Sequential Data Processing Research Articles

LipNet: End-to-End Lipreading

Lipreading is the task of decoding text from the movement of a speaker’s mouth. This research presents the development of an advanced end-to-end lipreading system. Leveraging deep learning architectures and multimodal fusion techniques, the proposed system interprets spoken language solely from visual cues, such as lip movements. Through meticulous data collection, annotation, preprocessing, model development, and evaluation, diverse datasets encompassing various speakers, accents, languages, and environmental conditions are curated to ensure robustness and generalization. Conventional methods divided the task into two phases: prediction and designing or learning visual characteristics. Most deep lipreading methods are trainable from end to end. In the past, lipreading has been tackled using tedious and sometimes unsatisfactory techniques that break down speech into smaller units like phonemes or visemes. But these methods often fail when faced with real-world problems, such contextual factors, accents, and differences in speech patterns. Nevertheless, current research on end-to-end trained models only carries out word classification; sentence-level sequence prediction is not included. LipNet is an end-to-end trained model that uses spatiotemporal convolutions, a recurrent network, and the connectionist temporal classification loss to translate a variable-length sequence of video frames to text. LipNet breaks from this traditional paradigm by using an all-encompassing, end-to-end approach supported by deep learning algorithms, Convolutional neural networks (CNNs) and recurrent neural networks (RNNs), which are skilled at processing sequential data and extracting high-level representations, are fundamental to LipNet's architecture.LipNet achieves 95.2% accuracy in sentence-level on the GRID corpus, overlapped speaker split task, outperforming experienced human lipreaders and the previous 86.4% word-level state-of-the-art accuracy.The results underscore the transformative potential of the lipreading system in real-world applications, particularly in domains such as assistive technology and human-computer interaction, where it can significantly improve communication accessibility and inclusivity for individuals with hearing impairments.

Evolutionary Deep Learning for Sequential Data Processing in Music Education

Evolutionary Deep Learning for Sequential Data Processing in Music Education

Task offloading with enhanced Deep Q-Networks for efficient industrial intelligent video analysis in edge–cloud collaboration

With the rapid development of 5G and IoT technologies, industrial video surveillance systems face increasingly significant challenges in processing large-scale data. Traditional cloud-centric video analysis methods often lead to processing delays and inefficiencies in resource allocation due to centralized data processing. In this context, the emergence of edge computing, especially in edge–cloud collaborative intelligent video analysis, offers an innovative solution. Within this architecture, task offloading strategies focus on intelligent computation task distribution to minimize latency and optimize resource utilization. However, existing methods frequently overlook key application-specific metrics in video analysis, such as accuracy and timeliness. This study introduces a novel Edge–Cloud Collaborative Industrial Video Intelligent Analysis Architecture (ECIVA) based on the GRU-Enhanced Deep Q-Network (GEDQN). This architecture integrates the strengths of Gated Recurrent Units (GRU) for processing sequential data with the efficiency of Deep Q-Networks in optimizing task-offloading decisions. By incorporating a token bucket mechanism, the architecture optimally manages the task offloading rate within the constraints of limited network bandwidth, thus ensuring high accuracy and real-time responsiveness in video analysis. Empirical evaluation with offshore drilling platform video data shows that the architecture achieves 87.5% mean Average Precision (mAP) at a 0.5 IoU threshold and increases system throughput by 39.2% over traditional cloud-centric methods.

Deep Learning-Driven E-Commerce Marketing Communication for Recommending Shopping System and Optimizing User Experience

As competition in the realm of e-commerce escalates, the provision of personalized and precise shopping recommendations emerges as a pivotal strategy for e-commerce platforms striving to engage users effectively. Traditional recommendation systems often grapple with challenges such as the inability to capture intricate relationships, limited personalization, and issues concerning diversity. In response to these challenges, this study introduces cutting-edge deep learning techniques, namely Transformer models, Generative Adversarial Networks (GANs), and reinforcement learning, with the aim of bolstering the recommendation accuracy and user experience within e-commerce shopping systems.Initially, we harness Transformer models, capitalizing on their exceptional performance in processing sequential data to adeptly extract and learn representations of both product and user features. This facilitates a more profound understanding of the correlations between products and user shopping behaviors, thus empowering the system to offer more tailored recommendations.

