Embedding Information Retrieval and Nearest-Neighbour Algorithm into Automated Essay Grading System
ICITA '05 Proceedings of the Third International Conference on Information Technology and Applications (ICITA'05) Volume 2 - Volume 02
Sentence Similarity Based on Semantic Nets and Corpus Statistics
IEEE Transactions on Knowledge and Data Engineering
The distributional similarity of sub-parses
EMSEE '05 Proceedings of the ACL Workshop on Empirical Modeling of Semantic Equivalence and Entailment
Recognizing textual entailment using sentence similarity based on dependency tree skeletons
RTE '07 Proceedings of the ACL-PASCAL Workshop on Textual Entailment and Paraphrasing
Hi-index | 0.00 |
This paper discusses an issue of identifying whether Malay sentences from two different resources are similar when they are compared. The problem arises when the sentences could not be identified by matching them directly as they are constructed using different words, phrases and structure. To overcome the problem, pola grammar algorithm is introduced to extract the grammatical relations and then used them as the elements to identify the sentences similarity. To evaluate the works, the sentences that are constructed by the students in a short answers essay typed examination are used as the sample. The sentences are compared to the sentences that are prepared for the marking scheme. The result has proven that the similarity can be identified using the comparison of the grammatical relations of the sentence.