A method to identify potential ambiguous Malay words through Ambiguity Attributes mapping: An exploratory Study

We describe here a methodology to identify a list of ambiguous Malay words that are commonly being used in Malay documentations such as Requirement Specification. We compiled several relevant and appropriate requirement quality attributes and sentence rules from previous literatures and adopt it to come out with a set of ambiguity attributes that most suit Malay words. The extracted Malay ambiguous words (potential) are then being mapped onto the constructed ambiguity attributes to confirm their vagueness. The list is then verified by Malay linguist experts. This paper aims to identify a list of potential ambiguous words in Malay as an attempt to assist writers to avoid using the vague words while documenting Malay Requirement Specification as well as to any other related Malay documentation. The result of this study is a list of 120 potential ambiguous Malay words that could act as guidelines in writing Malay sentences

💡 Research Summary

The paper presents a systematic approach for identifying potentially ambiguous Malay words that frequently appear in technical documentation such as software requirement specifications. Recognizing that linguistic ambiguity can lead to misinterpretation, design errors, and project delays, the authors first review existing quality attributes for requirements (clarity, consistency, verifiability, etc.) and sentence‑level rules from prior studies. They then adapt these concepts to the Malay language, constructing a set of “ambiguity attributes” that reflect Malay‑specific phenomena such as lexical polysemy, scope uncertainty, temporal/conditional vagueness, and subject‑object ambiguity.

To generate candidate words, the researchers compiled a Malay corpus drawn from publicly available requirement documents, specifications, and related texts. Using a Malay morphological analyzer and frequency analysis, they extracted high‑frequency terms that could be semantically vague. Each candidate was automatically mapped to the predefined ambiguity attributes via a rule‑based script, after which a manual review refined the mappings.

The resulting list of 250 initial candidates was subjected to expert validation by three Malay linguists. Through consensus discussion, the experts confirmed 120 words as genuinely ambiguous in the context of requirement writing. For each word, the paper provides the associated ambiguity attribute(s) and illustrative usage examples (e.g., “banyak” – scope uncertainty, “segera” – temporal vagueness).

The authors argue that this curated list can serve as a practical checklist for authors and reviewers, helping them avoid vague terminology and thereby improving the precision of Malay requirement specifications. They also discuss the broader applicability of their attribute‑mapping framework, suggesting that with appropriate linguistic adjustments it could be extended to other languages.

Limitations are acknowledged: the corpus size is modest, potentially omitting domain‑specific ambiguous terms; the validation relies on a small panel of experts, introducing subjectivity; and the current process still requires manual intervention. Future work is proposed in three directions: expanding the corpus to cover more domains, developing fully automated ambiguity detection algorithms that leverage the attribute schema, and conducting cross‑lingual studies to test the generalizability of the method.

In conclusion, the study delivers a concrete, linguistically grounded resource—a list of 120 potentially ambiguous Malay words—intended to guide writers of requirement specifications and related documentation toward clearer, less error‑prone communication.