Natural Turkish language processing based method for classifying R&D projects

Purpose: Classification procedures were made using text mining techniques and methods, and algorithms used and accuracy rates were compared in order to assign R&D projects applications to appropriate reviewer fields of activity. Theory and Methods: Convolutional Neural Network (CNN) models were tried to be created in order to classify the features by using the word representation method and automatic feature learning approach, which is one of the natural language techniques "Word2Vec".Results: The best results were obtained in the experimental studies conducted on the data set of R&D projects and articles with R&D project content selected from the TUBITAK Dergipark website. Conclusion: It has been ensured to eliminate the problems of reviewer assignments in the selection of R&D projects and to find the best project classification method. With the information obtained, the input of the decision support system to be created in the future was tried to be created.