Pub Date : 2022-03-31DOI: 10.22452/mjcs.sp2022no1.4
B. Devi, V. Vijayakumar, G. Suseela, B. P. Kavin, S. Sivaramakrishnan, Joel Rodrigues
Cloud computing is a new category of service that gives each customer access to a large-scale computing network. Since most cloud computing platforms provide services to a large number of people who aren't considered to be trustworthy, various cyber attacks may potentially target them. As a result, a cloud computing system must provide a security monitoring mechanism to protect the Virtual Machine from attacks. In this case, there is a tradeoff between the security level of the security system and its performance. If we need strong security, we'll need more laws or patterns, which means we'll need a lot more computational resources in proportion to the strength of security. As a result of the declining number of resources allocated to customers, we will add a new protection scheme in cloud environments to the VM in this report. Hence, the proposed system Proposed Elliptic curve – Diffie Hellman EC(DH)2 Algorithm is designed and deployed to improve the security in healthcare domain using hybrid computing. The most popular and recent technologies such as cloud computing and fog computing are integrated to explore data movement and stable medical data health-care information. Based on the experimental results, it is inferred that the proposed system offers high security and less operating time while handling the data making its deployment in the healthcare domain.
{"title":"AN IMPROVED SECURITY FRAMEWORK IN HEALTH CARE USING HYBRID COMPUTING","authors":"B. Devi, V. Vijayakumar, G. Suseela, B. P. Kavin, S. Sivaramakrishnan, Joel Rodrigues","doi":"10.22452/mjcs.sp2022no1.4","DOIUrl":"https://doi.org/10.22452/mjcs.sp2022no1.4","url":null,"abstract":"Cloud computing is a new category of service that gives each customer access to a large-scale computing network. Since most cloud computing platforms provide services to a large number of people who aren't considered to be trustworthy, various cyber attacks may potentially target them. As a result, a cloud computing system must provide a security monitoring mechanism to protect the Virtual Machine from attacks. In this case, there is a tradeoff between the security level of the security system and its performance. If we need strong security, we'll need more laws or patterns, which means we'll need a lot more computational resources in proportion to the strength of security. As a result of the declining number of resources allocated to customers, we will add a new protection scheme in cloud environments to the VM in this report. Hence, the proposed system Proposed Elliptic curve – Diffie Hellman EC(DH)2 Algorithm is designed and deployed to improve the security in healthcare domain using hybrid computing. The most popular and recent technologies such as cloud computing and fog computing are integrated to explore data movement and stable medical data health-care information. Based on the experimental results, it is inferred that the proposed system offers high security and less operating time while handling the data making its deployment in the healthcare domain.","PeriodicalId":49894,"journal":{"name":"Malaysian Journal of Computer Science","volume":" ","pages":""},"PeriodicalIF":0.6,"publicationDate":"2022-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43954763","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}
Pub Date : 2022-03-31DOI: 10.22452/mjcs.sp2022no1.5
S. A, K. B., A. S., S. P, S. V., Logesh Ravi
The use of blockchain technology in the financial contract management system leads many challenges for user authentication and key distribution. In traditional financial system, the core endeavor with strong letter of guarantee play an irreplaceable role in the supply chain management. In this context, fraud in financial contract management is severe problem for economical growth. To overcome the problems of the conventional transaction process, in this paper, we develop a user access control and key management scheme that uses blockchain decentralized network to manage the letter of guarantee. This decentralized network platform can helps the problem of no trust among the users, improves the efficiency of data transmission, reduces costs, and provides better financial services to the relevant parties in the supply chain. The proposed user authentication is developed based on the deterministic encryption algorithm to achieve decentralized security for trusted data transmission. Finally, the experimental results shows that the computation cost of the proposed authentication increases due to the rising of number of customer added to the blockchain network. Overall the proposed authentication scheme most suitable for banking system to issue the LoG contract in right time.
