Pub Date : 2023-09-18DOI: 10.21512/commit.v17i2.8503
Rindy Claudia Setiawan, Andry Chowanda
Recognizing the intensity of the emotions is a paramount task for an affective system. By recognizing the intensity of the emotions, the system can have better human-computer interaction. The research explores several machine learning approaches with several different feature extraction method combinations to solve the emotion intensity prediction task while also analyzing and comparing it with several previous related papers. The research uses the dataset provided through theWASSA 2017 and SemEval 2018 competition. The dataset utilizes four of the eight basic emotions that Plutchik defines (anger, fear, joy, and sadness). The total data result in 19,736 rows of entry, with a total of 10,715 (54.3%) for training, 1,811 (9.17%) for validation, and 7,210 (36.53%) for testing. Three feature extraction methods are used and compared: N-gram, TFIDF, and Bag-of-Words. Meanwhile, machine learning algorithms are Linear Regression, Ridge Regression, KNearest Neighbor for Regression, Regression Tree, and Support Vector Regression (SVR). The results show that SVR with TF-IDF features has the best result of all attempted experiments, with a Pearson correlation score of 0.755 for all data and 0.647 for gold labels data. The final model also accepts newly seen data and displays the corresponding emotion label and intensity.
{"title":"Emotion Intensity Value Prediction with Machine Learning Approach on Twitter","authors":"Rindy Claudia Setiawan, Andry Chowanda","doi":"10.21512/commit.v17i2.8503","DOIUrl":"https://doi.org/10.21512/commit.v17i2.8503","url":null,"abstract":"Recognizing the intensity of the emotions is a paramount task for an affective system. By recognizing the intensity of the emotions, the system can have better human-computer interaction. The research explores several machine learning approaches with several different feature extraction method combinations to solve the emotion intensity prediction task while also analyzing and comparing it with several previous related papers. The research uses the dataset provided through theWASSA 2017 and SemEval 2018 competition. The dataset utilizes four of the eight basic emotions that Plutchik defines (anger, fear, joy, and sadness). The total data result in 19,736 rows of entry, with a total of 10,715 (54.3%) for training, 1,811 (9.17%) for validation, and 7,210 (36.53%) for testing. Three feature extraction methods are used and compared: N-gram, TFIDF, and Bag-of-Words. Meanwhile, machine learning algorithms are Linear Regression, Ridge Regression, KNearest Neighbor for Regression, Regression Tree, and Support Vector Regression (SVR). The results show that SVR with TF-IDF features has the best result of all attempted experiments, with a Pearson correlation score of 0.755 for all data and 0.647 for gold labels data. The final model also accepts newly seen data and displays the corresponding emotion label and intensity.","PeriodicalId":31276,"journal":{"name":"CommIT Journal","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135257294","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 : 2023-09-18DOI: 10.21512/commit.v17i2.8425
Ilvico Sonata, Yaya Heryadi, Antoni Wibowo, Widodo Budiharto
The development of autonomous cars is currently increasing along with the need for safe and comfortable autonomous cars. The development of autonomous cars cannot be separated from the use of deep learning to determine the steering angle of an autonomous car according to the road conditions it faces. In the research, a Vision Transformer (ViT) model is proposed to determine the steering angle based on images taken using a front-facing camera on an autonomous car. The dataset used to train ViT is a public dataset. The dataset is taken from streets around Rancho Palos Verdes and San Pedro, California. The number of images is 45,560, which are labeled with the steering angle value for each image. The proposed model can predict steering angle well. Then, the steering angle prediction results are compared using the same dataset with existing models. The experimental results show that the proposed model has better accuracy regarding the resulting MSE value of 2,991 compared to the CNN-based model of 5,358 and the CNN-LSTM combination model of 4,065. From the results of this experiment, the ViT model can replace the existing model, namely the CNN model and the combination model between CNN and LSTM, in predicting the steering angle of an autonomous car.