Application of Music Retrieval & Generation

The research explores music generation through LSTM and VAEs neural network architectures, leveraging MIDI representations. LSTM specializes in sequential data processing, while VAEs compress music datasets into low-dimensional representations. Optimization of models is pursued by analyzing the relationship between training loss and epochs. Ultimately, a comparison between LSTM and VAEs determines the most effective algorithm for music generation. Traditional methods face issues like vanishing and exploding gradients, prompting exploration of deep learning approaches. The study aims to advance machine manipulation of music, facilitating new composition generation from existing MIDI files. Keywords: Music generation, LSTM, Variational Autoencoders (VAEs), MIDI representation, Sequential data, Neural network architectures, Training loss, Optimization, Deep learning, Vanishing gradient, Exploding gradient

Short-term wind speed interval prediction using improved quality-driven loss based gated multi-scale convolutional sequence model

Efficient estimation of the uncertainty associated with wind speed forecast is crucial for evaluating wind farms' power quality and operation. Typically, the performance of prediction interval (PI) generating models; is restricted in terms of computational efficiency inheritably from the sequential data processing; whereas restrictions in forecast quality are derived from the PI generation techniques. This paper presents an efficient Gated Multi-Scale Convolutional Sequence Model (GMSCSM) to forecast wind speed PIs. GMSCSM while conserving the ‘recurrence’ of LSTMs also offers ‘parallel input’ advantage of CNNs for better computational efficiency which is highly desirable for short-term forecasting models. In addition, capturing features at various scales in the sequence, GMSCSM learns both local details and global context which is vital for multi-horizon forecasts. The model generates quality PIs utilizing an improved quality-driven loss which we proposed by invoking calibration assessment in its existing definition. Forecasts generated for eight different datasets obtained from two different wind farms show an improvement of 30 % and 6 % in the average coverage width criterion index while reducing the model training time to nearly half and one third of the respective values obtained from traditional and LSTM based models which showcases the proposed model's excellent prediction capability and efficiency.

Cascade Residual Multiscale Convolution and Mamba-Structured UNet for Advanced Brain Tumor Image Segmentation

In brain imaging segmentation, precise tumor delineation is crucial for diagnosis and treatment planning. Traditional approaches include convolutional neural networks (CNNs), which struggle with processing sequential data, and transformer models that face limitations in maintaining computational efficiency with large-scale data. This study introduces MambaBTS: a model that synergizes the strengths of CNNs and transformers, is inspired by the Mamba architecture, and integrates cascade residual multi-scale convolutional kernels. The model employs a mixed loss function that blends dice loss with cross-entropy to refine segmentation accuracy effectively. This novel approach reduces computational complexity, enhances the receptive field, and demonstrates superior performance for accurately segmenting brain tumors in MRI images. Experiments on the MICCAI BraTS 2019 dataset show that MambaBTS achieves dice coefficients of 0.8450 for the whole tumor (WT), 0.8606 for the tumor core (TC), and 0.7796 for the enhancing tumor (ET) and outperforms existing models in terms of accuracy, computational efficiency, and parameter efficiency. These results underscore the model’s potential to offer a balanced, efficient, and effective segmentation method, overcoming the constraints of existing models and promising significant improvements in clinical diagnostics and planning.

DEVELOPMENT AND TRAINING OF LSTM MODELS FOR CONTROL OF VIRTUAL DISTRIBUTED SYSTEMS USING TENSORFLOW AND KERAS

DEVELOPMENT AND TRAINING OF LSTM MODELS FOR CONTROL OF VIRTUAL DISTRIBUTED SYSTEMS USING TENSORFLOW AND KERAS

Survey on Predicting the Risk of Heart Attack Through Retinal Eye Images Analysis

Cardiovascular diseases (CVDs) remain a leading cause of global morbidity and mortality. Early detection and intervention are crucial for improving patient outcomes and reducing the burden on healthcare systems. Recent research suggests a potential link between retinal vascular changes and cardiovascular health. Retinal images offer a non-invasive means to assess microvascular abnormalities, making them an attractive source of data for predictive modeling. This project focuses on developing a machine learning model, specifically using Recurrent Neural Networks (RNNs), to analyze retinal images and detect patterns indicative of heart diseases. RNNs are well-suited for processing sequential data, making them suitable for capturing temporal dependencies in the retinal images and improving the predictive accuracy of the model.