{"title":"BLOCKCHAIN-BASED DECENTRALIZED USER AUTHENTICATION SCHEME FOR LETTER OF GUARANTEE IN FINANCIAL CONTRACT MANAGEMENT","authors":"S. A, K. B., A. S., S. P, S. V., Logesh Ravi","doi":"10.22452/mjcs.sp2022no1.5","DOIUrl":"https://doi.org/10.22452/mjcs.sp2022no1.5","url":null,"abstract":"The use of blockchain technology in the financial contract management system leads many challenges for user authentication and key distribution. In traditional financial system, the core endeavor with strong letter of guarantee play an irreplaceable role in the supply chain management. In this context, fraud in financial contract management is severe problem for economical growth. To overcome the problems of the conventional transaction process, in this paper, we develop a user access control and key management scheme that uses blockchain decentralized network to manage the letter of guarantee. This decentralized network platform can helps the problem of no trust among the users, improves the efficiency of data transmission, reduces costs, and provides better financial services to the relevant parties in the supply chain. The proposed user authentication is developed based on the deterministic encryption algorithm to achieve decentralized security for trusted data transmission. Finally, the experimental results shows that the computation cost of the proposed authentication increases due to the rising of number of customer added to the blockchain network. Overall the proposed authentication scheme most suitable for banking system to issue the LoG contract in right time.","PeriodicalId":49894,"journal":{"name":"Malaysian Journal of Computer Science","volume":" ","pages":""},"PeriodicalIF":0.6,"publicationDate":"2022-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44852914","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}
Pub Date : 2022-03-31DOI: 10.22452/mjcs.sp2022no1.10
Ramesh Tr, U. Lilhore, P. M., Sarita Simaiya, Amandeep Kaur, Mounir Hamdi
Machine Learning (ML) is used in healthcare sectors worldwide. ML methods help in the protection of heart diseases, locomotor disorders in the medical data set. The discovery of such essential data helps researchers gain valuable insight into how to utilize their diagnosis and treatment for a particular patient. Researchers use various Machine Learning methods to examine massive amounts of complex healthcare data, which aids healthcare professionals in predicting diseases. In this research, we are using an online UCI dataset with 303 rows and 76 properties. Approximately 14 of these 76 properties are selected for testing, which is necessary to validate the performances of different methods. The isolation forest approach uses the data set’s most essential qualities and metrics to standardize the information for better precision. This analysis is based on supervised learning methods, i.e., Naive Bayes, SVM, Logistic regression, Decision Tree Classifier, Random Forest, and K- Nearest Neighbor. The experimental results demonstrate the strength of KNN with eight neighbours order to test the effectiveness, sensitivity, precision, and accuracy, F1-score; as compared to other methods, i.e., Naive Bayes, SVM (Linear Kernel), Decision Tree Classifier with 4 or 18 features, and Random Forest classifiers.
{"title":"PREDICTIVE ANALYSIS OF HEART DISEASES WITH MACHINE LEARNING APPROACHES","authors":"Ramesh Tr, U. Lilhore, P. M., Sarita Simaiya, Amandeep Kaur, Mounir Hamdi","doi":"10.22452/mjcs.sp2022no1.10","DOIUrl":"https://doi.org/10.22452/mjcs.sp2022no1.10","url":null,"abstract":"Machine Learning (ML) is used in healthcare sectors worldwide. ML methods help in the protection of heart diseases, locomotor disorders in the medical data set. The discovery of such essential data helps researchers gain valuable insight into how to utilize their diagnosis and treatment for a particular patient. Researchers use various Machine Learning methods to examine massive amounts of complex healthcare data, which aids healthcare professionals in predicting diseases. In this research, we are using an online UCI dataset with 303 rows and 76 properties. Approximately 14 of these 76 properties are selected for testing, which is necessary to validate the performances of different methods. The isolation forest approach uses the data set’s most essential qualities and metrics to standardize the information for better precision. This analysis is based on supervised learning methods, i.e., Naive Bayes, SVM, Logistic regression, Decision Tree Classifier, Random Forest, and K- Nearest Neighbor. The experimental results demonstrate the strength of KNN with eight neighbours order to test the effectiveness, sensitivity, precision, and accuracy, F1-score; as compared to other methods, i.e., Naive Bayes, SVM (Linear Kernel), Decision Tree Classifier with 4 or 18 features, and Random Forest classifiers.","PeriodicalId":49894,"journal":{"name":"Malaysian Journal of Computer Science","volume":" ","pages":""},"PeriodicalIF":0.6,"publicationDate":"2022-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44588574","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}
Pub Date : 2022-03-31DOI: 10.22452/mjcs.sp2022no1.7
