Abstract
The measurement of semantic similarity has been widely recognized as having a fundamental and key role in information science and information systems. Although various models have been proposed to measure semantic similarity, these models are not able effectively to quantify the weights of relevant factors that impact on the judgement of semantic similarity, such as the attributes of concepts, application context, and concept hierarchy. In this paper, we propose a novel approach that comprehensively considers the effects of various factors on semantic similarity judgment, which we name semantic similarity measurement based on a weighted concept lattice (SSMWCL). A feature model and network model are integrated together in SSMWCL. Based on the feature model, the combined weight of each attribute of the concepts is calculated by merging its information entropy and inclusion-degree importance in a specific application context. By establishing the weighted concept lattice, the relative hierarchical depths of concepts for comparison are computed according to the principle of the network model. The integration of feature model and network model enables SSMWCL to take account of differences in concepts more comprehensively in semantic similarity measurement. Additionally, a workflow of SSMWCL is designed to demonstrate these procedures and a case study of geo-information is conducted to assess the approach.
Highlights
In information science and systems, semantic similarity plays a major role in various fields such as information retrieval, data integration, data mining etc. [1,2,3]
With reference to various existing approaches, we propose a novel approach to semantic similarity measurement based on a weighted concept lattice (SSMWCL) in this paper, which combines both a feature model and network model
We propose semantic similarity measurement based on a weighted concept lattice (SSMWCL) as a new approach to measuring semantic similarity among concepts
Summary
In information science and systems, semantic similarity plays a major role in various fields such as information retrieval, data integration, data mining etc. [1,2,3]. With reference to various existing approaches, we propose a novel approach to semantic similarity measurement based on a weighted concept lattice (SSMWCL) in this paper, which combines both a feature model and network model. The absolute semantic similarity between concepts is measured by comparing the commonalities and differences of their weighted attributes, in which the relative hierarchical depth of the concepts in the lattice is taken into account based on the principle of the network model. For this construction, some mathematical tools are applied, including the rough set, information entropy, and formal concept analysis.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
