Using a Fuzzy-Based Cluster Algorithm for Recommending Candidates in E-Elections

Abstract

The use of the Internet now has a specific purpose: to find information. Unfortunately, the amount of data available on the Internet is growing exponentially, creating what can be considered a nearly infinite and ever-evolving network with no discernable structure. This rapid growth has raised the question of how to find the most relevant information.Many different techniques have been introduced to address the information overload, including search engines, semantic web, and recommender systems, among others.Recommender systems are computer-based techniques that are used to reduce information overload and recommend products likely to interest a user when given some information about the user's profile. This technique is mainly used in eCommerce to suggest items that fit a customer's purchasing tendencies.The use of recommender systems for eGovernment is a research topic that is intended to improve the interaction among public administrations, citizens, and the private sector through reducing information overload on eGovernment services. More specifically, eDemocracy aims to increase citizens' participation in democratic processes through the use of information and communication technologies.In this chapter, an architecture of a recommender system that uses fuzzy clustering methods for eElections is introduced. In addition, a comparison with the smartvote system, a Web-based Voting Assistance Application (VAA) used to aid voters in finding the party or candidate that is most in line with their preferences, is presented.

Election candidates fuzzy multi-agent recommender system

“…-Teran et al [19] solution from 2010 uses fuzzy c-means cluster algorithm, and our solution uses different algorithm for recommendation (type-2 fuzzy sets and similarity function). Also, our system is multi-agent based and uses flexible software development technology; -Herrera et al [10] solution from 2009 is decision making system which is based on knowledge base, compared to our solution which is a recommender system that works based on similarity of preferences of voters with candidates in a structured environment (questionnaire); -the Dutch Stemwijzer and the German Wahl-O-Mat have three possible answers on each question, and the Dutch Kieskompas has five possible answers on each question, with additional No opinion option.…”

Artificial intelligence and voting advice applications

“…From a mathematical (and ethical) standpoint, highdimensional algorithms should be preferable to algorithms based on summated scales in a low-dimensional space as used in the Kieskompas family of VAAs, since the latter tend to produce misleading results near the center of the dimensions as the summation of opposing responses falsely assumes all items of the scale to be interchangeable (for an example, see Gemenis, 2012). To improve the visualization of matching in a candidate-based VAA in Switzerland, Terán et al (2012) used fuzzy c-means (FCM) clustering to compute similarities based on distances in a high-dimensional space, but also to derive a two-dimensional space that included the percentage of similarity of the n-closest candidates. From a psychometric perspective, the matching using a low-dimensional solution can be improved through a careful examination of how different statements map onto different dimensions (if at all).…”

Tree Similarity Measure-based Recommender Systems

“…Increases in this information overload could clearly hamper the effectiveness of egovernment services, and difficulties in locating the right information for the right users will increasingly impact on loyalty of users. Recommender systems can overcome this problem and have been adopted in e-government applications Terán et al 2012;Terán & Meier 2010, 2011.…”

“…The proposed system identifies and suggests the most interesting services for a user by considering both the user's profile and the profile of the device being used. To assist voters to make decisions in the e-election process, a recommender system was proposed (Terán et al 2012;Terán & Meier 2010), which uses fuzzy clustering methods and provides information about candidates close to voters' preferences.…”