|
Title:
|
FINDING SEMANTIC GROUPS OF WEB CLIPS BY CAPITALIZING ON SOCIAL TAGGING |
|
Author(s):
|
lvaro Figueira, Henrique Alves |
|
ISBN:
|
978-972-8939-55-7 |
|
Editors:
|
Piet Kommers, Ji-ping Zhang, Tomayess Issa and Pedro Isaías |
|
Year:
|
2011 |
|
Edition:
|
Single |
|
Keywords:
|
Semantic grouping, Web clips, textual data, tagging and commenting, text mining. |
|
Type:
|
Full Paper |
|
First Page:
|
174 |
|
Last Page:
|
180 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Tagging is a frequent behavior of people consuming online information. Many well-known online systems take advantage of tags in order to improve the organization of the stored content, or to enhance the accuracy of search results from user queries. We created a system to collect and store text clips from online sources. The clips can be tagged or commented by users. Our system performs an automatic organization of this personal clipping collection by grouping clips with similar semantic value. We also provide to each user a means to vary the degree of freedom that the system uses to group clips. In this paper we describe the system and the algorithm that integrates the standard k-means for automatic clustering of texts with a new social classification metadata that comes from tags. As a result we created a system capable of not only providing automatic organization of clips, but also as a mean to identify eventual relations between otherwise unrelated clips. |
|
|
|
|
|
|