A. T. Mohamed, Sundar Santhoshkumar, Vijayakumar Varadarajan
Presently, process analytics extracts the knowledge from the past data to explore, monitor, and improve the processes. The recently developed deep learning (DL) models find it helpful to analyse medical data and make decisions. Among various diseases, type 2 diabetes mellitus (T2DM) becomes a widespread disease over the globe and it leads to severe outcomes. Chronic kidney disease (CKD) and coronary heart disease (CHD) are the major illness occurred in people with T2DM. Since the earlier prediction of the risk factors related to CKD and CHD on T2DM persons is necessary, this study focuses on the design of intelligent feature selection with deep learning based risk factor prediction (IFS-DLRFP) model. The proposed IFS-DLRFP technique intends to determine the early warning to the patients with T2DM to develop CKD or CHD. In addition, the IFS-DLRFP technique includes the design of fruit fly optimization algorithm (FFOA) based feature selection technique to choose an optimal set of features. Moreover, firefly optimization with gated recurrent unit (FF-GRU) based classification technique is derived to allocate appropriate class labels to the input data. The FF-GRU technique performs the hyperparameter tuning process using FF technique. In order to ensure the better performance of the IFS-DLRFP technique, a wide range of simulations take place on benchmark datasets and the simulation outcomes reported the supremacy of the IFS-DLRFP approach over the recent techniques.
{"title":"INTELLIGENT DEEP LEARNING BASED PREDICTIVE MODEL FOR CORONARY HEART DISEASE AND CHRONIC KIDNEY DISEASE ON PEOPLE WITH DIABETES MELLITUS","authors":"A. T. Mohamed, Sundar Santhoshkumar, Vijayakumar Varadarajan","doi":"10.22452/mjcs.sp2022no1.7","DOIUrl":"https://doi.org/10.22452/mjcs.sp2022no1.7","url":null,"abstract":"Presently, process analytics extracts the knowledge from the past data to explore, monitor, and improve the processes. The recently developed deep learning (DL) models find it helpful to analyse medical data and make decisions. Among various diseases, type 2 diabetes mellitus (T2DM) becomes a widespread disease over the globe and it leads to severe outcomes. Chronic kidney disease (CKD) and coronary heart disease (CHD) are the major illness occurred in people with T2DM. Since the earlier prediction of the risk factors related to CKD and CHD on T2DM persons is necessary, this study focuses on the design of intelligent feature selection with deep learning based risk factor prediction (IFS-DLRFP) model. The proposed IFS-DLRFP technique intends to determine the early warning to the patients with T2DM to develop CKD or CHD. In addition, the IFS-DLRFP technique includes the design of fruit fly optimization algorithm (FFOA) based feature selection technique to choose an optimal set of features. Moreover, firefly optimization with gated recurrent unit (FF-GRU) based classification technique is derived to allocate appropriate class labels to the input data. The FF-GRU technique performs the hyperparameter tuning process using FF technique. In order to ensure the better performance of the IFS-DLRFP technique, a wide range of simulations take place on benchmark datasets and the simulation outcomes reported the supremacy of the IFS-DLRFP approach over the recent techniques.","PeriodicalId":49894,"journal":{"name":"Malaysian Journal of Computer Science","volume":" ","pages":""},"PeriodicalIF":0.6,"publicationDate":"2022-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49397771","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}
Pub Date : 2022-03-31DOI: 10.22452/mjcs.sp2022no1.8
M. Gokiladevi, Sundar Santhoshkumar, Vijayakumar Varadarajan
In last decades, chronic kidney disease (CKD) becomes a global health problem that is steadily developing worldwide. It is a chronic illness highly related to increased morbidity and mortality, cardiovascular diseases, and high healthcare cost. Earlier identification and classification of CKD is treated as a major factor in controlling the mortality rate. Data mining (DM) techniques are used for the extraction of hidden details from the clinical and laboratory patient data that is used to aid doctors in enhancing diagnostic accuracy. Recently, machine learning (ML) techniques are commonly employed for the prediction and classification of diseases in healthcare sector. With this motivation, this study examines the performance of different ML algorithms to diagnose CKD at the earlier stages. The proposed model involves data pre-processing in two stages such as missing value replacement and data transformation. Besides, a set of five ML based classification models are involved such as support vector machine (SVM), random forest (RF), logistic regression (LR), K-nearest neighbor (KNN), and decision tree (DT). For investigating the performance of the different ML models, a benchmark CKD dataset from UCI repository is employed and the results are examined under different aspects. Among the different classifiers, the RF model has accomplished superior results with the maximum precision of 0.99, recall of 0.99, and F-score of 0.99 with a minimal error rate of 0.012.
