Factor Scoring and Machine Learning algorithm to Predict Student Counselling
Nusrat Jahan1, Saiful Islam2, Rezwana Sultana3
1Nusrat Jahan*, Computer Science and Engineering, Daffodil International University, Dhaka, Bangladesh, India.
2Saiful Islam, Computer Science and Engineering, Daffodil International University, Dhaka, Bangladesh, India.
3Rezwana Sultana, Computer Science and Engineering, Daffodil International University, Dhaka, Bangladesh, India.
Manuscript received on September 02, 2019. | Revised Manuscript received on September 27, 2019. | Manuscript published on October 30, 2019. | PP: 243-248 | Volume-9 Issue-1, October 2019 | Retrieval Number: A1131109119/2019©BEIESP | DOI: 10.35940/ijeat.A1131.109119
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Personal realization is one of the best things for a successful life. Sometimes, one needs help to realize about bad habits, career goals and accomplish mental health as well as to overcome other problems. This help is generally known as “Counselling”. To ensure effectiveness of counselling service, prime concern is to find out the target group of instances. Many researchers worked with student performance prediction based on academic attributes moreover students’ counselling is also needed to increase their performance. We addressed this issue for this paper work. Here, a model is proposed to predict a student who needs counselling. This study was mainly motivated by two main steps. The first was to investigate university students who feels an urge about having counselling for psychological help from their circumstances and second was to predict efficiently which group of students really needs counselling. This paper work was established with 498 instances and each comprised of 6 attributes. In the case of evaluate the result, paper shows superiority over state- of-the-art methods to predict student counselling through machine learning and factor scoring method. We applied 10 fold cross-validation and 66% dataset splits evaluation method to find out better algorithm among selected 5 algorithms which are Ibk, Naive Bayes, Multilayer, SMO and Random Forest. Weka 3.8.0 have been used for machine learning algorithms where Ibk (Instance Based Learning) was found best for our approach with 95.38% accuracy.
Keywords: Student Counselling, Factor Scoring, Machine Learning, Ibk, Weka.