Honours and Minor Thesis projects

Gen AI has taken the world by storm, it's been applied to many disciplines including in pure sciences.  Notably, Google Deepmind used graph based deep learning to discover millions of new materials.

Mental health challenges disproportionately affect vulnerable populations, often due to limited access to traditional healthcare services. The rise of Generative AI offers a groundbreaking opportunity to bridge this gap by providing personalized, scalable, and accessible mental health support. This project, led out of Action Lab, aims to harness the potential of Generative AI to develop innovative technologies tailored for mental health interventions.

Islam has become the current fastest growing religion in the world. It indicates that a lot of people starting to get interested to study Islam. The best way to study a religion is by reading its holy scripture, where in Islam is the Quran. People who embrace Islam as their religion are called Muslims, and according to them it requires to sincerely open your mind and heart to study Quran, and it will take a lifetime to understand and analyze the deep meaning of Quran. Therefore, most Muslims study Quran with guidance from Muslim scholars to get better understanding of what they read.

Despite the needs of stakeholders, which dental institutions as primary focus in this paper, the feasibility of integrating artificial intelligence (AI), specifically Language Models (LLMs), whether in a broad context or within specific environments such as dental education, is an active area of research. Studies related to AI applications, both in education or dental health institutions, suggest that ChatGPT's performance needs to be considered from various perspectives.

The integration of chatbots into customer service operations has undergone a transformative shift across diverse industries, becoming essential instruments for delivering assistance and personalized interactions. This development, spurred by the requirement for efficient, scalable, and 24/7 available customer service solutions, has become notable in various sectors, from retail to healthcare.

Word sense disambiguation (WSD), the process of computationally identifying the appropriate meaning of a word within its context, is a fundamental task in Natural Language Processing (NLP). Effective WSD is crucial for building accurate machine translation systems, information retrieval tools, and sentiment analysis applications, especially when dealing with diverse languages and linguistic variations.

The Javanese language, spoken by a population of over 98 million people, faces notable challenges in digital and technological applications, especially when compared to globally recognized languages. This disparity is highlighted in several studies that discuss the lack of deep learning research benefits due to data scarcity for Javanese. Additionally, other studies have pointed out the inaccessibility of data resources and benchmarks for Javanese, contrasting with languages like English and Mandarin Chinese.

Political polarization is a phenomenon that permeates societies worldwide, manifesting in divergent ideologies, entrenched viewpoints, and societal fragmentation. In the context of Indonesia, a diverse and populous nation with a complex socio-political landscape, understanding political polarization is crucial for fostering social cohesion and effective governance.

In the evolving landscape of data reporting, traditional text-based and quantitative methods are increasingly being supplemented by rich, community-generated qualitative data, including audio and video content. This shift presents unique challenges and opportunities in how non-profits, government bodies, and community organizations present and utilize this data.

Recent advancements in wearable technology have enabled continuous health monitoring, significantly expanding the capabilities of devices like the Apple Watch, Fitbit, and Samsung Watch in medical diagnostics. Among these, Electrocardiography (ECG) interpretation is a critical function, traditionally requiring expert analysis. This project proposes the use of deep learning (DL) algorithms to automate ECG interpretation on these devices, enhancing diagnostic accuracy and accessibility.