Unveiling Emotional Stress: Analyzing EEG Signals With Deep Learning Insights

This study focuses on a comprehensive exploration of EEG-based emotion recognition, with a particular focus on leveraging recurrent neural network (RNN) architectures, including long short-term memory (LSTM) and gated recurrent unit (GRU). The investigation is rooted in a publicly available EEG Brainwave Dataset, meticulously curated to capture human responses to emotional stimuli. The dataset features signals obtained from frontal and temporal brain lobes, categorized into distinct emotional states encompassing positive, neutral, and negative emotions. The analysis spans four distinct scenarios, each representing different combinations of features and preprocessing strategies. Notably primary emphasis is on LSTM and GRU architectures due to their proven effectiveness in sequential data processing tasks. In addition to evaluating the performance of LSTM and GRU models a thorough examination of all the four cases to assess their efficacy in accurately predicting emotional states from EEG signals. The study also emphasis on the importance of meticulous feature selection and preprocessing in optimizing model accuracy and robustness, highlighting the intricate interplay between data representation, model architecture, and task-specific requirements. Ultimately, the research findings contribute to advanced understanding of EEG-based emotion recognition and pave the way for further research in this exciting interdisciplinary field. Key Words: Electroencephalography(EEG), Long short-term memory (LSTM), Gated recurrent unit (GRU), Feature extraction, Preprocessing techniques, Statistical metrics, Frequency domain analysis.

Deep Learning in Cybersecurity: A Hybrid BERT–LSTM Network for SQL Injection Attack Detection

In the past decade, cybersecurity has become increasingly significant, driven largely by the increase in cybersecurity threats. Among these threats, SQL injection attacks stand out as a particularly common method of cyber attack. Traditional methods for detecting these attacks mainly rely on manually defined features, making these detection outcomes highly dependent on the precision of feature extraction. Unfortunately, these approaches struggle to adapt to the increasingly sophisticated nature of these attack techniques, thereby necessitating the development of more robust detection strategies. This paper presents a novel deep learning framework that integrates Bidirectional Encoder Representations from Transformers (BERT) and Long Short-Term Memory (LSTM) networks, enhancing the detection of SQL injection attacks. Leveraging the advanced contextual encoding capabilities of BERT and the sequential data processing ability of LSTM networks, the proposed model dynamically extracts word and sentence-level features, subsequently generating embedding vectors that effectively identify malicious SQL query patterns. Experimental results indicate that our method achieves accuracy, precision, recall, and F1 scores of 0.973, 0.963, 0.962, and 0.958, respectively, while ensuring high computational efficiency.

Emotion Detection in Text: Advances in Sentiment Analysis Using Deep Learning

In the modern era of digital communication, the analysis of sentiment has emerged as a crucial tool for understanding and inferring public sentiment as communicated through written text. This is particularly relevant in the context of social media platforms such as Twitter, Facebook and Instagram. The present study focuses on the urgent matter of public opinion regarding the practice of animal testing, employing advanced deep-learning methodologies for sentiment analysis. A dataset of 15,360 tweets about animal testing was collected using the Twitter API. The data was prepared for analysis by undergoing careful preprocessing and word embedding it through the utilization of Word2vec. To classify tweets into positive and negative sentiment categories, a Long Short-Term Memory (LSTM) model was employed, given its suitability for processing sequential data. Remarkably, an accuracy rate of 88.7 percent was achieved by the model. It was determined that around 80% of tweets expressed criticism towards animal testing, indicating the presence of a substantial negative sentiment majority. These results show the topic's continuing significance by emphasizing its highly emotional and controversial nature. It is concluded that deep learning, and in particular LSTM models, can be used to effectively analyze large amounts of social media data and yield insightful understandings of public opinion. This study underlines the significance of sentiment analysis for gaining insight into public opinion and for its applications in policymaking and discourse analysis.

