Data mining techniques for employability: Systematic literature review

The professional integration of young university graduates remains the main concern of all universities. As a result, these universities make enormous efforts to prepare laureates adapted to the job market. Therefore, it is considered appropriate to know the profiles of students employable in the labour market. Several models have been implemented to predict profiles acceptable by employers. In this paper, we present a Systematic Literature Review (SLR), from 2005 to 2019, about data mining (DM) techniques used to analyze and predict the professional integration of students and young graduates. We used several scientific databases and repositories: Scopus, WoS, IEEE, Springer, ScienceDirect and ACM to extract and analyze 157 references, with Zotero and NVIVO. After a meta-analyse, we focus on works that use and compare DM techniques. As a result, we found that the most famous DM techniques in our context are: SVM, Naïve Bayes, PCA, Logistic Regression, k-means algorithm, KNN, decision tree, Neural networks, text mining and Item Response Theory. From this SLR and according to the accuracy, we conclude that logistic regression, decision trees, ANN, Random Forest, Text-IRT are the best DM techniques for employability studies.