Pub Date : 2023-01-01DOI: 10.12720/jait.14.5.1019-1028
Dalia L. Elsheweikh
—Most models of automated web recommender systems depend on data mining algorithms to discover useful navigational patterns from the user’s previous browsing history. This paper presents a new model for developing a collaborative web recommendation system using a new technique for knowledge extraction. The proposed model introduces two techniques: cluster similarity-based technique and rule extraction technique to provide proper recommendations that meet the user’s needs. A cluster similarity-based technique groups the sessions that share common interests and behaviors according to a new similarity measure between the web users’ sessions. The rule extraction technique, which is based on a trained Artificial Neural Network (ANN) using a Genetic Algorithm (GA), is performed to discover groups of accurate and comprehensible rules from the clustering sessions. For extracting rules that belong to a specific cluster, GA can be applied to get the perfect values of the pages that maximize the output function of this cluster. A set of pruning schemes is proposed to decrease the size of the rule set and remove non-interesting rules. The resulting set of web pages recommended for a specific cluster is the dominant page in all rules that belong to this cluster. The experimental results indicate the proposed model’s efficiency in improving the classification’s precision and recall.
{"title":"A Novel Web Recommendation Model Based on the Web Usage Mining Technique","authors":"Dalia L. Elsheweikh","doi":"10.12720/jait.14.5.1019-1028","DOIUrl":"https://doi.org/10.12720/jait.14.5.1019-1028","url":null,"abstract":"—Most models of automated web recommender systems depend on data mining algorithms to discover useful navigational patterns from the user’s previous browsing history. This paper presents a new model for developing a collaborative web recommendation system using a new technique for knowledge extraction. The proposed model introduces two techniques: cluster similarity-based technique and rule extraction technique to provide proper recommendations that meet the user’s needs. A cluster similarity-based technique groups the sessions that share common interests and behaviors according to a new similarity measure between the web users’ sessions. The rule extraction technique, which is based on a trained Artificial Neural Network (ANN) using a Genetic Algorithm (GA), is performed to discover groups of accurate and comprehensible rules from the clustering sessions. For extracting rules that belong to a specific cluster, GA can be applied to get the perfect values of the pages that maximize the output function of this cluster. A set of pruning schemes is proposed to decrease the size of the rule set and remove non-interesting rules. The resulting set of web pages recommended for a specific cluster is the dominant page in all rules that belong to this cluster. The experimental results indicate the proposed model’s efficiency in improving the classification’s precision and recall.","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":"141 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":"136305458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.4.718-728
Vedika Jorika, Nagaratna Medishetty
—Different vertical domains have gained popularity in integrating Blockchain technology with their existing applications, because of its numerous benefits like immutable, transparency, privacy, persistence, and security. Blockchain technology is used in various circumstances, allows the applications to achieve higher security, improved traceability, and transparency. This paper reviewed most of the applications related to the different domains and the number of criteria met by each application in each domain requirement. This paper examines the advantages, disadvantages, and limitations of implementing the Blockchain in various applications in different domains. Furthermore, this paper describes the prerequisites for deploying Blockchain across multiple application fields.
{"title":"Demystifying Blockchain: A Critical Analysis of Application Characteristics in Different Domains","authors":"Vedika Jorika, Nagaratna Medishetty","doi":"10.12720/jait.14.4.718-728","DOIUrl":"https://doi.org/10.12720/jait.14.4.718-728","url":null,"abstract":"—Different vertical domains have gained popularity in integrating Blockchain technology with their existing applications, because of its numerous benefits like immutable, transparency, privacy, persistence, and security. Blockchain technology is used in various circumstances, allows the applications to achieve higher security, improved traceability, and transparency. This paper reviewed most of the applications related to the different domains and the number of criteria met by each application in each domain requirement. This paper examines the advantages, disadvantages, and limitations of implementing the Blockchain in various applications in different domains. Furthermore, this paper describes the prerequisites for deploying Blockchain across multiple application fields.","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":"1 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66333573","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.2.264-270
Jyoti Shetty, Karthik Cottur, G. Shobha, Y. Prajwal
—Forecasting resource usage values of a cloud service has ample applications such as service performance management, auto-scaling, capacity planning, and so on. While univariate forecasting techniques are the focus of current research, multivariate forecasting is rarely explored. This research work focuses on multivariate forecasting of resource usage values believing that there exists interdependency among the features of the underlying system that must be considered while forecasting. At first, the interdependency among the attributes is verified using Granger causality tests. Then the research explores various forecasting approaches — univariate Multi-Layer Perceptron (MLP), univariate Long Short Term Memory (LSTM), multivariate Vector Autoregression (VAR), and multivariate stacked LSTM. Further based on the observations of performances of these models the research proposes an implementation of a weighted ensemble of VAR and LSTM models to forecast key cloud resource usage metrics. The models thus proposed are implemented and validated using the publicly available GWA-T-12 Bitbrains time series dataset. The results show that the multivariate models outperform univariate models with lesser Normalised Root Mean Square Error (NRMSE) values. Also, the multivariate stacked LSTM outperforms VAR and the proposed ensemble forecasting model with lesser NRMSE values within a range of 1–5% for various resources across different lag values.
