A neural feature extraction model for classification of firms and prediction of outsourcing success: advantage of using relational sources of information for new suppliers

Abstract

Clustering is a family of classification techniques, often preceding further analysis or application in a number of fields like data analysis, strategy selection, supplier selection, etc. Data based neural techniques are gaining popularity in clustering applications due to flexibility and adaptability. Kohonen’s Self Organizing Map (SOM) is often used when the objects to be clustered have many attributes. In both supervised and un-supervised modes, Kohonen’s map exhibit good capability to extract a classification which assigns highest weight to the most important attribute. In this paper, we have applied SOM for classification of firms based on their sources of information for new suppliers/customers. Additional data regarding the outsourcing success of the firms’ is added to see if there is an association between a particular set of information sources and the probability of firms’ success to outsource to partner firms. Using data from World Bank BEEPS survey of German industries, we could produce three distinct clusters of industries. When successful outsourcing data were included, it still showed three clusters. The hits were obtained using specific support vector for identification of clusters. We found evidence of associations between relational sources and firms’ ability to outsource successfully.

Keywords:

• artificial intelligence
• neural network applications
• supplier selection
• SOM
• clustering