- The software provides a generic search for structured data(a database, a csv file etc) which does not require knowledge of any database languages(such as MySQL).
- However after the query has been entered, comes the most interesting part which is the ranked nature of the tuples returned in the result. Thus based on the query, the user gets a ranked list of tuples according to a relevancy score attached to each tuple.
- The relevancy score calculations are based on the characteristics (frequency, locality of occurence) of the query terms in the database and the a workload.
- The workload is a list of the queries made by users and keeps on growing( and improving the correctness of the ranked list of tuples ) as the software is used.
- What makes this project a bit different is that it provides ranked based results and does relevancy calculations on structured data, ideas which are mostly employed on unstructured data sets( such as the web pages on the internet ).