PROJECT TITLE :

Keyword Search on Temporal Graphs - 2017

ABSTRACT:

Archiving graph information over history is demanded in many applications, like social network studies, collaborative comes, scientific graph databases, and bibliographies. Typically people are fascinated by querying temporal graphs. Existing keyword search approaches for graph-structured information are insufficient for querying temporal graphs. This paper initiates the study of supporting keyword-based queries on temporal graphs. We tend to propose a probe syntax that's a moderate extension of keyword search, which allows casual users to simply search temporal graphs with optional predicates and ranking functions connected to timestamps. To generate results efficiently, we have a tendency to first propose a best path iterator, which finds the methods between two data nodes in every snapshot that is the “best” with respect to a few ranking factors. It prunes invalid or inferior ways and maximizes shared processing among different snapshots. Then, we develop algorithms that efficiently generate high-k query results. Intensive experiments verified the efficiency and effectiveness of our approach.