Matrix : Jurnal Manajemen Teknologi dan Informatika

Predicting financial default risks: A machine learning approach using smartphone data

2024-11-11T02:25:45+00:00

This study leverages machine learning (ML) techniques to predict financial default risks using smartphone data, providing a novel approach to financial risk assessment. Data were collected from 1,000 individuals who had taken personal loans, focusing on key behavioral parameters such as app usage frequency, GPS location data, and communication patterns over a six-month period prior to loan application. The analysis employed Logistic Regression, Decision Trees, and Random Forest models to determine correlations between these parameters and default risks. The Random Forest model demonstrated superior performance, achieving 85% accuracy. Key findings show that high usage of financial apps was associated with lower default risks, while irregular communication patterns and erratic mobility were significant indicators of higher risk. These results suggest that smartphone-derived behavioral data can significantly enhance traditional credit scoring methods. The study not only contributes to predictive analytics in financial risk management but also raises ethical considerations around privacy and data security.

Design and implementation of IoT-based motorcycle keyless ignition and starter using RFID and Blynk

2024-11-06T04:56:13+00:00

Motorcycles are one of the most widely used means of transportation in the community. However, this widespread usage has led to an increase in thefts, especially when motorcycles are not equipped with adequate security systems. Statistical data shows that motorcycles have the highest theft rate compared to other vehicles. To address this issue, an enhanced security system is necessary. This study proposes the use of Radio Frequency Identification (RFID) sensor and the Blynk application to provide additional security for motorcycles. The RFID system will only be accessible using the motorcycle owner's e-KTP (electronic Indonesian Identity Card) or other registered cards to start the vehicle. Additionally, the Blynk mobile application allows for vehicle control (ON and OFF) via Wi-Fi and provides real-time monitoring of the vehicle's status. This application enables motorcycle activation remotely via Wi-Fi, with a range of approximately 10 meters. Meanwhile, there is a 4-second delay to start the motor starter using e-KTP. Test results indicate that this IoT-based keyless ignition and starter system is effective in enhancing motorcycle security. With this system, motorcycles have a dual-layer security mechanism to minimize theft attempts.

Food Journal: An application for allergy early detection

2024-11-29T13:19:30+00:00

Allergies are now increasingly common, due to public awareness and lifestyle changes. Although antibody tests and skin pricks can detect allergies, their accuracy is not always 100%, causing some cases of false negatives. Observing the reactions caused by food is an alternative approach, but people often forget the food they consume, especially if the allergic reaction is indirect. Therefore, tools are needed to record foods and reactions, allowing traceability to undetected allergy triggers. This study is applied research that aims to develop a prototype application for recording daily food and tracking the history of eating activities based on the allergy symptoms entered. The software development method used is the Agile method with the Scrum framework. The Scrum framework is used considering that the development of this application requires speed in its provision, and is susceptible to change, so flexibility in development is an absolute must. The software testing was conducted by User Acceptance Testing (UAT) to ensure it meets the user’s needs and requirements. The UAT results are ease of use 3.8, functionality 3.99, user interface (UI) design 3.8, utility 4, and support and help 4. UAT results indicate that the ease of use and UI design have the worst scores and need to be improved, while the utility and support have the best results.

Ontology-based data framework for the digital preservation cultural heritage: A case of Subak

2024-12-02T04:10:36+00:00

Cultural heritage is increasingly fading due to the influence of globalization and social changes that shift traditional values. Digitalization is critical to addressing the degradation of cultural heritage knowledge by enabling more accessible storage and documentation and facilitating preservation and education efforts for future generations. Subak, as a unique traditional agricultural system in Bali, Indonesia, has a wealth of valuable knowledge and culture, but it is still scattered and unstructured. An ontology-based Knowledge Management System is designed to capture, organize, and utilize knowledge within a domain using ontologies as a fundamental component. Ontologies explicitly represent knowledge by providing meaning, properties, and relationships with related concepts. This paper proposes data modelling with the Ontology Development framework for the knowledge base. The methodological framework consists of knowledge externalization, including identifying and collecting knowledge based on data modelling through interviews and observations. The second steps are modelling and development of ontology, based on the knowledge that has been previously disclosed, including the hierarchical structure, properties, and relationships that form the knowledge base. The third stage is the Implementation and evaluation, which involves evaluating the quality of ontology using onto metrics and implementing the ontology into the KMS. The result of this study is data modeling for the knowledge base and the Ontology Development framework. The data modelling is a form of data mapping based on a combination of traditional philosophy and modern management principles that underlie Subak activities and artefacts, namely Tri Hita Karana, Desa Kala Patra, and POAC. The proposed framework can be a reference in developing new ontologies from scratch on unstructured cultural heritage data.

Designing an e-report system as a digital portfolio in early childhood education in Surabaya using the waterfall method

2024-11-29T12:21:29+00:00

This study focuses on the design and implementation of an E-Report system as a digital portfolio for Early Childhood Education in Genteng District, Surabaya. The current manual process using Excel and Canva for generating learning reports limits efficiency and accuracy in tracking children’s developmental progress. To address this, a web-based E-Report system was developed using the Waterfall method to streamline reporting processes, allowing educators to manage children’s data digitally while providing parents accessible, real-time progress reports. Development stages included requirement analysis, system design, and implementation, leveraging modern web technologies for optimal performance. User satisfaction assessments reveal high satisfaction levels, particularly in the system's format (average score of 3.80) and content (average score of 3.65), indicating a strong alignment with user needs. The implementation of the E-Report system is expected to elevate Early Childhood Education services in Genteng District by delivering a more efficient, accurate, and collaborative reporting tool.