{"title":"Guest editorial - Parallel and distributed computing and applications","authors":"Hong Shen, Hui Tian, Yingpeng Sang","doi":"10.2298/csis230100ixs","DOIUrl":"https://doi.org/10.2298/csis230100ixs","url":null,"abstract":"<jats:p>nema</jats:p>","PeriodicalId":50636,"journal":{"name":"Computer Science and Information Systems","volume":"1 1","pages":"ix"},"PeriodicalIF":1.4,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82766606","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
State of the art formal verification is based on formal methods and its goal is proving given correctness properties. For example, a PSTM scheduler was modeled in CSP in order to prove deadlock-freeness and starvation-freeness. However, as this paper shows, using solely formal methods is not sufficient. Therefore, in this paper we propose a complete formal verification of trustworthy software, which jointly uses formal verification and formal model testing. As an example, we first test the previous CSP model of PSTM transaction scheduler by comparing the model checker PAT results with the manually derived expected results, for the given test workloads. Next, according to the results of this testing, we correct and extend the CSP model. Finally, using PAT results for the new CSP model, we analyze the performance of the PSTM online transaction scheduling algorithms from the perspective of the relative speedup.
{"title":"Complete formal verification of the PSTM transaction Scheduler","authors":"M. Popovic, M. Popovic, B. Kordic, Huibiao Zhu","doi":"10.2298/csis210908058p","DOIUrl":"https://doi.org/10.2298/csis210908058p","url":null,"abstract":"State of the art formal verification is based on formal methods and its goal is proving given correctness properties. For example, a PSTM scheduler was modeled in CSP in order to prove deadlock-freeness and starvation-freeness. However, as this paper shows, using solely formal methods is not sufficient. Therefore, in this paper we propose a complete formal verification of trustworthy software, which jointly uses formal verification and formal model testing. As an example, we first test the previous CSP model of PSTM transaction scheduler by comparing the model checker PAT results with the manually derived expected results, for the given test workloads. Next, according to the results of this testing, we correct and extend the CSP model. Finally, using PAT results for the new CSP model, we analyze the performance of the PSTM online transaction scheduling algorithms from the perspective of the relative speedup.","PeriodicalId":50636,"journal":{"name":"Computer Science and Information Systems","volume":"42 1","pages":"307-327"},"PeriodicalIF":1.4,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76368517","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Recording anomalous traces in business processes diminishes an event log?s quality. The abnormalities may represent bad execution, security issues, or deviant behavior. Focusing on mitigating this phenomenon, organizations spend efforts to detect anomalous traces in their business processes to save resources and improve process execution. However, in many real-world environments, reference models are unavailable, requiring expert assistance and increasing costs. The con15 siderable number of techniques and reduced availability of experts pose an additional challenge for particular scenarios. In this work, we combine the representational power of encoding with a Meta-learning strategy to enhance the detection of anomalous traces in event logs towards fitting the best discriminative capability be tween common and irregular traces. Our approach creates an event log profile and recommends the most suitable encoding technique to increase the anomaly detetion performance. We used eight encoding techniques from different families, 80 log descriptors, 168 event logs, and six anomaly types for experiments. Results indicate that event log characteristics influence the representational capability of encodings. Moreover, we investigate the process behavior?s influence for choosing the suitable encoding technique, demonstrating that traditional process mining analysis can be leveraged when matched with intelligent decision support approaches.
