The purpose of this M. Sc. thesis proposal is to design and develop a multi-domain search application as the ones allowed by the Search Computing approach (www.search-computing.com), focused on the technologies and interactions typical of modern mobile systems (iPad, smartphones, and so on). In particular, the student will investigate how to build applications upon Android and iPhone platforms, how to specify requirements, design and implementation of search applications, and how to integrate advanced search services together.

This thesis aims at defining novel exploration and search interfaces for SeCo results and for applying the liquid query paradigm beyond the current state of the art. Starting from a thorough study of the state of the art in the context of user interaction models and interfaces for search engines, this work aims at 1) defining a new set of exploration and set interfaces for multi-domain search (possibly addressed to mobile devices) and 2) proposing a general-purpose, configurable framework that leverages the existing SeCo machinery. Finally, part of the thesis work will consist in performing user studies to evaluate the performance of the proposed interfaces with respect to state-of-the-art solutions.

Diversification is an important feature in modern search engines as it provides means to overcome query ambiguity and information overload. A common diversification approach involves selectively picking documents in a result set in order to cover as much as possible all the semantic domains underlying the user queries. The generation of highly significant search results for the users, though, can be on other important techniques, such as personalization, recommendation and context-awareness, at the purpose of anticipating diverse and interesting results in the result list.
This focus of this thesis is on the study, implementation and validation of result diversification techniques for multi-domain search results, possibly considering contextual information such as user profiles, preferences, etc.
 

In the last years the user information seeking process on the Web has shifted from document search to object search. Hence, the answers provided by Web search engines cannot consist any more in a mere list of pages. This can be experienced also on well known general purpose search engines; e.g., try to submit “Washington”, “Brad Pitt”, “Microsoft stocks”, “Katie Holmes” to Google and look at the result page layout). The problem can be generalized to multi-format and multi-domain search queries, that are not addressed at all by current search engines and that introduce new dimensions to the layout definition problem.
The aim of this thesis is to characterize the problem of the definition of the layout of such results. In particular, the thesis will focus on describing a conceptual definition of the Web search result layout problem, by identifying: the parameters involved in the layout design, the tuning dimensions available for optimizing the result layout, and the possible strategies that can be adopted for producing such layouts. Finally, we provide an outlook on the possible future research directions on this topic.