Guidance System for Scrutinizing the Students Performance using Random Forest Classifier
P. Satya Shekar Varma1, P. Shyam Sunder2, Koppula Sri Vasuki Reddy3

1P. Satya Shekar Varma*, Department of Computer Science and Engineering at Mahatma Gandhi Institute of Technology, Hyderabad, India.
2P. Shyam Sundar, Department of Computer Science and Engineering at Mahatma Gandhi Institute of Technology, Hyderabad, India.
3Koppula Sri Vasuki Reddy, Department of Computer Science and Engineering at Mahatma Gandhi Institute of Technology, Hyderabad, India.
Manuscript received on March 15, 2020. | Revised Manuscript received on March 31, 2020. | Manuscript published on April 10, 2020. | PP: 980-985 | Volume-9 Issue-6, April 2020. | Retrieval Number: F4063049620/2020©BEIESP | DOI: 10.35940/ijitee.F4063.049620
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: For today’s education leaders, the ongoing challenge is to assess the student’s academic performance that could probably affect their organization’s potential. So, the emergence of Educational Data Mining [EDM] became the solution. By utilizing the data mining methods, infused with a theory of understanding the application and elucidation of the education and learning experience, EDM practitioners are able to generate training models that interpret the results and spot the students who may show poor performance so as to help tutors to offer effective learning environment. This paper proposes a guidance system which aims to analyze student’s demographic data, academic details and extract all possible knowledge through surveys from students, parents and teachers with regard to latter state to configure whether the student is on the proper course of achieving the goals using the random forest classification algorithm. This model pursues highest possible accuracy comparison to the other previously related models proposed by authors. Furthermore, Anaconda3 data mining tool is used to develop this model which flourishes to draw the attention towards the pupils functioning based on their interests. In this study, we have accumulated the records of 480 students with 16 attributes. After contemplating all records of factors considered earlier for forecasting student’s academic participation, we cull the most consistent featured based on their hypothesis and association with the performance.
Keywords: Anaconda, Classification, Component, Decision Tree, EDM, Guidance System, Prediction, Random Forest.
Scope of the Article: Classification