{"title":"MACHINE LEARNING ALGORITHM SELECTION FOR CHRONIC KIDNEY DISEASE DIAGNOSIS AND CLASSIFICATION","authors":"M. Gokiladevi, Sundar Santhoshkumar, Vijayakumar Varadarajan","doi":"10.22452/mjcs.sp2022no1.8","DOIUrl":"https://doi.org/10.22452/mjcs.sp2022no1.8","url":null,"abstract":"In last decades, chronic kidney disease (CKD) becomes a global health problem that is steadily developing worldwide. It is a chronic illness highly related to increased morbidity and mortality, cardiovascular diseases, and high healthcare cost. Earlier identification and classification of CKD is treated as a major factor in controlling the mortality rate. Data mining (DM) techniques are used for the extraction of hidden details from the clinical and laboratory patient data that is used to aid doctors in enhancing diagnostic accuracy. Recently, machine learning (ML) techniques are commonly employed for the prediction and classification of diseases in healthcare sector. With this motivation, this study examines the performance of different ML algorithms to diagnose CKD at the earlier stages. The proposed model involves data pre-processing in two stages such as missing value replacement and data transformation. Besides, a set of five ML based classification models are involved such as support vector machine (SVM), random forest (RF), logistic regression (LR), K-nearest neighbor (KNN), and decision tree (DT). For investigating the performance of the different ML models, a benchmark CKD dataset from UCI repository is employed and the results are examined under different aspects. Among the different classifiers, the RF model has accomplished superior results with the maximum precision of 0.99, recall of 0.99, and F-score of 0.99 with a minimal error rate of 0.012.","PeriodicalId":49894,"journal":{"name":"Malaysian Journal of Computer Science","volume":" ","pages":""},"PeriodicalIF":0.6,"publicationDate":"2022-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44474788","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}
Pub Date : 2022-03-31DOI: 10.22452/mjcs.sp2022no1.1
D. R, S. V.
In recent times, Online Social Network (OSN) has accumulated a massive volume of user-generated data available in an unstructured format. It consists of user ideas, responses, and opinions on various topics. It extracts essential keywords in OSN, which is endowed with many exciting applications such as information recommendation or viral marketing. This paper emphasizes the importance of semantic graph-based methods for extracting vital keywords experimentally using a novel SKEM++ method. It is an innovative method for keyword extraction from OSN based on centrality measures. It utilizes a distributed computing approach to calculate the network Collective Centrality Measure (CCM) for each node and improve the semantics of keywords. The distributed approach is more scalable and computationally efficient than the conventional system, making it more suitable for large-scale real-time data sets such as the OSN. Experimental outcomes on the real-time Twitter Data set to infer the dominance of the proposed Collective Centrality Measure(CCM) method in evaluation with contemporary schemes in terms of F-score by 81% and recall by 80% and precision by 80% using Semantic Analysis.
