Data analysis for gathering and analysis of no structured and semi-structured information using data mining technique

In this study, a theoretical model for the combination of all individual questions from a questionnaire is investigated. For this implementation, a well-fitted Elman neural network is used in order to formulate the weighting relations between questions of the questionnaire which is produced for the project “Innovative Services to Strengthen Cooperation and Internationalization between SMEs” (I.S.C.I.) which has as objective the internationalization processes of the agro food industry of the Small and Medium-sized Enterprises (SMEs) (in Apulia Region, Italy, and in Ionian Islands, Greece). The aim of this study is the evaluation of the above questionnaire using a data mining technique in which the mental processing of an expert is linking with the Elman Neural Network. For this reason a proper questionnaire was used in order to classify the SMEs into three (3) classes according to their performance.