Semantic Relatedness Measure for Identifying Relevant Answers in Online Community Question Answering Services

• Main Authors: Lee, Jun Choi, Cheah, Yu-N
• Format: Conference or Workshop Item
• Language: English
• Published: 2015
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://ir.unimas.my/id/eprint/12080/7/Semantic%20relatedness%20measure%20for%20identifying%20relevant%20answers%20%28abstract%29.pdf; http://ir.unimas.my/id/eprint/12080/
• Institution: Universiti Malaysia Sarawak

This study introduces a new sentence-to-sentence semantic relatedness measure. The proposed measure optimized the word-to-word semantic relatedness that based on the depth of two concepts in WordNet. The study used Microsoft Research Paraphrases Corpus to validate the accuracy of the proposed method in identifying sentences with high semantic similarity. The result shows the proposed methods performed well compare to other unsupervised methods. At the end of the study, this paper also shows that the proposed semantic relatedness is able to identify relevant answers in Online Community Question Answering Services.