Siaw, Nyuk Hiong and Ranaivo-Malançon, Bali and Narayanan, Kulathuramaiyer and Jane, Labadin (2014) Linguistically Enhanced Collocate Words Model. In: Information Retrieval Technology. Lecture Notes in Computer Science (8870). Springer, pp. 230-243. ISBN 978-3-319-12843-6
|
PDF
Linguistically Enhanced Collocate Words Model (abstract).pdf Download (788kB) | Preview |
Abstract
Bag-of-word (BOW) or fixed size window approach for word extraction in natural language text has ignored text structure and context information. Similarly, word co-occurrence based on linear word proximity has also ignored the linguistic criteria of words. This paper aims to propose a semantic window of word to address the needs to provide a context for capturing the structure and context of word in a sentence for analysis. The semantic window of word has linguistic elements which can be injected for collocate word identification. Selected data has been used as case studies. Quantitative analysis has been conducted as well. The proposed approach is evaluated and compared to sliding window which is the baseline. Semantic window is found to perform better than sliding window for linguistically enhanced collocate word extraction.
| Item Type: | Book Chapter |
|---|---|
| Uncontrolled Keywords: | Semantic dependency parsing, linguistic, collocation, semantic window, unimas, university, universiti, Borneo, Malaysia, Sarawak, Kuching, Samarahan, ipta, education, research, Universiti Malaysia Sarawak |
| Subjects: | T Technology > T Technology (General) |
| Divisions: | Academic Faculties, Institutes and Centres > Faculty of Computer Science and Information Technology Faculties, Institutes, Centres > Faculty of Computer Science and Information Technology Academic Faculties, Institutes and Centres > Faculty of Computer Science and Information Technology |
| Depositing User: | Karen Kornalius |
| Date Deposited: | 23 May 2017 07:03 |
| Last Modified: | 23 May 2017 07:03 |
| URI: | http://ir.unimas.my/id/eprint/16386 |
Actions (For repository members only: login required)
 |
View Item |
