PeerJ Computer Science最新文献

{"title":"Prediction of stock price direction using the LASSO-LSTM model combines technical indicators and financial sentiment analysis.","authors":"Junwen Yang,&nbsp;Yunmin Wang,&nbsp;Xiang Li","doi":"10.7717/peerj-cs.1148","DOIUrl":"https://doi.org/10.7717/peerj-cs.1148","url":null,"abstract":"<p><p>Correctly predicting the stock price movement direction is of immense importance in the financial market. In recent years, with the expansion of dimension and volume in data, the nonstationary and nonlinear characters in finance data make it difficult to predict stock movement accurately. In this article, we propose a methodology that combines technical analysis and sentiment analysis to construct predictor variables and then apply the improved LASSO-LASSO to forecast stock direction. First, the financial textual content and stock historical transaction data are crawled from websites. Then transfer learning Finbert is used to recognize the emotion of textual data and the TTR package is taken to calculate the technical indicators based on historical price data. To eliminate the multi-collinearity of predictor variables after combination, we improve the long short-term memory neural network (LSTM) model with the Absolute Shrinkage and Selection Operator (LASSO). In predict phase, we apply the variables screened as the input vector to train the LASSO-LSTM model. To evaluate the model performance, we compare the LASSO-LSTM and baseline models on accuracy and robustness metrics. In addition, we introduce the Wilcoxon signed rank test to evaluate the difference in results. The experiment result proves that the LASSO-LSTM with technical and sentiment indicators has an average 8.53% accuracy improvement than standard LSTM. Consequently, this study proves that utilizing historical transactions and financial sentiment data can capture critical information affecting stock movement. Also, effective variable selection can retain the key variables and improve the model prediction performance.</p>","PeriodicalId":503620,"journal":{"name":"PeerJ Computer Science","volume":" ","pages":"e1148"},"PeriodicalIF":3.8,"publicationDate":"2022-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9680880/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40722404","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The proposal of a modeling methodology for an industrial internet information model.","authors":"Sicong Yu,&nbsp;Ying Huang,&nbsp;Tao Du,&nbsp;Yinglei Teng","doi":"10.7717/peerj-cs.1150","DOIUrl":"https://doi.org/10.7717/peerj-cs.1150","url":null,"abstract":"<p><p>With the large distributed, autonomous, diverse, and dynamic information sources generated in the Industrial Internet area, the information model becomes the critical technology for heterogeneous data interoperability. By establishing unified architecture, mutually agreed communication protocols and standardizing syntax and semantics, the potential of complex data can be released. However, most of the existing information models are isolated in the professional fields, and the interoperability and scope of standards are very limited. In this article, we design a uniform information model for the Industrial Internet, and present a general modeling method which aims to build a standardized organizational framework of information. Specifically, the Industrial Internet information model is first defined, where the seven key elements and value evaluation are devised for information extraction. Then, an optimization approach combining entropy and semantic distance theories is proposed that determines the information organization granularity. Next, as the cross-layer interaction of complex information is very tricky in a tree structure and its modeling cost is extremely high in a mesh topology, the underground root structure is invented for model representation. Finally, the modeling methodology is applied to the ordinary and precision machine tools demonstrating 18.75% and 18.18% modeling cost reduction, respectively, and these two information models are further implemented in a digital machining workshop to verify the effectiveness of the proposed modeling method.</p>","PeriodicalId":503620,"journal":{"name":"PeerJ Computer Science","volume":" ","pages":"e1150"},"PeriodicalIF":3.8,"publicationDate":"2022-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9680887/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40706893","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effect of visual imagery in COVID-19 social media posts on users' perception.","authors":"Waleed M Al-Nuwaiser","doi":"10.7717/peerj-cs.1153","DOIUrl":"https://doi.org/10.7717/peerj-cs.1153","url":null,"abstract":"<p><p>People receive a wide variety of news from social media. They especially look for information on social media in times of crisis with the desire to assess the risk they face. This risk assessment, and other aspects of user reactions, may be affected by characteristics of the social media post relaying certain information. Thus, it is critical to understand these characteristics to deliver information with the reactions in mind. This study investigated various types of imagery used as thumbnails in social media posts regarding news about the COVID-19 