Search of Data Patterns by Ranking Object and Constructing Bins Using Materialized Sub Graph
Ponnada Naga Ramya · 2014
To perform a key word search from the database the object rank algorithms and page rank algorithms are used. These are general search algorithms basically used in all search engines, these algorithms uses iterative computation over a full graph hence this computation becomes expensive for large graphs and also had expensive preprocessing. To make the computation simpler here the bin ranking and hub ranking are introduced. Here bin rank involves in generating the sub graphs by partitioning all the term based on their co- occurrence. The intuition is that a sub graph that contain all objects and links relevant to a set of related terms should have all the information needed to rank objects with respect to one of these terms. The bin rank can achieve sub second query execution time on the English Wikipedia data set. The general object rank search will approximate the search result on the graph. Thus this experimental evaluation in depth explains the trade-off between query execution, time and quality of results. The hub rank help to minimize the search engine traffic by ranking the recently viewed path.