{"title":"Matching business process behavior with encoding techniques via meta-learning: An anomaly detection study","authors":"G. Tavares, Sylvio Barbon Junior","doi":"10.2298/csis220110005t","DOIUrl":"https://doi.org/10.2298/csis220110005t","url":null,"abstract":"Recording anomalous traces in business processes diminishes an event log?s quality. The abnormalities may represent bad execution, security issues, or deviant behavior. Focusing on mitigating this phenomenon, organizations spend efforts to detect anomalous traces in their business processes to save resources and improve process execution. However, in many real-world environments, reference models are unavailable, requiring expert assistance and increasing costs. The con15 siderable number of techniques and reduced availability of experts pose an additional challenge for particular scenarios. In this work, we combine the representational power of encoding with a Meta-learning strategy to enhance the detection of anomalous traces in event logs towards fitting the best discriminative capability be tween common and irregular traces. Our approach creates an event log profile and recommends the most suitable encoding technique to increase the anomaly detetion performance. We used eight encoding techniques from different families, 80 log descriptors, 168 event logs, and six anomaly types for experiments. Results indicate that event log characteristics influence the representational capability of encodings. Moreover, we investigate the process behavior?s influence for choosing the suitable encoding technique, demonstrating that traditional process mining analysis can be leveraged when matched with intelligent decision support approaches.","PeriodicalId":50636,"journal":{"name":"Computer Science and Information Systems","volume":"91 1","pages":"1207-1233"},"PeriodicalIF":1.4,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74968710","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
While identifying specific user roles in social media -in particular bots or spammers- has seen significant progress, generic and all-encompassing user role classification remains elusive on the large data sets of today?s social media. Yet, such broad classifications enable a deeper understanding of user interactions and pave the way for longitudinal studies, capturing the evolution of users such as the rise of influencers. Studies of generic roles have been performed predominantly in a small scale, establishing fundamental role definitions, but relying mostly on ad-hoc, data set-dependent rules that need to be carefully hand-tuned. We build on those studies and provide a largely automated, scalable detection of a wide range of roles. Our approach clusters users hierarchically on salient, complementary features such as their actions, their ability to trigger reactions and their network positions. To associate these clusters with roles, we use supervised classifiers: trained on human experts on completely new media, but transferable on related data sets. Furthermore, we employ the combination of samples in order to improve scalability and allow probabilistic assignments of user roles. Our evaluation on Twitter indicates that a) stable and reliable detection of a wide range of roles is possible b) the labeling transfers well as long as the fundamental properties don?t strongly change between data sets and c) the approaches scale well with little need for human intervention.
{"title":"Detecting and analyzing fine-grained user roles in social media?","authors":"J. Kastner, Peter M. Fischer","doi":"10.2298/csis220110006k","DOIUrl":"https://doi.org/10.2298/csis220110006k","url":null,"abstract":"While identifying specific user roles in social media -in particular bots or spammers- has seen significant progress, generic and all-encompassing user role classification remains elusive on the large data sets of today?s social media. Yet, such broad classifications enable a deeper understanding of user interactions and pave the way for longitudinal studies, capturing the evolution of users such as the rise of influencers. Studies of generic roles have been performed predominantly in a small scale, establishing fundamental role definitions, but relying mostly on ad-hoc, data set-dependent rules that need to be carefully hand-tuned. We build on those studies and provide a largely automated, scalable detection of a wide range of roles. Our approach clusters users hierarchically on salient, complementary features such as their actions, their ability to trigger reactions and their network positions. To associate these clusters with roles, we use supervised classifiers: trained on human experts on completely new media, but transferable on related data sets. Furthermore, we employ the combination of samples in order to improve scalability and allow probabilistic assignments of user roles. Our evaluation on Twitter indicates that a) stable and reliable detection of a wide range of roles is possible b) the labeling transfers well as long as the fundamental properties don?t strongly change between data sets and c) the approaches scale well with little need for human intervention.","PeriodicalId":50636,"journal":{"name":"Computer Science and Information Systems","volume":"9 1","pages":"1263-1287"},"PeriodicalIF":1.4,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74751394","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mobile agents, which learn to optimize a task in real time, can adapt to dynamic environments and find the optimum locations with the navigation mechanism that includes a motion model. In this study, it is aimed to effectively cover points of interest (PoI) in a dynamic environment by modeling a group of unmanned aerial vehicles (UAVs) on the basis of a learning multi-agent system. Agents create an abstract rectangular plane containing the area to be covered, and then decompose the area into grids. An agent learns to locate on a center of grid that are closest to it, which has the largest number of PoIs to plan its path. This planning helps to achieve a high fairness index by reducing the number of common PoIs covered. The proposed method has been tested in a simulation environment and the results are presented by comparing with similar studies. The results show that the proposed method outperforms existing similar studies and is suitable for area coverage applications.
