Speeding Up Iterative Ontology Alignment using Block-Coordinate Descent

Authors: U. Thayasivam, P. Doshi

JAIR 2014 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We extensively evaluate this approach by integrating it into multiple ontology alignment algorithms. We selected Falcon-AO (Jian et al., 2005), Map PSO (Bock & Hettenhausen, 2010), OLA (Euzenat & Valtchev, 2004) and Optima (Doshi et al., 2009) as representative algorithms. Using a comprehensive testbed of several ontology pairs some of which are large spanning multiple domains, we show a significant reduction in the execution times of the alignment processes thereby converging faster.
Researcher Affiliation Academia Uthayasanker Thayasivam EMAIL Prashant Doshi EMAIL THINC Lab, Department of Computer Science, University of Georgia, Athens, GA 30602, USA
Pseudocode Yes Figure 3: General algorithms for iterative (a) update, and (b) search approaches toward aligning ontologies. Figure 5: General iterative algorithms of Fig. 3 are modified to obtain, (a) iterative update enhanced with BCD, and (b) iterative search enhanced with BCD.
Open Source Code No The paper states that the implementations and source codes for the *selected algorithms* (Falcon-AO, Map PSO, OLA, Optima) are freely accessible. However, it does not explicitly state that the authors' *own code* for the BCD-enhanced versions or their specific methodology is publicly available or provides a link to it.
Open Datasets Yes In order to comprehensively evaluate the efficiency of the BCD-enhanced and optimized algorithms, we contribute a novel biomedical ontology alignment testbed. ... The testbed with reference alignments is available for download at http://tinyurl.com/n4t2ns3.
Dataset Splits No The paper mentions evaluating algorithms on ontology pairs and that
Hardware Specification Yes One of these is a Red Hat machine with Intel Xeon Core 2, processor speed of about 3 GHz with 8GB of memory (anatomy ontology pair) and the other one is a Windows 7 machine with Intel Core i7, 1.6 GHz processor and 4GB of memory (benchmark and conference ontology pairs).
Software Dependencies No The paper does not provide specific version numbers for any software libraries, frameworks, or programming languages used in their experimental setup. It mentions general software concepts like OWL and description logics but no concrete, versioned dependencies.
Experiment Setup No The paper describes the general iterative approach and how BCD is integrated, including conditions for convergence (e.g., 'While δ η'). However, it does not provide specific hyperparameter values (e.g., learning rates, batch sizes, specific numerical values for termination parameters other than 'η') or detailed training configurations for their experiments beyond the conceptual algorithmic steps.