A support vector machine is a supervised learning algorithm mainly used to classify and regression problems. It works by mapping input features with output in a given dimension space to segregate the data points in a suitable class. Self-efficacy plays a critical and crucial role in an individual’s ability to face challenges and achieve goals in case of difficulties in life. In the paper, six factors responsible for encouraging or deterring the learners’ self-efficacy are studied. A multiclass strategy classifying self-efficacy is measured by introducing One-versus-One (OVO) and One-vs-Rest (OVR) using SVM to derive the self-efficacy of a learner. The results show that OVO and OVR derive an accuracy of 88% and 76%, respectively, stating that the OVO strategy shows more accuracy than OVR in finding learner’s self-efficacy. Further, the paper also discusses that OVO is the most suitable strategy for predicting the self-efficacy class label for a new learner.