Gunadi Harry Sulistyo

ANFIS application to infer student performance in English learning using affective factors

Teaching and learning practices are essentially meant to equip students with necessary competences. In assessing student learning, few teaching and learning practices consider the role of affective factors. Assessment relying solely on cognitive factors could impact teaching and learning effectiveness although marginal work exists exploring affective factors to enhance teaching and learning effectiveness. The current study considers using affective factors to infer student learning of English skills. Student affective factors, such as motivation, attitude, introversion, extroversion, anxiety, and self-esteem are measured using questionnaires. The data obtained is then used to model student in learning using Adaptive Neuro-Fuzzy Inference System (ANFIS) with a different set of parameters. Evaluation of the models is based on Mean Absolute Error (MAE) and Mean Squared Error (MSE). The result from the experiment shows that the model is able to infer student learning by 0.076 (MAE) and 0.008 (MSE) respectively. In addition, the proposed model that is not too general nor too complex and with fewer training iteration numbers shows the lowest error among other proposed models.

Fuzzy Inference System Based on a Model of Affective-Cognitive Criteria

Criterion-referenced assessment (CRA) uses a narrowly-defined set of skill criteria that can guide teacher to assess student grade by comparing student learning score with the criteria. However, the use of CRA is regarded as incomplete as most of the criteria in general based on student skill of knowledge. Affective factors also need to be considered in the assessment. Nonetheless, measuring affective factors is not as straightforward a task as measuring cognitive factors because the quantification is often represented in linguistic terms. In this study, affective factors and cognitive factors based on CRA are modeled together for a combined assessment. A questionnaire to measure student affective attributes is developed. A novel method of fuzzy inference system (FIS) to infer student achievement in English learning based on CRA is proposed. The method was applied to an analysis of data collected from students studying English as a second language. The result indicates the usefulness of the FIS based on CRA as a basis to assess student performances by using affective and cognitive factors.