{"title":"Point of interest coverage with distributed multi-unmanned aerial vehicles on dynamic environment","authors":"Fatih Aydemir, Aydın Çetin","doi":"10.2298/csis221222037a","DOIUrl":"https://doi.org/10.2298/csis221222037a","url":null,"abstract":"Mobile agents, which learn to optimize a task in real time, can adapt to dynamic environments and find the optimum locations with the navigation mechanism that includes a motion model. In this study, it is aimed to effectively cover points of interest (PoI) in a dynamic environment by modeling a group of unmanned aerial vehicles (UAVs) on the basis of a learning multi-agent system. Agents create an abstract rectangular plane containing the area to be covered, and then decompose the area into grids. An agent learns to locate on a center of grid that are closest to it, which has the largest number of PoIs to plan its path. This planning helps to achieve a high fairness index by reducing the number of common PoIs covered. The proposed method has been tested in a simulation environment and the results are presented by comparing with similar studies. The results show that the proposed method outperforms existing similar studies and is suitable for area coverage applications.","PeriodicalId":50636,"journal":{"name":"Computer Science and Information Systems","volume":"16 1","pages":"1061-1084"},"PeriodicalIF":1.4,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72638814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The version control system of every software product can provide important information about how the system is connected. In this study, we first propose a language-independent method to collect and filter dependencies from the version control, and second, we use the results obtained in the first step to identify key classes from three software systems. To identify the key classes, we are using the dependencies extracted from the version control system together with dependencies from the source code, and also separate. Based on the results obtained we can say that compared with the results obtained by using only dependencies extracted from code, the mix between both types of dependencies provides small improvements. And, by using only dependencies from the version control system, we obtained results that did not surpass the results previously mentioned, but are still acceptable. We still consider this an important result because this might open an important opportunity for software systems that use dynamically typed languages such as JavaScript, Objective-C, Python, and Ruby, or systems that use multiple languages. These types of systems, for which the code dependencies are harder to obtain, can use the dependencies extracted from the version control to gain better knowledge about the system.
{"title":"Logical dependencies: Extraction from the versioning system and usage in key classes detection","authors":"A. Stana, Ioana Sora","doi":"10.2298/csis220518025s","DOIUrl":"https://doi.org/10.2298/csis220518025s","url":null,"abstract":"The version control system of every software product can provide important information about how the system is connected. In this study, we first propose a language-independent method to collect and filter dependencies from the version control, and second, we use the results obtained in the first step to identify key classes from three software systems. To identify the key classes, we are using the dependencies extracted from the version control system together with dependencies from the source code, and also separate. Based on the results obtained we can say that compared with the results obtained by using only dependencies extracted from code, the mix between both types of dependencies provides small improvements. And, by using only dependencies from the version control system, we obtained results that did not surpass the results previously mentioned, but are still acceptable. We still consider this an important result because this might open an important opportunity for software systems that use dynamically typed languages such as JavaScript, Objective-C, Python, and Ruby, or systems that use multiple languages. These types of systems, for which the code dependencies are harder to obtain, can use the dependencies extracted from the version control to gain better knowledge about the system.","PeriodicalId":50636,"journal":{"name":"Computer Science and Information Systems","volume":"60 1","pages":"1015-1035"},"PeriodicalIF":1.4,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85653463","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Guest editorial - Engineering of computer based systems","authors":"Miodrag Djukic, M. Popovic","doi":"10.2298/csis230100vd","DOIUrl":"https://doi.org/10.2298/csis230100vd","url":null,"abstract":"<jats:p>nema</jats:p>","PeriodicalId":50636,"journal":{"name":"Computer Science and Information Systems","volume":"32 1","pages":"v-vi"},"PeriodicalIF":1.4,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83899505","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Maria Sousa, Ana Mendes, Dora Almeida, Álvaro Rocha
The purpose of this study is to describe and analyze whether digital remote work in times of Covid-19 is influencing the satisfaction of Public Administration employees. Based on the objective of this study, an online survey was conducted in the Portuguese Public Administration, for a sample of 70 individuals, working at home due to the situation of Public Health caused by the Coronavirus. Digital remote work is being applied massively worldwide and is a specific form of work organization supported by information and knowledge. Digital remote workers carry out their activities at home and using digital technologies, depending on the nature of the tasks and work situations. To understand the satisfaction of Public Administration employees, an empirical study was carried out, supported by data collection through an online survey. The main conclusions were that despite the constraints (resistance of top management, organizational culture, autonomy, and flexibility of workers, among others) that existed before the health and socioeconomic crisis caused by the Coronavirus pandemic, digital remote work is a given in the life of organizations, public or private, and of workers with reflection at various levels in society and particularly in the professional fulfillment and satisfaction of employees. According to the analysis carried out on the data collected to support the conclusions of this study, the degree of satisfaction of Public Administration employees is influenced in different ways by the influencing factors studied: autonomy at work, conditions at work, and income. However, regarding the factor of quality of life at work, this link has not been established. Thus, it was possible to conclude that satisfaction increases positively and strongly with autonomy at work. Technological specialization and productivity still have a positive influence, but with low intensity contribute to the satisfaction of AP employees. Working conditions also negatively influence satisfaction, although at an average intensity. However, the average degree of job satisfaction varies according to the different age groups, with employees aged 35 or more having a higher satisfaction average than employees whose ages vary between 34 and the beginning of their working lives.