pandemic. In an experimental design, 300 participants viewed social media posts containing each of the three types of imagery: data visualization (directly about risk information), advisory (not containing direct risk information, but instead help on how to lower risk), or clickbait (containing no risk-related information, just generic visuals). After observing the social media posts, they answered questionnaires measuring their emotions (valence, arousal, and dominance), risk perception, perceived credibility of the post, and engagement. The participants also indicated their emotions towards and interest in COVID-19 news coverage, age, gender, and how often and actively they use social media. These variables acted as controls. The data were analysed using mixed linear models. Results indicated that advisory imagery positively influenced valence, arousal, dominance, credibility, and (lower) risk perception. Alternatively, imagery showing data visualizations yielded low valence, arousal, dominance, credibility, and high risk perception. Clickbait-styled thumbnails which carry no information are usually measured between the other two types. The type of imagery did not affect the motivation to engage with a post. Aside from visual imagery, most variables were affected by COVID sentiment and the usual activity on social media. These study results indicate that one should use advisory imagery for more comfortable news delivery and data visualization when the poster wishes to warn users of existing risks.</p>","PeriodicalId":503620,"journal":{"name":"PeerJ Computer Science","volume":" ","pages":"e1153"},"PeriodicalIF":3.8,"publicationDate":"2022-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9680878/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40722402","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Decoupling facial motion features and identity features for micro-expression recognition.","authors":"Tingxuan Xie,&nbsp;Guoquan Sun,&nbsp;Hao Sun,&nbsp;Qiang Lin,&nbsp;Xianye Ben","doi":"10.7717/peerj-cs.1140","DOIUrl":"https://doi.org/10.7717/peerj-cs.1140","url":null,"abstract":"<p><strong>Background: </strong>Micro-expression is a kind of expression produced by people spontaneously and unconsciously when receiving stimulus. It has the characteristics of low intensity and short duration. Moreover, it cannot be controlled and disguised. Thus, micro-expression can objectively reflect people's real emotional states. Therefore, automatic recognition of micro-expressions can help machines better understand the users' emotion, which can promote human-computer interaction. What's more, micro-expression recognition has a wide range of applications in fields like security systems and psychological treatment. Nowadays, thanks to the development of artificial intelligence, most micro-expression recognition algorithms are based on deep learning. The features extracted by deep learning model from the micro-expression video sequences mainly contain facial motion feature information and identity feature information. However, in micro-expression recognition tasks, the motions of facial muscles are subtle. As a result, the recognition can be easily interfered by identity feature information.</p><p><strong>Methods: </strong>To solve the above problem, a micro-expression recognition algorithm which decouples facial motion features and identity features is proposed in this paper. A Micro-Expression Motion Information Features Extraction Network (MENet) and an Identity Information Features Extraction Network (IDNet) are designed. By adding a Diverse Attention Operation (DAO) module and constructing divergence loss function in MENet, facial motion features can be effectively extracted. Global attention operations are used in IDNet to extract identity features. A Mutual Information Neural Estimator (MINE) is utilized to decouple facial motion features and identity features, which can help the model obtain more discriminative micro-expression features.</p><p><strong>Results: </strong>Experiments on the SDU, MMEW, SAMM and CASME II datasets were conducted, which achieved competitive results and proved the superiority of the proposed algorithm.</p>","PeriodicalId":503620,"journal":{"name":"PeerJ Computer Science","volume":" ","pages":"e1140"},"PeriodicalIF":3.8,"publicationDate":"2022-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9680898/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40722408","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting no-shows for dental appointments.","authors":"Yazeed Alabdulkarim,&nbsp;Mohammed Almukaynizi,&nbsp;Abdulmajeed Alameer,&nbsp;Bassil Makanati,&nbsp;Riyadh Althumairy,&nbsp;Abdulaziz Almaslukh","doi":"10.7717/peerj-cs.1147","DOIUrl":"https://doi.org/10.7717/peerj-cs.1147","url":null,"abstract":"<p><p>Patient no-shows is a significant problem in healthcare, reaching up to 80% of booked appointments and costing billions of dollars. Predicting no-shows for individual patients empowers clinics to implement better mitigation strategies. Patients' no-show behavior varies across health clinics and the types of appointments, calling for fine-grained studies to uncover these variations in no-show patterns. This article focuses on dental appointments because they are notably longer than regular medical appointments due to the complexity of dental procedures. We leverage machine learning techniques to develop predictive models for dental no-shows, with the best model achieving an Area Under the Curve (AUC) of 0.718 and an F1 score of 66.5%. Additionally, we propose and evaluate a novel method to represent no-show history as a binary sequence of events, enabling the predictive models to learn the associated future no-show behavior with these patterns. We discuss the utility of no-show predictions to improve the scheduling of dental appointments, such as reallocating appointments and reducing their duration.