Showing papers in "International Journal of Computational Intelligence Systems in 2023"
TL;DR: Wang et al. as mentioned in this paper used multi-scale feature extraction and fusion methods in the image feature characterization and text information representation sections of the VQA system, respectively to improve its accuracy.
Abstract: Abstract The Visual Question Answering (VQA) system is the process of finding useful information from images related to the question to answer the question correctly. It can be widely used in the fields of visual assistance, automated security surveillance, and intelligent interaction between robots and humans. However, the accuracy of VQA has not been ideal, and the main difficulty in its research is that the image features cannot well represent the scene and object information, and the text information cannot be fully represented. This paper used multi-scale feature extraction and fusion methods in the image feature characterization and text information representation sections of the VQA system, respectively to improve its accuracy. Firstly, aiming at the image feature representation problem, multi-scale feature extraction and fusion method were adopted, and the image features output of different network layers were extracted by a pre-trained deep neural network, and the optimal scheme of feature fusion method was found through experiments. Secondly, for the representation of sentences, a multi-scale feature method was introduced to characterize and fuse the word-level, phrase-level, and sentence-level features of sentences. Finally, the VQA model was improved using the multi-scale feature extraction and fusion method. The results show that the addition of multi-scale feature extraction and fusion improves the accuracy of the VQA model.
26 citations
TL;DR: Wang et al. as mentioned in this paper proposed a new recommendation model, DRR-Max, based on deep reinforcement learning (DRL), which adopted a state generation module specially designed to obtain users' long-term and short-term preferences from user profiles and user history score item information.
Abstract: Abstract With the development of Internet technology, the problem of information overload has increasingly attracted attention. Nowadays, the recommendation system with excellent performance in information retrieval and filtering would be widely used in the business field. However, most existing recommendation systems are considered a static process, during which recommendations for internet users are often based on pre-trained models. A major disadvantage of these static models is that they are incapable of simulating the interaction process between users and their systems. Moreover, most of these models only consider users’ real-time interests while ignoring their long-term preferences. This paper addresses the abovementioned issues and proposes a new recommendation model, DRR-Max, based on deep reinforcement learning (DRL). In the proposed framework, this paper adopted a state generation module specially designed to obtain users’ long-term and short-term preferences from user profiles and user history score item information. Next, Actor-Critical algorithm is used to simulate the real-time recommendation process.Finally, this paper uses offline and online methods to train the model. In the online mode, the network parameters were dynamically updated to simulate the interaction between the system and users in a real recommendation environment. Experimental results on the two publicly available data sets were used to demonstrate the effectiveness of our proposed model.
4 citations
TL;DR: In this article , a hybrid approach based on a fully dense fusion neural network (FD-CNN) and dual preprocessing was proposed to identify retinal diseases, such as choroidal neovascularization, diabetic macular edema, drusen from OCT images.
Abstract: Abstract Retinal issues are crucial because they result in visual loss. Early diagnosis can aid physicians in initiating treatment and preventing visual loss. Optical coherence tomography (OCT), which portrays retinal morphology cross-sectionally and noninvasively, is used to identify retinal abnormalities. The process of analyzing OCT images, on the other hand, takes time. This study has proposed a hybrid approach based on a fully dense fusion neural network (FD-CNN) and dual preprocessing to identify retinal diseases, such as choroidal neovascularization, diabetic macular edema, drusen from OCT images. A dual preprocessing methodology, in other words, a hybrid speckle reduction filter was initially used to diminish speckle noise present in OCT images. Secondly, the FD-CNN architecture was trained, and the features obtained from this architecture were extracted. Then Deep Support Vector Machine (D-SVM) and Deep K-Nearest Neighbor (D-KNN) classifiers were proposed to reclassify those features and tested on University of California San Diego (UCSD) and Duke OCT datasets. D-SVM demonstrated the best performance in both datasets. D-SVM achieved 99.60% accuracy, 99.60% sensitivity, 99.87% specificity, 99.60% precision and 99.60% F1 score in the UCSD dataset. It achieved 97.50% accuracy, 97.64% sensitivity, 98.91% specificity, 96.61% precision, and 97.03% F1 score in Duke dataset. Additionally, the results were compared to state-of-the-art works on the both datasets. The D-SVM was demonstrated to be an efficient and productive strategy for improving the robustness of automatic retinal disease classification. Also, in this study, it is shown that the unboxing of how AI systems' black-box choices is made by generating heat maps using the local interpretable model-agnostic explanation method, which is an explainable artificial intelligence (XAI) technique. Heat maps, in particular, may contribute to the development of more stable deep learning-based systems, as well as enhancing the confidence in the diagnosis of retinal disease in the analysis of OCT image for ophthalmologists.
2 citations
TL;DR: In this article , an approach to enrich K-Nearest Neighbors (KNN) to deal with Implicit Aspect Identification task (IAI), through the use of WordNet semantic relations, was proposed.