{"title":"Digital remote work influencing public administration employees satisfaction in public health complex contexts","authors":"Maria Sousa, Ana Mendes, Dora Almeida, Álvaro Rocha","doi":"10.2298/csis230110060s","DOIUrl":"https://doi.org/10.2298/csis230110060s","url":null,"abstract":"The purpose of this study is to describe and analyze whether digital remote work in times of Covid-19 is influencing the satisfaction of Public Administration employees. Based on the objective of this study, an online survey was conducted in the Portuguese Public Administration, for a sample of 70 individuals, working at home due to the situation of Public Health caused by the Coronavirus. Digital remote work is being applied massively worldwide and is a specific form of work organization supported by information and knowledge. Digital remote workers carry out their activities at home and using digital technologies, depending on the nature of the tasks and work situations. To understand the satisfaction of Public Administration employees, an empirical study was carried out, supported by data collection through an online survey. The main conclusions were that despite the constraints (resistance of top management, organizational culture, autonomy, and flexibility of workers, among others) that existed before the health and socioeconomic crisis caused by the Coronavirus pandemic, digital remote work is a given in the life of organizations, public or private, and of workers with reflection at various levels in society and particularly in the professional fulfillment and satisfaction of employees. According to the analysis carried out on the data collected to support the conclusions of this study, the degree of satisfaction of Public Administration employees is influenced in different ways by the influencing factors studied: autonomy at work, conditions at work, and income. However, regarding the factor of quality of life at work, this link has not been established. Thus, it was possible to conclude that satisfaction increases positively and strongly with autonomy at work. Technological specialization and productivity still have a positive influence, but with low intensity contribute to the satisfaction of AP employees. Working conditions also negatively influence satisfaction, although at an average intensity. However, the average degree of job satisfaction varies according to the different age groups, with employees aged 35 or more having a higher satisfaction average than employees whose ages vary between 34 and the beginning of their working lives.","PeriodicalId":50636,"journal":{"name":"Computer Science and Information Systems","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135402202","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
C. Marco-Detchart, J.A. Rincon, C. Carrascosa, V. Julian
In recent years, several proposals have been based on Artificial Intelligence techniques for automatically detecting the presence of pests and diseases in crops from images usually taken with a camera. By training with pictures of affected crops and healthy crops, artificial intelligence techniques learn to distinguish one from the other. Furthermore, in the long term, it is intended that the tools developed from such approaches will allow the automation and increased frequency of plant analysis, thus increasing the possibility of determining and predicting crop health and potential biotic risks. However, the great diversity of proposed solutions leads us to the need to study them, present possible situations for their improvement, such as image preprocessing, and analyse the robustness of the proposals examined against more realistic pictures than those existing in the datasets typically used. Taking all this into account, this paper embarks on a comprehensive exploration of various AI techniques leveraging leaf images for the autonomous detection of plant diseases. By fostering a deeper understanding of the strengths and limitations of these methodologies, this research contributes to the vanguard of agricultural disease detection, propelling innovation, and fostering the maturation of AI-driven solutions in this critical domain.
