Convolutional Recurrent Neural Network Research Articles

Joint Inversion of Surface-Wave Dispersions and Receiver Functions Based on Deep Learning

Abstract We proposed a deep learning (DL) method to derive VS models from joint inversion of Rayleigh-wave dispersions and receiver functions, which is based on multilabel convolutional neural network and recurrent neural network. We used a spline-based approach to generate synthetic models instead of directly using existing models to build the training data set, which improves the generalization of the method. Unlike the traditional methods, which usually set a fixed VP/VS ratio, our method makes full use of the powerful data mining ability of DL to invert the VS models assuming different VP/VS ratios. A loss function is specially designed that focuses on key features of the model space, for example, the shape and depth of Moho. Synthetic tests demonstrate that the proposed method is accurate and fast. Application to the southeast margin of the Tibetan Plateau shows results consistent with the previous joint inversion with P constraints, indicating the proposed method is reliable and robust.

Forecasting Citywide Crowd Transition Process via Convolutional Recurrent Neural Networks

Forecasting Citywide Crowd Transition Process via Convolutional Recurrent Neural Networks

Prediction of Parking Space Availability Using Improved MAT-LSTM Network

The prediction of parking space availability plays a crucial role in information systems providing parking guidance. However, controversy persists regarding the efficiency and accuracy of mainstream time series prediction methods, such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs). In this study, a comparison was made between a temporal convolutional network (TCN) based on CNNs and a long short-term memory (LSTM) network based on RNNs to determine an appropriate baseline for predicting parking space availability. Subsequently, a multi-head attention (MAT) mechanism was incorporated into an LSTM network, attempting to improve its accuracy. Experiments were conducted on three real and two synthetic datasets. The results indicated that the TCN achieved the fastest convergence, whereas the MAT-LSTM method provided the highest average accuracy, namely 0.0330 and 1.102 × 10−6, on the real and synthetic datasets, respectively. Furthermore, the improved MAT-LSTM model accomplished an increase of up to 48% in accuracy compared with the classic LSTM model. Consequently, we concluded that RNN-based networks are better suited for predicting long-time series. In particular, the MAT-LSTM method proposed in this study holds higher application value for predicting parking space availability with a higher accuracy.

Discerning Deception: A Face-Centric Deepfake Detection Approach with ResNeXt-50 and LSTMs

Abstract: A.I. has grown to epidemic proportions over the last years as its applied in almost all sectors to allocate workload from humans but end up being done effectively with no human intervention. A branch of A.I. called deep learning, which operates by mimicking human judgment and action through neural network systems. Nonetheless, with the increase height of the two platforms have been experienced sufficient cases of misguided individuals using tools to recycle videos, audios, and texts to achieve their agendas. This insinuates a due assumption that Generative Adversarial Networks, GANs, are central to the development of believable deepfakes. GANs have developed a crucial ability to generate videos that replace frames with material from another video source to create deepfakes videos. While GANs serves various purposes such as entertainment, teaching, and experimentation, malicious actors can misuse these deep learning techniques to manipulate videos, impacting the privacy of individuals in society. This paper conducts an analysis of different deepfake detection models, comparing their efficacy and discussing potential future extensions of deepfake technology. The study presents a novel deepfake detection approach utilizing a combination of Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN). This method utilizes ResNext50 for extracting features at the frame level, while employing LSTM (Long Short-Term Memory) for video classification based on these extracted features. Various datasets are incorporated, including the deepfake detection challenge dataset (DFDC) and Face Forensics deepfake collections (FF++), combining them to achieve a high-accuracy model capable of accurately discerning between real and deepfake videos. The results of this study make a valuable contribution to the continuous endeavors aimed at improving deepfake detection abilities and ensuring privacy protection in a time heavily influenced by artificial intelligence

Comparative Analysis Of Machine Learning And Deep Learning Models Integrated With Altman Z-Score For Financial Distress Prediction In Companies Listed On The Indonesia Stock Exchange (IDX)

