deep-MAD: Deep learning based multiple attack detection for secure device- to-device communication in FOG environment1

An exponential growth of users demands ubiquitous connectivity, which requires the integration of new technology. Therefore, Device to Device (D2D) communication has been considered a promising technology that utilizes effective and efficient communication. Even though numerous studies have been conducted for establishing secure D2D communication, however, existing techniques face challenges like privacy, security threats, and poor generality. To overcome these issues a novel Deep-MAD model is proposed to preserve data privacy along with its access control in the D2D network and multiple attack detection in a fog environment. A Fully Homomorphic Elliptic Curve Cryptography (FHECC) is introduced to transmit data securely in a D2D network. The data owner uses FHECC algorithm to encrypt the plain text into cipher text before storing it on the fog. Whenever the user requests data from the fog, the fog service provider confirm the user’s access control. Furthermore, the deep learning-based Bi-LSTM is used to differentiate the device as an authorized or unauthorized user. If the IP address is genuine then the inverse FHECC is used to decrypt the data for authorized users. Otherwise, the particular device is blocked and it is sent for further verification for classifying the types of attacks. The effectiveness of the proposed strategy is examined using several parameters, such as computational complexity, scalability, accuracy, and Execution time. The proposed technique improves the overall computational overhead of 31.77, 9.34, and 4.67 better than AKA protocol, lightweight cipher, and FHEEP respectively.

Denoising sparker seismic data with Deep BiLSTM in fractional Fourier transform

Sparkers are widely-used sources of high-resolution images for shallow sub-bottom seismic monitoring. Sparker seismic data contain high random noise levels caused by the high voltage and electrical current required to power the sparker source. Random noise is distributed over the entire frequency and wavenumber, and is thereby difficult to suppress. Numerous studies have been conducted to develop random noise suppression by using convolutional neural networks (CNNs). However, because CNNs learn common characteristics of random noise without considering the sequence while training, learning the sequence relationship between useful signal and random noise in time-series data is difficult. As an alternative, we proposed a recurrent neural network (RNN)-based denoising method that considers seismic data sequences. Effective denoising of random noise using RNNs requires features that distinguish between useful signals and random noise during network training. We proposed a denoising method for sparker seismic data that uses bidirectional long short-term memory RNN in fractional Fourier transform (FrFT). The time-frequency distribution in the FrFT domain was utilized for network training, and useful signals were effectively separated from random noise. The proposed method was applied to multi-channel sparker seismic data acquired in Yeongil, Pohang (offshore eastern Korea) by the Korea Maritime & Ocean University. Numerical results confirmed that the proposed method effectively suppressed random noise while maintaining a useful signal in sparker seismic data.

A RUL prediction method of rolling bearings based on degradation detection and deep BiLSTM

<abstract> <p>In order to grasp the degradation of rolling bearings and prevent the failure of mechanical equipment, a remaining useful life (RUL) prediction method of rolling bearings based on degradation detection and deep bidirectional long short-term memory networks (BiLSTM) was proposed, considering the incomplete degradation feature extraction and low prediction accuracy of existing methods. By extracting the characteristics of time domain, frequency domain, and time-frequency domain of the full-life bearing vibration signal, the monotonicity, trend, and robustness measurement indexes of each feature were calculated. The best feature set that can fully reflect the degradation information was constructed by ranking the weighted comprehensive indexes of the features. A degradation detection strategy was used to determine the degradation starting time for setting piecewise linear network label. The RUL prediction model based on deep BiLSTM was established and optimized through Dropout technology and piecewise learning rate. The model was verified by the full-life data set of rolling bearings. The results showed that compared with the support vector machine (SVM), the traditional recurrent neural network (RNN), the single-layer BiLSTM, and long short-term memory networks (LSTM) model without Dropout, the proposed method fitted the degradation trend best, and the root mean square error (RMSE) was the smallest and only 0.0281, which improved the accuracy of RUL prediction of rolling bearings, helped prevent bearing failure, and ensured the safe and reliable operation of rotating machinery.</p> </abstract>

Recognition of Wh-Question Sign Gestures in Video Streams using an Attention Driven C3D-BiLSTM Network