{"title":"A Weighted Ensemble of VAR and LSTM for Multivariate Forecasting of Cloud Resource Usage","authors":"Jyoti Shetty, Karthik Cottur, G. Shobha, Y. Prajwal","doi":"10.12720/jait.14.2.264-270","DOIUrl":"https://doi.org/10.12720/jait.14.2.264-270","url":null,"abstract":"—Forecasting resource usage values of a cloud service has ample applications such as service performance management, auto-scaling, capacity planning, and so on. While univariate forecasting techniques are the focus of current research, multivariate forecasting is rarely explored. This research work focuses on multivariate forecasting of resource usage values believing that there exists interdependency among the features of the underlying system that must be considered while forecasting. At first, the interdependency among the attributes is verified using Granger causality tests. Then the research explores various forecasting approaches — univariate Multi-Layer Perceptron (MLP), univariate Long Short Term Memory (LSTM), multivariate Vector Autoregression (VAR), and multivariate stacked LSTM. Further based on the observations of performances of these models the research proposes an implementation of a weighted ensemble of VAR and LSTM models to forecast key cloud resource usage metrics. The models thus proposed are implemented and validated using the publicly available GWA-T-12 Bitbrains time series dataset. The results show that the multivariate models outperform univariate models with lesser Normalised Root Mean Square Error (NRMSE) values. Also, the multivariate stacked LSTM outperforms VAR and the proposed ensemble forecasting model with lesser NRMSE values within a range of 1–5% for various resources across different lag values.","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":"1 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66330210","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.3.532-542
Thabiso N. Khosa, Topside E. Mathonsi, D. D. Plessis
—Over the past few years, Mobile Ad-hoc Networks (MANET) has been playing an important role in ubiquitous networks based on its ability to support mobility without depending on infrastructure-based design, dynamic topology, and thus, are known as decentralized environment. One of the advantages of MANET is that its nodes can act both as routers and hosts. This, therefore, implies that its nodes can transmit packets between source to destination nodes. As a result of such and many more advantages, these networks are more vulnerable to different types of network attacks. In the recent past, several secured routing protocols were proposed and implemented for MANET. However, those protocols cannot fully guarantee security within these networks in terms of Denial of Services (DoS) attacks such as black hole and gray hole attacks. The review of the literature showed that existing solutions cannot always ensure true node classification. This is because MANET’s cooperative existence sometimes leads to the false exclusion of innocent nodes and/or proper classification of malicious nodes. A new Gray Hole Prevention (GRAY-HP) algorithm for the detection of malicious nodes with the actual high accuracy ratio of node classification is proposed in this paper. The proposed algorithm employs and modifies the gray-attack prevention technique known as Secure Detection Prevention and Elimination Gray Hole (SDPEGH), and the proactive scheme. It has been confirmed by Network Simulator 2 (NS2) computer simulation that the proposed algorithm outperforms the Genetic Algorithm to Bacterial Foraging Optimization (GA-BFO) and Rough Set Theory (RSetTheory) algorithms in terms of throughput, routing overhead and delivery ratio. The proposed GRAY-HP algorithm guarantees the successful elimination of Gray hole nodes, while it also ensures that no legitimate nodes are excluded
{"title":"A Model to Prevent Gray Hole Attack in Mobile Ad-Hoc Networks","authors":"Thabiso N. Khosa, Topside E. Mathonsi, D. D. Plessis","doi":"10.12720/jait.14.3.532-542","DOIUrl":"https://doi.org/10.12720/jait.14.3.532-542","url":null,"abstract":"—Over the past few years, Mobile Ad-hoc Networks (MANET) has been playing an important role in ubiquitous networks based on its ability to support mobility without depending on infrastructure-based design, dynamic topology, and thus, are known as decentralized environment. One of the advantages of MANET is that its nodes can act both as routers and hosts. This, therefore, implies that its nodes can transmit packets between source to destination nodes. As a result of such and many more advantages, these networks are more vulnerable to different types of network attacks. In the recent past, several secured routing protocols were proposed and implemented for MANET. However, those protocols cannot fully guarantee security within these networks in terms of Denial of Services (DoS) attacks such as black hole and gray hole attacks. The review of the literature showed that existing solutions cannot always ensure true node classification. This is because MANET’s cooperative existence sometimes leads to the false exclusion of innocent nodes and/or proper classification of malicious nodes. A new Gray Hole Prevention (GRAY-HP) algorithm for the detection of malicious nodes with the actual high accuracy ratio of node classification is proposed in this paper. The proposed algorithm employs and modifies the gray-attack prevention technique known as Secure Detection Prevention and Elimination Gray Hole (SDPEGH), and the proactive scheme. It has been confirmed by Network Simulator 2 (NS2) computer simulation that the proposed algorithm outperforms the Genetic Algorithm to Bacterial Foraging Optimization (GA-BFO) and Rough Set Theory (RSetTheory) algorithms in terms of throughput, routing overhead and delivery ratio. The proposed GRAY-HP algorithm guarantees the successful elimination of Gray hole nodes, while it also ensures that no legitimate nodes are excluded","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":"1 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66331787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.1.122-129
