Journal Published Online: 28 February 2017
Volume 1, Issue 1

Concept-Based Text Mining Technique for Semantic Classification of Manufacturing Suppliers

CODEN: SSMSCY

Abstract

Small-to-medium sized enterprises (SMEs) in the manufacturing sector are increasingly strengthening their web presence in order to improve their visibility and remain competitive in the global market. With the explosive growth of unstructured content on the Web, more advanced methods for information organization and retrieval are needed to improve the intelligence and efficiency of the supplier discovery and evaluation process. In this paper, a technique for automated characterization and classification of manufacturing suppliers based on their textual portfolios was presented. A probabilistic technique that adopts Naïve Bayes method was used as the underlying mathematical model of the proposed text classifier. To improve the semantic relevance of the results, classification was conducted at the conceptual level rather than at the term level that is typically used by conventional text classifiers. The necessary steps for training data preparation and representation related to manufacturing supplier classification problem are delineated. The proposed classifier is capable of forming both simple and complex classes of manufacturing SMEs based on their advertised capabilities. The performance of the proposed classifier wass evaluated experimentally based on the standard metrics used in information retrieval such as precision, recall, and F-measure. It was concluded that the proposed concept-based classification technique outperforms the traditional term-based methods with respect to accuracy, robustness, and cost.

Author Information

Shotorbani, P.
Engineering Informatics Research Group, Texas State Univ., San Marcos, TX, US
Ameri, F.
Engineering Informatics Research Group, Texas State Univ., San Marcos, TX, US
Pages: 24
Price: Free
Related
Reprints and Permissions
Reprints and copyright permissions can be requested through the
Copyright Clearance Center
Details
Stock #: SSMS20160005
ISSN: 2520-6478
DOI: 10.1520/SSMS20160005