Pub Date : 2024-01-01DOI: 10.1016/j.procs.2024.09.458
Nidhal Balti , Baptiste Vrigneau , Pascal Scalart
Non-Intrusive Load Monitoring (NILM) attempts to break down the aggregated electrical consumption signal into the power consumption of each individual appliance, which can provide helpful understanding on energy consumption patterns and helps reduce overall energy usage and costs. This paper proposes an occupancy-aided energy disaggregation approach to address the NILM problem. Our methodology encompasses three key steps: firstly, features extraction from environmental sensors through the training of a DAE model; secondly, inference of occupancy information using the K-means algorithm; and finally, the disaggregation process using a Recurrent Neural Network (RNN) model, incorporating the detected occupancy status alongside power data. Experiments conducted on our real-world dataset demonstrate that our method significantly outperforms the state-of-the-art models while having good generalization capacity, achieving roughly 40% Mean Absolute Error (MAE) gain and 30% Root Mean Squared Error (RMSE) gain on a specific appliances disaggregation compared to the conventional NILM approach where only the power data is used.
{"title":"Occupancy detection for enhanced energy disaggregation","authors":"Nidhal Balti , Baptiste Vrigneau , Pascal Scalart","doi":"10.1016/j.procs.2024.09.458","DOIUrl":"10.1016/j.procs.2024.09.458","url":null,"abstract":"<div><div>Non-Intrusive Load Monitoring (NILM) attempts to break down the aggregated electrical consumption signal into the power consumption of each individual appliance, which can provide helpful understanding on energy consumption patterns and helps reduce overall energy usage and costs. This paper proposes an occupancy-aided energy disaggregation approach to address the NILM problem. Our methodology encompasses three key steps: firstly, features extraction from environmental sensors through the training of a DAE model; secondly, inference of occupancy information using the K-means algorithm; and finally, the disaggregation process using a Recurrent Neural Network (RNN) model, incorporating the detected occupancy status alongside power data. Experiments conducted on our real-world dataset demonstrate that our method significantly outperforms the state-of-the-art models while having good generalization capacity, achieving roughly 40% Mean Absolute Error (MAE) gain and 30% Root Mean Squared Error (RMSE) gain on a specific appliances disaggregation compared to the conventional NILM approach where only the power data is used.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"246 ","pages":"Pages 529-537"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142746307","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}
Traffic safety measures are essential for addressing traffic accident clusters, which denote specific locations and times at which traffic accidents are prone. This study introduces a novel approach for identifying these clusters through hypothesis testing of the spatiotemporal network of previous traffic accidents. The experimental results across various regions in Aichi Prefecture are more accurate than those of a previous method for detecting these clusters.
{"title":"Cluster Detection for Traffic Accidents on Spatiotemporal Networks","authors":"Keisuke Ando , Yusuke Kuniyoshi , Natsuki Onogi , Takeshi Uchitane , Naoto Mukai , Kazunori Iwata , Nobuhiro Ito , Yong Jiang","doi":"10.1016/j.procs.2024.09.416","DOIUrl":"10.1016/j.procs.2024.09.416","url":null,"abstract":"<div><div>Traffic safety measures are essential for addressing traffic accident clusters, which denote specific locations and times at which traffic accidents are prone. This study introduces a novel approach for identifying these clusters through hypothesis testing of the spatiotemporal network of previous traffic accidents. The experimental results across various regions in Aichi Prefecture are more accurate than those of a previous method for detecting these clusters.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"246 ","pages":"Pages 371-380"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142746393","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}
Pub Date : 2024-01-01DOI: 10.1016/j.procs.2024.09.236
Alicja Fandrejewska , Witold Chmielarz , Marek Zborowski