Abstract: Abstract Opinion mining or sentiment analysis (SA) is a key component of real-world applications for e-commerce organizations, manufacturers, and customers. SA deals with the computational evaluation of people’s views, thoughts, and feelings in the text, whether they are visible or concealed. The Aspect based SA level is becoming one of the most active phases in this area. In this paper, we propose an approach to enrich K-Nearest Neighbors (KNN) to deal with Implicit Aspect Identification task (IAI). Through the use of WordNet semantic relations, we propose an enhancement for KNN distance computation to support the IAI task. For a conclusive empirical evaluation, experiments are conducted on two datasets of electronic products and restaurant reviews and the effects of our approach are examined and analyzed according to three criteria: KNN distance used to compute the similarity, the number of nearest neighbors (K) and the KNN behavior towards Overfitting and Underfitting. The experimental results show that our approach helps KNN improve its performance and better deal with Overfitting and Underfitting for Implicit Aspect Identification.
2 citations
TL;DR: In this paper , a 12-layer deep 1D Convolutional Neural Network (CNN) is introduced for better features extraction, and finally, SoftMax and machine learning classifiers are applied to classify the heartbeats.
Abstract: Abstract In India, over 25,000 people have died from cardiovascular annually over the past 4 years , and over 28,000 in the previous 3 years. Most of the deaths nowadays are mainly due to cardiovascular diseases (CVD). Arrhythmia is the leading cause of cardiovascular mortality. Arrhythmia is a condition in which the heartbeat is abnormally fast or slow. The current detection method for diseases is analyzing by the electrocardiogram (ECG), a medical monitoring technique that records heart activity. Since actuations in ECG signals are so slight that they cannot be seen by the human eye, the identification of cardiac arrhythmias is one of the most difficult undertakings. Unfortunately, it takes a lot of medical time and money to find professionals to examine a large amount of ECG data . As a result, machine learning-based methods have become increasingly prevalent for recognizing ECG features. In this work, we classify five different heartbeats using the MIT-BIH arrhythmia database . Wavelet self-adaptive thresholding methods are used to first denoise the ECG signal. Then, an efficient 12-layer deep 1D Convolutional Neural Network (CNN) is introduced for better features extraction, and finally, SoftMax and machine learning classifiers are applied to classify the heartbeats. The proposed method achieved an average accuracy of 99.40%, precision of 98.78%, recall of 98.78%, and F1 score of 98.74%, which clearly show that it outperforms with the exiting model . Architecture of proposed work is simple but effective in remote cardiac diagnosis paradigm that can be implemented on e-health devices.
1 citations
TL;DR: In this paper , the authors investigated the performance of existing multi-modal fusion pre-training algorithms and medical multidimensional fusion methods and compared their key characteristics, such as supported medical data, diseases, target samples, and implementation performance.
Abstract: Abstract In recent years, deep learning has been applied in the field of clinical medicine to process large-scale medical images, for large-scale data screening, and in the diagnosis and efficacy evaluation of various major diseases. Multi-modal medical data fusion based on deep learning can effectively extract and integrate characteristic information of different modes, improve clinical applicability in diagnosis and medical evaluation, and provide quantitative analysis, real-time monitoring, and treatment planning. This study investigates the performance of existing multi-modal fusion pre-training algorithms and medical multi-modal fusion methods and compares their key characteristics, such as supported medical data, diseases, target samples, and implementation performance. Additionally, we present the main challenges and goals of the latest trends in multi-modal medical convergence. To provide a clearer perspective on new trends, we also analyzed relevant papers on the Web of Science. We obtain some meaningful results based on the annual development trends, country, institution, and journal-level research, highly cited papers, and research directions. Finally, we perform co-authorship analysis, co-citation analysis, co-occurrence analysis, and bibliographic coupling analysis using the VOSviewer software.
1 citations
TL;DR: Wang et al. as discussed by the authors proposed an improved algorithm based on the YOLO-v4 algorithm, which added three attention mechanism modules to the appropriate network level to enhance the key feature points of face wearing masks and suppress useless information.
Abstract: Abstract To reduce the chance of being infected by the COVID-19, wearing masks correctly when entering and leaving public places has become the most feasible and effective ways to prevent the spread of the virus. It is a concern to how to quickly and accurately detect whether a face is worn a mask correctly while reduce missed detection and false detection in practical applied scenarios. In this paper, an improved algorithm is proposed based on the YOLO-v4 algorithm. The attention mechanism module is added to the appropriate network level to enhance the key feature points of face wearing masks and suppress useless information. Apart from that, three attention mechanism modules are added to different layers of the YOLO-v4 network for ablation experiments, including CBAM (convolutional block attention module), SENet (squeeze-and-excitation networks) and CANet (coordinate attention networks). The path-aggregation network and feature pyramid are used to extract features from images. Two network models were compared and improved in the experiment, and it is found that adding the dual-channel attention mechanism CBAM before the three YOLO heads of YOLOv4 and in the neck network had better detection performance than the single channel attention mechanism SENet and the coordinated attention mechanism CANet. The experimental results show that when the attention module CBAM and the YOLO-v4 model are integrated, the accuracy of the selected MAFA + WIDER Face dataset reaches the highest value of 93.56%, which is 4.66% higher than that of the original YOLO-v4.
