Abstract
AbstractThis paper reports on creating virtual assistants (VA) that enable users to query a database in the natural language. Building SQL queries from the natural language is a complicated task. We build the query via a conversation between the user and the virtual assistant allowing the users to describe their needs during a more detailed conversation. The VA uses information about the schema of the data source to guide the user. The query is built incrementally. To test the proposed method, we implemented a dialogue system for querying a part of the Open Food Facts database. The evaluation results show that users successfully completed the task in most cases. The easiest task was completed by 72% of users, the most sophisticated task was completed by 58% of users. To finish the tasks, users had to provide parameters that the VA prompted for, to sort the records, and to add filtering conditions using natural language. The proposed approach allows the building of similar VAs for different databases.KeywordsVirtual assistantsSemantic parsingMachine learning
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