The primary objective of this article is to attempt to assess the role of ICT in shaping more environmentally friendly and sustainable consumer behaviour in buying, selling, exchanging and recycling clothes, shoes and accessories. The research questions were related to the impact of ICT technologies, particularly mobile applications and websites offering vintage items, on the development of sustainable behaviour related to online shopping and increasing awareness regarding environmentally friendly consumer behaviour. This research problem is particularly relevant given the economic, energy, political and environmental crises in selected countries: Poland, Spain and Turkey. The comparison was made at the turn of December 2022 and January 2023, in the situation of a deepening economic crisis in the examined countries, stimulated by progressive political crises, inflation and the war in Ukraine. The data were collected using a survey distributed using CAWI (Computer-Assisted Website Interview) technique on the servers of the Faculty of Management of the University of Warsaw. In total, there were 353 respondents who completed the entire survey, out of which 114 survey participants were from Poland, 151 from Türkiye and 88 from Spain. The limitation of the work was that the survey was conducted with the participation of the representatives of an academic environment and the results were used for a comparison of only three selected countries. Nevertheless, its innovation consists in the fact that the authors compare, for the first time in the literature, the opinions of respondents on Internet applications concerning supporting and shaping more environmentally friendly and sustainable consumer behaviour in culturally diverse countries, under different crisis conditions. The results of the conducted survey concerning clothing and footwear consumption can be useful for practitioners to point to possible strategies to raise consumer awareness and shape more sustainable consumer behaviour in culturally diverse countries located in different geographical regions using ICT technologies.
{"title":"Internet Applications for Vintage Shopping Supporting Sustainable Development: A Comparative Analysis in Selected Countries","authors":"Alicja Fandrejewska , Witold Chmielarz , Marek Zborowski","doi":"10.1016/j.procs.2024.09.236","DOIUrl":"10.1016/j.procs.2024.09.236","url":null,"abstract":"<div><div>The primary objective of this article is to attempt to assess the role of ICT in shaping more environmentally friendly and sustainable consumer behaviour in buying, selling, exchanging and recycling clothes, shoes and accessories. The research questions were related to the impact of ICT technologies, particularly mobile applications and websites offering vintage items, on the development of sustainable behaviour related to online shopping and increasing awareness regarding environmentally friendly consumer behaviour. This research problem is particularly relevant given the economic, energy, political and environmental crises in selected countries: Poland, Spain and Turkey. The comparison was made at the turn of December 2022 and January 2023, in the situation of a deepening economic crisis in the examined countries, stimulated by progressive political crises, inflation and the war in Ukraine. The data were collected using a survey distributed using CAWI (Computer-Assisted Website Interview) technique on the servers of the Faculty of Management of the University of Warsaw. In total, there were 353 respondents who completed the entire survey, out of which 114 survey participants were from Poland, 151 from Türkiye and 88 from Spain. The limitation of the work was that the survey was conducted with the participation of the representatives of an academic environment and the results were used for a comparison of only three selected countries. Nevertheless, its innovation consists in the fact that the authors compare, for the first time in the literature, the opinions of respondents on Internet applications concerning supporting and shaping more environmentally friendly and sustainable consumer behaviour in culturally diverse countries, under different crisis conditions. The results of the conducted survey concerning clothing and footwear consumption can be useful for practitioners to point to possible strategies to raise consumer awareness and shape more sustainable consumer behaviour in culturally diverse countries located in different geographical regions using ICT technologies.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"246 ","pages":"Pages 138-150"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142746648","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}
Pub Date : 2024-01-01DOI: 10.1016/j.procs.2024.09.413
Haruhiko Kaiya , Yuuki Koga , Soichiro Mori , Shinpei Ogata , Hiroyuki Nakagawa , Hironori Takeuchi
Processes of some business or life activities are sometimes changed due to some reasons, such as the emergence of new technologies and the change of the human behavior caused by a seasonal event, e.g. Christmas. Such changes are called concept drift. Detecting concept drift is useful for many reasons. For example, we can update existing out-of-date business rules. Many methods of concept drift detection in processes have been already proposed. However, most of them are a little bit complex because sliding widows should be defined on a log of business process during its analysis. We thus propose a light-weight method for its detection by using heuristic miner, which is a famous algorithm for process discovery. In our method, we simple observe the discovered model to identify the infrequent actions and transitions between actions. Our method helps us to identify several types of concept drift although some types cannot be detected. We discuss how to overcome current limitations of our method.