1 citations
TL;DR: Wang et al. as mentioned in this paper proposed an improved adaptive lion swarm optimization (LSO) algorithm integrating the chaotic search strategy and information entropy to address the problem that the standard LSO algorithm has slow convergence and easily falls into the local optimum in later iterations.
Abstract: Abstract This paper proposes an improved adaptive lion swarm optimization (LSO) algorithm integrating the chaotic search strategy and information entropy to address the problem that the standard LSO algorithm has slow convergence and easily falls into the local optimum in later iterations. At first, an adaptive factor is introduced to improve tent chaotic mapping and used for population position initialization to enhance population diversity and realize uniform traversal while ensuring random distribution, ultimately improving the global search ability. Second, to address the problem that the cub selection strategy is blind, resulting in insufficient traversal in the early stage, a dynamic step-size perturbation factor is established using the second-order norm and information entropy. Adaptive parameters are used to dynamically adjust the selection probability of different cub behaviors based on the number of iterations to suppress the premature convergence of the algorithm. Finally, tent chaotic search is employed to adaptively adjust the search range and improve the individuals with poor fitness through multiple neighborhood points of the local optimal solution, further improving the algorithm’s search speed and accuracy. Experimental results on 18 benchmark functions revealed that the proposed algorithm yields superior performance in terms of convergence speed, optimization accuracy, and ability to jump out of the local optimal solution compared with the standard LSO, gray wolf optimizer, and particle swarm optimization algorithms. Furthermore, the improved LSO algorithm was used to optimize the initial weights and thresholds of the BP neural network, and the effectiveness of the proposed algorithm was further verified by studying the house price prediction problem using two real-world datasets.
1 citations
TL;DR: In this article , an interval-valued Atanassov-intuitionistic fuzzy Aczel-Alsina power averaging (IVA-IFAAPA) was proposed.
Abstract: Abstract Aczel–Alsina t-norm and t-conorm are important t-norm and t-conorm, and they are extended from algebraic t-norm and t-conorm. Obviously, Aczel–Alsina t-norm and t-conorm are more general than some existing t-norm and t-conorm. Furthermore, the power aggregation (PA) operator is also a very famous and valuable operator which can consider the power relation between any two input parameters. In addition, Interval-valued Atanassov-intuitionistic fuzzy set (IVA-IFS) can easily express uncertain information. In order to fully use their advantages, in this analysis, we extend the PA operators based on Aczel–Alsina t-norm and t-conorm to IVA-IFS and propose the interval-valued Atanassov-intuitionistic fuzzy Aczel–Alsina power averaging (IVA-IFAAPA), interval-valued Atanassov-intuitionistic fuzzy Aczel–Alsina power ordered averaging (IVA-IFAAPOA), interval-valued Atanassov-intuitionistic fuzzy Aczel–Alsina power geometric (IVA-IFAAPG) and interval-valued Atanassov-intuitionistic fuzzy Aczel–Alsina power ordered geometric (IVA-IFAAPOG) operators. Moreover, we discuss the properties of the presented operators such as idempotency, monotonicity, and boundedness. In addition, a multi-attribute decision-making (MADM) procedure is proposed to process the IVA-IF information. Finally, a practical example is used to show the effectiveness and superiority of the proposed method by comparing it with some existing operators.
1 citations
TL;DR: In this paper , a convolutional neural network (CNN) model with 16 layers was designed for noise reduction, considering the strong similarity between noisy signals and denoising signals.
Abstract: Abstract Microseismic signals contain various information for oil and gas developing. Increasing the signal-to-noise ratio of microseismic signals can successfully improve the effectiveness of oil and gas resource exploration. The lack of sufficient labeled microseismic signals makes it difficult to train neural network model. Transfer learning can solve this problem using image data sets to pre-train the denoising model and the learned knowledge can be transferred into microseismic signals denoising. In addition, a convolutional neural network (CNN) model with 16 layers is designed for noise reduction. Considering the strong similarity between noisy signals and denoising signals, residual learning is utilized to optimize the denoising model. The simulation experiment results show that the proposed denoising model eliminates the noise in the microseismic signals effectively and quickly, restores the amplitude of the microseismic signals with high accuracy, and has excellent effect in denoising on the information at the edge.
1 citations
TL;DR: In this article , the authors proposed a Synthetic Multi-minority Oversampling (SMMO) framework by integrating with a collaborative feature selection (CoFS) approach in network intrusion detection systems, which aims to increase the detection accuracy of the extreme minority classes by improving the dataset's class distribution and selecting relevant features.
