Abstract:
The Web has become the most important information source for most of us. Unfortunately, there is no guarantee for the correctness of information on the Web. Moreover, different websites often provide conflicting information on a subject. Several truth discovery methods have been proposed for various scenarios, and they have been successfully applied in diverse application domains. In this paper, we have attempted to answer the question whether the truth is relevant. We conducted an experimental study in which we analyzed and compared the results of two different truth discovery methods: Relevance-based sources ranking and Majority vote. We have found that the truth is not always held by the most relevant sources on the web. Sometimes the truth is given by the majority vote of the crowd. In addition, we have proposed a method of presenting the results of truth discovery with gradual degrees of belief. A method that allows to configure and target the desired level of trust.
Machine summary:
Exploring Relevance as Truth Criterion on the Web and Classifying Claims in Belief Levels Fairouz Zendaoui* *Corresponding author, Laboratoire de la Communication dans les Systèmes Informatiques, Ecole Nationale Supérieure d’Informatique, BP 68M, 16309, Oued-Smar, Alger, Algérie.
We conducted an experimental study in which we analyzed and compared the results of two different truth discovery methods: Relevance-based sources ranking and Majority vote.
They make different assumptions about aspects of truth discovery like input data, source reliability, identified truths, object, claimed value and output.
Conflicting claimed values regarding the location of Heritage Site “Timgad” {مراجعه شود به فایل جدول الحاقی} On the other hand, there have been many studies on ranking web pages according to relevance based among others on domain authority, page authority and citation flow (Roa- Valverde & Sicilia, 2014).
All truth discovery methods present the inferred values with the same level of confidence, that is, the same degree of belief.
Related Works Several methods have been proposed for truth discovery among conflicting information from the Web. This problem of resolving conflicts from multiple sources has been extensively studied.
Another observation is that all the works dealing with the truth discovery have presented the inferred values with the same level of confidence, while the methods used, each time, provide scores that differ from one value to another.
Lastly, we show how to classify the inferred values in belief levels according to their computed confidence masses, and this for each of the implemented methods.