Avaliações online para nivelamento e formação de classificadores humanos

Abstract

The effectiveness of teaching and learning of classification of documents in the Library Science degree courses has been a challenge because, every year, make up classes ever larger and more uneven. It has been thus a difficult task for a teacher alone manage a large number of students with different levels of knowledge and individual learning difficulties. Moreover, there is a great need for methodologies and educational technologies that support this complex context of teaching and learning. Since the classification is a knowledge that requires different skills, the process of teaching and learning requires a more detailed monitoring. This paper presents thus a methodology, supported by computer technology, to diagnose, monitor and regulate learning by individual evaluation continuous actions. For the application of this methodology was developed a online system of diagnostic and formative assessments and modeled a system of summative assessment to support the teaching-learning in classification of documents. These systems are based mainly on activities of classification. The diagnostic evaluation is performed by techniques that enable automatic cluster grouping students by similar profiles and identify the individual learning difficulties. Formative assessment is carried out by resources of Online System for Classification of Activities (SOAC), which allow the regulation of learning by both the teacher as the students themselves. The summative evaluation, for its time, will be held by the Online System of Evaluation Distance (SOAD) that automatically generate tests consist of questions designed to measure skills and identify competences. This approach aims to add more quality teaching and learning of classification in order to promote leveling of students and success of learning. The experimental results of applying our methodology in real classrooms show that is possible, even in very large classes, reduce inequalities and promote collective successes of learning. The objective of this work is to ensure better formation of human classifiers, and therefore better quality of services in area of classification of documents.