Abstract: Abstract Researchers publish various studies to improve the performance of network intrusion detection systems. However, there is still a high false alarm rate and missing intrusions due to class imbalance in the multi-class dataset. This imbalanced distribution of classes results in low detection accuracy for the minority classes. This paper proposes a Synthetic Multi-minority Oversampling (SMMO) framework by integrating with a collaborative feature selection (CoFS) approach in network intrusion detection systems. Our framework aims to increase the detection accuracy of the extreme minority classes (i.e., user-to-root and remote-to-local attacks) by improving the dataset’s class distribution and selecting relevant features. In our framework, SMMO generates synthetic data and iteratively over-samples multi-minority classes. And the collaboration of correlation-based feature selection with an evolutionary algorithm selects essential features. We evaluate our framework with a random forest, J48, BayesNet, and AdaBoostM1. In a multi-class NSL-KDD dataset, the experimental results show that the proposed framework significantly improves the detection accuracy of the extreme minority classes compared with other approaches.
TL;DR: Wang et al. as mentioned in this paper proposed a traffic congestion index and devise a new traffic congestion prediction model spatio-temporal transformer (STTF) based on transformer, a deep learning model.
Abstract: Abstract With the rapid development of economy, the sharp increase in the number of urban cars and the backwardness of urban road construction lead to serious traffic congestion of urban roads. Many scholars have tried their best to solve this problem by predicting traffic congestion. Some traditional models such as linear models and nonlinear models have been proved to have a good prediction effect. However, with the increasing complexity of urban traffic network, these models can no longer meet the higher demand of congestion prediction without considering more complex comprehensive factors, such as the spatio-temporal correlation information between roads. In this paper, we propose a traffic congestion index and devise a new traffic congestion prediction model spatio-temporal transformer (STTF) based on transformer, a deep learning model. The model comprehensively considers the traffic speed of road segments, road network structure, the spatio-temporal correlation between road sections and so on. We embed temporal and spatial information into the model through the embedding layer for learning, and use the spatio-temporal attention module to mine the hidden spatio-temporal information within the data to improve the accuracy of traffic congestion prediction. Experimental results based on real-world datasets demonstrate that the proposed model significantly outperforms state-of-the-art approaches.
TL;DR: In this article , a multi-level random forest algorithm for intrusion detection using a fuzzy inference system was developed, which combines the strengths of the filter and wrapper approaches to create a more advanced multilevel feature selection technique, which strengthened network security.
Abstract: Abstract Intrusion detection ( ID ) methods are security frameworks designed to safeguard network information systems. The strength of an intrusion detection method is dependent on the robustness of the feature selection method. This study developed a multi-level random forest algorithm for intrusion detection using a fuzzy inference system. The strengths of the filter and wrapper approaches are combined in this work to create a more advanced multi-level feature selection technique, which strengthens network security. The first stage of the multi-level feature selection is the filter method using a correlation-based feature selection to select essential features based on the multi-collinearity in the data. The correlation-based feature selection used a genetic search method to choose the best features from the feature set. The genetic search algorithm assesses the merits of each attribute, which then delivers the characteristics with the highest fitness values for selection. A rule assessment has also been used to determine whether two feature subsets have the same fitness value, which ultimately returns the feature subset with the fewest features. The second stage is a wrapper method based on the sequential forward selection method to further select top features based on the accuracy of the baseline classifier. The selected top features serve as input into the random forest algorithm for detecting intrusions. Finally, fuzzy logic was used to classify intrusions as either normal, low, medium, or high to reduce misclassification. When the developed intrusion method was compared to other existing models using the same dataset, the results revealed a higher accuracy, precision, sensitivity, specificity, and F1-score of 99.46%, 99.46%, 99.46%, 93.86%, and 99.46%, respectively. The classification of attacks using the fuzzy inference system also indicates that the developed method can correctly classify attacks with reduced misclassification. The use of a multi-level feature selection method to leverage the advantages of filter and wrapper feature selection methods and fuzzy logic for intrusion classification makes this study unique.
TL;DR: In this article , the authors apply XGBoost algorithm to knowledge tracing model to improve the prediction performance, which can accurately evaluate the learning state of students and conduct personalized instruction according to the characteristics of different students.
Abstract: Abstract The knowledge tracing (KT) model is an effective means to realize the personalization of online education using artificial intelligence methods. It can accurately evaluate the learning state of students and conduct personalized instruction according to the characteristics of different students. However, the current knowledge tracing models still have problems of inaccurate prediction results and poor features utilization. The study applies XGBoost algorithm to knowledge tracing model to improve the prediction performance. In addition, the model also effectively handles the multi-skill problem in the knowledge tracing model by adding the features of problem and knowledge skills. Experimental results show that the best AUC value of the XGBoost-based knowledge tracing model can reach 0.9855 using multiple features. Furthermore, compared with previous knowledge tracing models used deep learning, the model saves more training time.
