Arif Djunaidy
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Arif Djunaidy has a total of 45 co-authors. Here are the top co-authors by number of shared publications:
- Daniel Siahaan
Department of Informatics, Institut Teknologi Sepuluh Nopember, Surabaya, Indonesia
11 shared publications - Evi Triandini
Institut Teknologi dan Bisnis STIKOM Bali
4 shared publications - Ahmad Muklason
Institut Teknologi Sepuluh Nopember
2 shared publications Andi Tenriawaru
2 shared publicationsEdwin Riksakomara
Department of Information Systems, Institut Teknologi Sepuluh Nopember, Surabaya, Indonesia
2 shared publicationsEva Hariyanti
Department of Information Systems, Universitas Airlangga, Surabaya, East Java, 60115
2 shared publications
Latest Publications
- Open Access
- 20 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 108284-108305
Multidimensional Readiness Evaluation of Smart Tourism Destinations: A Natural Language Processing and Thematic Analysis Approach
The rapid expansion of the tourism sector, driven by technological progress, has led to smart tourism, improving visitor experiences and streamlining destination administration. In underdeveloped nations, such as Indonesia, destinations often are not adequately prepared to use smart technology, resulting in inefficiencies and unsuccessful projects. Existing models for evaluating tourism progress typically concentrate on specific regions or isolated dimensions, lacking a comprehensive assessment of governance, socioeconomic conditions, and information technology (IT) awareness. This research addresses this gap by introducing a multidimensional framework for evaluating smart tourism readiness based on the Technology-Organization-Environment (TOE) paradigm. The study starts with a thorough review of existing literature using the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) method to find key factors for readiness, and it uses a combination of both qualitative and quantitative research methods. We conducted a quantitative examination using natural language processing (NLP) to cluster the dimensions of the existing evaluation models. In the qualitative phase, two validation processes were conducted. The first process involved expert interviews with stakeholders to validate the dimensions identified in the previous step. In the second validation process, we simplified the information by matching the proposed dimensions with IT governance components from well-known frameworks like COBIT and ITIL 4. This mapping ensures the proposed dimensions are aligned with IT governance components, supporting the effective management of smart technology in smart tourism destinations. The study identifies seven critical readiness dimensions as a key finding: governance, sustainability, online tourism services, smart technology, accessibility, innovation culture, and socioeconomic conditions. The findings contribute to a holistic readiness evaluation framework, offering actionable insights for policymakers and tourism stakeholders to inform strategic planning and policy development. This comprehensive model supports the successful implementation of smart tourism initiatives, addressing the unique challenges of developing countries and fostering long-term sustainability.
- 20 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 108284-108305
Open AccessMultidimensional Readiness Evaluation of Smart Tourism Destinations: A Natural Language Processing and Thematic Analysis Approach
The rapid expansion of the tourism sector, driven by technological progress, has led to smart tourism, improving visitor experiences and streamlining destination administration. In underdeveloped nations, such as Indonesia, destinations often are not adequately prepared to use smart technology, resulting in inefficiencies and unsuccessful projects. Existing models for evaluating tourism progress typically concentrate on specific regions or isolated dimensions, lacking a comprehensive assessment of governance, socioeconomic conditions, and information technology (IT) awareness. This research addresses this gap by introducing a multidimensional framework for evaluating smart tourism readiness based on the Technology-Organization-Environment (TOE) paradigm. The study starts with a thorough review of existing literature using the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) method to find key factors for readiness, and it uses a combination of both qualitative and quantitative research methods. We conducted a quantitative examination using natural language processing (NLP) to cluster the dimensions of the existing evaluation models. In the qualitative phase, two validation processes were conducted. The first process involved expert interviews with stakeholders to validate the dimensions identified in the previous step. In the second validation process, we simplified the information by matching the proposed dimensions with IT governance components from well-known frameworks like COBIT and ITIL 4. This mapping ensures the proposed dimensions are aligned with IT governance components, supporting the effective management of smart technology in smart tourism destinations. The study identifies seven critical readiness dimensions as a key finding: governance, sustainability, online tourism services, smart technology, accessibility, innovation culture, and socioeconomic conditions. The findings contribute to a holistic readiness evaluation framework, offering actionable insights for policymakers and tourism stakeholders to inform strategic planning and policy development. This comprehensive model supports the successful implementation of smart tourism initiatives, addressing the unique challenges of developing countries and fostering long-term sustainability.
