Research

Home | Teaching | Research | Publications

🗨️ Explainable Information Retrieval

This research aims to make search systems more transparent by generating explanations that show why certain documents are relevant to a user’s query.

1) Generating Search Explanations using Large Language Models
Arif Laksito and Mark Stevenson. Presented at the Workshop on Explainability in Information Retrieval, SIGIR 2025. [poster]. [dataset]

Aspect-oriented explanations in search results are typically concise text snippets placed alongside retrieved documents to serve as explanations that assist users in efficiently locating relevant information. While Large Language Models (LLMs) have demonstrated exceptional performance for a range of problems, their potential to generate explanations for search results has not been explored. This study addresses that gap by leveraging both encoder-decoder and decoder-only LLMs to generate explanations for search results. The explanations generated are consistently more accurate and plausible explanations than those produced by a range of baseline models.

---

🔥 Social media sensing & Fire spread modelling

In this innovative project, we integrate social media sensing with fire spread modelling. We aim to enhance early fire detection and prediction by harnessing real-time data from social platforms. This interdisciplinary approach combines cutting-edge technology with environmental science, fostering a proactive strategy for wildfire management and community safety.

1) Machine Learning and Social Media Harvesting for Wildfire Prevention
AD Laksito, K Kusrini, A Setyanto, MZF Johari, ZR Maruf, KA Yuana, et al. Published in 2023 IEEE 13th International Conference on Pattern Recognition Systems (ICPRS). [pdf]

2) Monte Carlo method for map area calculation in wildland fire map management
Kumara Ari Yuana, Arief Setyanto, Arif Dwi Laksito, Zauvik Rizaldi Maruf, et al. Published in 2022 6th International Conference on Information Technology, Information Systems and Electrical Engineering (ICITISEE). [pdf]