|
Title:
|
BUILDING USERS PROFILES DYNAMICALLY THROUGH ANALYSIS OF MESSAGES IN COLLABORATIVE SYSTEMS |
|
Author(s):
|
Gustavo Fernandes, Sean Siqueira |
|
ISBN:
|
978-989-8533-16-6 |
|
Editors:
|
Bebo White and Pedro Isaías |
|
Year:
|
2013 |
|
Edition:
|
Single |
|
Keywords:
|
User profile, User modeling, Collaborative systems, Social networks, Named entities, Natural language processing |
|
Type:
|
Full Paper |
|
First Page:
|
75 |
|
Last Page:
|
82 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Over the years the number of people accessing the web has considerably grown. Most of the hits come from collaborative systems or social networks like Facebook. In order to help these systems to evolve and provide better functionalities, its crucial to know users deeply. Its possible to extract users information when he signs up on a system, but this information is rarely updated. This paper presents an approach to better understand the users interests and preferences by dynamically mapping their profiles from text messages that are inserted spontaneously in a collaborative system. These messages are automatically processed to build three distinct types of user profiles based on the elements that are presented in these messages: the message text, links and hashtags. After conducting qualitative research through interviews with some users who have their profiles mapped on Twitter social network, results show that the approach was well accepted among these users and they believe that the acquired profile represent them correctly. |
|
|
|
|
|
|
