By Topic

A novel ranking framework for linked data from relational databases

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

The purchase and pricing options are temporarily unavailable. Please try again later.
6 Author(s)
Zhang, Jing ; IBM China Development Laboratory, Beijing 100193, China ; Ma, Chune ; Zhao, Chenting ; Zhang, Jun
more authors

This paper investigates the problem of ranking linked data from relational databases using a ranking framework. The core idea is to group relationships by their types, then rank the types, and finally rank the instances attached to each type. The ranking criteria for each step considers the mapping rules and heterogeneous graph structure of the data web. Tests based on a social network dataset show that the linked data ranking is effective and easier for people to understand. This approach benefits from utilizing relationships deduced from mapping rules based on table schemas and distinguishing the relationship types, which results in better ranking andvisualization of the linked data.

Published in:

Tsinghua Science and Technology  (Volume:15 ,  Issue: 6 )