{"title":"A Light-Weight Method of Concept Drift Detection using Heuristic Miner","authors":"Haruhiko Kaiya , Yuuki Koga , Soichiro Mori , Shinpei Ogata , Hiroyuki Nakagawa , Hironori Takeuchi","doi":"10.1016/j.procs.2024.09.413","DOIUrl":"10.1016/j.procs.2024.09.413","url":null,"abstract":"<div><div>Processes of some business or life activities are sometimes changed due to some reasons, such as the emergence of new technologies and the change of the human behavior caused by a seasonal event, e.g. Christmas. Such changes are called concept drift. Detecting concept drift is useful for many reasons. For example, we can update existing out-of-date business rules. Many methods of concept drift detection in processes have been already proposed. However, most of them are a little bit complex because sliding widows should be defined on a log of business process during its analysis. We thus propose a light-weight method for its detection by using heuristic miner, which is a famous algorithm for process discovery. In our method, we simple observe the discovered model to identify the infrequent actions and transitions between actions. Our method helps us to identify several types of concept drift although some types cannot be detected. We discuss how to overcome current limitations of our method.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"246 ","pages":"Pages 343-352"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142746725","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}
Pub Date : 2024-01-01DOI: 10.1016/j.procs.2024.09.515
Ahmed Faris Alsayyad , Alaa Abid Muslam Abid Ali , Mohamed Mabrouk , Ahmed Al-Shammari , Mounir Zrigui
The difficulties about user security and privacy have appeared as significant concerns in recent years. The number of cyber-attacks grows at a concerning velocity, hence rendering internet users susceptible to malicious activities perpetrated by hackers. Data mining approaches are employed to extract accurate results from massive and complex databases. Furthermore, the utilization of Blockchain (BC) approaches is increasingly popular in current Internet of Things (IoT) applications as an opportunity to address issues related to privacy and security. Lots of studies have been performed on algorithms for data mining and techniques concerning blockchain. Time series data is a commonly used form of data. Time Series Classification (TSC) refers to the creation of predictive models that generate a target variable or label based on linear or sequential data inputs across a considerable duration. The possible results may be presented in either ordinal or numerical form. Even so, previous studies have shown major limitations when it comes to handling privacy and security issues that can’t be applicable in dynamic instances, as well as the substantial computational cost necessary. Moreover, correctly determining the amount of sensitive parameters required to complete the classification process remains a challenge. We have put forth a comprehensive survey on the classification of blockchain data. In the first phase of our study, we conducted an analysis and categorization of both conventional data classification approaches and contemporary time series data classification techniques. We further discussed limitations and strengths of existing techniques. Finally, we highlight future research problems and directions.