TL;DR: In this article , a new similarity operator under intuitionistic fuzzy sets (IFSs) is presented to ameliorate the itemized setbacks noticed with the hitherto similarity operators, including imprecise results, omission of hesitation information, misleading interpretations and outright violations of metric axioms of similarity operator.
Abstract: Abstract Many complex real-world problems have been resolved based on similarity operators under intuitionistic fuzzy sets (IFSs). Numerous authors have developed intuitionistic fuzzy similarity operators (IFSOs) but with some setbacks, which include imprecise results, omission of hesitation information, misleading interpretations, and outright violations of metric axioms of similarity operator. To this end, this article presents a newly developed similarity operator under IFSs to ameliorate the itemized setbacks noticed with the hitherto similarity operators. To buttress the validity of the new similarity operator, we discuss its properties in alliance with the truisms of similarity. In addition, we discuss some complex decision-making situations involving car purchase selection process, pattern recognition, and emergency management using the new similarity operator based on multiple criteria decision making (MCDM) technique and recognition principle, respectively. Finally, comparative studies are presented to argue the justification of the new similarity operator. In short, the novelty of this work includes the evaluation of the existing IFSOs to isolate their fault lines, development of a new IFSO technique with the capacity to resolve the fault lines in the existing techniques, elaboration of some properties of the newly developed IFSO, and its applications in the solution of disaster control, pattern recognition, and the process of car selection for purchasing purpose based on the recognition principle and MCDM.
TL;DR: In this article , the maximal product of two cubic fuzzy graph structures is investigated, and the degree and the total degree of a vertex in the product of at most two cubic-fuzzy graph structures are calculated based on the membership functions of vertices and edges.
Abstract: Abstract The cubic fuzzy graph structure, as a combination of cubic fuzzy graphs and fuzzy graph structures, shows better capabilities in solving complex problems, especially in cases where there are multiple relationships. The quality and method of determining the degree of vertices in this type of fuzzy graphs simultaneously supports fuzzy membership and interval-valued fuzzy membership, in addition to the multiplicity of relations, motivated us to conduct a study on the maximal product of cubic fuzzy graph structures. In this research, upon introducing the cubic fuzzy graph structure, some properties of the maximal product and its characteristics have been investigated. By introducing the degree and the total degree of a vertex in the product of at most two cubic fuzzy graph structures, its calculation methods are categorized based on different conditions among the membership functions of vertices and edges. The results show that all features of two cubic fuzzy graph structures do not appear in their maximal product and vice versa. Finally, an application of cubic fuzzy graph structure in project management is presented.
TL;DR: Zhang et al. as discussed by the authors proposed an adaptive directed denoising filter (ADD filter) based on a neural network, which consists of three major stages: training, filtering, and enhancing.
Abstract: Abstract In this modern era, visual data transmission, processing, and analysis play a vital role in daily life. Image denoising is the process of approximately estimating the original version of a degraded image. The presence of unexpected noise (e.g., fixed, random, and Gaussian) is the root cause of degradation, which has been reduced to some extent by many linear and non-linear filters based on a median value. The real issue is developing a strategy that should be generalized enough to effectively restore an image corrupted with multi-nature noise. Many researchers have developed novel concepts, but their tactics must acquire the highest performance in this area. This article proposes a constrained strategy for this problem, i.e., an adaptively directed denoising filter (ADD filter) based on a neural network. It consists of three major stages: training, filtering, and enhancing. First, we train a feed-forward back-propagation neural network on noisy and noise-free pixels for effective differentiation. Second, we apply a one-pass selective filter to the noisy image. The objective of this one-pass filter is to minimize noise using an adaptive median or directional filter based on density. Finally, the iterative directional filter is applied to the pre-processed image to enhance its visual quality. The extensive experiments depict that the proposed system has achieved better subjective results and improved local (structural similarity) and global (peak signal-to-noise ratio or mean square error) statistical measures.
TL;DR: In this article , the authors compared the performance of a football player's injury full-cycle management and monitoring system based on blockchain and machine learning algorithm with the traditional football players' injury management, and the experimental results showed that the average self-processing capacity of the football player injury MMS based on the blockchain and ML algorithm was 70%.
Abstract: Abstract Football injuries are the most common factor affecting a football player's performance, and the last thing a football player wants. To understand the causes of football players’ injuries and how to recover sports injuries most efficiently, the football players’ injuries were managed and monitored throughout the whole cycle. However, the traditional football player injury cycle management and monitoring system are not only insecure in data storage, but more importantly, it lacks intelligent analysis of the collected data. With the continuous development of blockchain and machine learning technologies, blockchain technology is used to collect, store, clean, mine and visualize the full-cycle data of football players' injuries, and machine learning is used to provide intelligent solutions for football players' injury recovery. This paper compared the football player's injury full-cycle management and monitoring system based on blockchain and machine learning algorithm with the traditional football player's injury management and monitoring system. The experimental results showed that the average self-processing capacity of the football player injury MMS based on blockchain and ML algorithm was 70%, while the average self-processing capacity of the traditional football player injury management and monitoring system was 50%. Therefore, the application of blockchain and machine learning algorithm in the football player’s injury full-cycle management and monitoring system can effectively improve the system’s self-processing ability.
