Séminaire de Seanghort BORN, doctorant au LIUM-IEIAH

Date : 7/04/2025
Heure : 11h00
Lieu : IC2, Salle des conseils
Intervenant : Seanghort BORN
 
 

Towards a Smarter Homophone Correction Tool: A Case Study in Khmer Writing

Homophone errors are a common challenge in written communication, affecting both high-resource languages like English and low-resourced languages such as Khmer. These errors are often difficult to detect, as they require contextual understanding rather than simple spelling corrections. While existing spelling correction tools enhance text accuracy, they do little to improve users’ long-term writing skills, often leading to over-reliance on automated corrections.

This study aims to bridge the gap by investigating the challenges of homophone usages, specifically among the Khmer users, and proposing an innovative solution. Through a questionnaire-based survey, we analyze the prevalence of homophone errors and their impact on Khmer speakers. Additionally, we conduct an experimental study using Typing Tracker, where participants transcribe audio-recorded articles to determine their ability to correctly use homophones in context.

From these insights, we present Sor-Ser which is a nouvel model that integrate Natural Language Processing (NLP) with Learning Analytics (LA) techniques. This model model is designed to correct homophone errors while enhancing users’ writing proficiency. By addressing both error correction and skill development, Sor-Ser aims to provide a comprehensive solution that improves Khmer writing accuracy while fostering long-term proficiency and confidence.