Framework of sentiment annotation for document specification in Indonesian language base on topic modeling and machine learning

Tata Sutabri, Tata Sutabri and Miftah Ardiansyah, Miftah Ardiansyah Framework of sentiment annotation for document specification in Indonesian language base on topic modeling and machine learning. IEEE.

[img]
Preview
Text
Paper.TataSutabri.CITSM.Bali.2017.pdf

Download (176kB) | Preview

Abstract

Reservation service users and or purchase online based on the marketplace often face difficulties in determining the object or service selected closest to the criteria of potential users. Aside from the rating or rating which features conventional, potential customers can make decisions with the customer review feature that has to wear or purchase items or services. The availability of these features provide a new task for prospective customers to get a thorough analysis, prospective customers are advised to read and analyze each comment related to the amount not less diverse language and style of Indonesian. The difficulty will be growing and time-consuming for prospective users when there are objects or services that are the same in different online services. This study proposes a framework to overcome the difficulties prospective customers. This framework implements a blend of approaches topic models, machine learning to perform sentiment analysis on services and purchase of objects or services based on online. The proposed framework has relevance or context of user reviews. Outcome future of this framework, including the form of the model ranking or rating based every existing review; due to the nature of the framework offered is specific to have a specific domain which minimizes missing review.

Item Type: Article
Subjects: H Social Sciences > H Social Sciences (General)
L Education > L Education (General)
Q Science > Q Science (General)
T Technology > T Technology (General)
Depositing User: Dr. Tata Sutabri
Date Deposited: 30 Mar 2023 00:40
Last Modified: 30 Mar 2023 00:40
URI: http://eprints.binadarma.ac.id/id/eprint/17429

Actions (login required)

View Item View Item