TL;DR: In this article , the impact of the COVID-19 pandemic and market sentiment on the dynamics of the exchange rate of USD/JPY, GBP/USD, and USD/CNY was investigated.
Abstract: Abstract This paper attempts to investigate the impact of the COVID-19 pandemic and market sentiment on the dynamics of USD/JPY, GBP/USD, and USD/CNY. We compose the market sentiment variable and incorporate the newly confirmed COVID-19 cases and sentiment variable into the traditional exchange rate forecasting model. We find that confirmed COVID-19 cases and sentiment variables in the US, Japan, UK, and China in the period of January 23rd, 2020 to September 14th, 2021 are significant in explaining the bilateral exchange rate movement. Recurrent neural network (RNN) and long short-term memory (LSTM) models outperform the other deep learning models and vector autoregressive (VAR) model in forecasting the bilateral exchange rate movement during the COVID-19 pandemic period. Further analysis using high-frequency intraday data and ensemble models shows that ensemble models significantly improve the accuracy of exchange rate prediction, as they are better at coping with the nonlinear and nonstationary features of exchange rate time series.
TL;DR: Wang et al. as mentioned in this paper used the Changchun Changsheng Vaccine Incident as an example, and the degree of attention to emergency-related keyword searches in the Baidu Index as a descriptive variable for the development of network public opinion.
Abstract: Abstract Internet public opinion is a complex and changeable system, and its trend development is characterized by explosive, evolutionary uncertainty, concealment and interactivity due to the participation of the vast number of Internet users. Today, with the rapid development of network information technology, public opinion has an increasing influence on the stable development of society. Computational intelligence is the frontier field of artificial intelligence development, and computational intelligence is used to mine and analyze public opinion text information and study the evolution of online public opinion. This paper uses the Changchun Changsheng Vaccine Incident as an example, and the netizens’ degree of attention to emergency-related keyword searches in the Baidu Index as a descriptive variable for the development of network public opinion. After applying the optimal segmentation algorithm, the development of public opinion is divided into phases. On this basis, a social network analysis is adopted to analyze the spatial and topological structure of each phase of network public opinion, using data from the Sina Weibo platform. Based on optimal segmentation, the development of network public opinion of the Changchun Changsheng Vaccine Incident can be divided into four phases, namely latent, spreading, control, and stable; each phase has different spatial and topological characteristics. Corresponding policy suggestions on network public opinion governance are put forward for each phase.
TL;DR: In this article , an adversarial training method is proposed to assist the model in strengthening general representation learning by making a classification model as a generator and introducing an unsupervised discriminator D to distinguish the hidden feature of the classification model from real images to limit their spatial distance.
Abstract: Abstract Over-fitting is a significant threat to the integrity and reliability of deep neural networks with generous parameters. One problem is that the model learned more specific features than general features in the training process. To solve the problem, we propose an adversarial training method to assist the model in strengthening general representation learning. In this method, we make a classification model as a generator G and introduce an unsupervised discriminator D to distinguish the hidden feature of the classification model from real images to limit their spatial distance. Notably, the D will fall into the trap of a perfect discriminator resulting in the gradient of confrontation loss of 0 after overtraining. To avoid this situation, we train the D with a probability $$P_{c}$$ P c . Our proposed method is easy to incorporate into existing frameworks. It has been evaluated under various network architectures over different fields of datasets. Experiments show that this method, under low computational cost, outperforms the benchmark by 1.5–2 points on different datasets. For semantic segmentation on VOC, our proposed method achieves 2.2 points higher mAP.
TL;DR: In this article , a transfer learning-based Chinese named entity recognition model is proposed to improve the performance of NER in the lack of well-annotated entity data, which utilizes the BiLSTM+CRF as the main structure and integrates character boundary information to assist the attention network.
Abstract: Abstract To improve the performance of named entity recognition in the lack of well-annotated entity data, a transfer learning-based Chinese named entity recognition model is proposed in this paper. The specific tasks are as follows: (1) first/, a data transfer method based on entity features is proposed. By calculating the similarity of feature distribution between low resource data and high resource data, the most representative entity features are selected for feature transfer mapping, and the distance of entity distribution between the two domains is calculated to make up the gap between the data of the two domains then model is trained by high resource data. (2) Then, an entity boundary detection method is proposed. This method utilizes the BiLSTM+CRF as the main structure and integrates character boundary information to assist the attention network to improve the model’s ability to recognize entity boundaries. (3) Finally, multiple named entity recognition methods are selected as baseline methods for comparison, and experiments are conducted on several datasets. The results show that the model proposed in this paper improves the accuracy of named entity recognition by 1%, the recall rate by 2%, and the F 1 value by 2% on average in the field with low-resource.