{"title":"Evaluation of deep learning techniques for plant disease detection","authors":"C. Marco-Detchart, J.A. Rincon, C. Carrascosa, V. Julian","doi":"10.2298/csis221222073m","DOIUrl":"https://doi.org/10.2298/csis221222073m","url":null,"abstract":"In recent years, several proposals have been based on Artificial Intelligence techniques for automatically detecting the presence of pests and diseases in crops from images usually taken with a camera. By training with pictures of affected crops and healthy crops, artificial intelligence techniques learn to distinguish one from the other. Furthermore, in the long term, it is intended that the tools developed from such approaches will allow the automation and increased frequency of plant analysis, thus increasing the possibility of determining and predicting crop health and potential biotic risks. However, the great diversity of proposed solutions leads us to the need to study them, present possible situations for their improvement, such as image preprocessing, and analyse the robustness of the proposals examined against more realistic pictures than those existing in the datasets typically used. Taking all this into account, this paper embarks on a comprehensive exploration of various AI techniques leveraging leaf images for the autonomous detection of plant diseases. By fostering a deeper understanding of the strengths and limitations of these methodologies, this research contributes to the vanguard of agricultural disease detection, propelling innovation, and fostering the maturation of AI-driven solutions in this critical domain.","PeriodicalId":50636,"journal":{"name":"Computer Science and Information Systems","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135445068","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In light of advancements in information technology and the widespread impact of the COVID-19 pandemic, consumer behavior has undergone a significant transformation, shifting from traditional in-store shopping to the realm of online retailing. This shift has notably accelerated the growth of the online retail sector. An essential advantage offered by e-commerce lies in its ability to accumulate and analyze user data, encompassing browsing and purchase histories, through its recommendation systems. Nevertheless, prevailing methodologies predominantly rely on historical user data, which often lack the dynamism required to comprehend immediate user responses and emotional states during online interactions. Recognizing the substantial influence of visual stimuli on human perception, this study leverages eye-tracking technology to investigate online consumer behavior. The research captures the visual engagement of 60 healthy participants while they engage in online shopping, while also taking note of their preferred items for purchase. Subsequently, we apply statistical analysis and machine learning models to unravel the impact of visual complexity, consumer considerations, and preferred items, thereby providing valuable insights for the design of e-commerce platforms. Our findings indicate that the integration of eye-tracking data into e-commerce recommendation systems is conducive to enhancing their performance. Furthermore, machine learning algorithms exhibited remarkable classification capabilities when combined with eye-tracking data. Notably, during the purchase of hedonic products, participants primarily fixated on product images, whereas for utilitarian products, equal attention was dedicated to images, prices, reviews, and sales volume. These insights hold significant potential to augment the effectiveness of e-commerce marketing endeavors.
{"title":"Machine learning based approach for exploring online shopping behavior and preferences with eye tracking","authors":"Zhenyao Liu, Wei-Chang Yeh, Ke-Yun Lin, Chia-Sheng Lin, Chuan-Yu Chang","doi":"10.2298/csis230807077l","DOIUrl":"https://doi.org/10.2298/csis230807077l","url":null,"abstract":"In light of advancements in information technology and the widespread impact of the COVID-19 pandemic, consumer behavior has undergone a significant transformation, shifting from traditional in-store shopping to the realm of online retailing. This shift has notably accelerated the growth of the online retail sector. An essential advantage offered by e-commerce lies in its ability to accumulate and analyze user data, encompassing browsing and purchase histories, through its recommendation systems. Nevertheless, prevailing methodologies predominantly rely on historical user data, which often lack the dynamism required to comprehend immediate user responses and emotional states during online interactions. Recognizing the substantial influence of visual stimuli on human perception, this study leverages eye-tracking technology to investigate online consumer behavior. The research captures the visual engagement of 60 healthy participants while they engage in online shopping, while also taking note of their preferred items for purchase. Subsequently, we apply statistical analysis and machine learning models to unravel the impact of visual complexity, consumer considerations, and preferred items, thereby providing valuable insights for the design of e-commerce platforms. Our findings indicate that the integration of eye-tracking data into e-commerce recommendation systems is conducive to enhancing their performance. Furthermore, machine learning algorithms exhibited remarkable classification capabilities when combined with eye-tracking data. Notably, during the purchase of hedonic products, participants primarily fixated on product images, whereas for utilitarian products, equal attention was dedicated to images, prices, reviews, and sales volume. These insights hold significant potential to augment the effectiveness of e-commerce marketing endeavors.","PeriodicalId":50636,"journal":{"name":"Computer Science and Information Systems","volume":"156 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135445973","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：