</p>","PeriodicalId":503620,"journal":{"name":"PeerJ Computer Science","volume":" ","pages":"e1147"},"PeriodicalIF":3.8,"publicationDate":"2022-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9680883/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40705482","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Lightweight multi-scale network for small object detection.","authors":"Li Li,&nbsp;Bingxue Li,&nbsp;Hongjuan Zhou","doi":"10.7717/peerj-cs.1145","DOIUrl":"https://doi.org/10.7717/peerj-cs.1145","url":null,"abstract":"<p><p>Small object detection is widely used in the real world. Detecting small objects in complex scenes is extremely difficult as they appear with low resolution. At present, many studies have made significant progress in improving the detection accuracy of small objects. However, some of them cannot balance the detection speed and accuracy well. To solve the above problems, a lightweight multi-scale network (LMSN) was proposed to exploit the multi-scale information in this article. Firstly, it explicitly modeled semantic information interactions at every scale via a multi-scale feature fusion unit. Secondly, the feature extraction capability of the network was intensified by a lightweight receptive field enhancement module. Finally, an efficient channel attention module was employed to enhance the feature representation capability. To validate our proposed network, we implemented extensive experiments on two benchmark datasets. The mAP of LMSN achieved 75.76% and 89.32% on PASCAL VOC and RSOD datasets, respectively, which is 5.79% and 11.14% higher than MobileNetv2-SSD. Notably, its inference speed was up to 61 FPS and 64 FPS, respectively. The experimental results confirm the validity of LMSN for small object detection.</p>","PeriodicalId":503620,"journal":{"name":"PeerJ Computer Science","volume":" ","pages":"e1145"},"PeriodicalIF":3.8,"publicationDate":"2022-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9680894/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40706829","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Feature selection enhancement and feature space visualization for speech-based emotion recognition.","authors":"Sofia Kanwal,&nbsp;Sohail Asghar,&nbsp;Hazrat Ali","doi":"10.7717/peerj-cs.1091","DOIUrl":"https://doi.org/10.7717/peerj-cs.1091","url":null,"abstract":"<p><p>Robust speech emotion recognition relies on the quality of the speech features. We present speech features enhancement strategy that improves speech emotion recognition. We used the INTERSPEECH 2010 challenge feature-set. We identified subsets from the features set and applied principle component analysis to the subsets. Finally, the features are fused horizontally. The resulting feature set is analyzed using t-distributed neighbour embeddings (t-SNE) before the application of features for emotion recognition. The method is compared with the state-of-the-art methods used in the literature. The empirical evidence is drawn using two well-known datasets: Berlin Emotional Speech Dataset (EMO-DB) and Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) for two languages, German and English, respectively. Our method achieved an average recognition gain of 11.5% for six out of seven emotions for the EMO-DB dataset, and 13.8% for seven out of eight emotions for the RAVDESS dataset as compared to the baseline study.</p>","PeriodicalId":503620,"journal":{"name":"PeerJ Computer Science","volume":" ","pages":"e1091"},"PeriodicalIF":3.8,"publicationDate":"2022-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9680882/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40722407","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"eGAC3D: enhancing depth adaptive convolution and depth estimation for monocular 3D object pose detection.","authors":"Duc Tuan Ngo,&nbsp;Minh-Quan Viet Bui,&nbsp;Duc Dung Nguyen,&nbsp;Hoang-Anh Pham","doi":"10.7717/peerj-cs.1144","DOIUrl":"https://doi.org/10.7717/peerj-cs.1144","url":null,"abstract":"<p><p>Many alternative approaches for 3D object detection using a singular camera have been studied instead of leveraging high-precision 3D LiDAR sensors incurring a prohibitive cost. Recently, we proposed a novel approach for 3D object detection by employing a ground plane model that utilizes geometric constraints named GAC3D to improve the results of the deep-based detector. GAC3D adopts an adaptive depth convolution to replace the traditional 2D convolution to deal with the divergent context of the image's feature, leading to a significant improvement in both training convergence and testing accuracy on the KITTI 3D