Universal approximation property of stochastic configuration networks for time series

For the purpose of processing sequential data, such as time series, and addressing the challenge of manually tuning the architecture of traditional recurrent neural networks (RNNs), this paper introduces a novel approach-the Recurrent Stochastic Configuration Network (RSCN). This network is constructed based on the random incremental algorithm of stochastic configuration networks. Leveraging the foundational structure of recurrent neural networks, our learning model commences with a modest-scale recurrent neural network featuring a single hidden layer and a solitary hidden node. Subsequently, the node parameters of the hidden layer undergo incremental augmentation through a random configuration process, with corresponding weights assigned structurally. This iterative expansion continues until the network satisfies predefined termination criteria. Noteworthy is the adaptability of this algorithm to handle time series data, exhibiting superior performance compared to traditional recurrent neural networks with similar architectures. The experimental results presented in this paper underscore the efficacy of the proposed RSCN for sequence data processing, showcasing its advantages over conventional recurrent neural networks in the context of the performed experiments.

Learning fused lasso parameters in portfolio selection via neural networks

In recent years, fused lasso models are becoming popular in several fields, such as computer vision, classification and finance. In portfolio selection, they can be used to penalize active positions and portfolio turnover. Despite efficient algorithms and software for solving non-smooth optimization problems have been developed, the amount of regularization to apply is a critical issue, especially if we have to achieve a financial aim. We propose a data-driven approach for learning the regularization parameters in a fused lasso formulation of the multi-period portfolio selection problem, able to realize a given financial target. We design a neural network architecture based on recurrent networks for learning the functional dependence between the regularization parameters and the input data. In particular, the Long Short-Term Memory networks are considered for their ability to process sequential data, such as the time series of the asset returns. Numerical experiments performed on market data show the effectiveness of our approach.

Next word prediction for Urdu language using deep learning models

Deep learning models are being used for natural language processing. Despite their success, these models have been employed for only a few languages. Pretrained models also exist but they are mostly available for the English language. Low resource languages like Urdu are not able to benefit from these pre-trained deep learning models and their effectiveness in Urdu language processing remains a question. This paper investigates the usefulness of deep learning models for the next word prediction and suggestion model for Urdu. For this purpose, this study considers and proposes two word prediction models for Urdu. Firstly, we propose to use LSTM for neural language modeling of Urdu. LSTMs are a popular approach for language modeling due to their ability to process sequential data. Secondly, we employ BERT which was specifically designed for natural language modeling. We train BERT from scratch using an Urdu corpus consisting of 1.1 million sentences thus paving the way for further studies in the Urdu language. We achieved an accuracy of 52.4% with LSTM and 73.7% with BERT. Our proposed BERT model outperformed two other pre-trained BERT models developed for Urdu. Since this is a multi-class problem and the number of classes is equal to the vocabulary size, this accuracy is still promising. Based on the present performance, BERT seems to be effective for the Urdu language, and this paper lays the groundwork for future studies.

Generalisation of Feed-Forward Neural Networks and Recurrent Neural Networks

This paper presents an in-depth analysis of Feed-Forward Neural Networks (FNNs) and Recurrent Neural Networks (RNNs), two powerful models in the field of artificial intelligence. Understanding these models and their applications is crucial for harnessing their potential. The study addresses the need to comprehend the unique characteristics and architectures of FNNs and RNNs. These models excel at processing sequential and temporal data, making them indispensable in tasks. Furthermore, the paper emphasises the importance of variables in FNNs and proposes a novel method to rank the importance of independent variables in predicting the output variable. By understanding the relationship between inputs and outputs, valuable insights can be gained into the underlying patterns and mechanisms driving the system being modelled. Additionally, the research explores the impact of initial weights on model performance. Contrary to conventional beliefs, the study provides evidence that neural networks with random weights can achieve competitive performance, particularly in situations with limited training datasets. This finding challenges the traditional notion that careful initialization is necessary for neural networks to perform well. In summary, this paper provides a comprehensive analysis of FNNs and RNNs while highlighting the importance of understanding the relationship between variables and the impact of initial weights on model performance. By shedding light on these crucial aspects, this research contributes to the advancement and effective utilisation of neural networks, paving the way for improved predictions and insights in various domains.