{"title":"SKEM++: SEMANTIC KEYWORD EXTRACTION MODEL USING COLLECTIVE CENTRALITY MEASURE ON BIG SOCIAL DATA","authors":"D. R, S. V.","doi":"10.22452/mjcs.sp2022no1.1","DOIUrl":"https://doi.org/10.22452/mjcs.sp2022no1.1","url":null,"abstract":"In recent times, Online Social Network (OSN) has accumulated a massive volume of user-generated data available in an unstructured format. It consists of user ideas, responses, and opinions on various topics. It extracts essential keywords in OSN, which is endowed with many exciting applications such as information recommendation or viral marketing. This paper emphasizes the importance of semantic graph-based methods for extracting vital keywords experimentally using a novel SKEM++ method. It is an innovative method for keyword extraction from OSN based on centrality measures. It utilizes a distributed computing approach to calculate the network Collective Centrality Measure (CCM) for each node and improve the semantics of keywords. The distributed approach is more scalable and computationally efficient than the conventional system, making it more suitable for large-scale real-time data sets such as the OSN. Experimental outcomes on the real-time Twitter Data set to infer the dominance of the proposed Collective Centrality Measure(CCM) method in evaluation with contemporary schemes in terms of F-score by 81% and recall by 80% and precision by 80% using Semantic Analysis.","PeriodicalId":49894,"journal":{"name":"Malaysian Journal of Computer Science","volume":" ","pages":""},"PeriodicalIF":0.6,"publicationDate":"2022-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43655122","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}
Pub Date : 2022-03-31DOI: 10.22452/mjcs.sp2022no1.9
Arodh Lal Karn, V. Sachin, Sudhakar Sengan, I. V., Logesh Ravi, Dilip Kumar Sharma, S. V.
In the banking business, Machine Learning (ML) is critical for averting financial losses. Credit risk evaluation is perhaps the most important prediction task that may result in billions of dollars in damages each year (i.e., the risk of default on debt). Gradient Boosted Decision Tree (GBDT) models are now responsible for a large portion of the improvements in ML for predicting credit risk. However, these improvements begin to stagnate without adding pricey new data sources or carefully designed features. In this work, we describe our efforts to develop a unique Deep Learning (DL)-based technique for assessing credit risk that does not rely on additional model inputs. We present a new credit decision support approach with Gated Recurrent Unit (GRU) and Convolutional Neural Networks (CNN) that uses lengthy historical sequences of financial data while requiring few resources. We show that our DL technique, which uses Term Frequency-Inverse Document Frequency (TF-IDF) pre-classifiers, outperforms the benchmark models, resulting in considerable cost savings and early credit risk identification. We also show how our method may be utilized in a production setting, where our sampling methodology allows sequences to be effectively kept in memory and used for quick online learning and inference.
{"title":"DESIGNING A DEEP LEARNING-BASED FINANCIAL DECISION SUPPORT SYSTEM FOR FINTECH TO SUPPORT CORPORATE CUSTOMER’S CREDIT EXTENSION","authors":"Arodh Lal Karn, V. Sachin, Sudhakar Sengan, I. V., Logesh Ravi, Dilip Kumar Sharma, S. V.","doi":"10.22452/mjcs.sp2022no1.9","DOIUrl":"https://doi.org/10.22452/mjcs.sp2022no1.9","url":null,"abstract":"In the banking business, Machine Learning (ML) is critical for averting financial losses. Credit risk evaluation is perhaps the most important prediction task that may result in billions of dollars in damages each year (i.e., the risk of default on debt). Gradient Boosted Decision Tree (GBDT) models are now responsible for a large portion of the improvements in ML for predicting credit risk. However, these improvements begin to stagnate without adding pricey new data sources or carefully designed features. In this work, we describe our efforts to develop a unique Deep Learning (DL)-based technique for assessing credit risk that does not rely on additional model inputs. We present a new credit decision support approach with Gated Recurrent Unit (GRU) and Convolutional Neural Networks (CNN) that uses lengthy historical sequences of financial data while requiring few resources. We show that our DL technique, which uses Term Frequency-Inverse Document Frequency (TF-IDF) pre-classifiers, outperforms the benchmark models, resulting in considerable cost savings and early credit risk identification. We also show how our method may be utilized in a production setting, where our sampling methodology allows sequences to be effectively kept in memory and used for quick online learning and inference.","PeriodicalId":49894,"journal":{"name":"Malaysian Journal of Computer Science","volume":" ","pages":""},"PeriodicalIF":0.6,"publicationDate":"2022-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42486664","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}