{"title":"End-to-End Steering Angle Prediction for Autonomous Car Using Vision Transformer","authors":"Ilvico Sonata, Yaya Heryadi, Antoni Wibowo, Widodo Budiharto","doi":"10.21512/commit.v17i2.8425","DOIUrl":"https://doi.org/10.21512/commit.v17i2.8425","url":null,"abstract":"The development of autonomous cars is currently increasing along with the need for safe and comfortable autonomous cars. The development of autonomous cars cannot be separated from the use of deep learning to determine the steering angle of an autonomous car according to the road conditions it faces. In the research, a Vision Transformer (ViT) model is proposed to determine the steering angle based on images taken using a front-facing camera on an autonomous car. The dataset used to train ViT is a public dataset. The dataset is taken from streets around Rancho Palos Verdes and San Pedro, California. The number of images is 45,560, which are labeled with the steering angle value for each image. The proposed model can predict steering angle well. Then, the steering angle prediction results are compared using the same dataset with existing models. The experimental results show that the proposed model has better accuracy regarding the resulting MSE value of 2,991 compared to the CNN-based model of 5,358 and the CNN-LSTM combination model of 4,065. From the results of this experiment, the ViT model can replace the existing model, namely the CNN model and the combination model between CNN and LSTM, in predicting the steering angle of an autonomous car.","PeriodicalId":31276,"journal":{"name":"CommIT Journal","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135257295","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 : 2023-09-18DOI: 10.21512/commit.v17i2.8872
Afiyah Rifkha Rahmika, Zulkifli Tahir, Ady Wahyudi Paundu, Zahir Zainuddin
Demands for information over the Internet massively increase through the continuous expansion of website applications. Therefore, generating powerful and efficient server architecture for web servers is a must to satisfy Internet users and avoid the overloaded system. The research focuses on developing a new mechanism for load balancing to distribute incoming HTTP requests in website applications by combining the Least Connection algorithm and Multi-Agent System (LC-MAS). The proposed mechanism distributes the request based on load condition and the fewest number of active connections. The research applies virtualization technology to build servers on this proposed mechanism. The architecture is built inside a physical server with Proxmox as virtualization management and Linux Debian 7.11 as an operating system. Then, the research is tested in two scenarios (LCMAS and LC) using 500, 1,000, and 1,500 requests. The performance of this proposed mechanism is measured through the values of average response time, throughput, and error percentage. The results show that the proposed mechanism (LC-MAS) distributes the workload more equally than LC, with an average response time for 1,500 requests of 1338.8 milliseconds, 20.07% error, and 125 transactions per second. The LC-MAS makes the website application performance much better when the request increases. The LC-MAS helps in the utilization of system resources and improves system robustness.
{"title":"Web Server Load Balancing Mechanism with Least Connection Algorithm and Multi-Agent System","authors":"Afiyah Rifkha Rahmika, Zulkifli Tahir, Ady Wahyudi Paundu, Zahir Zainuddin","doi":"10.21512/commit.v17i2.8872","DOIUrl":"https://doi.org/10.21512/commit.v17i2.8872","url":null,"abstract":"Demands for information over the Internet massively increase through the continuous expansion of website applications. Therefore, generating powerful and efficient server architecture for web servers is a must to satisfy Internet users and avoid the overloaded system. The research focuses on developing a new mechanism for load balancing to distribute incoming HTTP requests in website applications by combining the Least Connection algorithm and Multi-Agent System (LC-MAS). The proposed mechanism distributes the request based on load condition and the fewest number of active connections. The research applies virtualization technology to build servers on this proposed mechanism. The architecture is built inside a physical server with Proxmox as virtualization management and Linux Debian 7.11 as an operating system. Then, the research is tested in two scenarios (LCMAS and LC) using 500, 1,000, and 1,500 requests. The performance of this proposed mechanism is measured through the values of average response time, throughput, and error percentage. The results show that the proposed mechanism (LC-MAS) distributes the workload more equally than LC, with an average response time for 1,500 requests of 1338.8 milliseconds, 20.07% error, and 125 transactions per second. The LC-MAS makes the website application performance much better when the request increases. The LC-MAS helps in the utilization of system resources and improves system robustness.","PeriodicalId":31276,"journal":{"name":"CommIT Journal","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135258144","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 : 2023-09-08DOI: 10.21512/commit.v17i2.9053
Djoni Haryadi Setiabudi, Michael Santoso
Educational data mining is a technique to evaluate educational process of university students, especially in their early stages. Most preliminary studies focus on observing courses undertaken by students from one semester to the next to predict their success rate. However, besides studying, many students are also involved in non-academic activities, which tends to affect their grades. Therefore, the research aims to determine the effect of student activities on grades while taking into account their academic activities. The method used for clustering is K-Means. Data are collected by observing students’ activity patterns in lectures. The research is conducted in two study programs at Petra Christian University: Business Management and Architecture. The results show that the K-Means method gives good results. The clusters formed from the data show non-homogenous groups and produce insights from several groups. The results show a tendency for students’ performance to increase along with the number of activities and points earned. Most students have increased activities during busy times in the third, fourth, fifth, and sixth semesters. The peak is between the fifth and sixth semesters. Then, it starts to decrease in the seventh and eighth semesters. Therefore, students’ activities in the Business Management study program affect performance significantly. Meanwhile, in the Architecture study program, it has an insignificant effect on performance.