Financial distress is a condition in which a company experiences a decline in its financial situation, which is typically temporary. However, financial distress can worsen if not promptly addressed, leading to bankruptcy. Early identification of potential financial distress in a company is crucial for stakeholders such as investors, creditors, and regulators. In practice, predicting financial distress in a company is not an easy task. One of the methods commonly used to identify early symptoms of financial distress is the method introduced by Altman in 1986. Altman's research model, known as the Z-Score, determines a value based on standard calculations of financial ratios to indicate the likelihood of a company experiencing bankruptcy. The use of artificial intelligence, such as deep learning, can enhance the scope of research on the analysis and prediction of financial distress. This study aims to conduct a comparative analysis of machine learning models, such as the K-Nearest Neighbor Classifier and Random Forest Classifier, and deep learning models, such as Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN). This analysis is carried out to evaluate the success rate of the proposed deep learning models in predicting financial distress in companies operating in Indonesia. Based on the conducted research, the K-Nearest Neighbor Classifier algorithm achieved an accuracy of 77.42% during testing and 88.89% during validation, the Random Forest Classifier algorithm achieved an accuracy of 87.09% during testing and 95.24% during validation, the CNN model achieved an accuracy of 95.16% during testing and 96.83% during validation, and the RNN model achieved an accuracy of 93.55% during testing and 96.83% during validation. Based on these results, the deep learning method has a higher average success rate than machine learning models in predicting financial distress.

A Deep Approach to Deep Fake Detection

In recent months, the proliferation of free deep learning-based software tools has facilitated the creation of credible face exchanges in videos, resulting in what are known as "DeepFake" (DF) videos. While manipulations of digital videos have been demonstrated for several decades through the use of visual effects, recent advances in deep learning have significantly increased the realism of fake content and the accessibility with which it can be created. These AI-synthesized media, popularly referred to as DF, pose a significant challenge for detection. Detecting DF is a major challenge due to the complexity of training algorithms to spot them. In this work, we propose a detection approach using Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs). Our system leverages a CNN to extract features at the frame level, which are then used to train an RNN. The RNN learns to classify whether a video has been manipulated, detecting temporal inconsistencies between frames introduced by DF creation tools. We evaluate our approach against a large set of fake videos collected from standard datasets and demonstrate competitive results using a simple architecture.

Emotion Symphony: Deep Learning Journey Through Human Facial Emotions

Abstract: In today's technologically advanced world, understanding human emotions holds paramount importance in various domains ranging from human-computer interaction to mental health diagnosis. Emotions are primarily conveyed through facial expressions, making facial emotion recognition a crucial area of research. This project, "Emotion Symphony," embarks on a deep learning journey aimed at developing an efficient system for recognizing and interpreting human facial emotions. The project employs state-of-the-art deep learning techniques to extract meaningful features from facial images and map them to corresponding emotional states. Leveraging convolutional neural networks (CNNs) and recurrent neural networks (RNNs), the system is designed to analyse subtle nuances in facial expressions, capturing both spatial and temporal dependencies for accurate emotion recognition. To train the model, extensive datasets comprising diverse facial expressions are utilized, ensuring robustness and generalization of the proposed system. Additionally, data augmentation techniques are employed to enhance the model's ability to handle variations in lighting conditions, poses, and facial expressions.

Interpreting Doctor Notes using Handwriting Recognition

Abstract: Handwriting recognition of medical prescriptions has been a challenging problem over the recent years with constant research in providing possible accurate solutions. Indecipherable handwritten prescription and inefficiency of Pharmacist to understand the medical prescription can lead to serious and harmful effect to the patients. Even in the recognition of handwriting, mainly doctors notes, they are very difficult for everyone to understand and it takes time for a person to analyse it. So, this idea mainly focused on interpreting doctor’s notes using handwritten recognition and deep learning techniques. The handwritten or printed document pictures are transformed into their electronic counterparts using an optical character recognition (OCR) system. Due to individuals' inconsistent writing styles, dealing with handwritten texts is significantly more difficult than dealing with printed ones. Handwritten text recognition could be done by Image processing, Machine Learning or Deep Learning Techniques. Out of these Deep Learning remains to be the most popular and prominent. Some of the Deep Learning techniques includes Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs). This gives a review of the various recognition methodologies used for interpreting handwritten texts. It includes the most important algorithms that could be used for detecting the handwritten word/text/character by using various approaches for the recognition process. In the end we are thus comparing the accuracies provided by these systems.

