Abstract. With large datasets such as Linked Open Data available, there is a need for more user-friendly interfaces which will bring the advantages of these data closer to the casual users. Several recent studies have shown user preference to Natural Language Interfaces (NLIs) in comparison to others. Although many NLIs to ontologies have been developed, those that have reasonable performance are domain-specific and tend to require customisation for each new domain which, from a developer's perspective, makes them expensive to maintain. We present our system FREyA, which combines syntactic parsing with the knowledge encoded in ontologies in order to reduce the customisation effort. If the system fails to automatically derive an answer, it will generate clarification dialogs for the user. The user's selections are saved and used for training the system in order to improve its performance over time. FREyA is evaluated using Mooney Geoquery dataset with very high precision and recall.

Abstract. Natural Language Interfaces are increasingly relevant for information systems fronting rich structured data stores such as RDF and OWL repositories, mainly because of the conception of them being intuitive for human. In the previous work, we developed FREyA, an interactive Natural Language Interface for querying ontologies. It uses syntactic parsing in combination with the ontology-based lookup in order to interpret the question, and involves the user if necessary. The user's choices are used for training the system in order to improve its performance over time. In this paper, we discuss the suitability of FREyA to query the Linked Open Data. We report its performance in terms of precision and recall using the MusicBrainz and DBpedia datasets.

No abstract

When researching new product ideas or filing new patents, inventors need to retrieve all relevant pre-existing know-how and/or to exploit and enforce patents in their technological domain. However, this process is hindered by lack of richer metadata, which if present, would allow more powerful concept-based search to complement the current keywordbased approach. This paper presents our approach to automatic patent enrichment, tested in large-scale, parallel experiments on USPTO and EPO documents. It starts by defining the metadata annotation task and examines its challenges. The text analysis tools are presented next, including details on automatic annotation of sections, references and measurements. The key challenges encountered were dealing with ambiguities and errors in the data; creation and maintenance of large, domain-independent dictionaries; and building an efficient, robust patent analysis pipeline, capable of dealing with terabytes of data. The accuracy of automatically created metadata is evaluated against a human-annotated gold standard, with results of over 90% on most annotation types.

No abstract