{"title":"Effect of Students’ Activities on Academic Performance Using Clustering Evolution Analysis","authors":"Djoni Haryadi Setiabudi, Michael Santoso","doi":"10.21512/commit.v17i2.9053","DOIUrl":"https://doi.org/10.21512/commit.v17i2.9053","url":null,"abstract":"Educational data mining is a technique to evaluate educational process of university students, especially in their early stages. Most preliminary studies focus on observing courses undertaken by students from one semester to the next to predict their success rate. However, besides studying, many students are also involved in non-academic activities, which tends to affect their grades. Therefore, the research aims to determine the effect of student activities on grades while taking into account their academic activities. The method used for clustering is K-Means. Data are collected by observing students’ activity patterns in lectures. The research is conducted in two study programs at Petra Christian University: Business Management and Architecture. The results show that the K-Means method gives good results. The clusters formed from the data show non-homogenous groups and produce insights from several groups. The results show a tendency for students’ performance to increase along with the number of activities and points earned. Most students have increased activities during busy times in the third, fourth, fifth, and sixth semesters. The peak is between the fifth and sixth semesters. Then, it starts to decrease in the seventh and eighth semesters. Therefore, students’ activities in the Business Management study program affect performance significantly. Meanwhile, in the Architecture study program, it has an insignificant effect on performance.","PeriodicalId":31276,"journal":{"name":"CommIT Journal","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136364049","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 : 2023-09-08DOI: 10.21512/commit.v17i2.8518
Yashella Tirana, Sfenrianto Sfenrianto
Many users complain about using the largest mobile Internet Service Provider (ISP) application in Indonesia, MyIndihome, such as difficulties in verifying, logging in, and changing cell phone numbers and emails. With these complaints, the satisfaction of the MyIndihome application users decreases. The research aims to determine the effect of information quality, system quality, service quality, ease of use, usefulness, and chatbot effectiveness on user satisfaction with MyIndihome. Chatbot effectiveness is a novelty of the research because it has not been studied in previous research. The research applies a quantitative approach. Then the sampling technique used is probability sampling, and the method is simple random sampling with 417 respondents. Data collection techniques are carried out by distributing online questionnaires, and the data are statistically processed with SmartPLS and analyzed by Structural Equation Model (SEM). After carrying out several stages of testing from validity tests, reliability tests, and structural models, the results show that information quality, system quality, ease of use, usability, and chatbot effectiveness have a significant effect on user satisfaction. However, the service quality has no effect. These results can help companies to increase user satisfaction with the MyIndihome application. They can increase the variables that influence user satisfaction with the MyIndihome application.
