Concept Discovery and Argument Bundles in the Web of Experiences

Abstract

Millions of people interact and share interesting information every day in the Social Web. From_x000D_ daily conversations to comments about products in e-commerce sites, the content generated by_x000D_ people in these sites is huge and diverse. Among the wide diversity of user-contributed content_x000D_ on the web, there is a particular kind that has the potential of being put to good use by intelligent_x000D_ systems: human experiences. People very often use other people's experiences before making_x000D_ decisions, and when these kind of human experiences are expressed and recorded on the web,_x000D_ they can be shared with by large number of people._x000D_ Nevertheless sometimes this content is not easily accessible, so a person trying to book a hotel_x000D_ may read a few reviews over a few hotels - but cannot possibly read them all. There is a clear_x000D_ need for an in-depth analysis of this kind of information, based on textual expressions of human_x000D_ particular experiences._x000D_ Our approach, in the framework of the Web of Experiences, aims at acquiring practical_x000D_ knowledge from individual experiences with entities in the real world expressed in textual form._x000D_ Moreover, this knowledge has to be represented in a way that facilitates the reuse of the_x000D_ experiential knowledge by other individuals with different preferences. Our approach has three_x000D_ stages: First, we extract the most salient set of aspects used by the individuals to describe their_x000D_ experiences with the entities in a domain. Second, using the set of extracted aspects, we group_x000D_ them in concepts to create a concept vocabulary that models the set of issues addressed in the_x000D_ reviews. Third, using the vocabulary of concepts, we create a bundle of arguments for each_x000D_ entity. An argument bundle characterizes the pros and cons of an entity, aggregating practical_x000D_ knowledge from judgments written by individuals with different biases and preferences._x000D_ Moreover, we show how argument bundles allow us to define the notions of user query and the_x000D_ satisfaction degree of a bundle by a user query, proving that argument bundles are not only_x000D_ capable of representing practical knowledge but they are also useful to perform inference given_x000D_ a set of user preferences specified in a query._x000D_ We evaluate the argument bundles of our approach with the Amazon score ratings and the_x000D_ camera characterizations of Dpreview. We show that pro and con arguments are very close to_x000D_ those listed in Dpreview. Evaluating entity rankings, we show that Dpreview and our approach_x000D_ give congruent rankings, while Amazon's is not congruent neither with Dpreview's or ours.

Date of Award	7 Jun 2017
Original language	Undefined/Unknown
Supervisor	Ricardo Juan Toledo Morales (Tutor) & Enric Plaza Cervera (Director)