TL;DR: In this paper , the authors show the evolution of the innovation and development of the financial digital industry driven by information technology in recent years, summarizes and discusses the key technologies emerging in it, and analyzes its limitations and shortcomings.
Abstract: Abstract With the development and application of Internet technology, the world has entered a new era of the Internet economy. Information resources worldwide are no longer limited by time or space, which significantly improves the operational efficiency of production and commerce. As an essential part of the digital economy, digital finance is developing rapidly. Compared with traditional finance, Internet finance and digital finance have the advantages of low informational and transactional costs, and efficient services, which bring more benefits to consumers and investors. In the traditional small and fragmented long-tail market, the availability of finance has been effectively improved through technology-driven development. However, the sound development of the financial digital industry is inseparable from the self-generated sense of innovation and the exploration of green and sustainable development within the industry. This paper shows the evolution of the innovation and development of the financial digital industry driven by information technology in recent years, summarizes and discusses the key technologies emerging in it, and analyzes its limitations and shortcomings. This work will provide some references for the green development of the current industry.
TL;DR: Wang et al. as discussed by the authors used least absolute shrinkage and selection operator penalized Cox analysis for constructing a risk score prognostic model and drew receiver operating characteristic (ROC) curves to predict overall survival.
Abstract: Abstract With the rapid development of information technology, many medical systems have emerged one after another with the support of continuous learning. A method of medical data privacy protection and resource utilization based on continuous learning is proposed to initialize the depth model of specific medical tasks. The depth model includes feature sampling model, data review model and task expression model, Finally, the depth model is trained according to the data from n institutions in turn. This method can overcome the obstacles of data sharing. The intelligent medical system of medical knowledge sharing will greatly improve the level of existing medical technology. An increasing body of evidence suggests that long non-coding RNAs (lncRNAs) participate in various physiological processes and pathological diseases. Esophageal adenocarcinoma develops rapidly with poor prognosis and high mortality in the near and long term. Immunotargeted therapy is a research hotspot. However, it is necessary to explore the immunomodulatory molecules of esophageal adenocarcinoma and analyze their relationship with clinicopathological characteristics and prognosis. We aimed to construct a robust immune-related lncRNA signature associated with survival outcomes in esophageal adenocarcinoma. We identified an immune-related lncRNA pairs signature with prognostic value from The Cancer Genome Atlas. Differentially expressed immune-related lncRNAs (DEirlncRNAs) were identified and paired, followed by prognostic assessment using univariate Cox regression analysis. We used least absolute shrinkage and selection operator penalized Cox analysis for constructing a risk score prognostic model and drew receiver operating characteristic (ROC) curves to predict overall survival. Then, we evaluated our signature in several settings: chemotherapy, tumor-infiltrating immune cells, and immune-mediated gene expression. In total, 339 DEirlncRNA pairs were identified, 11 of which were involved in the risk score prognostic signature. The area under ROC curves representing the predictive effect for 1-, 2-, and 3-year survival rates were 0.942, 0.987, and 0.977, respectively. The risk score model was confirmed as an independent prognostic factor and was significantly superior to clinicopathological characteristics. Correlation analyses showed disparities in drug sensitivity, tumor-infiltrating immune cells, and immune-related gene expression. We identified a novel prognostic immune-related lncRNA pair signature for esophageal adenocarcinoma. The risk score-based groups displayed different immune statuses, drug sensitivity, and immune-mediated gene expression. These findings may offer insights into the prognostic evaluation of esophageal adenocarcinoma and may provide a basis for creating personalized treatment plans.
TL;DR: Wang et al. as discussed by the authors investigated the impact of the digital economy on green total factor productivity (GTFP) and its transmission mechanism using panel data from 256 cities in China from 2009 to 2020, and examined a directional distance function and the Malmquist-Luenberger productivity index to estimate the GTFP growth and constructs an ordinary least squares model to explore the impact effect and mechanism.
Abstract: Abstract This study investigates the impact of the digital economy (DE) on green total factor productivity (GTFP) and its transmission mechanism. Using panel data from 256 cities in China from 2009 to 2020, the study examines a directional distance function and the Malmquist–Luenberger productivity index to estimate the GTFP growth and constructs an ordinary least squares model to explore the impact effect and mechanism. Three findings are drawn from the estimation results: (1) The DE has significantly promoted GTFP. (2) Technological innovation has significantly aided in the promotion of GTFP. (3) By encouraging technological innovation, the DE further enhances the promotion of GTFP, verifying the DE → technology innovation → green conduction mechanism of total factor productivity.
