Journal XixiLuo

From Semantic Portal Wiki

Presentation given at CSCI 6966 Advanced Semantic Web (Fall 2008)#Lesson 10

• Speaker: Xixi Luo
• Title: Towards content trust of web resources
• Authors: Yolanda Gil, Donovan Artz
• Conference: Journal of Web Semantics, 2007
• URL: http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B758F-4PSK8X8-1&_user=10&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_version=1&_urlVersion=0&_userid=10&md5=f32c3fb637b5f878bbdf047affec64e1
• Date of Presentation: 2008/11/06

Questions

ID	Question	Name	Answer
Gil2007towards question 1 by lebo	The results of the simulation compare the Mean Squared Error, k-sum, and Edit Distance to a 'baseline', which is "a ... search engine ... that ranks search results by topic and popularity, ... without taking trust into account." A lot of assumptions are made to model the trust users have for associations and resources, and the simulation used 1,000 queries generated from 1,000 generated resources, 10,000 generated associations, and 1,000 generated users. Although the paper is trying to compare their trust-based ordering against 'traditional' search engines, are they not just comparing their technique to random performance instead?	Tim Lebo
Journal XixiLuo GTW 1	Table 3 is said to show that "in the majority of the cases (72%) the user was able to determine bias," but it seems that what is shown is that the users (who are said to have no expertise in the relevant areas) believe there to be bias. Was this belief verified in any way? Is it possible that users with no domain-expertise might mis-interpret the data and mis-judge whether there is actual bias? Similarly, it's not clear to me that the provenance association data is grounded in fact. Are enough details given to know in what manner these associations are given (is it enough for a website to claim that the data it presents comes from a source, or does an examination of the data need to indicate that the data actually aligns with the claimed source)?	Gregory Todd Williams
Journal XixiLuo Jesse Weaver	At the end of section 8, the authors state: "We do not know if these parameters accurately reflect the Web, .... We intend to incorporate real-world characteristics of the Web in our simulator in future work." What would be some examples of "real-world characteristics of the Web" that are not used in this simulation?	Jesse Weaver

Attendees

Tim Lebo