Pub Date : 2022-01-31DOI: 10.22452/mjcs.vol35no1.3
Mojtaba Shokohinia, A. Dideban, F. Yaghmaee
Recently, many methods have been developed for representing knowledge, reasoning, and result extraction extracting results based on the respective domain knowledge in question. Despite the ontological success in knowledge representation, the reasoning method has faces some challenges. The main challenge in ontology reasoning methods is the failure in solving realization problems in the reasoning process. Apart from the complexity of solving realization problems, this already daunting challenge is compounded by computational complexity the time complexity of the solving realization problem solving process problems is equal to that of NEXP TIME. This important issue problem is achieved solved by solving the subsumption and satisfiability problems. Thus, to solve the realization problem, we first partition the ontology or extract partitions related to the query. Then, the satisfiability problem is solved by extracting partitions, and all concepts related to the query are extracted. This study proposes a method to overcome this problem, where a new solution is proposed with an appropriate time position. Finally, the efficiency of the proposed method, is evaluated against other reasoning engines, and the results show optimized performance vis-a-vis previous studies.
{"title":"A METHOD FOR IMPROVING REASONING AND REALIZATION PROBLEM SOLVING IN DESCRIPTIVE LOGIC- BASED AND ONTOLOGY-BASED REASONERS","authors":"Mojtaba Shokohinia, A. Dideban, F. Yaghmaee","doi":"10.22452/mjcs.vol35no1.3","DOIUrl":"https://doi.org/10.22452/mjcs.vol35no1.3","url":null,"abstract":"Recently, many methods have been developed for representing knowledge, reasoning, and result extraction extracting results based on the respective domain knowledge in question. Despite the ontological success in knowledge representation, the reasoning method has faces some challenges. The main challenge in ontology reasoning methods is the failure in solving realization problems in the reasoning process. Apart from the complexity of solving realization problems, this already daunting challenge is compounded by computational complexity the time complexity of the solving realization problem solving process problems is equal to that of NEXP TIME. This important issue problem is achieved solved by solving the subsumption and satisfiability problems. Thus, to solve the realization problem, we first partition the ontology or extract partitions related to the query. Then, the satisfiability problem is solved by extracting partitions, and all concepts related to the query are extracted. This study proposes a method to overcome this problem, where a new solution is proposed with an appropriate time position. Finally, the efficiency of the proposed method, is evaluated against other reasoning engines, and the results show optimized performance vis-a-vis previous studies.","PeriodicalId":49894,"journal":{"name":"Malaysian Journal of Computer Science","volume":" ","pages":""},"PeriodicalIF":0.6,"publicationDate":"2022-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43414768","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}
Pub Date : 2022-01-31DOI: 10.22452/mjcs.vol35no1.5
S. S, Karpagam G R, V. B
In the current era, as there is an abundant expansion of functionally similar web services, it becomes a prodigious issue for the web service discovery process. The service classification plays a significant role to greatly reduce the search space and retrieves the desirable service quickly and accurately. The classification is performed using the functional values. Recent research activities recommend RNN (Recurrent Neural Network) deep learning algorithms for efficient classification process. The state-of-the-art of GRU (Gated Recurrent Unit) one of the RNN model, provides a proficient classification process. However, the ratio of training and testing dataset, and hyperparameters namely neural network size, and batch size etc, affects the classification accuracy. The objective of the paper is to incorporate GRU model for efficient classification process. The novelty of the proposed model lies in implementing the GRU model for semantic web service classification. Furthermore, the genetic algorithm is used to predict the optimal ratio of training and testing dataset and optimal hidden neural Network units of GRU model in order to attain optimal classification. The experimental results exemplifies that the semantic web service classification is efficiently deliberated using the proposed GA-GRU model that outperforms the classification process as compared with the conventional semantic extraction using accuracy, precision, F-measure, recall and FDR (False Date Rate) rate.
