This article is a transcript of the lightning talk we presented this week at Haystack - the Search and Relevance Conference. We showed a method allowing to enrich and rewrite user queries using Wikidata and the Elasticsearch percolator.
Haystack offered us the opportunity to present a lightning talk during the conference. We presented an innovative method that can be used in order to rewrite user queries. This method has the advantage to require only an Elasticsearch instance and Open Data coming from Wikidata.
When dealing with voice search, a chatbot or even with a certain type of users or use cases, the user queries can be quite verbose. A rewriting process is then necessary to extract the meaningful keywords. A user query like How to prepare athletes to be present on D-Day taking into account all the complexity of the competition? can be eventually rewritten with just the following keywords: present, D-Day, competition.
For this task we are using Wikidata - an Open Source database maintained by the Wikimedia Foundation and the Elasticsearch percolator - a feature available in Elasticsearch since the early days but still not so well known.
The following slides present in detail the process we put in place.
You can easily implement the same framework on your infrastructure but if you need assistance, do not hesitate to contact us.