SST-CRAM: spatial-spectral-temporal based convolutional recurrent neural network with lightweight attention mechanism for EEG emotion recognition

SST-CRAM: spatial-spectral-temporal based convolutional recurrent neural network with lightweight attention mechanism for EEG emotion recognition

Edge Computing Paradigms: Bridging the Gap between Software Engineering and IoT

This exploration presents a linked profound learning-based asset booking strategy to work on the general execution of edge-incorporated edge IoT networks. For an IoT network to complete a responsibility productively and inside the dispensed time, it should get the best assets from the edge layer. The choice and appropriation of ideal assets rely upon cautious asset booking. Previously, profound learning methods were made to coordinate edge networks with Internet of Things applications while bringing down information transmission inactivity. To further develop an IoT application's general viability and administration quality, it is important to think about a couple of extra measurements, for example, response time, holding up time, and data transmission necessities. A linked procedure using a gated recurrent unit and convolutional neural network is provided to accomplish this improved exhibition. To pick the best assets from the asset pool and disseminate them to the IoT networks, the proposed asset planning model thinks about the qualities and necessities of the assets. This work presents an intensive examination of the connected strategy and exploratory perceptions.

Transforming oral cancer care: The promise of deep learning in diagnosis

The diagnosis and treatment of oral cancer present significant challenges, including delayed diagnosis at advanced stages and limited access to healthcare. Deep learning (DL), a subset of artificial intelligence, holds promise for transforming medical image analysis and predictive analytics. In this perspective, we examine the applications of DL in oral cancer. Specifically, we explore the efficacy of convolutional neural networks (CNNs) and recurrent neural networks (RNNs) in diagnosing and predicting the prognosis of oral cancer in the last five years. Additionally, we underscore the importance of oral cancer databases in advancing research and clinical practice. However, DL methods face constraints related to input variability and model interpretability. Addressing these issues is crucial to harnessing the full potential of DL in oral cancer treatment. In summary, this article underscores the innovative contributions of DL in revolutionizing oral cancer management and advocating for precision medicine in oncology.

Text Classification in Natural Language Processing

This paper presents an overview of text classification techniques, focusing on the pre-processing steps, feature extraction methods, and model selection strategies employed in the process. Algorithms such as Naive Bayes, Support Vector Machines (SVM), logistic regression, and neural networks are used. Furthermore, recent advancements in deep learning models for text classification, including Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) are used. Comprehensive understanding of text classification methodologies in NLP and insights into current trends and challenges in the field are mentioned

Vision Saver Deep Learning Approaches for Diabetic Retinopathy Detection

One of the main causes of adult blindness and a frequent consequence of diabetes is diabetic retinopathy. In order to preserve eyesight, diabetic retinopathy must be identified and treated early. In this study, "Vision Saver," a deep learning-based method for the automated detection and treatment of diabetic retinopathy, is introduced. Vision Saver provides a comprehensive solution for the detection of retinal problems by utilizing deep neural networks, enabling prompt patient care and intervention. Convolutional neural networks (CNNs) and recurrent neural networks (RNNs), two cutting-edge deep learning approaches, are used by the proposed system to evaluate retinal images and provide precise diagnoses. Additionally, we investigate how generative adversarial networks (GANs) might improve image quality and produce artificial data for training. The architecture of Vision Saver is built to handle both fundus photos and optical coherence tomography (OCT) images, enabling a flexible method of evaluating diabetic retinopathy. Vision Saver's management component includes not only detection but also monitoring of therapy progression. The system aids medical personnel in making educated judgments about interventions like laser therapy, injections, or surgical operations through ongoing monitoring and follow-up. The development and assessment of Vision Saver are covered in this essay, with a focus on how well it performs in terms of sensitivity, specificity, and accuracy. We also discuss concerns with model interpretability, data protection, and integration into clinical workflows. Ultimately, the objective is to improve patient outcomes and lessen the burden of blindness brought on by diabetic retinopathy by offering a comprehensive solution that improves screening and management of the illness. Deep learning in healthcare has a lot of potential, and Vision Saver is an example of how it can significantly improve the lives of people with diabetic retinopathy.