{"title":"A Survey on Time Series Data Classification: Blockchain Technologies and Security Concerns","authors":"Ahmed Faris Alsayyad , Alaa Abid Muslam Abid Ali , Mohamed Mabrouk , Ahmed Al-Shammari , Mounir Zrigui","doi":"10.1016/j.procs.2024.09.515","DOIUrl":"10.1016/j.procs.2024.09.515","url":null,"abstract":"<div><div>The difficulties about user security and privacy have appeared as significant concerns in recent years. The number of cyber-attacks grows at a concerning velocity, hence rendering internet users susceptible to malicious activities perpetrated by hackers. Data mining approaches are employed to extract accurate results from massive and complex databases. Furthermore, the utilization of Blockchain (BC) approaches is increasingly popular in current Internet of Things (IoT) applications as an opportunity to address issues related to privacy and security. Lots of studies have been performed on algorithms for data mining and techniques concerning blockchain. Time series data is a commonly used form of data. Time Series Classification (TSC) refers to the creation of predictive models that generate a target variable or label based on linear or sequential data inputs across a considerable duration. The possible results may be presented in either ordinal or numerical form. Even so, previous studies have shown major limitations when it comes to handling privacy and security issues that can’t be applicable in dynamic instances, as well as the substantial computational cost necessary. Moreover, correctly determining the amount of sensitive parameters required to complete the classification process remains a challenge. We have put forth a comprehensive survey on the classification of blockchain data. In the first phase of our study, we conducted an analysis and categorization of both conventional data classification approaches and contemporary time series data classification techniques. We further discussed limitations and strengths of existing techniques. Finally, we highlight future research problems and directions.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"246 ","pages":"Pages 961-970"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142746764","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}
Pub Date : 2024-01-01DOI: 10.1016/j.procs.2024.09.491
Andrei-Robert Alexandrescu, Laura Dioşan
In autonomous driving, a tremendous amount of imagery data is collected at all times. Manual annotation of such high-resolution images represents a costly and inefficient process. Active Learning comes to aid annotators in their process to focus on labelling meaningful samples which leads to competitive Machine Learning models, useful for various prediction tasks. In this paper, we introduce a novel Active Learning sampling technique, inspired by the Ant Colony Optimization algorithm, that considers both uncertainty and diversity features. We also introduce two hybrid sampling techniques that use weighted sums. We validate the proposed method on the Semantic Segmentation task, on a popular dataset from the railway domain. We also showcase the effectiveness of Active Learning in the scenario of Rail Semantic Segmentation by using only a quarter of the data to obtain competitive results of up to 78% mean Intersection over Union.
{"title":"Active Learning for Railway Semantic Segmentation through Ant Colony Optimization","authors":"Andrei-Robert Alexandrescu, Laura Dioşan","doi":"10.1016/j.procs.2024.09.491","DOIUrl":"10.1016/j.procs.2024.09.491","url":null,"abstract":"<div><div>In autonomous driving, a tremendous amount of imagery data is collected at all times. Manual annotation of such high-resolution images represents a costly and inefficient process. Active Learning comes to aid annotators in their process to focus on labelling meaningful samples which leads to competitive Machine Learning models, useful for various prediction tasks. In this paper, we introduce a novel Active Learning sampling technique, inspired by the Ant Colony Optimization algorithm, that considers both uncertainty and diversity features. We also introduce two hybrid sampling techniques that use weighted sums. We validate the proposed method on the Semantic Segmentation task, on a popular dataset from the railway domain. We also showcase the effectiveness of Active Learning in the scenario of Rail Semantic Segmentation by using only a quarter of the data to obtain competitive results of up to 78% mean Intersection over Union.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"246 ","pages":"Pages 724-733"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142746849","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}
Pub Date : 2024-01-01DOI: 10.1016/j.procs.2024.09.461
Ali Alfatemi , Sarah A.L. Jamal , Nasim Paykari , Mohamed Rahouti , Abdellah Chehri
This study delves into the challenge of classifying visually similar bird species, an area of significant interest in the field of fine-grained image classification. Utilizing a substantial dataset comprising images of ten bird species which was selected carefully to challenge the model to classify species of extreme similarities. To achieve this, we were keen to collect the data with subtle visual dissimilarities and of different positions taken for these birds. The research explores the potential of deep learning techniques to differentiate species based on subtle inter-species variations. This task is particularly demanding due to the minimal yet critical differences between these closely related species. Our research leveraged a unique deep learning model using convolutional neural networks (CNNs) to accurately classify birds with minimal visual differences. This innovative approach marks a significant step forward in machine learning for biological classification, with implications for biodiversity and ecological conservation. Our study demonstrates the effectiveness of our deep learning model in accurately classifying bird species, showcasing the potential of advanced techniques in complex Classification tasks. This research enhances the use of computational methods in biodiversity and ecological conservation. Additionally, it underscores the importance of birds as indicators of environmental changes, such as climate shifts, aiding in early detection of potential ecological issues.
