Evaluating a Faceted Search Approach for Efficient News Event Filtering

Abstract

The aim is to investigate the user's behavior from the three types of search interface and compare the effectiveness of the user's point of view. The goal of the thesis is to find out to what extent the faceted search approach developed is efficient. To do so, I compare to other baseline approaches typically found on the Web to search for news content. User evaluation was conducted through one experiment to delve into the behavior of the users and the performance of these three interfaces. A baseline search interface includes a search only option. The second search interface shows the summary and which category each article is entitled to. The last search interface (faceted filtering) uses the same establishment as the latter one with addition facets functions. The three prototypes were evaluated with quantitative and qualitative methods for data gathering. Semi-structured interview, questionnaires, observation and System Usability Scale (SUS). The feedback was valuable, as a result, indicate a significant difference between the search interfaces from user performance, where the faceted filtering search interface is more efficient compared to the other two baseline approaches search interfaces. The faceted filtering search interface exceeds the baseline and the text summary search interface in performance and outperforms the other two baselines in terms of user satisfaction.