- Open Access
- 12 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 144193-144207
Development of Dynamic Personalized Learning Paths Based on Knowledge Preferences and the Ant Colony Algorithm
The importance of personalized learning paths is acknowledged as an effective solution for students with diverse learning characteristics. However, implementing learning paths in real classrooms presents challenges, particularly in assigning teachers to create a unique learning path for each student, considering their diverse preferences and the difficulty levels of learning objects, which will be time-consuming. To overcome manual learning path design, we propose an ant colony to create automatic, personalized learning paths based on user preferences and the difficulty level of the learning objects. Furthermore, we have implemented a system capable of generating dynamic personalized learning paths based on knowledge preferences to address issues related to the over- and under-prediction of student ability. Additionally, the system has the ability to automatically classify learning objects according to their difficulty levels, guaranteeing regular updates. We compared user experience perception and the impact of personalized learning paths with conventional e-learning to evaluate the effectiveness of the proposed method. The experiments showed that integrating personalized learning paths resulted in an impressive 71% increase in student performance. In comparison, conventional e-learning only improved student performance by 54%. These findings suggest that the proposed system presents promising opportunities for enhanced learning outcomes.
- 12 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 144193-144207
Open AccessDevelopment of Dynamic Personalized Learning Paths Based on Knowledge Preferences and the Ant Colony Algorithm
The importance of personalized learning paths is acknowledged as an effective solution for students with diverse learning characteristics. However, implementing learning paths in real classrooms presents challenges, particularly in assigning teachers to create a unique learning path for each student, considering their diverse preferences and the difficulty levels of learning objects, which will be time-consuming. To overcome manual learning path design, we propose an ant colony to create automatic, personalized learning paths based on user preferences and the difficulty level of the learning objects. Furthermore, we have implemented a system capable of generating dynamic personalized learning paths based on knowledge preferences to address issues related to the over- and under-prediction of student ability. Additionally, the system has the ability to automatically classify learning objects according to their difficulty levels, guaranteeing regular updates. We compared user experience perception and the impact of personalized learning paths with conventional e-learning to evaluate the effectiveness of the proposed method. The experiments showed that integrating personalized learning paths resulted in an impressive 71% increase in student performance. In comparison, conventional e-learning only improved student performance by 54%. These findings suggest that the proposed system presents promising opportunities for enhanced learning outcomes.
- Open Access
- 17 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 13878-13892
Concept–Effect Relationship Weighting Based on Frequency of Concept’s Co-Occurrence for Developing Personalized Remedial Learning Path
Notably, a personalized learning path is advantageous for students who need remedial assistance due to their struggle to master concepts and poor exam performance. Personalized learning paths produced from conceptual maps have proven highly effective. Nevertheless, in establishing personalized learning paths using conceptual maps, the manual assignment of weights that depict the degree of relationship between test items and concepts is time-consuming and ineffective. To get around the problems with giving weights by hand, we suggest an auto-generated concept relationship weighting method that is based on how often primary concepts and supporting concepts appear together in test items in this study. Moreover, we build a system that generates a personalized remedial learning path for students using the proposed weighting method and a conceptual map. To evaluate the effectiveness of the proposed approach, we carried out an experiment on high school students. As shown by the experimental results, the personalized remedial learning path generated can significantly enhance the learning achievement of underachieving students. In addition, the new auto-generated weighting method eliminated the need for experts to assign weights and resolved conflicting weight values.
