HIEDS: A Generic and Efficient Approach to Hierarchical Dataset Summarization

Authors: Gong Cheng, Cheng Jin, Yuzhong Qu

IJCAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We systematically experiment with our approach on real-world RDF datasets. In this section, we empirically study the quality of summaries for real-world RDF datasets generated by HIEDS under various configurations, compare it with a baseline method, and report its running time.
Researcher Affiliation Academia Gong Cheng , Cheng Jin, Yuzhong Qu National Key Laboratory for Novel Software Technology, Nanjing University, Nanjing 210023, China
Pseudocode Yes Algorithm 1: Computing CPV 0(G) and Algorithm 2: Finding Relations that Satisfy Eq. (11)
Open Source Code No The paper states 'We have implemented an online prototype of our HIEDS approach.1' with a footnote '1http://ws.nju.edu.cn/hieds/', which links to a project page demonstrating the prototype, but does not provide explicit access to the source code for the methodology.
Open Datasets Yes Two extensively used RDF datasets were tested. SWDF2 (Semantic Web Dog Food) offers 200K entity-property-value triples describing 20K entities in the research domain (e.g., papers, researchers). Linked MDB3 offers 6M entity-property-value triples describing 0.6M movie-related entities (e.g., actors). Footnotes: 2http://data.semanticweb.org/ 3http://data.linkedmdb.org/
Dataset Splits No The paper discusses how groups are iteratively subdivided and evaluated, but does not specify train, validation, or test splits of the datasets themselves, nor does it refer to cross-validation or predefined splits.
Hardware Specification Yes We tested the running time of HIEDS on an Intel E3-1225 v3 with 30G memory for our Java program.
Software Dependencies No The paper mentions 'our Java program' and 'Lucene (lucene.apache.org)' but does not provide specific version numbers for these software dependencies.
Experiment Setup Yes In this experiment, groups were iteratively subdivided until each leaf group contained not more than 50 entities. We fixed µE = +1 and E = 0.01 (c.f. Algorithm 1), to focus on the effects of other parameters. The following parameters were fixed for simplicity: = β = 0.5, δ = 0, E = R = 0.01, k = 7, and not introducing constraint 4 .