8.13

Suppose you wish to perform keyword querying on a set of tuples in a database, where each tuple has only a few attributes, each containing only a few words. Does the concept of term frequency make sense in this context? And that of inverse document frequency? Explain your answer. Also suggest how you can define the similarity of two tuples using TF-IDF concepts.

---

Although querying on structured data are typically done using query languages such as SQL, users who are not familiar with the schema or the query language find it difficult to get information from such data. Based on the success of keyword querying in the context of information retrieval from the web, techniques have been developed to support keyword queries on structured and semi-structured data.

The above paragraph is taken from section 8.3.4 of the book.

So yes, the concepts of term frequency and inverse document frequency do make sense when performing keyword querying on a set of tuples in a database.

The tuples will be considered as the documents.