Keerthi Kethineni, G. Pradeepini
Diagnosing plant disease is the foundation for effective and accurate plant disease prevention in a complicated environment. Smart farming is one of the fast-growing processes in the agricultural system, with the identification of disease in plants being a major one to help farmers. The processed data is saved in a database and used in making decisions in advance support, analysis of plants, and helps in crop planning. Plants are one of the essential resources for avoiding global warming. However, diseases such as blast, canker, black spot, brown spot, and bacterial leaf damage the plants. In this paper, image processing integration is developed to identify the type of disease and help automatically inspect all the leaf batches by storing the processed data. In some places, farmers are unaware of the experts and do not have proper facilities. In such conditions, one technique can be beneficial in keeping track and monitoring more crops. This technique makes it much easier and cheaper to detect disease. Machine learning can provide a method and algorithm to detect the disease. There should be training in images of all types of leaves, including healthy and disease leaf images. Five-stage detection processes are done in this paper. The stages are preprocessing, segmentation using k-Mean, feature extraction, features optimization using Firefly optimization Algorithm (FA), and classification using Support Vector Machine (SVM). The accuracy rate achieved using the proposed technique, i.e., GA-SVM is 91.3%, sensitivity is 90.72%, specificity 91.88, and precision is 92%. The results are evaluated using the matlab software tool.
{"title":"Identification of Leaf Disease Using Machine Learning Algorithm for Improving the Agricultural System","authors":"Keerthi Kethineni, G. Pradeepini","doi":"10.12720/jait.14.1.122-129","DOIUrl":"https://doi.org/10.12720/jait.14.1.122-129","url":null,"abstract":"Diagnosing plant disease is the foundation for effective and accurate plant disease prevention in a complicated environment. Smart farming is one of the fast-growing processes in the agricultural system, with the identification of disease in plants being a major one to help farmers. The processed data is saved in a database and used in making decisions in advance support, analysis of plants, and helps in crop planning. Plants are one of the essential resources for avoiding global warming. However, diseases such as blast, canker, black spot, brown spot, and bacterial leaf damage the plants. In this paper, image processing integration is developed to identify the type of disease and help automatically inspect all the leaf batches by storing the processed data. In some places, farmers are unaware of the experts and do not have proper facilities. In such conditions, one technique can be beneficial in keeping track and monitoring more crops. This technique makes it much easier and cheaper to detect disease. Machine learning can provide a method and algorithm to detect the disease. There should be training in images of all types of leaves, including healthy and disease leaf images. Five-stage detection processes are done in this paper. The stages are preprocessing, segmentation using k-Mean, feature extraction, features optimization using Firefly optimization Algorithm (FA), and classification using Support Vector Machine (SVM). The accuracy rate achieved using the proposed technique, i.e., GA-SVM is 91.3%, sensitivity is 90.72%, specificity 91.88, and precision is 92%. The results are evaluated using the matlab software tool.","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":"31 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66329209","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.1.39-45
Bleron Zherka, Zhilbert Tafa
Air Pollution (AP) is one of the main threats to global health. Real-time dynamic mapping of pollution distribution is of a crucial importance to the AP reduction and management. Conventional air quality monitoring relies on expensive and cumbersome monitoring stations. Such stations are sparsely deployed over a region – typically one to a few per city. The extrapolation of the dynamic spatiotemporal data away from these stations might be inaccurate. In this paper, we present a participatory Vehicle Sensor Network (VSN) based on low-cost mobile nodes deployed on public (taxi) vehicles. The system enables continuous real-time data acquisition, transmission, and utilization. As compared to the conventional approaches, our system greatly improves sensing coverage. The proposed platform enables the acquisition of a large amount of georeferenced and time-stamped data. It provides real time pollution mapping and historical data view. The system’s operational stability and continuity are examined and confirmed through the analysis of background data collected during 15 days of experimental implementation.