TL;DR: In this article , the authors proposed a clustering algorithm based on ROCK that improves the link rather than the previously used distance as a criterion for cluster split and can successfully address the problem of data scarcity based on collaborative suggestion.
Abstract: Abstract The rapid development of network technology has revolutionized information dissemination and made it possible for various new communication channels to gradually seep into people’s daily lives. The primary task within the context of new media is the creation of agricultural products live e-commerce. This study first analyses and categorizes the preferences of product purchasers using a collaborative filtering process in order to achieve this. It then suggests a clustering algorithm based on ROCK that improves the link rather than the previously used distance as a criterion for cluster split and can successfully address the problem of data scarcity based on collaborative suggestion. The difficulties of live e-commerce entrepreneurship and the need of fostering new farmers’ live e-commerce entrepreneurship skills are also discussed in this essay. The case study illustrates that the research’s strategy can successfully examine the path towards developing new farmers’ live e-commerce skills.
TL;DR: In this paper , a nonparametric nearest neighbor classification method based on global variance differences is proposed, where the difference in variance is calculated before and after adding the sample to be the subject, then the difference is divided by the variance before adding the sampled to be tested, and the resulting quotient serves as the objective function.
Abstract: Abstract As technology improves, how to extract information from vast datasets is becoming more urgent. As is well known, k-nearest neighbor classifiers are simple to implement and conceptually simple to implement. It is not without its shortcomings, however, as follows: (1) there is still a sensitivity to the choice of k -values even when representative attributes are not considered in each class; (2) in some cases, the proximity between test samples and nearest neighbor samples cannot be reflected accurately due to proximity measurements, etc. Here, we propose a non-parametric nearest neighbor classification method based on global variance differences. First, the difference in variance is calculated before and after adding the sample to be the subject, then the difference is divided by the variance before adding the sample to be tested, and the resulting quotient serves as the objective function. In the final step, the samples to be tested are classified into the class with the smallest objective function. Here, we discuss the theoretical aspects of this function. Using the Lagrange method, it can be shown that the objective function can be optimal when the sample centers of each class are averaged. Twelve real datasets from the University of California, Irvine are used to compare the proposed algorithm with competitors such as the Local mean k-nearest neighbor algorithm and the pseudo-nearest neighbor algorithm. According to a comprehensive experimental study, the average accuracy on 12 datasets is as high as 86.27 $$\%$$ % , which is far higher than other algorithms. The experimental findings verify that the proposed algorithm produces results that are more dependable than other existing algorithms.
TL;DR: In this article , the authors investigate the distance measure between any two conventional type trapezoidal-valued intuitionistic fuzzy sets (CTrVIFSs) whose membership and non-membership grades of an element are expressed as conventional trapezoid intuistic fuzzy numbers.
Abstract: Abstract The article aims to investigate the distance measure between any two conventional type trapezoidal-valued intuitionistic fuzzy sets (CTrVIFSs) whose membership and non-membership grades of an element are expressed as conventional trapezoidal intuitionistic fuzzy numbers (CTrIFN). Using the proposed distance measure, the similarity measure of CTrVIFSs is determined and its efficiency is shown by applying it to pattern recognition problems and MCDM problems. The similarity measure propounded in this article can be used to tackle real-world problems involving CTrVIFS as parameters, such as clustering, machine learning, and DNA matching. The application section discusses that this research can help decision-makers to recognize patterns and categorize samples with those patterns. Furthermore, the model of a real-world problem is given which utilizes the suggested similarity measure to solve MCDM problems, demonstrate the usability of the new technique and comprehend its applied intelligence above other methods. Finally, a general conclusion and future scope on this topic are discussed.
TL;DR: In this paper , a real-time tracking control problem for an autonomous underwater vehicle based on an acoustic-based positioning method, i.e., the so-called GPS intelligent buoy system, which causes inevitable measurement delay.
Abstract: Abstract This paper deals with the real-time tracking control problem for an autonomous underwater vehicle based on an acoustic-based positioning method, i.e., the so-called GPS intelligent buoy system, which causes inevitable measurement delay. The measurement delay increases the control difficulty and degrades the tracking accuracy. Additionally, the exact modeling for an autonomous underwater vehicle is difficult due to uncertain hydrodynamic parameters. Based on these findings, a model-free control scheme is proposed. In the proposed scheme, the GPS intelligent buoy system provides the position signals without velocity measurements. Considering the measurement noise, a robust exact differentiator is used instead of the traditional numerical differentiation method to obtain the derivatives of position signals, which saves the limited actuator energy of autonomous underwater vehicles. Simulations are performed to verify the validity of the proposed control scheme. The results demonstrate that the proposed control scheme can achieve high timeliness and high tracking accuracy for autonomous underwater vehicles. Compared to the conventional model predictive control, the proposed controller requires 89.7% less average calculation time. In addition, the proposed controller outperforms the conventional proportion-differentiation controller in root-mean-square error by approximately 62.3−80.7%.