Skip to content

자연어처리와 기계학습을 이용한 리뷰 진위 판별

Notifications You must be signed in to change notification settings

constmoon/Restaurant-review

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

27 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Restaurant-review

산업화 사회를 넘어 빅데이터 사회로 오면서 소비자들은 수많은 리뷰를 생산하고 공유하고 있다. 하지만 이 과정에서 일반적인 소비자들이 작성하는 리뷰와 공급자의 협찬을 받아 기업의 의도가 들어간 리뷰가 뒤섞이면서 리뷰의 신뢰성이 떨어지게 되었다. 이에 따라 소비자들은 ‘거짓 리뷰’를 가려내기 위해서 비속어 등 특정 키워드들을 함께 검색하는 방법 등을 고안해 사용해왔다. 본 프로젝트에서는 보다 효율적이고 정확하게 판단하기 위하여 거짓, 진짜 리뷰 간의 통계적으로 유의미한 언어적 차이가 있는지 검증한다. 또한 다양한 기계학습을 이용해 리뷰의 진위를 판별할 수 있는지 확인하고, 학습 방법에 따른 성능 차이를 비교함으로써 어떤 방식이 가장 효과적인지 알아보고자 한다.

This project aims to make sure that there is a difference between genuine reviews and fake reviews through natural language processing techniques and to create a program that can identify them. The types of reviews to be analyzed are limited to restaurant reviews.

About

자연어처리와 기계학습을 이용한 리뷰 진위 판별

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages