DocumentCode :
3746169
Title :
Keynote speech III controlled imprecision: Fuzzy technology for exploration processes
Author :
Marek Reformat
Author_Institution :
Department of Electrical and Computer Engineering, University of Alberta, Edmonton, Canada
fYear :
2015
Firstpage :
27
Lastpage :
27
Abstract :
Everyday, the users search the web for things of their interest. On multiple occasions they expect precise, as much as possible, results. However, human´s curiosity and a need for being exposed to different things, things that provide some levels of novelty is an important part of the exploration processes. Existing systems supporting the users in their search activities provide them with some variations, but it is not a controlled process. Diversity is accidental. The users do not have any influence on a degree or extensiveness of this diversity. In this presentation we state that application of fuzziness in systems supporting users in their search will allow them to guide and control mechanisms that identify alternatives, and influence recommendations. The fuzzy-based methodologies are capable of dealing with imprecise and vague statements, and can be applied to scenarios where the users want to relax their requirements. This means the users are able to browse and explore variety of things that do not match the search criteria to the highest possible degree. The two applications of fuzzy methods that ensure controllable selection processes and illustrate benefits of fuzzy-based processing of available information are illustrated. Firstly, we concentrate on social networks. A methodology for selecting groups of individuals that satisfy linguistically described requirements regarding the degree of matching between users´ interests and collective interests of groups is presented. Secondly, we offer a novel recommending approach that provides users with a fuzzy-based process aiming at construction of lists of suggested items. This is accomplished via explicit control of requirements regarding rigorousness of identifying users who become a reference base for generating suggestions. A new way of ranking items rated by multiple users based on Pythagorean fuzzy sets and taking into account not only assigned rates but also their number is described.
Publisher :
ieee
Conference_Titel :
Technologies and Applications of Artificial Intelligence (TAAI), 2015 Conference on
Electronic_ISBN :
2376-6824
Type :
conf
DOI :
10.1109/TAAI.2015.7407048
Filename :
7407048
Link To Document :
بازگشت