Understanding queries in a search database system

  • Authors:

  • Ronald Fagin;Benny Kimelfeld;Yunyao Li;Sriram Raghavan;Shivakumar Vaithyanathan

  • Affiliations:

  • IBM Research, San Jose, CA, USA;IBM Research, San Jose, CA, USA;IBM Research, San Jose, CA, USA;IBM Research, San Jose, CA, USA;IBM Research, San Jose, CA, USA

  • Venue:
  • Proceedings of the twenty-ninth ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

It is well known that a search engine can significantly benefit from an auxiliary database, which can suggest interpretations of the search query by means of the involved concepts and their interrelationship. The difficulty is to translate abstract notions like concept and interpretation into a concrete search algorithm that operates over the auxiliary database. To surpass existing heuristics, there is a need for a formal basis, which is realized in this paper through the framework of a search database system, where an interpretation is identified as a parse. It is shown that the parses of a query can be generated in polynomial time in the combined size of the input and the output, even if parses are restricted to those having a nonempty evaluation. Identifying that one parse is more specific than another is important for ranking answers, and this framework captures the precise semantics of being more specific; moreover, performing this comparison between parses is tractable. Lastly, the paper studies the problem of finding the most specific parses. Unfortunately, this problem turns out to be intractable in the general case. However, under reasonable assumptions, the parses can be enumerated in an order of decreasing specificity, with polynomial delay and polynomial space.