Current Research Projects
Metacognition-driven framework for green programming (Metafor)
This research explores the integration of sustainability principles into programming education at the higher education level through the design of a metacognition-driven framework. It is motivated by the urgent need to incorporate environmental considerations into software engineering practices due to the significant impact of application-based digital technologies on the environment. Current programming learning gaps result in insufficient awareness of code composition’s influence on energy consumption and network usage. Sustainable programming practices, such as minimizing inefficient code and optimizing source code, are crucial for reducing environmental impact. Cultivating these related skills should be encouraged by integrating green thinking through metacognitive processes in programming education to understand the long-term implications of programming decisions on the environment. Through metacognition processes, students can be deeply encouraged to acquire these skills.
By analyzing the current landscape of programming education at the higher education level, this study identifies gaps in sustainability awareness and proposes a Metafor (metacognition-driven framework for green programming) to address them. The research analyzes thousands of lines of code using a static code analyzer server to identify clean code compositions and problematic code produced by students, then applies grounded theory, surveys, and focus group discussions (FGDs) to formulate the framework. The expected outcome of this research is to provide guidance for clean and sustainable programming practices and to promote a deeper understanding of the environmental implications of programming decisions through integrated metacognitive skills.
Part of this research was further carried out in collaboration through a visiting researcher program at the Norwegian University of Science and Technology; further details can be found [Here]
NutrySmart is a mobile application developed to enable the early detection of various malnutrition problems among pregnant women, adolescents, and toddlers. The application operates in two distinct environments: the development environment and the consultation environment.
The development environment is designed to build the intelligent capabilities of the NutrySmart system by incorporating knowledge from nutrition experts, as well as data and factual information obtained from nutritional assessments. These data are collected through the integration of several nutritional measurement methods, including the Estimated Food Record, the Food Frequency Questionnaire (FFQ), and the Recommended Dietary Allowance (AKG – Angka Kecukupan Gizi). Additionally, the system’s intelligence is enhanced using updated daily food intake calculations for users, which are designed in Microsoft Excel. The acquired knowledge is then structured and formatted for use by the inference engine. Expert rules, factual data, and other knowledge structures are developed and stored in the expert system’s knowledge base. The inference engine, a core component responsible for evaluating rules and generating outputs, is developed and customized to meet the specific needs of the expert system.
The consultation environment is developed to allow end-users to access nutritional consultations through the NutrySmart system. These users are typically members of the general public seeking information on early indications of various nutrition-related disorders caused by poor dietary patterns, inadequate nutrient intake, or unsuitable food choices. In this environment, the inference engine analyzes information provided by users and generates conclusions or recommendations based on the knowledge established during the development phase. The expert system offers personalized recommendations or diagnoses to users according to the predefined rules and user-provided inputs.
Details can be accessed by this link https://nutrysmart.id/
Previous Research Projects
Fostering Argumentative Writing and Metacognition Awareness of Undergraduate Students with Support of Annotation, Conversational Agents and Collaborative Concept Map
Argumentative writing is fundamental for undergraduate students’ academic life and scientific writing related to critical thinking and problem-solving. However, previous studies investigated that students have various difficulties in argumentative writing both at the macro level, such as illogical and unclear ideas, less-structured arguments, and unbalance interpretation of issues, data, and evidence; and micro-level related to students’ involvement cognition skills and self-regulatory procedures during the writing process. Therefore, this study aims to foster argumentative writing and metacognition awareness of undergraduate students by integrating computer-supported argumentative writing tools, including annotation, conversational agents (CAs), and collaborative concept maps, into an online learning management system. Since the study was conducted during the COVID-19 pandemic, these tools can support meaningful learning activities and investigation in argumentative writing. An experiment was conducted for eight weeks at an Indonesian university for 60 participants, including 30 for the experimental group and 30 for the control group, and their argumentative writing qualities were also evaluated. The result showed that these three tools effectively foster students’ five elements of argumentative writing, including claims, grounds, warrants, and backings, rebuttal. Furthermore, the deep analysis found that the number of annotations can significantly predict the students’ argumentative writing development, and a collaborative concept map can significantly predict students’ metacognition awareness. Moreover, students’ perceptions of the three proposed tools can effectively help argumentative writing and metacognition awareness. Finally, we proposed some recommendations for future research both in educational and technical aspects.
Some of the findings from this study have been previously published in: https://doi.org/10.1177/07356331241242437
