Ranking target objects of navigational queries

Abstract

Web navigation plays an important role in exploring public interconnected data sources such as life science data. A navigational query in the life science graph produces a result graph which is a layered directed acyclic graph (DAG). Traversing the result paths in this graph reaches a target object set (TOS). The challenge for ranking the target objects is to provide recommendations that reflect the relative importance of the retrieved object, as well as its relevance to the specific query posed by the scientist. We present a metric layered graph PageRank (lgPR) to rank target objects based on the link structure of the result graph. LgPR is a modification of PageRank; it avoids random jumps to respect the path structure of the result graph. We also outline a metric layered graph ObjectRank (lgOR) which extends the metric ObjectRank to layered graphs. We then present an initial evaluation of lgPR. We perform experiments on a real-world graph of life sciences objects from NCBI and report on the ranking distribution produced by lgPR. We compare lgPR with PageRank. In order to understand the characteristics of lgPR, an expert compared the Top K target objects (publications in the PubMed source) produced by lgPR and a word-based ranking method that uses text features extracted from an external source (such as Entrez Gene) to rank publications.

BioNav: An Ontology-Based Framework to Discover Semantic Links in the Cloud of Linked Data

“…The most important terms or publications correspond to the highly ranked Mesh terms or publications in the last layer of the lgODG. In this paper we focus on an extension of the Object Rank [1,2] and Path Count [3] metrics for layered graphs or layered graph Weighted Path Count (lgWP), which is defined as follows:…”

Ranked Search on Data Graphs

“…[RWL+06] proposed the lgOR ranking, a variant of ObjectRank, to answer such queries. These queries apply authority flow ranking on an acyclic directed layered graph produced by the hard path filter.…”

“…The graphsampling technique probabilistically selects a subset of the paths using a Bayesian network. It is applied to approximating lgOR queries (introduced in [RWL+06]), which are equivalent to a hard path hard filter followed by an authority-flow soft filter. This approximation is indispensable when the data graph is large.…”

“…The Hubs of Knowledge project [SIY06] applies the PageRank algorithm on a query-dependent subgraph of the original biological graph. Raschid et al [RWL+06] apply PageRank and ObjectRank to answer navigational queries on biological data. Conceptually, the ranking is produced in the following way: Myriads of random surfers are initially found at the objects containing the keyword -OLAP‖, which we call the base set, and then they traverse the database graph.…”

“…GID emulates domain graph query languages such as lgOR, lgPR [RWL+06] and filter queries in PubMed [PM07]. GID can be combined with more general graph languages to support complex queries.…”