Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/455861
Title: | Machine learning models for Predicting students performance |
Researcher: | Vinithkumar, E S |
Guide(s): | Appavu alias balamurugan, S and Sasikala, S |
Keywords: | Engineering and Technology Computer Science Computer Science Software Engineering Machine Learning Models Students Performance Prediction Clustering |
University: | Anna University |
Completed Date: | 2022 |
Abstract: | In recent scenario of Student Education, evaluating the student performance is the significant thing for the Institutions and the Universities, since it makes to develop effective models to enhance the student results. These are effectively processed by the mechanization of several features involved in the basic student characteristics and behaviours that handle large volume of data. The organizations used classification models with mining conceits to process the student records containing the student s behaviour and the connectivity between them and academic performances. Moreover, the Educational Data Mining (EDM) and learning analytics are effectively used for enhancing the quality of result classification. The educational institutions are involved in efforts to reduce the poor results of students. With that concern, many techniques are developed for evaluating the student performances for making the respective faculties to mediate to improve the overall results. For developing Accurate Student Classification Model, this work comprises of three phases of work, as, i. Cluster based Student Record Classification ii. Multi-Tier Student Performance Evaluation Model (MTSPEM) iii. Behaviour based Student Classification System (SCS-B) The first phase of work use K-Mean Clustering model for student record classification under three classes such as Low performance student, Average Student and Smart Student. For performing the classifications, the attributes such as, Grade point, number of arrears, student attendance, Medium of education and Board of education are considered. And, WEKA tool is used here for model implementation and result evaluations. In the second phase of work, traditional and ensemble classifiers are used for classification. The developed multi-tier model uses two layers as, primary and secondary classification levels, in which traditional classifier, Naive Bayes Classification, and ensemble classifiers, Boosting, Stacking and Random Forest (RF) are used, respectively. newline |
Pagination: | xvii,143p. |
URI: | http://hdl.handle.net/10603/455861 |
Appears in Departments: | Faculty of Information and Communication Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 154.15 kB | Adobe PDF | View/Open |
02_prelim pages.pdf | 2.5 MB | Adobe PDF | View/Open | |
03_content.pdf | 167.08 kB | Adobe PDF | View/Open | |
04_abstract.pdf | 195.06 kB | Adobe PDF | View/Open | |
05_chapter 1.pdf | 720.88 kB | Adobe PDF | View/Open | |
06_chapter 2.pdf | 933.04 kB | Adobe PDF | View/Open | |
07_chapter 3.pdf | 1.13 MB | Adobe PDF | View/Open | |
08_chapter 4.pdf | 1.59 MB | Adobe PDF | View/Open | |
09_chapter 5.pdf | 1.3 MB | Adobe PDF | View/Open | |
10_chapter 6.pdf | 263.5 kB | Adobe PDF | View/Open | |
11_annexures.pdf | 230.63 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 155.7 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: