Indonesian Journal of Information Systems

Artificial Intelligence Chatbots in Education: Academics Beliefs, Concerns and Pathways for Integration

Belinda Ndlovu — 2025-02-28

Although globally there are mixed perceptions regarding the academic integrity of chatbots, existing research has mainly focused on developed nations, neglecting the unique perspectives of academics in developing countries, with different contextual, environmental, and technological settings. This study presents lecturers’ perceptions of using Artificial Intelligence (AI) chatbots in education. Guided by the Unified Theory of Acceptance and Use of Technology 2 (UTAUT2), this research collected quantitative and qualitative data from 140 lecturers and three administrators from a STEM-based Zimbabwean university. The research confirmed that performance expectancy (belief in improved efficiency and personalised learning) and perceived value and social influence drive adoption. Contrary to previous studies, there was no significant link between effort expectancy (reduced workload) and chatbot use. Demographics like gender, age, and qualifications did not impact chatbot use. Academics were cautiously optimistic, recognising benefits like personalised learning and routine task management but concerned about ease of use, technical expertise, and ethical considerations. To effectively integrate AI chatbots into higher education processes, there is a need for funding, technical support, training, strengthening IT infrastructure and establishing frameworks for responsible use. Emphasising efficiency, personalisation, and robust support will help overcome barriers and maximise AI chatbots’ potential in education.

An exploration of students' cyber threats perception in the digital age

Donald Mothisi — 2025-02-28

This study aims to investigate cyber threat awareness among students from a rural-based university and propose a model to enhance their awareness. Students rely on information and communication technologies (ICTs) for educational and personal activities. Students in rural areas may have less cybersecurity education and awareness than their urban counterparts. This can affect their awareness of malware, social engineering, and other cyber threats. It also heightens the challenges students face in mitigating security breaches. Data was collected using a survey to assess students' awareness of cyber threats. This assisted in determining students' knowledge, attitude, and behaviour (KAB) when engaged in online activities. The results indicated that less than 20 per cent of the students are aware of threats like Trojan horses, phishing, and keyloggers. The limited awareness of these threats could negatively impact students' ability to protect their information resources. It is recommended that rural-based students are continuously made aware of cyber threats. This study proposes the student online threat awareness model (SOTAM) to enhance cyber threat awareness among students.

Chargeback as an ICT Cost Reduction Strategy

George Ajufo — 2025-02-28

A leading financial institution in Nigeria, hereinafter referred to as "the Bank," has deployed Information and Communications Technology (ICT) systems to drive the Bank's strategy and operations with significant success, albeit with massive investments. However, despite the benefits derived, there have been concerns, especially with the dwindling revenues of the Bank, that the ever-increasing cost of ICT could become unsustainable. These concerns have led the Bank's management to request the ICT Department to find ways of reducing costs. This study investigated the adoption of ICT chargeback to reduce ICT costs in the Bank without impacting ICT service quality. The study utilized variables identified by prior researchers on ICT chargeback. Data was gathered from the Bank's staff using online surveys. The findings from the analysis of data provided sufficient evidence to support the assertion that ICT chargeback adoption would lead to ICT cost reduction in the Bank, consistent with the results of previous studies. The study also indicated that chargeback adoption would facilitate decision-making and more responsible usage of ICT infrastructure in the Bank. However, the study also found some negative consequences which would result from its adoption. For instance, the study showed that ICT Chargeback would discourage innovation due to cost consciousness and foster an unhealthy relationship between ICT and the business. In conclusion, the study recommended the adoption of ICT chargeback with the caveat that the negative consequences identified should be minimized to ensure that they do not vitiate the gains from the adoption.

Enhancing Eye Health Diagnosis through Deep Transfer Learning: Unveiling Insights from Low Quality Fundus Images

S. Pariselvam — 2025-02-28

Due to the frequency of eye illnesses, effective and precise diagnostic instruments are required. This work suggests an approach that uses low quality fundus images with deep transfer learning more precisely, the EfficientNetB0 architecture to improve eye health diagnosis. We tackle the problem caused by the quality of fundus photographs that are commonly found in clinical settings, which frequently display noise and abnormalities. Our methodology consists of pretraining the EfficientNetB0 model on a sizable dataset of excellent fundus photos, followed by fine-tuning it on a dataset of poor fundus photos. By employing this transfer learning technique, the model enhances its diagnostic capabilities by learning to identify significant features from the low-quality images. We ran tests on a variety of datasets that included fundus photos of varying degrees of deterioration in order to assess our approach. As compared to conventional techniques, the results reveal a significant improvement in diagnostic accuracy, demonstrating the effectiveness of deep transfer learning for improving eye health diagnosis from difficult fundus images. With fused features from MobileNet and DenseNet-121 models, the ANN specifically achieved accuracies of 98.5% for cataracts, 99.1% for diabetic retinopathy, 99% for glaucoma, and 99.5% for normal conditions.