{"title":"A Vehicle Sensor Network for Real-Time Air Pollution Analysis","authors":"Bleron Zherka, Zhilbert Tafa","doi":"10.12720/jait.14.1.39-45","DOIUrl":"https://doi.org/10.12720/jait.14.1.39-45","url":null,"abstract":"Air Pollution (AP) is one of the main threats to global health. Real-time dynamic mapping of pollution distribution is of a crucial importance to the AP reduction and management. Conventional air quality monitoring relies on expensive and cumbersome monitoring stations. Such stations are sparsely deployed over a region – typically one to a few per city. The extrapolation of the dynamic spatiotemporal data away from these stations might be inaccurate. In this paper, we present a participatory Vehicle Sensor Network (VSN) based on low-cost mobile nodes deployed on public (taxi) vehicles. The system enables continuous real-time data acquisition, transmission, and utilization. As compared to the conventional approaches, our system greatly improves sensing coverage. The proposed platform enables the acquisition of a large amount of georeferenced and time-stamped data. It provides real time pollution mapping and historical data view. The system’s operational stability and continuity are examined and confirmed through the analysis of background data collected during 15 days of experimental implementation.","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":"1 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66329998","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.4.685-693
Kamal Uddin Sarker, R. Hasan, A. Deraman, Salman Mahmmod
—The software industry is enjoying the permeable and trans-border flow of software markets and can access resources from all corners of the world. Software engineers gain international work experience through a distributed working environment. It involves participation from individuals with different cultures, languages, and geographic time zones to work on a single project. In addition to providing global opportunities for software experts and businessmen, it also introduces new project management challenges. Barriers exist in trust, communication, monitoring, languages, cultures, and time zones. Distance mode management faces more technical challenges due to stakeholders’ ambiguous understanding and various documentation. This study addresses an in-depth analysis of challenges and currently practicing methods. Moreover, a new virtual project management framework is proposed to minimize issues and maximize the virtual project management team’s throughput. The framework is compared with commonly used methodologies by experts who have experience in global software project management, and the analysis is performed using the analytical hierarchy process. The evaluation matrix has shown that the proposed model is adequate for distance project management with better score in virtual scope and virtual management). Its excellency is in standard documentation practice, change management, and improving re-usability practice that will enhance business goals and stakeholder’s satisfaction.
{"title":"A Distributed Software Project Management Framework","authors":"Kamal Uddin Sarker, R. Hasan, A. Deraman, Salman Mahmmod","doi":"10.12720/jait.14.4.685-693","DOIUrl":"https://doi.org/10.12720/jait.14.4.685-693","url":null,"abstract":"—The software industry is enjoying the permeable and trans-border flow of software markets and can access resources from all corners of the world. Software engineers gain international work experience through a distributed working environment. It involves participation from individuals with different cultures, languages, and geographic time zones to work on a single project. In addition to providing global opportunities for software experts and businessmen, it also introduces new project management challenges. Barriers exist in trust, communication, monitoring, languages, cultures, and time zones. Distance mode management faces more technical challenges due to stakeholders’ ambiguous understanding and various documentation. This study addresses an in-depth analysis of challenges and currently practicing methods. Moreover, a new virtual project management framework is proposed to minimize issues and maximize the virtual project management team’s throughput. The framework is compared with commonly used methodologies by experts who have experience in global software project management, and the analysis is performed using the analytical hierarchy process. The evaluation matrix has shown that the proposed model is adequate for distance project management with better score in virtual scope and virtual management). Its excellency is in standard documentation practice, change management, and improving re-usability practice that will enhance business goals and stakeholder’s satisfaction.","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":"1 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66333308","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.4.701-717
O. Stitini, I. García-Magariño, S. Kaloun, O. Bencharef
.