{"title":"Factors on Mobile Application User Satisfaction in the Largest Indonesian Internet Service Provider (ISP)","authors":"Yashella Tirana, Sfenrianto Sfenrianto","doi":"10.21512/commit.v17i2.8518","DOIUrl":"https://doi.org/10.21512/commit.v17i2.8518","url":null,"abstract":"Many users complain about using the largest mobile Internet Service Provider (ISP) application in Indonesia, MyIndihome, such as difficulties in verifying, logging in, and changing cell phone numbers and emails. With these complaints, the satisfaction of the MyIndihome application users decreases. The research aims to determine the effect of information quality, system quality, service quality, ease of use, usefulness, and chatbot effectiveness on user satisfaction with MyIndihome. Chatbot effectiveness is a novelty of the research because it has not been studied in previous research. The research applies a quantitative approach. Then the sampling technique used is probability sampling, and the method is simple random sampling with 417 respondents. Data collection techniques are carried out by distributing online questionnaires, and the data are statistically processed with SmartPLS and analyzed by Structural Equation Model (SEM). After carrying out several stages of testing from validity tests, reliability tests, and structural models, the results show that information quality, system quality, ease of use, usability, and chatbot effectiveness have a significant effect on user satisfaction. However, the service quality has no effect. These results can help companies to increase user satisfaction with the MyIndihome application. They can increase the variables that influence user satisfaction with the MyIndihome application.","PeriodicalId":31276,"journal":{"name":"CommIT Journal","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136364048","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}
Digitalization is inevitable, including in the health sector. The iPosyandu, a mobile digital platform, is introduced to help the report of Community Health Workers (CHWs) and monitor the Pos Pelayanan Terpadu (Posyandu - Integrated Healthcare Center) data online. Unfortunately, CHWs still report data manually using paper, which takes a long time to store because some are still reluctant to change to digital services. Therefore, it is necessary to study CHWs’ intention to create new values and accept technology to sustain the application. The research aims to determine the factor influencing the intention to participate in value co-creation and use iPosyandu by extending the Unified Theory of Acceptance and Use of Technology (UTAUT) among CHW. A Partial Least Square-Structural Equation Modelling (PLS-SEM) is conducted with a cross-sectional survey involving 222 CHWs in Purwakarta, Indonesia. The research finds that effort expectancy and perceived policy support significantly affect the intention to participate in value co-creation and usage of iPosyandu. The findings highlight that the critical role of intention to participate in value co-creation significantly affects the intention to use iPosyandu. The findings also suggest that policymakers and application developers should increase the use of iPosyandu by improving the effort systems, providing policy support, and facilitating CHWs to cocreate the value of the application to encourage them to use iPosyandu.
{"title":"Understanding Participation in Value Co-Creation and Acceptance of iPosyandu by Extending UTAUT among Community Health Workers","authors":"Azmii Lathifah, Utomo Sarjono Putro, Fedri Ruluwedrata Rinawan, Santi Novani, Valid Hasyimi, Adhya Rare Tiara","doi":"10.21512/commit.v17i2.8567","DOIUrl":"https://doi.org/10.21512/commit.v17i2.8567","url":null,"abstract":"Digitalization is inevitable, including in the health sector. The iPosyandu, a mobile digital platform, is introduced to help the report of Community Health Workers (CHWs) and monitor the Pos Pelayanan Terpadu (Posyandu - Integrated Healthcare Center) data online. Unfortunately, CHWs still report data manually using paper, which takes a long time to store because some are still reluctant to change to digital services. Therefore, it is necessary to study CHWs’ intention to create new values and accept technology to sustain the application. The research aims to determine the factor influencing the intention to participate in value co-creation and use iPosyandu by extending the Unified Theory of Acceptance and Use of Technology (UTAUT) among CHW. A Partial Least Square-Structural Equation Modelling (PLS-SEM) is conducted with a cross-sectional survey involving 222 CHWs in Purwakarta, Indonesia. The research finds that effort expectancy and perceived policy support significantly affect the intention to participate in value co-creation and usage of iPosyandu. The findings highlight that the critical role of intention to participate in value co-creation significantly affects the intention to use iPosyandu. The findings also suggest that policymakers and application developers should increase the use of iPosyandu by improving the effort systems, providing policy support, and facilitating CHWs to cocreate the value of the application to encourage them to use iPosyandu.","PeriodicalId":31276,"journal":{"name":"CommIT Journal","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136364988","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 : 2023-09-06DOI: 10.21512/commit.v17i2.8446
Natasya Elora Carissa, Muhammad Erlangga, Cindy Sonesha Evik, Putu Wuri Handayani
The rapid development of Subscription Video on Demand (SVoD) services in Indonesia makes it promising. The change in consumers’ behavior from watching movies through television channels and cinemas to online streaming has encouraged industry players to look for innovation. The research aims to analyze factors influencing the intention to continue using SVoD with Netflix as the case study. The research combines two theories, namely Information System Success (ISS) Model and Expectation Confirmation Theory (ECT). The research also adds an aspect of personalization which is one of the characteristics of SVoD services. There are 623 respondents who have used Netflix’s SVoD service at least once (purposive sampling) to participate in the research. The data are analyzed using the covariancebased structural equation model and facilitated using AMOS 26 program. The results indicate that service quality has a positive effect on confirmation. Then, system quality has a positive effect on perceived usefulness, and confirmation has a positive effect on satisfaction. Moreover, satisfaction, perceived usefulness, and personalization positively affect continuance intention to use SVoD services. Based on these results, the research is expected to contribute to SVoD service providers to evaluate their services so that users have the intention to continue using the SVoD services.