- 17 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 13878-13892
Open AccessConcept–Effect Relationship Weighting Based on Frequency of Concept’s Co-Occurrence for Developing Personalized Remedial Learning Path
Notably, a personalized learning path is advantageous for students who need remedial assistance due to their struggle to master concepts and poor exam performance. Personalized learning paths produced from conceptual maps have proven highly effective. Nevertheless, in establishing personalized learning paths using conceptual maps, the manual assignment of weights that depict the degree of relationship between test items and concepts is time-consuming and ineffective. To get around the problems with giving weights by hand, we suggest an auto-generated concept relationship weighting method that is based on how often primary concepts and supporting concepts appear together in test items in this study. Moreover, we build a system that generates a personalized remedial learning path for students using the proposed weighting method and a conceptual map. To evaluate the effectiveness of the proposed approach, we carried out an experiment on high school students. As shown by the experimental results, the personalized remedial learning path generated can significantly enhance the learning achievement of underachieving students. In addition, the new auto-generated weighting method eliminated the need for experts to assign weights and resolved conflicting weight values.
- Open Access
- 24 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 58-63
Comparison of Deep Learning Methods in Detecting Hate Speech in Indonesian Tweets
Hate speech has negative effects on both the targeted victims and the listeners. The dissemination of hate speech can occur not only physically or verbally, but also in writing on social media. The emergence of hate speech on social media platforms can be difficult to identify in written communication. Currently, hate speech detection relies on machine learning. This study generates a vector representation of words using three pre-trained word insertion models: Global Vectors (GloVe), FastText, and Bidirectional Encoder Representations from Transformers (BERT). Synthetic Minority Oversampling Technique (SMOTE) and Random Over Sampling (ROS) were utilized as balancing methods to rectify data imbalance between classes. In addition, three distinct deep learning architectures were used to identify sentence-level hate speech in Indonesian tweets: Bidirectional Long Sort-Term Memory (BiLSTM), Convolution Neural Network (CNN), and Recurrent Neural Network (RNN). The dataset was collected by crawling the data via the Twitter API. After data underwent preprocessing, characteristics were extracted. Based on experimental results, classifiers employing RNN and BERT embedding and utilizing SMOTE produced the most accurate results (95.5%).
- 24 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 58-63
Open AccessComparison of Deep Learning Methods in Detecting Hate Speech in Indonesian Tweets
Hate speech has negative effects on both the targeted victims and the listeners. The dissemination of hate speech can occur not only physically or verbally, but also in writing on social media. The emergence of hate speech on social media platforms can be difficult to identify in written communication. Currently, hate speech detection relies on machine learning. This study generates a vector representation of words using three pre-trained word insertion models: Global Vectors (GloVe), FastText, and Bidirectional Encoder Representations from Transformers (BERT). Synthetic Minority Oversampling Technique (SMOTE) and Random Over Sampling (ROS) were utilized as balancing methods to rectify data imbalance between classes. In addition, three distinct deep learning architectures were used to identify sentence-level hate speech in Indonesian tweets: Bidirectional Long Sort-Term Memory (BiLSTM), Convolution Neural Network (CNN), and Recurrent Neural Network (RNN). The dataset was collected by crawling the data via the Twitter API. After data underwent preprocessing, characteristics were extracted. Based on experimental results, classifiers employing RNN and BERT embedding and utilizing SMOTE produced the most accurate results (95.5%).
- Open Access
- 3 May 2023
- Published by F1000 Research Ltd in F1000Research
- Vol. 12, 462
Adaptive modeling for security vulnerability propagation to predict the impact of business process redesign
Background: Business process redesign (BPR) is typical in organizations and is followed by adaptive maintenance on supporting applications. However, BPR leads to information security vulnerabilities that can propagate to its supporting applications. Methods: This study proposes a new method called Node Strength-based Vulnerability Modeling (NSVM) for modeling security vulnerability propagation in the business processes and IT service layers. We applied the concept of social network strength to build our propagation model. The propagation model is needed to predict the impact of BPR on application vulnerabilities. We chose e-commerce applications as a case study. We evaluated the vulnerability propagation model by comparing the predicted vulnerability scores from the model with the actual scores of e-commerce applications in the National Vulnerability Database. Results: Our experimentation indicates that the propagation strength between nodes is influenced by Common Weakness Enumerations (CWEs) between them. Thus, the vulnerability propagation model can predict vulnerability scores at module nodes in the IT service layer. In the NSVM, the best prediction scores were obtained by aggregating the adjacency and initial scores using the maximum principle approach. The best evaluation results yield mean absolute error (MAE), root mean squared error (RMSE), and mean squared error (MSE) scores of 0.60, 1.44, and 1.16, respectively. Conclusion: Our study shows that the vulnerability propagation model with an adaptive mechanism based on BPR can be used to predict security vulnerability scores as the impact of business process redesign.