Memorability shapes perceived time (and vice versa).

Visual stimuli are known to vary in their perceived duration. Some visual stimuli are also known to linger for longer in memory. Yet, whether these two features of visual processing are linked is unknown. Despite early assumptions that time is an extracted or higher-order feature of perception, more recent work over the past two decades has demonstrated that timing may be instantiated within sensory modality circuits. A primary location for many of these studies is the visual system, where duration-sensitive responses have been demonstrated. Furthermore, visual stimulus features have been observed to shift perceived duration. These findings suggest that visual circuits mediate or construct perceived time. Here we present evidence across a series of experiments that perceived time is affected by the image properties of scene size, clutter and memorability. More specifically, we observe that scene size and memorability dilate time, whereas clutter contracts it. Furthermore, the durations of more memorable images are also perceived more precisely. Conversely, the longer the perceived duration of an image, the more memorable it is. To explain these findings, we applied a recurrent convolutional neural network model of the ventral visual system, in which images are progressively processed over time. We find that more memorable images are processed faster, and that this increase in processing speed predicts both the lengthening and the increased precision of perceived durations. These findings provide evidence for a link between image features, time perception and memory that can be further explored with models of visual processing.

Multi-resolution sequence and structure feature extraction for binding site prediction

Circular ribonucleic acids (circRNAs) are single-stranded RNA molecules that form loops and are widely expressed in various cells and tissues. They interact with RNA-binding proteins (RBPs) and play a vital regulatory role in the onset and development of several diseases. Researchers have proposed various hybrid architecture prediction methods based on convolutional neural networks and recurrent neural networks to recognize the interactions and sites between circRNAs and RBPs and thus reveal the biological functions of circRNAs. However, existing methods usually ignore the structural information of circRNA, which may affect the modeling of circRNA and RBP binding modes. To address these problems, we propose a prediction model based on multi-resolution feature extraction. First, it generates sequence features using unsupervised word embedding and nucleotide density. Then, it uses implicit and explicit pseudo-secondary structure hybrid encoding to fuse sequence and structural information and better simulate circRNA-RBP binding patterns. Second, it uses an enhanced bidirectional sample convolution and interaction network encoder to capture and integrate high-order features of distinct resolutions from the multi-scale convolution module. This provides rich semantic input to the downstream bidirectional long short-term memory network to improve prediction accuracy. Experimental results on 37 circRNA and 31 linear RNA datasets show that our method has significant advantages in identifying RNA-RBP interactions. Furthermore, the four motifs learned by our method are verified against existing motif databases, indicating that it can discover biologically meaningful circRNA-RBP binding patterns.

A GRU–CNN model for auditory attention detection using microstate and recurrence quantification analysis

Attention as a cognition ability plays a crucial role in perception which helps humans to concentrate on specific objects of the environment while discarding others. In this paper, auditory attention detection (AAD) is investigated using different dynamic features extracted from multichannel electroencephalography (EEG) signals when listeners attend to a target speaker in the presence of a competing talker. To this aim, microstate and recurrence quantification analysis are utilized to extract different types of features that reflect changes in the brain state during cognitive tasks. Then, an optimized feature set is determined by employing the processes of significant feature selection based on classification performance. The classifier model is developed by hybrid sequential learning that employs Gated Recurrent Units (GRU) and Convolutional Neural Network (CNN) into a unified framework for accurate attention detection. The proposed AAD method shows that the selected feature set achieves the most discriminative features for the classification process. Also, it yields the best performance as compared with state-of-the-art AAD approaches from the literature in terms of various measures. The current study is the first to validate the use of microstate and recurrence quantification parameters to differentiate auditory attention using reinforcement learning without access to stimuli.

