A natural language ontology-driven query interface

Recent research showed that adopting formal ontologies as a means for accessing heterogeneous data sources has many benefits, in that not only does it provide a uniform and flexible approach to integrating and describing such sources, but it can also support the final user in querying them, thus improving the usability of the integrated system. We introduce a framework that enables access to heterogeneous data sources by means of a conceptual schema and supports the users in the task of formulating a precise query over it. In de- scribing a specific domain, the ontology defines a vocabulary which is often richer than the logical schema of the underlying data and usually closer to the user’s own vocabulary. The ontology can thus be effectively exploited by the user in order to formulate a query that best captures their information need. The user is constantly guided and assisted in this task by an intuitive visual interface, whose intelligence is dynamically driven by reasoning over the ontology. The inferences drawn on the conceptual schema help the user in choosing what is more appropriate with respect to their information need, restricting the possible choices to only those parts of the ontology which are relevant and meaningful in a given context. The most powerful and innovative feature of our framework lies in the fact that not only do not users need to be aware of the underlying organisation of the data, but they are also not required to have any specific knowledge of the vocabulary used in the ontology. In fact, such knowledge can be gradually acquired by using the tool itself, gaining confidence with both the vocabulary and the ontology. Users may also decide to just explore the ontology without actually querying the information system, with the aim of discovering general information about the modelled domain. Another important aspect is that only queries that are logically consistent with the context and the constraints imposed by the ontology can be formulated, since contradictory or redundant pieces of information are not presented to the user at all. This makes user’s choices clearer and simpler, by ruling out irrelevant information that might be distracting and even generate confusion. Furthermore, it also eliminates the often frustrating and time-consuming process of finding the right combination of parts that together constitute a meaningful query. For this reason, the user is free to explore the ontology without the worry of making a “wrong” choice at some point and can concentrate on expressing their information need. Queries can be specified through a refinement process consisting in the iteration of few basic op- erations: the user first specifies an initial request starting with generic terms, then refines or deletes some of the previously added terms or introduces new ones, and iterates the process until the resulting query satisfies their information need. The available operations on the current query include addition, substitution and deletion of pieces of in- formation, and all of them are supported by the reasoning services running over the ontology. In this paper we summarise only the NL aspects of a tool based on those ideas, Quelo; for a complete picture of our ideas and of the tool re- fer to our papers (Franconi et al., 2011; Dongilli et al., 2004; Catarci et al., 2004; Catarci et al., 2005; Dongilli and Franconi, 2006; Franconi et al., 2010). Quelo relies on a web-based client- server architecture.....