{"title":"GENETIC ALGORITHM - OPTIMIZED GATED RECURRENT UNIT (GRU) NETWORK FOR SEMANTIC WEB SERVICES CLASSIFICATION","authors":"S. S, Karpagam G R, V. B","doi":"10.22452/mjcs.vol35no1.5","DOIUrl":"https://doi.org/10.22452/mjcs.vol35no1.5","url":null,"abstract":"In the current era, as there is an abundant expansion of functionally similar web services, it becomes a prodigious issue for the web service discovery process. The service classification plays a significant role to greatly reduce the search space and retrieves the desirable service quickly and accurately. The classification is performed using the functional values. Recent research activities recommend RNN (Recurrent Neural Network) deep learning algorithms for efficient classification process. The state-of-the-art of GRU (Gated Recurrent Unit) one of the RNN model, provides a proficient classification process. However, the ratio of training and testing dataset, and hyperparameters namely neural network size, and batch size etc, affects the classification accuracy. The objective of the paper is to incorporate GRU model for efficient classification process. The novelty of the proposed model lies in implementing the GRU model for semantic web service classification. Furthermore, the genetic algorithm is used to predict the optimal ratio of training and testing dataset and optimal hidden neural Network units of GRU model in order to attain optimal classification. The experimental results exemplifies that the semantic web service classification is efficiently deliberated using the proposed GA-GRU model that outperforms the classification process as compared with the conventional semantic extraction using accuracy, precision, F-measure, recall and FDR (False Date Rate) rate.","PeriodicalId":49894,"journal":{"name":"Malaysian Journal of Computer Science","volume":" ","pages":""},"PeriodicalIF":0.6,"publicationDate":"2022-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49267590","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}
Pub Date : 2022-01-31DOI: 10.22452/mjcs.vol35no1.4
R. Atallah, A. Kamsin, M. Ismail, A. S. Al-Shamayleh
Face recognition is one of the most approachable and accessible authentication methods. It is also accepted by users, as it is non-invasive. However, aging results in changes in the texture and shape of a face. Hence, age is one of the factors that decreases the accuracy of face recognition. Face aging, or age progression, is thus a significant challenge in face recognition methods. This paper presents the use of artificial neural network with model-agnostic meta-learning (ANN-MAML) for face-aging recognition. Model-agnostic meta-learning (MAML) is a meta-learning method used to train a model using parameters obtained from identical tasks with certain updates. This study aims to design and model a framework to recognize face aging based on artificial neural network. In addition, the face-aging recognition framework is evaluated against previous frameworks. Furthermore, the performance and the accuracy of ANN-MAML was evaluated using the CALFW (Cross-Age LFW) dataset. A comparison with other methods showed superior performance by ANN-MAML.
{"title":"NEURAL NETWORK WITH AGNOSTIC META-LEARNING MODEL FOR FACE-AGING RECOGNITION","authors":"R. Atallah, A. Kamsin, M. Ismail, A. S. Al-Shamayleh","doi":"10.22452/mjcs.vol35no1.4","DOIUrl":"https://doi.org/10.22452/mjcs.vol35no1.4","url":null,"abstract":"Face recognition is one of the most approachable and accessible authentication methods. It is also accepted by users, as it is non-invasive. However, aging results in changes in the texture and shape of a face. Hence, age is one of the factors that decreases the accuracy of face recognition. Face aging, or age progression, is thus a significant challenge in face recognition methods. This paper presents the use of artificial neural network with model-agnostic meta-learning (ANN-MAML) for face-aging recognition. Model-agnostic meta-learning (MAML) is a meta-learning method used to train a model using parameters obtained from identical tasks with certain updates. This study aims to design and model a framework to recognize face aging based on artificial neural network. In addition, the face-aging recognition framework is evaluated against previous frameworks. Furthermore, the performance and the accuracy of ANN-MAML was evaluated using the CALFW (Cross-Age LFW) dataset. A comparison with other methods showed superior performance by ANN-MAML.","PeriodicalId":49894,"journal":{"name":"Malaysian Journal of Computer Science","volume":" ","pages":""},"PeriodicalIF":0.6,"publicationDate":"2022-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42912571","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：