Open Educational Resources (OER) repositories stores a large amount and variety of data from multiple sources, and these presents relevant information to the educational learning process; but nowadays, thinking computing takes great importance, and that's why a search system of specialized educational resources in this area is of great need. This work presents the use of Data Mining in OER repositories, for the construction of a multi-label search platform. The process of extracting information is based on Web Scraping techniques, and the use of a multi-label classifier based on Multi Layer Perceptron. This work contributes to a search of OERs based on purposes learning, using as a case study the area of thinking computing.