A decision support system for career recommendation

In the last years, the increasing number of artificial intelligence applications have been continuously invading our daily lives. This thesis, as a result of the introduction of machine learning approaches to the career planning domain, has been undertaken in order to develop a recommender system that counsels and proposes a work industry to university graduates. A system based on machine learning algorithms that recommends to new graduates an industry to work at, based on the education history, grades and personal information of previous graduates is designed in this study. The Cross Industry Standard Process for Data Mining (CRISP-DM), which is one of the most common data mining processes, is employed after reviewing the characteristics of the problem at hand. The six steps of CRISP-DM, namely understanding the business, understanding the data, preparing the data, modelling, evaluation and setting out, have guided the research methodology. In the modeling phase KNN, Random Forest, Naive Bayes, Support Vector Machines and Decision Tree machine learning algorithms have been utilized. In order to answer the research questions set by this thesis, a case study based on the data collected by Hacettepe University Department of Industrial Engineering and Hacettepe University Student Affairs Office (ÖİDB) has been designed and executed. At the end of the research, the accuracy of supervised machine learning algorithms has been examined with the use of a confusion matrix, and the best compared result has been obtained from Random Forest (with a 67,46% accuracy).