Semantic search processing in natural language interface

Natural language interface (NLI) allows user to communicate with the semantic knowledge base through natural language (NL). However, several challenges exist such as understanding user´s intention in the query and matching the elements in the query with the knowledge base (KB). Several solutions exist but were developed differently in terms of query processing components, query matching style and expressivity support. The focus of the autoSPARQL application introduced in this paper is the soft queries processing and the result is compared against FREyA, an NLI equipped with clarification dialogue component. Result has demonstrated that there is no significant difference between these applications. The results also open rooms for future works such as re-testing in bigger scale and more concrete numerical and semantic expression processing.