Human sign language is a visual and gestural means of communication used by people with hearing impairments to interact with others. It has the potential to enable interaction between individuals who struggle with verbal communication and those who lack the essential proficiency to comprehend sign language. Wh-questions constitute a substantial portion of daily sign language interactions. Automatic recognition of Wh-question words in sign language, encompassing interrogative words, has received limited attention in gesture recognition. In this context, a novel dataset named the American Question Sign Video dataset (AQSVd) has been introduced as a significant contribution. Recognizing Wh-question signs in video streams empowers individuals to effectively convey messages through hand movements and gestures, fostering inclusivity and accessibility in communication for deaf and hearing-impaired individuals. The paper proposes a Deep Convolutional-3D BiLSTM Multi-head Attention network for recognizing American Wh-Question word sign gestures in video streams. Incorporating Multi-head attention in the proposed model enhances its ability to capture intricate spatial and temporal features, essential for accurate gesture recognition. The proposed model is thoroughly evaluated across a range of datasets, such as INCLUDE50, WLASL-100, and our AQSVd dataset, achieving remarkable results. Specifically, our model demonstrated an outstanding 98.91% validation accuracy on the AQSVd dataset. The C3D-BiLSTM MHAttention model outperforms other state-of-the-art models, demonstrating its superiority in sign language recognition tasks. The proposed dataset and C3D-BiLSTM Multi-head Attention model contribute significantly to this field, offering potential benefits in education, human-robot interaction, and overall communication for the hearing-impaired community.

Arabic Idioms Detection by Utilizing Deep Learning and Transformer-based Models

Arabic language resources and natural language processing technologies have seen significant advancements in recent years. The detection of idiomatic expressions is a crucial problem in Arabic natural language processing. Unfortunately, there have been few advancements in this field due to the scarcity of datasets and computational models that still struggle to handle Arabic idiomatic phrases. Although idioms have significance in Arabic because of their distinctive relation to history and cultural knowledge, they also offer a language tool for conveying concepts and thoughts that may diverge from their literal interpretation. Hence, to accurately detect their incongruent language, a model requires the ability to interpret the idioms’ contextual meaning. To tackle this issue, this study entails comprehensive work on detecting Arabic idiomatic and literal statements. For that, we introduce a large Arabic idiomatic and literal statements. The dataset includes idioms from several Arabic idiomatic platforms balanced with literal statements from titles in news platforms with similar keywords found in the idioms. This study adopts four deep learning models, namely, CNN, LSTM, Bi-LSTM, and GRU trained with word embeddings, FastText and Skip-Gram Word2Vec, and three transformer-based models, BERT, RoBERTa, and DistilBERT were compiled and assessed via empirical evaluations. Our experiments demonstrated that transformer-based models outperformed deep learning models trained by word embedding to detect Arabic idioms and literal statements, reaching 97% in favor of DistilBERT. We also enhanced the accuracy through the ensemble stacking method, boosting it by 0.8 points. Furthermore, we propose a novel Arabic idiom interpreter that offers simple explanations for selected idiomatic contents within the literal text. Our work enriches the challenge of detecting incongruities found in similar figurative genres, specifically in the rich morphological language, Arabic.

Exploring Sequential Feature Selection in Deep Bi-LSTM Models for Speech Emotion Recognition

Exploring Sequential Feature Selection in Deep Bi-LSTM Models for Speech Emotion Recognition

Adaptive Ensembled Fusion Based Deep CNN-Bilstm Model For Heart Disease Prediction In IoT

Internet-of-Things (IoT)-based heart disease prediction is a complex task and processing the real collected data directly for remote patient monitoring suffers from the limitations due to the irrelevant data features, affecting the prediction accuracy and raising the security concerns. Hence, the efficient Adaptive ensembled deep Convolution neural network –Bidirectional Long Short Term Memory (Adaptive ensembled deep CNN-BiLSTM ) classifier model is proposed via the fusion of interactive hunt-based CNN and Whale on Marine optimization (WoM)-based deep BiLSTM. The Adaptive optimization developed from the standard hybrid characteristics such as random searching, seeking, attack prohibition, following, and waiting characteristics optimized the fusion parameters of the developed classifier for attaining high detection accuracy. Additionally, the modified Elliptic Curve Cryptography (ECC) based Diffi-Huffman encryption algorithm provides the authentication and security of sensitive patient data in heart disease prediction. The developed model is evaluated with other competent methods in terms of accuracy, sensitivity, specificity as well as F-measure, which are reported as 97.573%, 98.012%, 97.592%, and 97.705% respectively.