- 3 May 2023
- Published by F1000 Research Ltd in F1000Research
- Vol. 12, 462
Open AccessAdaptive modeling for security vulnerability propagation to predict the impact of business process redesign
Background: Business process redesign (BPR) is typical in organizations and is followed by adaptive maintenance on supporting applications. However, BPR leads to information security vulnerabilities that can propagate to its supporting applications. Methods: This study proposes a new method called Node Strength-based Vulnerability Modeling (NSVM) for modeling security vulnerability propagation in the business processes and IT service layers. We applied the concept of social network strength to build our propagation model. The propagation model is needed to predict the impact of BPR on application vulnerabilities. We chose e-commerce applications as a case study. We evaluated the vulnerability propagation model by comparing the predicted vulnerability scores from the model with the actual scores of e-commerce applications in the National Vulnerability Database. Results: Our experimentation indicates that the propagation strength between nodes is influenced by Common Weakness Enumerations (CWEs) between them. Thus, the vulnerability propagation model can predict vulnerability scores at module nodes in the IT service layer. In the NSVM, the best prediction scores were obtained by aggregating the adjacency and initial scores using the maximum principle approach. The best evaluation results yield mean absolute error (MAE), root mean squared error (RMSE), and mean squared error (MSE) scores of 0.60, 1.44, and 1.16, respectively. Conclusion: Our study shows that the vulnerability propagation model with an adaptive mechanism based on BPR can be used to predict security vulnerability scores as the impact of business process redesign.
- Open Access
- 10 February 2023
- Published by Universitas Nusantara PGRI Kediri in INTENSIF: Jurnal Ilmiah Penelitian dan Penerapan Teknologi Sistem Informasi
- Vol. 7  (1) , 134-145
A Systematic Review of Indonesia’s Heritage Tourism in Perspective of Smart Tourism Conceptual Model
Currently, smart tourism has been adopted by cities in many countries. But this concept has various meaning built on different point of view of the smart tourism initiatives. This paper aims to give contribution on conceptual model formulation for smart tourism in process approach, using heritage tourism as case study. A literature review using PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) was carried out to achieve an understanding of the term smart tourism itself, with reference to various literature from qualified journals. From the literature review, various terms related to the definition, objectives, characteristics, inputs and outputs of smart tourism will be obtained. A semantic process with content analysis method to help determine the dominant factors/variables as important objects of Smart Tourism. The model to be built describes the components that influence the smart tourism process. This model also considers some local values that are used in general and particular in tourism policies that have been implemented in Indonesia. These values are believed to have an influence and characteristic of the smart tourism model in Indonesia. In the end, the model also describes the expected results of the smart tourism process.
- 10 February 2023
- Published by Universitas Nusantara PGRI Kediri in INTENSIF: Jurnal Ilmiah Penelitian dan Penerapan Teknologi Sistem Informasi
- Vol. 7  (1) , 134-145
Open AccessA Systematic Review of Indonesia’s Heritage Tourism in Perspective of Smart Tourism Conceptual Model
Currently, smart tourism has been adopted by cities in many countries. But this concept has various meaning built on different point of view of the smart tourism initiatives. This paper aims to give contribution on conceptual model formulation for smart tourism in process approach, using heritage tourism as case study. A literature review using PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) was carried out to achieve an understanding of the term smart tourism itself, with reference to various literature from qualified journals. From the literature review, various terms related to the definition, objectives, characteristics, inputs and outputs of smart tourism will be obtained. A semantic process with content analysis method to help determine the dominant factors/variables as important objects of Smart Tourism. The model to be built describes the components that influence the smart tourism process. This model also considers some local values that are used in general and particular in tourism policies that have been implemented in Indonesia. These values are believed to have an influence and characteristic of the smart tourism model in Indonesia. In the end, the model also describes the expected results of the smart tourism process.