{"title":"Multi-Label Classification with Deep Learning and Manual Data Collection for Identifying Similar Bird Species","authors":"Ali Alfatemi , Sarah A.L. Jamal , Nasim Paykari , Mohamed Rahouti , Abdellah Chehri","doi":"10.1016/j.procs.2024.09.461","DOIUrl":"10.1016/j.procs.2024.09.461","url":null,"abstract":"<div><div>This study delves into the challenge of classifying visually similar bird species, an area of significant interest in the field of fine-grained image classification. Utilizing a substantial dataset comprising images of ten bird species which was selected carefully to challenge the model to classify species of extreme similarities. To achieve this, we were keen to collect the data with subtle visual dissimilarities and of different positions taken for these birds. The research explores the potential of deep learning techniques to differentiate species based on subtle inter-species variations. This task is particularly demanding due to the minimal yet critical differences between these closely related species. Our research leveraged a unique deep learning model using convolutional neural networks (CNNs) to accurately classify birds with minimal visual differences. This innovative approach marks a significant step forward in machine learning for biological classification, with implications for biodiversity and ecological conservation. Our study demonstrates the effectiveness of our deep learning model in accurately classifying bird species, showcasing the potential of advanced techniques in complex Classification tasks. This research enhances the use of computational methods in biodiversity and ecological conservation. Additionally, it underscores the importance of birds as indicators of environmental changes, such as climate shifts, aiding in early detection of potential ecological issues.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"246 ","pages":"Pages 558-565"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142746909","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}
Models for artificial intelligence, machine learning, and neural networks are implemented on digital computers with a von Neumann architecture. Few studies have considered analog neural networks. Our previous study used multipliers to represent connecting weights in a neural network. The multipliers calculate the product of input signals and their corresponding connecting weights. However, using MOSFET multipliers, their operating range is limited by semiconductor characteristics. The input and output ranges for networks that use these multipliers are thus limited. Furthermore, the circuit operation becomes unstable. Here, we propose a logarithmic four-quadrant multiplier for representing connecting weights. The output of this multiple circuit is a more accurate value compared to the previous circuit. Experiments show that this multiplier exhibits stable operation over a wide range. Therefore, this model can be used directly for input/output of an analog control unit. A model that uses only analog electronic circuits is presented. Its learning time is quite short compared to that for models implemented on a digital computer. We increased the number of units and network layers. We suggest the possibility of a hardware implementation of a deep learning model. Furthermore, this model expects the elucidation of the biomedical learning mechanism.
{"title":"Learning Neural Network Circuit based on Logarithmic Multipliers","authors":"Masashi Kawaguchi , Naohiro Ishii , Masayoshi Umeno","doi":"10.1016/j.procs.2024.09.272","DOIUrl":"10.1016/j.procs.2024.09.272","url":null,"abstract":"<div><div>Models for artificial intelligence, machine learning, and neural networks are implemented on digital computers with a von Neumann architecture. Few studies have considered analog neural networks. Our previous study used multipliers to represent connecting weights in a neural network. The multipliers calculate the product of input signals and their corresponding connecting weights. However, using MOSFET multipliers, their operating range is limited by semiconductor characteristics. The input and output ranges for networks that use these multipliers are thus limited. Furthermore, the circuit operation becomes unstable. Here, we propose a logarithmic four-quadrant multiplier for representing connecting weights. The output of this multiple circuit is a more accurate value compared to the previous circuit. Experiments show that this multiplier exhibits stable operation over a wide range. Therefore, this model can be used directly for input/output of an analog control unit. A model that uses only analog electronic circuits is presented. Its learning time is quite short compared to that for models implemented on a digital computer. We increased the number of units and network layers. We suggest the possibility of a hardware implementation of a deep learning model. Furthermore, this model expects the elucidation of the biomedical learning mechanism.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"246 ","pages":"Pages 253-262"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142746917","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}
Pub Date : 2024-01-01DOI: 10.1016/j.procs.2024.09.478
Zofia Matusiewicz
Discovering knowledge from data has become one of the most critical problems in computer science in the last decades. Many methods and solutions to this issue have been created. It is not only the collection and analysis of data that is becoming an indispensable part of our lives but also the continuous process of improving detection methods for discovering knowledge from data. In the presented work, we modify the study of the relationship between attributes and specific ones, which are fuzzy relational equations. E. Sanchez, one of the pioneers of work on fuzzy relational equations, started research on using this method to study the relationship between input and output data, indicating it as a tool for analysing medical data. Since the 1970s, these equations have been studied with different types of compositions. The author of this work deals with this subject, examining the assumptions regarding the operations that can be used in max - relations’ composition in fuzzy relation equations A o x = b to have the solution set. In this work, we use a new way of compositing relations. It enables the use of various types of decision-attribute dependencies. We note that various dependencies may exist between individual data attributes and the decision. Undoubtedly, it is another stage of work on relational equations and provides new opportunities to discover the relationships between input and output data.