{"title":"The Influence of Perceived Usefulness, Satisfaction, and Personalization on Subscription Video on Demand Continuance Intentions","authors":"Natasya Elora Carissa, Muhammad Erlangga, Cindy Sonesha Evik, Putu Wuri Handayani","doi":"10.21512/commit.v17i2.8446","DOIUrl":"https://doi.org/10.21512/commit.v17i2.8446","url":null,"abstract":"The rapid development of Subscription Video on Demand (SVoD) services in Indonesia makes it promising. The change in consumers’ behavior from watching movies through television channels and cinemas to online streaming has encouraged industry players to look for innovation. The research aims to analyze factors influencing the intention to continue using SVoD with Netflix as the case study. The research combines two theories, namely Information System Success (ISS) Model and Expectation Confirmation Theory (ECT). The research also adds an aspect of personalization which is one of the characteristics of SVoD services. There are 623 respondents who have used Netflix’s SVoD service at least once (purposive sampling) to participate in the research. The data are analyzed using the covariancebased structural equation model and facilitated using AMOS 26 program. The results indicate that service quality has a positive effect on confirmation. Then, system quality has a positive effect on perceived usefulness, and confirmation has a positive effect on satisfaction. Moreover, satisfaction, perceived usefulness, and personalization positively affect continuance intention to use SVoD services. Based on these results, the research is expected to contribute to SVoD service providers to evaluate their services so that users have the intention to continue using the SVoD services.","PeriodicalId":31276,"journal":{"name":"CommIT Journal","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135205064","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}
Analyzing what consumer needs remains every day’s challenge for every business. Every business entity requires continuous effort as consumers become more demanding and have more access to product/service offerings, leading to more competitive market dynamics and the necessity for more innovative ways of offering products/services. The research aims to recommend a set of customer attributes for the studied company and analyze the selected attributes using a combination of Importance Performance Analysis (IPA) and fuzzy Kano. The research is a case study of a company selling gift vouchers for individual and corporate consumers. The research combines literature study and affinity diagram workshop to identify the required consumer attributes, which are analyzed using the integration of IPA and fuzzy Kano. The results suggest that the studied company should concentrate on several attributes, such as A7-simple requirement during the purchasing process, A10-no administration fee during purchase, A14-cross promotion with various sister brands, and A15-no minimum purchase. The attributes fall under “concentrate here” in the IPA grid while at the same time, those are considered as “effective improving area” in the fuzzy Kano grid. The studied company is also recommended to keep their good work on the attribute of A5-expiry date longer than one year so that it remains their competitive attribute and does not fall into the other inferior quadrants.