object detection benchmark. This article presents an alternative architecture named eGAC3D that adopts a revised depth adaptive convolution with variant guidance to improve detection accuracy. Additionally, eGAC3D utilizes the pixel adaptive convolution to leverage the depth map to guide our model for detection heads instead of using an external depth estimator like other methods leading to a significant reduction of time inference. The experimental results on the KITTI benchmark show that our eGAC3D outperforms not only our previous GAC3D but also many existing monocular methods in terms of accuracy and inference time. Moreover, we deployed and optimized the proposed eGAC3D framework on an embedded platform with a low-cost GPU. To the best of the authors' knowledge, we are the first to develop a monocular 3D detection framework on embedded devices. The experimental results on Jetson Xavier NX demonstrate that our proposed method can achieve nearly real-time performance with appropriate accuracy even with the modest hardware resource.</p>","PeriodicalId":503620,"journal":{"name":"PeerJ Computer Science","volume":" ","pages":"e1144"},"PeriodicalIF":3.8,"publicationDate":"2022-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9680869/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40705483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Computer model of a spectrum analyzer for a virtual laboratory: development and introduction to the educational process.","authors":"Dmitry Gubsky,&nbsp;Yevgeniya Daineko,&nbsp;Madina Ipalakova,&nbsp;Anatoly Kleschenkov,&nbsp;Dana Tsoy","doi":"10.7717/peerj-cs.1130","DOIUrl":"https://doi.org/10.7717/peerj-cs.1130","url":null,"abstract":"<p><p>The worldwide COVID-19 pandemic has changed the development plans of every country. Instead, governments had to constantly deal with ever-emerging issues in healthcare, education, economics and industry. As a result, there has been an accelerated introduction of digitalization in these spheres. Thus, an increasing number of people have started using electronic services that have improved their digital literacy. This feature has had a positive impact on society and helped to create new interaction tools between populations and governments, students and institutions, customers and companies. The article aims to analyze how studying radio electronics can be improved by involving new tools and how they can be applied in distance learning. This work presents the results of the development and application of a virtual radio signal simulation in the educational process in the form of a laboratory practicum. Working on this approach required specific research in the field; the foreign experience was observed and studied. The review allowed us to find out how digitalization and the application of digital tools affect the behaviour, cognition, and overall performance of students during the pandemic. The authors conducted a questionnaire among students to evaluate the features of the virtual laboratory work and their effect on the educational process. The results analyzed are given in the article. They showed that students highly appreciate the introduction of such tools in learning. Moreover, like the entire laboratory, the proposed model can be used in the educational process offline and with distance learning. Finally, the article describes the experience and results of the software package's development and integration for the spectrum analyzer's computer model and virtual laboratory work using the MS VS environment in C ++. The results of the conducted work demonstrate the versatility of the proposed approach, its positive impact on the educational process, high potential in the other spheres.</p>","PeriodicalId":503620,"journal":{"name":"PeerJ Computer Science","volume":" ","pages":"e1130"},"PeriodicalIF":3.8,"publicationDate":"2022-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9680885/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40706892","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Relational graph convolutional networks: a closer look.","authors":"Thiviyan Thanapalasingam,&nbsp;Lucas van Berkel,&nbsp;Peter Bloem,&nbsp;Paul Groth","doi":"10.7717/peerj-cs.1073","DOIUrl":"https://doi.org/10.7717/peerj-cs.1073","url":null,"abstract":"<p><p>In this article, we describe a reproduction of the Relational Graph Convolutional Network (RGCN). Using our reproduction, we explain the intuition behind the model. Our reproduction results empirically validate the correctness of our implementations using benchmark Knowledge Graph datasets on node classification and link prediction tasks. Our explanation provides a friendly understanding of the different components of the RGCN for both users and researchers extending the RGCN approach. Furthermore, we introduce two new configurations of the RGCN that are more parameter efficient. The code and datasets are available at https://github.com/thiviyanT/torch-rgcn.</p>","PeriodicalId":503620,"journal":{"name":"PeerJ Computer Science","volume":" ","pages":"e1073"},"PeriodicalIF":3.8,"publicationDate":"2022-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9680895/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40705481","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}