- Open Access
- 22 March 2022
- Published by UNIMAS Publisher in Journal of Computing and Social Informatics
- Vol. 1  (1) , 14-28
Modified PRDG Model for Caregiver Segmentation Using Zarit Burden Interview Instrument
The increasing demand for Indonesian workers in Taiwan has an impact on caregiver problems which can be triggered by the burden of caring for the elderly. Therefore, the aim of this study is to identify the characteristics of caregivers who are resilient to burdens based on Indonesian female caregivers who work in Taiwan data to be a guide for selecting prospective caregivers. The process includes analyzing the personal characteristics that have the most influence on the burden using multiple regression and then clustering caregiver data using K-Means with the Elbow Method and Silhouette Index. Then, segmentation in each cluster based on a comparison of the average values. The results of clustering accuracy on dimensions (PRDG) and modified dimensions (S+PRDG) were compared and the smallest error cluster was in case 4 in the S+PRDG dimension with the Elbow Method of 3.6%. Based on segmentation on that dimension, cluster 2 is a resilient caregiver cluster. Then the results of the multiple regression analysis (Number of Children, Education and Work Location) were studied further for each caregiver in cluster 2 and the conclusions are, their average number of children is 1, final education is in junior high school and their work location is in the capital of Taiwan.
- 22 March 2022
- Published by UNIMAS Publisher in Journal of Computing and Social Informatics
- Vol. 1  (1) , 14-28
Open AccessModified PRDG Model for Caregiver Segmentation Using Zarit Burden Interview Instrument
The increasing demand for Indonesian workers in Taiwan has an impact on caregiver problems which can be triggered by the burden of caring for the elderly. Therefore, the aim of this study is to identify the characteristics of caregivers who are resilient to burdens based on Indonesian female caregivers who work in Taiwan data to be a guide for selecting prospective caregivers. The process includes analyzing the personal characteristics that have the most influence on the burden using multiple regression and then clustering caregiver data using K-Means with the Elbow Method and Silhouette Index. Then, segmentation in each cluster based on a comparison of the average values. The results of clustering accuracy on dimensions (PRDG) and modified dimensions (S+PRDG) were compared and the smallest error cluster was in case 4 in the S+PRDG dimension with the Elbow Method of 3.6%. Based on segmentation on that dimension, cluster 2 is a resilient caregiver cluster. Then the results of the multiple regression analysis (Number of Children, Education and Work Location) were studied further for each caregiver in cluster 2 and the conclusions are, their average number of children is 1, final education is in junior high school and their work location is in the capital of Taiwan.
- 1 November 2021
- Published by Elsevier in Computers & Security
- Vol. 110
Information security vulnerability prediction based on business process model using machine learning approach
- 1 November 2021
- Published by Elsevier in Computers & Security
- Vol. 110
Information security vulnerability prediction based on business process model using machine learning approach
- 31 December 2019
- Published by Medwell Publications in Journal of Engineering and Applied Sciences
- Vol. 14  (4) , 1182-1188
Geovisualization Way for Exploiting Customer’s Emotions on Twitter
Geovisualization Way for Exploiting Customer’s Emotions on Twitter
- 31 December 2019
- Published by Medwell Publications in Journal of Engineering and Applied Sciences
- Vol. 14  (4) , 1182-1188
Geovisualization Way for Exploiting Customer’s Emotions on Twitter
Geovisualization Way for Exploiting Customer’s Emotions on Twitter