{"title":"The concept and method of determining the relation between data using relational equations with multi-operations composition","authors":"Zofia Matusiewicz","doi":"10.1016/j.procs.2024.09.478","DOIUrl":"10.1016/j.procs.2024.09.478","url":null,"abstract":"<div><div>Discovering knowledge from data has become one of the most critical problems in computer science in the last decades. Many methods and solutions to this issue have been created. It is not only the collection and analysis of data that is becoming an indispensable part of our lives but also the continuous process of improving detection methods for discovering knowledge from data. In the presented work, we modify the study of the relationship between attributes and specific ones, which are fuzzy relational equations. E. Sanchez, one of the pioneers of work on fuzzy relational equations, started research on using this method to study the relationship between input and output data, indicating it as a tool for analysing medical data. Since the 1970s, these equations have been studied with different types of compositions. The author of this work deals with this subject, examining the assumptions regarding the operations that can be used in max - relations’ composition in fuzzy relation equations <em>A</em> o <em>x = b</em> to have the solution set. In this work, we use a new way of compositing relations. It enables the use of various types of decision-attribute dependencies. We note that various dependencies may exist between individual data attributes and the decision. Undoubtedly, it is another stage of work on relational equations and provides new opportunities to discover the relationships between input and output data.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"246 ","pages":"Pages 646-655"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142746918","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}
In the field of Arabic handwriting recognition, the search for models that perfectly combine efficiency and accuracy is still ongoing. This paper introduces a novel approach that harnesses the synergy between autoencoders and convolutional neural networks (CNNs) to set a new benchmark in the recognition of Arabic handwritten characters. Our method centers around an autoencoder that meticulously learns a compact representation of the characters, followed by the integration of its encoder into a CNN architecture, dubbed the Encoder-CNN. The prowess of our model is demonstrated through rigorous experiments on the Arabic Handwritten Characters Dataset (AHCD), where it achieved a best accuracy of 98.87%. These results not only underscore the model’s ability to capture the intricate nuances of Arabic script but also its robustness in generalizing to unseen data.
{"title":"Encoder-Embedded Feature Enhancement in Convolutional Neural Networks for Arabic Handwritten Recognition","authors":"Oussama Alkayed , Marwa Amara , Nadia Smairi , Abdelmalek Zidouri","doi":"10.1016/j.procs.2024.09.484","DOIUrl":"10.1016/j.procs.2024.09.484","url":null,"abstract":"<div><div>In the field of Arabic handwriting recognition, the search for models that perfectly combine efficiency and accuracy is still ongoing. This paper introduces a novel approach that harnesses the synergy between autoencoders and convolutional neural networks (CNNs) to set a new benchmark in the recognition of Arabic handwritten characters. Our method centers around an autoencoder that meticulously learns a compact representation of the characters, followed by the integration of its encoder into a CNN architecture, dubbed the Encoder-CNN. The prowess of our model is demonstrated through rigorous experiments on the Arabic Handwritten Characters Dataset (AHCD), where it achieved a best accuracy of 98.87%. These results not only underscore the model’s ability to capture the intricate nuances of Arabic script but also its robustness in generalizing to unseen data.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"246 ","pages":"Pages 676-685"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142746937","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号