Search As You Type in Database
Parvathi R1, Syama R2
1Parvathi R, Department of Information Technology, University of Kerala, SCT College of Engineering, Trivandrum (Kerala), India.
2Syama R, Department of Computer Science, University of Kerala, SCT College of Engineering, Trivandrum (Kerala), India.
Manuscript received on 15 August 2015 | Revised Manuscript received on 25 August 2015 | Manuscript Published on 30 August 2015 | PP: 269-273 | Volume-4 Issue-6, August 2015 | Retrieval Number: F4233084615/15©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: A search-as-you-type system computes answers on-the-fly as a user types in a keyword query character by character. Search-as-you-type support study on data residing in a relational DBMS. And also focus on how to support this type of search using the native database language, SQL. A main challenge is how to leverage existing database functionalities to meet the high-performance requirement to achieve an interactive speed. Study on how to use auxiliary indexes is stored as tables to increase the search performance. Solutions for both single-keyword queries and multi-keyword queries are presented, and develop novel techniques for fuzzy search using SQL by allowing mismatches between query keywords and answers. Experiments on large, real data sets show that techniques enable DBMS systems on a commodity computer to support search-as-you-type on tables with millions of records. The main consideration was to increase the speed by using auxiliary indexes stored as tables. The search is done based on both single and multi-keyword. Exact search for single keyword queries are done using UDF, LIKE predicate and inverted-index table and the prefix table. Exact search for multi keyword queries are done using UDF, LIKE predicate, full-text indexes and UDF (called “FI+UDF”), full-text indexes and the LIKE predicate (called “FI+LIKE”), the inverted-index table with prefix table and word-level incremental method. Fuzzy search for single keyword queries are implemented using UDF, gram-based method, neighborhood-generation-based method, character-level incremental algorithms. Fuzzy search for multi keyword queries are implemented using word-level incremental algorithms, called NGB+ and Incre+. The approach using inverted index tables and prefix tables supports prefix, fuzzy search and achieve the best performance. The experimental results on large, real data sets showed that the proposed techniques can enable DBMS systems to support search-as-you-type on large tables.
Keywords: Fuzzy Search, Type Ahead, Prefix Search, Edit Distance
Scope of the Article: Fuzzy Logics