Thair Hamtini

Preprocessing and analyzing educational data set using X-API for improving student's performance

Educational data mining concerns of developing methods to discover hidden patterns from educational data. The quality of data mining techniques depends on the collected data and features. In this paper, we proposed a new student performance model with a new category of features, which called behavioral features. This type of features is related to the learner interactivity with e-learning system. We collect the data from an e-Learning system called Kalboard 360 using Experience API Web service (XAPI). After that, we use some data mining techniques such as Artificial Neural Network, Naïve Bayesian, and Decision Tree classifiers to evaluate the impact of such features on students academic performance. The results reveal that there is a strong relationship between learner behaviors and its academic achievement. Results with different classification methods using behavioral features achieved up to 29% improvement in the classification accuracy compared to the same data set when removing such features.

A proposed dynamic technique for detecting learning style using literature based approach

This paper describes a dynamic technique for identifying learners learning styles based on their behavior in the learning environment influenced by literature approach. The technique was suggested based on VAK Learning Styles model and the behavioral patterns. First we defined the learning material features (contents, case studies, examples, exercises and assessments) and the behavioral patterns (staying time, visits, and answers); then we connected these features and patterns with the VAK learning styles determining the effect of each learning style on each pattern. Then, we applied some general rules and algorithms to estimate the learning style. We prepared a pilot e-learning environment to test and validate the proposed technique, the estimated learning styles from the detection module in the learning environment were compared to the results of a VAK questionnaire; which learners answered before entering the learning environment, the behavior patterns of eighteen participants in the pilot environment were monitored in order to estimate the learning style, eight were detected correctly, four balanced and seven in-correctly, the accuracy of the system was measured to about 52.78 %.

Tutors and university teachers' perception about quality assurance in distance education: The case of the undergraduate management course in Minas Gerais - Brazil

This paper aims to describe the perception of tutors and university teachers about quality assurance in a distance undergraduate management course in Minas Gerais Brazil. The results provided a useful amount of information to improve the quality of the course and even the platform used.

Quality Assurance in Distance Education in Brazil

Distance Education constitutes one of the education fields that are evolving rapidly around the world. So, it is possible to consider it as an important instrument to reach youngsters and adults whom learning needs were not satisfactorily met by the traditional educational system. Distance Education has a great potential in several levels and niches in the society. This educational methodology has shown a new paradigm that permits access to much more people at the universities, as well as the improvements of the qualitative level of the professors who haven’t much time to upgrade themselves, mainly in Brazil where most primary and high school teachers dwell in more than one school to complement their wages which are very low.