A study of concept semantic similarity measure optimization method
Date
2016
Authors
Cao, R.
Wu, L.
Wang, L.
Wang, R.
Yang, C.
Editors
Advisors
Journal Title
Journal ISSN
Volume Title
Type:
Journal article
Citation
Journal of Computational and Theoretical Nanoscience, 2016; 13(1):547-557
Statement of Responsibility
Conference Name
Abstract
A novel scheme based on neural networks for optimizing semantic similarity measure is proposed in this study. This scheme is investigated by using three representative neural networks, i.e., the Back Propagation (BP) neural networks optimized by Genetic Algorithm (GA), the BP neural networks optimized by the Mind Evolutionary Algorithm (MEA), and the generalized regression neural network (GRNN). The performance of our proposed scheme is examined on the problem of semantic similarity measure for Special Crop. To this end, the Special Crops domain ontology is established and a set of factors having major influence on semantic similarity measure for neural networks training process is identified. Extensive experiments and comparison demonstrate that optimizing semantic similarity measure based on neural network is both feasible and effective. The promising experimental results also suggest that the method based on the generalized regression neural network performs the best.
School/Discipline
Dissertation Note
Provenance
Description
Access Status
Rights
Copyright 2016 American Scientific Publishers