．
{"title":"Towards Ideal and Efficient Recommendation Systems Based on the Five Evaluation Concepts Promoting Serendipity","authors":"O. Stitini, I. García-Magariño, S. Kaloun, O. Bencharef","doi":"10.12720/jait.14.4.701-717","DOIUrl":"https://doi.org/10.12720/jait.14.4.701-717","url":null,"abstract":".","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":"1 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66333392","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.4.830-837
Izzati Mohaimin, R. Apong, A. R. Damit
—As online information increases over the years, text mining researchers developed Natural Language Processing tools to extract relevant and useful information from textual data such as online news articles. The Malay language is widely spoken, especially in the Southeast Asian region, but there is a lack of Natural Language Processing (NLP) tools such as Malay corpora and Part-of-Speech (POS) taggers. Existing NLP tools are mainly based on Standard Malay of Malaysia and Indonesian language, but there is none for the Bruneian Malay. We addressed this issue by designing a Standard Brunei Malay corpus consisting of over 114,000 lexical tokens, annotated using 17 Malay POS tagsets. Furthermore, we implemented two commonly used POS tagging techniques, Conditional Random Field (CRF) and Bi-directional Long Short-Term Memory (BLSTM), to develop Bruneian POS taggers and compared their performances. The results showed that both CRF and BLSTM models performed well in predicting POS tags on Bruneian texts. However, CRF models outperform BLSTM, where CRF using all features achieved an F-Measure of 92.06% on news articles and 90.71% of F-Measure on crime articles. Adding a batch normalization layer to the BLSTM model architecture increased the performance by 7.13%. To further improve the BLSTM models, we suggested increasing the training data and experimenting with different hyperparameter settings. The findings also indicated that modelling BLSTM with fastText has improved the POS prediction of Bruneian words.
{"title":"Part-of-Speech (POS) Tagging for Standard Brunei Malay: A Probabilistic and Neural-Based Approach","authors":"Izzati Mohaimin, R. Apong, A. R. Damit","doi":"10.12720/jait.14.4.830-837","DOIUrl":"https://doi.org/10.12720/jait.14.4.830-837","url":null,"abstract":"—As online information increases over the years, text mining researchers developed Natural Language Processing tools to extract relevant and useful information from textual data such as online news articles. The Malay language is widely spoken, especially in the Southeast Asian region, but there is a lack of Natural Language Processing (NLP) tools such as Malay corpora and Part-of-Speech (POS) taggers. Existing NLP tools are mainly based on Standard Malay of Malaysia and Indonesian language, but there is none for the Bruneian Malay. We addressed this issue by designing a Standard Brunei Malay corpus consisting of over 114,000 lexical tokens, annotated using 17 Malay POS tagsets. Furthermore, we implemented two commonly used POS tagging techniques, Conditional Random Field (CRF) and Bi-directional Long Short-Term Memory (BLSTM), to develop Bruneian POS taggers and compared their performances. The results showed that both CRF and BLSTM models performed well in predicting POS tags on Bruneian texts. However, CRF models outperform BLSTM, where CRF using all features achieved an F-Measure of 92.06% on news articles and 90.71% of F-Measure on crime articles. Adding a batch normalization layer to the BLSTM model architecture increased the performance by 7.13%. To further improve the BLSTM models, we suggested increasing the training data and experimenting with different hyperparameter settings. The findings also indicated that modelling BLSTM with fastText has improved the POS prediction of Bruneian words.","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":"1 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66333721","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.12720/jait.14.2.204-211
Dominic Kern, Matthias Tessmann
—The most common video conferencing topologies are mesh and star topologies. The star topology requires a powerful server which leads to high costs. In the mesh topology, this is not the case, as each participant is directly connected to every other participant. However, due to the load caused by the numerous connections, the mesh topology is not suitable for larger video conferences. In this paper, we propose a video conferencing service that combines the advantages of the mesh and star topologies to enable larger video conferences without the need for powerful servers. This is achieved by distributing the video streams over the most powerful participants instead of a server. The resulting system achieves an improvement in video quality compared to a reference test in the mesh topology, which was determined based on the transmission rate and frame rate.
{"title":"Client-Based Distributed Video Conferencing via WebRTC","authors":"Dominic Kern, Matthias Tessmann","doi":"10.12720/jait.14.2.204-211","DOIUrl":"https://doi.org/10.12720/jait.14.2.204-211","url":null,"abstract":"—The most common video conferencing topologies are mesh and star topologies. The star topology requires a powerful server which leads to high costs. In the mesh topology, this is not the case, as each participant is directly connected to every other participant. However, due to the load caused by the numerous connections, the mesh topology is not suitable for larger video conferences. In this paper, we propose a video conferencing service that combines the advantages of the mesh and star topologies to enable larger video conferences without the need for powerful servers. This is achieved by distributing the video streams over the most powerful participants instead of a server. The resulting system achieves an improvement in video quality compared to a reference test in the mesh topology, which was determined based on the transmission rate and frame rate.","PeriodicalId":36452,"journal":{"name":"Journal of Advances in Information Technology","volume":"1 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"66330539","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","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：