2024-03-29T14:33:06Z
https://meral.edu.mm/oai
oai:meral.edu.mm:recid/3986
2021-12-13T00:40:36Z
1582963302567:1597824273898
user-ucsy
Improved Feature-based Summarizing and Mining from Hotel Customer Reviews
Nyaung, Dim En
Thein, Thin Lai Lai
Due to the rapid increase of Internet, webopinion sources dynamically emerge which is useful forboth potential customers and product manufacturersfor prediction and decision purposes. These are theuser generated contents written in natural languagesand are unstructured-free-texts scheme. Therefore,opinion mining techniques become popular toautomatically process customer reviews for extractingproduct features and user opinions expressed overthem. Since customer reviews may contain bothopinionated and factual sentences, a supervisedmachine learning technique applies for subjectivityclassification to improve the mining performance. Inthis paper, we dedicate our work to the main subtask ofopinion summarization. The task of product featureand opinion extraction is critical to opinionsummarization, because its effectiveness significantlyaffects the identification of semantic relationships. Thepolarity and numeric score of all the features aredetermined by Senti-WordNet Lexicon how intense theopinion is for both positive and negative features. Theproblem of opinion summarization refers how to relatethe opinion words with respect to a certain feature.Probabilistic based model of supervised learning willimprove the result that is more flexible and effective.
2015-02-05
http://hdl.handle.net/20.500.12678/0000003986
https://meral.edu.mm/records/3986