Authors:
Afef Walha
1
;
Faiza Ghozzi
2
and
Faiez Gargouri
2
Affiliations:
1
Multimedia and InfoRmation Systems and Advanced Computing Laboratory, Tunisia
;
2
Multimedia, InfoRmation Systems and Advanced Computing Laboratory and Sfax University, Tunisia
Keyword(s):
ETL, Sentiment Analysis, Social Network.
Related
Ontology
Subjects/Areas/Topics:
Databases and Datawarehouses
;
e-Business and e-Commerce
;
Internet Technology
;
Metadata and Metamodeling
;
Social Media Analytics
;
Social Networks and Organizational Culture
;
Society, e-Business and e-Government
;
System Integration
;
User Modeling
;
Web Information Systems and Technologies
;
Web Interfaces and Applications
;
Web Programming
Abstract:
Considered as a rich source of information, social networking sites have been created lot of buzz because people share and discuss their opinions freely. Sentiment analysis is used for knowing voice or response of crowd for products, services, organizations, individuals, events, etc. Due to their importance, people opinions are analyzed in several domains including information retrieval, semantic web, text mining. These researches define new classification techniques to assign positive or negative opinion. Decisional systems like WeBhouse, known by their data-consuming must be enriched by this kind of pertinent opinions to give better help to decision makers. Nevertheless, cleaning and transformation processes recognized by several approaches as a key of WeBhouse development, don’t deal with sentiment analysis. To fulfill this gap, we propose a new analysis algorithm which determines user’s sentiment score of a post shared on the social network Facebook. This algorithm analyzes user’
s opinion depending on opinion terms and emoticons included in his comments. This algorithm is integrated in transformation process of ETL approach.
(More)