Implementation of ERP and SCM System to Improve Productivity and Competitive Advantage of Veneer Sales at PT Karya Megah Indowood

Bayu Setyo Nugroho — 2025-02-28

PT Karya Megah Indowood is a company engaged in veneer manufacturing. The company does not utilize renewable technology to create a system that can assist the company in managing its operations. By not using adequate technology, the company takes more time for its administrative activities, thus affecting the company's productivity. The lack of technology and integrated management systems also leads to a lack of efficiency in inventory management, distribution, and internal communication, hindering the company's growth. It is expected that an integrated technology system can assist the company in increasing its productivity. The problems faced by the company can be improved through the implementation of technology in the Supply Chain Management process and Enterprise Resource Planning using the odoo system. The use of Purchase, Inventory, Sales, and CRM modules is an application in odoo that is applied to purchase goods, check inventory, and strengthen the company's relationship with customers at PT Karya Megah Indowood so that the company's productivity will increase.

Improved Banking Customer Retention Prediction Based on Advanced Machine Learning Models

Linda Wahyu Widianti — 2025-02-28

The quick growth of the banking sector is reflected in the rise in the number of banks. In addition to the intense competition among banks for new customers, efforts to keep existing ones are essential to minimizing potential losses for the company. To ascertain whether customers will leave the bank or remain customers, this study will employ churn forecasts. A 1,750,036-customer demographic dataset, which includes data on bank customers who have left or are still customers, is used in the training process to compare five machine learning technology models in order to investigate the improvement of binary classification prediction accuracy. These models are Decision Tree, Random Forest, Gradient Boost, Cat Boost, and Light Gradient Boosting Machine (LGBM). According to the study's results, LGBM performs better than the other four models since it has the highest recall and accuracy and the fewest False Negatives. The LGBM model's corresponding accuracy, precision, recall, f1 score, and AUC are 0.8789, 0.8978, 0.8553, 0.8758, and 0.9694. This demonstrates that, in comparison to traditional methods, machine learning optimization can produce notable advantages in churn risk classification. This study offers compelling proof that sophisticated machine learning modeling can revolutionize banking industry client retention management.

Application Of Expert System In Determining Diseases In Potato Plants

Ali Ikhwan — 2025-02-28

This research aims to develop an expert system in diagnosing diseases in potato plants using the Case Based Reasoning (CBR) method approach combined with the K-Nearest Neighbor (K-NN) algorithm. The system is designed to help farmers identify the type of disease based on the symptoms that appear, as well as provide relevant solutions to increase crop productivity. In previous research, the CBR method showed a limited accuracy rate of 74% because it only relied on one algorithm. Through the application of two methods in data analysis, namely CBR and K-NN, this study succeeded in increasing the diagnosis accuracy to be higher than the previous approach of 80%. The system is implemented in the form of a web-based application that is easily accessible by farmers. The results show that the integration of these two methods provides more optimal, effective, and accurate results in detecting potato plant diseases based on symptom data. The findings are expected to contribute significantly to the development of agricultural technology, especially in improving the harvest success of potato farmers in Indonesia.

Importance IT Role of IT Self-Efficacy towards Actual Competency, IT Usage, and Productivity: A Case Study on University Students

Tony Dwi Susanto — 2025-02-28

Teaching how to use technology and encouraging students to improve their learning productivity by using technology are challenging. This study aims to investigate the important role of IT self-efficacy towards IT actual competency, actual usage of IT, and productivity in the case of university students to propose a teaching strategy that can improve students’ competency from “know-what” to “know-how” continued to real usage and productivity. It produces a model of the relationships between IT self-efficacy, IT actual competency, actual usage of information technology, and productivity. The relationships were quantitively measured using data from 89 students. The construct validity of the measurement model was examined using convergent and discriminant validity analysis. The hypothesized model was tested using structural equation modeling and bootstrap analysis. The findings suggest that IT self-efficacy does not directly impact actual usage and productivity. Still, it directly affects IT actual competency, leading to actual usage and productivity. In terms of novelty, this study examines a comprehensive relationship between 4 variables: IT self-efficacy, IT actual competency, IT usage, and Productivity, while the other existing studies just examined the relationships between 2 variables: IT self-efficacy and IT actual competency, IT self-efficacy and attitude towards IT usage, IT self-efficacy and IT Usage. For practical contributions, this study highlights the importance of improving students’ IT self-efficacy as an initial strategy to motivate students to use technology in their learning process which can lead to improved productivity.

Comparison of Automatic and Manual Regression Testing on Mobile Application Health Technology with Black Box Testing Method

Yoanna Fransisca Putri — 2025-02-28

The opening of opportunities for health tech services in Indonesia can lead to increased competition. This makes the company continue to innovate by updating or adding functionality to software systems that have been developed. Because of this, software testing with a shorter time is required so that test performance remains maximum. The object of this research is the Teman diabetes application developed by PT. Global Urban Essentials. In this research, the software testing conducted includes automatic regression testing and manual regression testing. The two testing approaches will be compared to evaluate their effectiveness in detecting bugs and their efficiency in terms of time usage. The method used in automatic regression testing is the black box testing method, implemented within the Software Testing Life Cycle (STLC) that is include requirement analysis, test planning, test case development, test environment setup, and test execution. The primary focus of this research is to asses the effectiveness of automatic and manual regression testing in identifying bugs/errors and measuring efficiency in time utilization, based on the analysis of test reports generated from both methods. In this study, it was found that testing carried out by automation is 9.77% faster. In addition, a bug was found in the shopping feature as a result of the new features being released. This bug was discovered after running automated regression testing and was not found in manual regression testing because the bug testing was missed.