Unveiling the Power of Deep Learning: Insights into Advanced Neural n Network Architectures

Deep learning has emerged as a powerful paradigm within artificial intelligence, revolutionizing various fields with its ability to automatically discover intricate patterns and representations from vast amounts of data. In this article, we delve into the latest advancements in deep learning, focusing on the exploration of advanced neural network architectures. Through a comprehensive review and analysis, we uncover the underlying principles and methodologies driving the success of these architectures, elucidating their capabilities and limitations. From convolutional neural networks (CNNs) for image recognition to recurrent neural networks (RNNs) for sequential data processing, we highlight key innovations and breakthroughs that have propelled the field forward. Furthermore, we discuss recent trends such as attention mechanisms, transformer models, and graph neural networks, showcasing their potential applications and impact across diverse domains. By unraveling the power of deep learning and its advanced neural network architectures, this article aims to provide valuable insights into the state-of-the-art techniques driving AI research and development.

Deep-Learning-based Cryptanalysis through Topic Modeling

Neural cryptography is a technique that uses neural networks for secure data encryption. Cryptoanalysis, on the other hand, deals with analyzing and decrypting ciphers, codes, and encrypted text without using a real key. Chosen-plaintext cryptanalysis is a subfield of cryptanalysis where both plain text and ciphertext are available and the goal is either to find the encryption technique, the encryption key, or both. This study addresses chosen plaintext cryptanalysis within public key cryptography, to categorize topics of encrypted text. Using a fixed encryption technique and key, the focus was placed on creating a framework that identifies the topic associated with ciphertext, using diverse plaintexts and their corresponding cipher texts. To our knowledge, this is the first time that chosen-plaintext cryptanalysis has been discussed in the context of topic modeling. The paper used deep learning techniques such as CNNs, GRUs, and LSTMs to process sequential data. The proposed framework achieved up to 67% precision, 99% recall, 80% F1-score, and 71% AUPR on a dataset, showcasing promising results and opening avenues for further research in this cryptanalysis subarea.

Handwritten digit recognition based on deep learning techniques

The identification of handwritten digits in images recognition and machine learning is a prominent research area. In order to create a handwritten digit recognition model for this investigation, deep learning is introduced. The proposed approach integrates Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and EfficientNetB0, three separate deep learning models. Specifically, the CNN model utilizes pooling layers and data augmentation techniques to enhance its classification ability, the RNN model takes advantage of its ability to process sequential data, and the EfficientNetB0 model benefits from a deeper and more complex network structure. These models are trained and evaluated using the Modified National Institute of Standards and Technology (MNIST) dataset. The experimental results demonstrate the efficacy of the proposed approach: the CNN model attains a remarkable accuracy of 98.9% on the test set, thereby showcasing its exceptional classification performance. Similarly, the RNN model achieves an accuracy of 96.7%, underscoring its suitability for analyzing sequential data. Furthermore, the EfficientNetB0 model attains an accuracy of 98.1%, thereby elucidating the benefits of the deeper network architecture. The models constructed in this study have significant real-world implications, such as improved object recognition systems, medical diagnostics, and autonomous driving. The EfficientNetB0 model produced an accuracy of 98.1% with its complex network architecture when applied to recognise handwritten digits.

Recurrent Neural Networks and Three-Point Bending Test on the Identification of Material Hardening Parameters

The continuous evolution of metallic alloys in the automotive industry has led to the development of more advanced and flexible constitutive models that attempt to accurately describe the various fundamental properties and behavior of these materials. These models have become increasingly complex, incorporating a larger number of parameters that require an accurate calibration procedure to fit the constitutive parameters with experimental data. In this context, machine learning (ML) methodologies have the potential to advance material constitutive modeling, enhancing the efficiency of the material parameter calibration procedure. Recurrent neural networks (RNNs) stand out among various learning algorithms due to their ability to process sequential data and overcome limitations imposed by nonlinearities and multiple parameters involved in phenomenological models. This study explores the modeling capabilities of long short-term memory (LSTM) structures, a type of RNN, in predicting the hardening behavior of a sheet metal material using the results of a standardized experimental three-point bending test, with the aim of extending this methodology to other experimental tests and constitutive models. Additionally, a variable analysis is performed to select the most important variables for this experimental test and assess the influence of friction, material thickness, and elastic and plastic properties on the accuracy of predictions made by neural networks. The required data for designing and training the network solutions are collected from numerical simulations using finite element methodology (FEM), which are subsequently validated by experiments. The results demonstrate that the proposed LSTM-based approach outperforms traditional identification techniques in predicting the material hardening parameters. This suggests that the developed procedure can be effectively applied to efficiently characterize different materials, especially those extensively used in industrial applications, ranging from mild steels to advanced high-strength steels.