Efficient IR-Style Keyword Search over Relational Databases

Abstract

Applications in which plain text coexists with structured data are pervasive. Commercial relational database management systems (RDBMSs) generally provide querying capabilities for text attributes that incorporate state-of-the-art information retrieval (IR) relevance ranking strategies, but this search functionality requires that queries specify the exact column or columns against which a given list of keywords is to be matched. This requirement can be cumbersome and inflexible from a user perspective: good answers to a keyword query might need to be "assembled" -in perhaps unforeseen ways-by joining tuples from multiple relations. This observation has motivated recent research on free-form keyword search over RDBMSs. In this paper, we adapt IR-style document-relevance ranking strategies to the problem of processing free-form keyword queries over RDBMSs. Our query model can handle queries with both AND and OR semantics, and exploits the sophisticated single-column text-search functionality often available in commercial RDBMSs. We develop query-processing strategies that build on a crucial characteristic of IR-style keyword search: only the few most relevant matches -according to some definition of "relevance"-are generally of interest. Consequently, rather than computing all matches for a keyword query, which leads to inefficient executions, our techniques focus on the top-k matches for the query, for moderate values of k. A thorough experimental evaluation over real data shows the performance advantages of our approach. *

Finding Top-k Min-Cost Connected Trees in Databases

“…The relation-based approaches aim at processing a keyword query with SQL, by utilizing the schema information in RDBMS. Such systems include DBXPlore [1], IR-Style [17], and DISCOVER [18]. On the other hand, the tuple-based approaches aim at processing a keyword query by utilizing the weights associated with nodes (tuples) and edges (foreign-key reference) in a graph.…”

“…There are l groups, V 1 , · · · , V l . All groups can be obtained with either the symboltable techniques [1] or the full text index techniques [18]. Minimum group Steiner tree problem (GST-1): Given a l-keyword query, to find the minimum cost connected tree T , such as V (T ) ∩ V i = ∅ for i = 1, · · · , l. For brevity, the cost of such a tree T , is given below.…”

“…It is widely realized that the integration of information retrieval (IR) and database (DB) techniques will provide users with a wide range of high quality services [3,2,10]. The recent studies on supporting IR style queries in RDBMS include DBXPlore [1], IR-Style [17], DISCOVER [18], ObjectRank [4], BANKS-I [6], and BANKS-II [25]. All consider a RDBMS as a graph where nodes represent tuples/relations and edges represent foreign key references among tuples cross relations.…”

Keyword Query Expansion Paradigm Based on Recommendation and Interpretation in Relational Databases

“…According to the different modeling methods of databases, the existing query methods can be divided into two main categories: schema graph-based methods and data graph-based methods. In [16][17][18][19][20][21], the database is modeled as a schema graph, where nodes represent tables and edges represent primary-foreign-key relationships. These methods enumerate all possible CNs (Candidate Networks) based on the schema graph.…”

An Effective Object-Level XML Keyword Search

“…BLINKS [8] improves it by designing a bi-level index for result pruning, with the tradeoffs in index size and maintenance cost. XKeyword [11] uses schema information to reduce search space, but its query evaluation is based on the method of DISCOVER [10] built on RDBMS, which cannot distinguish the containment and reference edges to further reduce search space. [5] builds a tree+IDRef model to capture ID references by avoiding the NP-hard complexity.…”