Accommodation recommendation system from user reviews based on feature-based weighted non-negative matrix factorization method

Author

Lerttripinyo, Thanatcha ; Jatukannyaprateep, Pariyawit ; Prompoon, Nakornthip ; Pattanothai, Chate

Author_Institution

Dept. of Comput. Eng., Chulalongkorn Univ., Bangkok, Thailand

fYear

2015

fDate

22-24 July 2015

Firstpage

108

Lastpage

113

Abstract

Nowadays, online information dominates people decision making, especially in a daily life basis, as well as traveling. Many websites provide accommodation information and reviews. Hence, this information overload makes choosing the right accommodation difficult and time-consuming for users. We propose ACRoSS, a system that collects data from websites and chooses representative reviews by feature-based weighted non-negative matrix factorization method regarding predetermined topics. The benefit of this project is to help users to choose an accommodation easily. The system can classify the topic with the precision of 61%. The percentages of users that have high satisfaction in representative reviews are 55.48%.

Keywords

Internet; data mining; matrix decomposition; recommender systems; ACRoSS; Web sites; accommodation recommendation system; feature-based weighted nonnegative matrix factorization method; opinion mining; representative user reviews; text mining; Data processing; Databases; Feature extraction; Frequency measurement; Text mining; User interfaces; Opinion mining; Recommendation System; Text mining;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Science and Software Engineering (JCSSE), 2015 12th International Joint Conference on

Conference_Location

Songkhla

Type

conf

DOI

10.1109/JCSSE.2015.7219780

Filename

7219780