The uncertainty mapping of ontologies based on three-dimensional combination weight vector space model

Abstract

Purpose This paper aims to apply vector space model (VSM)-PCR model to compute the similarity of Fault zone ontology semantics, which verified the feasibility and effectiveness of the application of VSM-PCR method in uncertainty mapping of ontologies. Design/methodology/approach The authors first define the concept of uncertainty ontology and then propose the method of ontology mapping. The proposed method fully considers the properties of ontology in measuring the similarity of concept. It expands the single VSM of concept meaning or instance set to the “meaning, properties, instance” three-dimensional VSM and uses membership degree or correlation to express the level of uncertainty. Findings It provides a relatively better accuracy which verified the feasibility and effectiveness of VSM-PCR method in treating the uncertainty mapping of ontology. Research limitations/implications The future work will focus on exploring the similarity measure and combinational methods in every dimension. Originality/value This paper presents an uncertain mapping method of ontology concept based on three-dimensional combination weighted VSM, namely, VSM-PCR. It expands the single VSM of concept meaning or instance set to the “meaning, properties, instance” three-dimensional VSM. The model uses membership degree or correlation which is used to express the degree of uncertainty; as a result, a three-dimensional VSM is obtained. The authors finally provide an example to verify the feasibility and effectiveness of VSM-PCR method in treating the uncertainty mapping of ontology.

Author Profiling in Social Media with Multimodal Information

“…Although there are different models to represent the text, the Vector Space Model (VSM) (Sidorov et al, 2014) is the principal model for many textual tasks, which is used to represent the text documents and define the similarity among them (Liu et al, 2005;Han et al, 2018). Bag of Word (BOW) (Wallach, 2006) is an approach used to represent each document as a histogram of words under the VSM.…”