Enhances the recognition accuracy of the complex text for online production labels by improving convolutional recurrent neural network

Enhances the recognition accuracy of the complex text for online production labels by improving convolutional recurrent neural network

GLOBAL PATTERNS AND EXTREME EVENTS IN SOVEREIGN RISK PREMIA: A FUZZY S DEEP LEARNING COMPARATIVE

Investment in foreign countries has become more common nowadays and this implies that there may be risks inherent to these investments, being the sovereign risk premium the measure of such risk. Many studies have examined the behaviour of the sovereign risk premium, nevertheless, there are limitations to the current models and the literature calls for further investigation of the issue as behavioural factors are necessary to analyse the investor’s risk perception. In addition, the methodology widely used in previous research is the regression model, and the literature shows it as scarce yet. This study provides a model for a new of the drivers of the government risk premia in developing countries and developed countries, comparing Fuzzy methods such as Fuzzy Decision Trees, Fuzzy Rough Nearest Neighbour, Neuro-Fuzzy Approach, with Deep Learning procedures such as Deep Recurrent Convolution Neural Network, Deep Neural Decision Trees, Deep Learning Linear Support Vector Machines. Our models have a large effect on the suitability of macroeconomic policy in the face of foreign investment risks by delivering instruments that contribute to bringing about financial stability at the global level.

Detection of Deepfake Media Using a Hybrid CNN–RNN Model and Particle Swarm Optimization (PSO) Algorithm

Deepfakes are digital audio, video, or images manipulated using machine learning algorithms. These manipulated media files can convincingly depict individuals doing or saying things they never actually did. Deepfakes pose significant risks to our lives, including national security, financial markets, and personal privacy. The ability to create convincing deep fakes can also harm individuals’ reputations and can be used to spread disinformation and fake news. As such, there is a growing need for reliable and accurate methods to detect deep fakes and prevent their harmful effects. In this paper, a hybrid convolutional neural network (CNN) and recurrent neural network (RNN) with a particle swarm optimization (PSO) algorithm is utilized to demonstrate a deep learning strategy for detecting deepfake videos. High accuracy, sensitivity, specificity, and F1 score were attained by the proposed approach when tested on two publicly available datasets: Celeb-DF and the Deepfake Detection Challenge Dataset (DFDC). Specifically, the proposed method achieved an average accuracy of 97.26% on Celeb-DF and an average accuracy of 94.2% on DFDC. The results were compared to other state-of-the-art methods and showed that the proposed method outperformed many. The proposed method can effectively detect deepfake videos, which is essential for identifying and preventing the spread of manipulated content online.

Deep Reinforcement Learning Based Optimization and Risk Control of Trading Strategies

Deep reinforcement learning (DRL) based optimization leverages advanced machine learning techniques to solve complex decision-making problems in various domains. Deep learning in trading involves the application of sophisticated neural network architectures to analyze financial data and make predictions in stock markets, forex, commodities, and other trading domains. Deep learning algorithms, such as convolutional neural networks (CNNs) or recurrent neural networks (RNNs), traders can extract valuable insights from vast amounts of historical market data, including price movements, trading volumes, and market sentiment. These models can learn complex patterns and relationships within the data, enabling them to forecast future market trends, identify potential trading opportunities, and manage risks more effectively. Deep learning in trading has shown promise in improving decision-making processes, enhancing trading strategies, and achieving higher returns, although challenges such as data scarcity, model interpretability, and overfitting remain areas of ongoing research and development. In this paper evaluated the stock market to achieve the financial gain to achieve the significant improvement in prediction with the stock trading. The data related to the stock market are optimized with stock trend data of 7,000 stocks situated in the United States. The estimated stock trade data were computed and processed with the Multiagent Q-learning model for the data detection and classification. A deep Q-network is developed to adaptively pick distinct action subsets and train the market making agent based on the inventory states. The experimental findings demonstrate the effectiveness of our suggested methodology in feature learning and superiority in accuracy improvement when compared to deep learning-based modelling of frequency trading patterns and standard signal processing approaches for stock trend prediction.