Detecting Semantically Similar Constructs using Google Trigram Sentence Similarity Measures
| dc.contributor.author | Linda Wong Lin Juan | |
| dc.date.accessioned | 2026-10-16T07:41:45Z | |
| dc.date.issued | 2018 | |
| dc.description | In order to assess unobserved behaviours or traits, psychometric measurements are being used. Generating a valid psychometric measurement for a study requires massive effort. This can be caused by bias in judgement. If the same constructs are re-used with different name. Therefore, this study address and attempt to resolve this issue with computational approaches to automatically detect if two constructs are semantically similar. In this study, constructs properties are being concatenated and being treated as sentence so that sentence similarity measures can be employed to detect the semantic similarity of the constructs. Different sentence similarity measures are examined to address this issue. Google Trigram Measure (GTM), as the best of the sentence similarity measures are investigated and later revealed its pitfall of which in turn lead to proposal of a hybrid GTM measure to improve the efficacy of detecting semantically similar constructs. Through experiments, GtmMi has been selected as the best sentence similarity among the proposed measures. With the use of real use case, Unified Theory of Acceptance and Use of Technology (UTAUT), the proposed solution is applied to locate similar constructs approximately to each other. The result from the use case has proved that the proposed solution is able to determine similarity between constructs. | |
| dc.identifier.uri | https://scholarhub.unimas.my/handle/123456789/3129 | |
| dc.language.iso | English | |
| dc.publisher | UNIVERSITI MALAYSIA SARAWAK | |
| dc.relation.ispartofseries | Faculty of Computer Science and Information Technology | |
| dc.subject | Construct similarity, Hybrid google trigram measure, Semantic similarity | |
| dc.title | Detecting Semantically Similar Constructs using Google Trigram Sentence Similarity Measures | |
| dc.type | Masters |