{"title":"Classifying Customer Attributes with Importance Performance Analysis and Fuzzy Kano","authors":"Elia Oey, Nyimas Revita Permaisuri Putri, Benyamin Suwito Rahardjo","doi":"10.21512/commit.v17i2.8534","DOIUrl":"https://doi.org/10.21512/commit.v17i2.8534","url":null,"abstract":"Analyzing what consumer needs remains every day’s challenge for every business. Every business entity requires continuous effort as consumers become more demanding and have more access to product/service offerings, leading to more competitive market dynamics and the necessity for more innovative ways of offering products/services. The research aims to recommend a set of customer attributes for the studied company and analyze the selected attributes using a combination of Importance Performance Analysis (IPA) and fuzzy Kano. The research is a case study of a company selling gift vouchers for individual and corporate consumers. The research combines literature study and affinity diagram workshop to identify the required consumer attributes, which are analyzed using the integration of IPA and fuzzy Kano. The results suggest that the studied company should concentrate on several attributes, such as A7-simple requirement during the purchasing process, A10-no administration fee during purchase, A14-cross promotion with various sister brands, and A15-no minimum purchase. The attributes fall under “concentrate here” in the IPA grid while at the same time, those are considered as “effective improving area” in the fuzzy Kano grid. The studied company is also recommended to keep their good work on the attribute of A5-expiry date longer than one year so that it remains their competitive attribute and does not fall into the other inferior quadrants.","PeriodicalId":31276,"journal":{"name":"CommIT Journal","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135205783","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}
In court, criminal investigations and identity management tools, like check-in and payment logins, face videos, and photos, are used as evidence more frequently. Although deeply falsified information may be found using deep learning classifiers, block-box decisionmaking makes forensic investigation in criminal trials more challenging. Therefore, the research suggests a three-step classification technique to classify the deceptive deepfake image content. The research examines the visual assessments of an EfficientNet and Shifted Window Transformer (SWinT) hybrid model based on Convolutional Neural Network (CNN) and Transformer architectures. The classifier generality is improved in the first stage using a different augmentation. Then, the hybrid model is developed in the second step by combining the EfficientNet and Shifted Window Transformer architectures. Next, the GradCAM approach for assessing human understanding demonstrates deepfake visual interpretation. In 14,204 images for the validation set, there are 7,096 fake photos and 7,108 real images. In contrast to focusing only on a few discrete face parts, the research shows that the entire deepfake image should be investigated. On a custom dataset of real, Generative Adversarial Networks (GAN)-generated, and human-altered web photos, the proposed method achieves an accuracy of 98.45%, a recall of 99.12%, and a loss of 0.11125. The proposed method successfully distinguishes between real and manipulated images. Moreover, the presented approach can assist investigators in clarifying the composition of the artificially produced material.
{"title":"Classification of Deepfake Images Using a Novel Explanatory Hybrid Model","authors":"Sudarshana Kerenalli, Vamsidhar Yendapalli, Mylarareddy Chinnaiah","doi":"10.21512/commit.v17i2.8761","DOIUrl":"https://doi.org/10.21512/commit.v17i2.8761","url":null,"abstract":"In court, criminal investigations and identity management tools, like check-in and payment logins, face videos, and photos, are used as evidence more frequently. Although deeply falsified information may be found using deep learning classifiers, block-box decisionmaking makes forensic investigation in criminal trials more challenging. Therefore, the research suggests a three-step classification technique to classify the deceptive deepfake image content. The research examines the visual assessments of an EfficientNet and Shifted Window Transformer (SWinT) hybrid model based on Convolutional Neural Network (CNN) and Transformer architectures. The classifier generality is improved in the first stage using a different augmentation. Then, the hybrid model is developed in the second step by combining the EfficientNet and Shifted Window Transformer architectures. Next, the GradCAM approach for assessing human understanding demonstrates deepfake visual interpretation. In 14,204 images for the validation set, there are 7,096 fake photos and 7,108 real images. In contrast to focusing only on a few discrete face parts, the research shows that the entire deepfake image should be investigated. On a custom dataset of real, Generative Adversarial Networks (GAN)-generated, and human-altered web photos, the proposed method achieves an accuracy of 98.45%, a recall of 99.12%, and a loss of 0.11125. The proposed method successfully distinguishes between real and manipulated images. Moreover, the presented approach can assist investigators in clarifying the composition of the artificially produced material.","PeriodicalId":31276,"journal":{"name":"CommIT Journal","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135205784","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 : 2023-09-06DOI: 10.21512/commit.v17i2.8272
Yohan Adhi Styoutomo, Yova Ruldeviyani
XYZ financial institution is a government institution that receives and processes transaction reports from banks and remittances, so its data classification is very confidential. However, during the Work from Home (WFH) policy in the Covid-19 pandemic, XYZ financial institution has received many spam/phishing attacks. Hence, this incident shows that some employees need an awareness of information security. The research offers a different Information Security Awareness (ISA) questionnaire using the Human Aspects of the Information Security Questionnaire (HAIS-Q) and ISO/IEC 27001:2013 as focus areas. The research uses the theory of Knowledge, Attitude, and Behavior (KAB) to determine the dimensions that need improvement and priority ranking using Fuzzy Analytical Hierarchy Process (FAHP). Furthermore, the research conducts a Focus Group Discussion (FGD) to explore the root causes of employee behavior. The FGD results show that there are still employees who do not know about information security, such as password combinations and length, so limited knowledge affects employees’ attitudes and behaviors. The research results from 34 respondents show that the employees’ information security awareness level is in the moderate category (78.8%). They still need to increase their awareness of information security, especially in managing passwords, using email and the Internet, and reporting incidents. Recommendations have been prepared to improve the dimensions and areas that have yet to be categorized as good. In the future, the ISA questionnaire is expected to be used in other organizations.
