Malay lexical simplification model for non-native speaker

• Published in: 2022 International Conference on Intelligent Systems and Computer Vision (ISCV) pp. 1 - 6
• Main Authors: Omar, Salehah, Bakar, Juhaida Abu, Nadzir, Maslinda Mohd, Harun, Nor Hazlyna, Yusoff, Nooraini
• Format: Conference Proceeding
• Language: English
• Published: IEEE 18.05.2022
• Subjects: Analytical models; complex word identification model; Computational modeling; Computer vision; Machine learning; Morphology; NLP component; text analysis; text simplification; Vocabulary
• ISSN: 2768-0754
• DOI: 10.1109/ISCV54655.2022.9806133

Vocabulary is an important language skill that can affect a person's understanding of a sentence. Thus, lexical simplification is the task of converting difficult words into simpler words. It is to make it easier for the reader to understand the sentences. The biggest challenge in lexical simplification is to simplify the words needed without changing the meaning of the sentence. Past studies have shown that there are weaknesses in this task, where simple words are also identified as complex words. This issue has led to the simplification of unnecessary words. The purpose of the study is to produce a complex word identification model for the Malay language into words that are more easily understood by nonnative speakers. Experiments was performed on the appropriate features to obtain the required results. Machine learning was used to ensure the results were more accurate. This study is a novelty in text simplification of the Malay language in the field of Natural Language Processing (NLP) and may be used as a preprocessing tool to improve other tasks in NLP.