From Search Engines to Search Services: An End-User Driven Approach

From Search Engines to Search Services: An End-User Driven Approach Gabriela Bosetti1, Sergio Firmenich1,2, Alejandro Fernandez1, Marco Winckler3, Gustavo Rossi1,2 1LIFIA, CIC, Facultad de Informática, Universidad Nacional de La Plata {gabriela.bosetti, sergio.firmenich, alejandro.fernandez, gustavo}@lifia.info.unlp.edu.ar 2 CONICET, Argentina 3 ICS-IRIT, University of Toulouse 3, France winckler@irit.fr Abstract. The World Wide Web is a vast and continuously changing source of information where searching is a frequent, and sometimes critical, user task. Searching is not always the user’s primary goal but an ancillary task that is per- formed to find complementary information allowing to complete another task. In this paper, we explore primary and/or ancillary search tasks and propose an ap- proach for simplifying the user interaction during search tasks. Rather than fo- cusing on dedicated search engines, our approach allows the user to abstract search engines already provided by Web applications into pervasive search ser- vices that will be available for performing searches from any other Web site. We also propose to allow users to manage the way in which searching results are displayed and the interaction with them. In order to illustrate the feasibility of this approach, we have built a support tool based on a plug-in architecture that allows users to integrate new search services (created by themselves by means of visual tools) and execute them in the context of both kinds of searches. A case study illustrates the use of these tools. We also present the results of two evalua- tions that demonstrate the feasibility of the approach and the benefits in its use. Keywords: Web Search, Client-Side Adaptation 1 Introduction Searching is one of the main important tasks that users perform when using Web brows- ers. In terms of user aims, searches in the Web may occur as a main task (primary search) or as a secondary one (ancillary search) [2]. According to this work [2], primary search corresponds to the user’s primary need for information, usually involving a sin- gle cycle question-answer (e.g. looking for something in Google or Amazon). In oppo- sition, ancillary searches are aimed at providing details (under demand) about the cur- rent information been displayed in a Web site the user is accessing; this task can be performed in different ways, e.g. by using an external Web page’s search engine or the contextual search menus that Web browsers provide. In order to delve deeper into the search process, we show in Figure 1 two task models: one for primary searches and other for the ancillary ones. Figure 1.a depicts how primary searches are independent of each other (e.g. finding the Web site of ICWE2016 or the WWW2016, in both cases, they are independent primary searches). In opposition, ancillary searches may become a recursive task, but in the background the main focus is still the original Web site. Figure 1.b depicts an example of this kind of search. Here, a researcher is looking for authors and articles in the ICWE2016 Web site. If this user requires further information about one of the authors, he can look for him at DBLP. There, the user may be interested in some specific resulting article, which triggers another ancillary search in Google Scholar to find the file or see the amount of citations. After that, the user may require to repeat a similar cycle of ancillary information seeking, i.e. search for another au- thor/article listed in ICWE2106’s Web site. Note that this last scenario may imply a detriment of user experience because it requires moving information and its context from one search Web site to another one. This get worse given the proliferation of Web applications managing millions of information items from diverse domains (cinema, tourism, wikis, Web searches, etc.) that could be queried by users.

a) Primary Search

b) Ancillary Search
Fig. 1: Primary and ancillary search task models and scenarios It must be clear that the information required by users in both kind of searches may be really different [13]. In ancillary searches, users need to move among Web sites just for obtaining complementary information that is frequently domain-specific (e.g. com- puter science literature), and consequently they would choose to use a specific Web site to look for such kind of information (e.g. DBLP or GoogleScholar). A second aspect in this kind of search is how and where the interaction with search results occurs. Usually, the interaction with search results is in the external Web page where the search was actually performed and not in the search-execution context (e.g. the Web page from where the user was motivated to perform the ancillary search). Here, it is important to differentiate between generic Web searchers (based on Web content scrapers) and domain-specific ones (based on domain objects). For performing primary search, it is

…(Full text truncated)…

This content is AI-processed based on ArXiv data.