XYZ金融机构是一家政府机构，接收和处理来自银行和汇款的交易报告，因此其数据分类是非常机密的。然而，在Covid-19大流行期间，在家工作(WFH)政策期间，XYZ金融机构收到了许多垃圾邮件/网络钓鱼攻击。因此，这次事件表明，一些员工需要有信息安全意识。该研究提供了一个不同的信息安全意识(ISA)问卷，使用信息安全问卷(HAIS-Q)和ISO/IEC 27001:2013作为重点领域。本研究运用知识、态度和行为理论(Knowledge, Attitude, and Behavior, KAB)确定需要改进的维度，并运用模糊层次分析法(FAHP)进行优先级排序。此外，本研究还通过焦点小组讨论(Focus Group Discussion, FGD)来探讨员工行为的根本原因。FGD结果显示，仍然有员工不了解信息安全，如密码组合和长度，有限的知识影响了员工的态度和行为。34名受访者的调查结果显示，员工的信息安全意识水平处于中等水平(78.8%)。他们仍然需要提高他们的信息安全意识，特别是在管理密码、使用电子邮件和互联网以及报告事件方面。已经提出了建议，以改进尚未归类为良好的方面和领域。将来，预计内部审查制度调查表将在其他组织中使用。
{"title":"Information Security Awareness Raising Strategy Using Fuzzy AHP Method with HAIS-Q and ISO/IEC 27001:2013: A Case Study of XYZ Financial Institution","authors":"Yohan Adhi Styoutomo, Yova Ruldeviyani","doi":"10.21512/commit.v17i2.8272","DOIUrl":"https://doi.org/10.21512/commit.v17i2.8272","url":null,"abstract":"XYZ financial institution is a government institution that receives and processes transaction reports from banks and remittances, so its data classification is very confidential. However, during the Work from Home (WFH) policy in the Covid-19 pandemic, XYZ financial institution has received many spam/phishing attacks. Hence, this incident shows that some employees need an awareness of information security. The research offers a different Information Security Awareness (ISA) questionnaire using the Human Aspects of the Information Security Questionnaire (HAIS-Q) and ISO/IEC 27001:2013 as focus areas. The research uses the theory of Knowledge, Attitude, and Behavior (KAB) to determine the dimensions that need improvement and priority ranking using Fuzzy Analytical Hierarchy Process (FAHP). Furthermore, the research conducts a Focus Group Discussion (FGD) to explore the root causes of employee behavior. The FGD results show that there are still employees who do not know about information security, such as password combinations and length, so limited knowledge affects employees’ attitudes and behaviors. The research results from 34 respondents show that the employees’ information security awareness level is in the moderate category (78.8%). They still need to increase their awareness of information security, especially in managing passwords, using email and the Internet, and reporting incidents. Recommendations have been prepared to improve the dimensions and areas that have yet to be categorized as good. In the future, the ISA questionnaire is expected to be used in other organizations.","PeriodicalId":31276,"journal":{"name":"CommIT Journal","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135205264","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：