Building a rating model for webblogs
- Pathum Thani, Thailand : Asian Institute of Technology, 2007
- 72 leaves : ill.
- Thesis ; no. ICT-07-15 .
- Asian Institute of Technology. Thesis ; no. ICT-07-15 .
A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Information & Communications technologies, School of Engineering and Technology
Thesis (M.Eng.) - Asian Institute of Technology, 2007
The number of blogs is increasing quickly. It is difficult to find blogs that are worth reading. This thesis describes an overview of some existing rating methods and some blog's characteristics, and then discusses how to design a new rating model for webblogs. The purpose of the new rating model is to compute credibility automatically for data source based on the distribution of mean rating, rater's information, and blog reader's comments given to blogs. The rating data is collected from the blog rating system and from the rated blogs through trackback systems. Statistical method is then used to convert confidence intervals of the distribution of rating means to affirm the credibility score; the AHP technique is also applied to calculate the weight of users' rating. The results of evaluation show that the new rating model could be used to detect spam in rating and reduce the impact of unknown raters in overall credibility score