Visitors: 1383

  MAZEDAN DIGITAL LIBRARY    

  JOURNAL MANAGEMENT SYSTEM

MDL
JMS  
MAZEDAN DIGITAL LIBRARY

The development of a model of augmented ensemble learning to enhance student performance in behaviour analysis

JOURNAL:MAZEDAN INTERNATIONAL JOURNAL OF APPLIED MATHEMATICS

Download pdf
# Authors First Online DOI Downloads Citations
1. Akash A Raipure1, Yogendra Sharma2, Dyuti Banerjee3, Sadik Abdul Hamid Khan4 27 Dec 2021 NA 13 0
Abstract

Deep learning techniques classify acquired data into 1 of "N" behavioural groups. Pattern analysis and post-processing are used by these models. Their accuracy for certain behavioural classes restricts their therapeutic use. This book presents an improved ensemble learning methodology to scale student behaviour analysis. The proposed approach intelligently blends different deep learning architectures to identify student behaviour. Models include VGGNet-19, ResNet101, Inception Net, and Xception Net. VGGNet-19 and ResNet101 are used for short-range data sequences, whereas Inception Net and Xception Net are utilised for intermediate to long sequences. Long-range sequences include academic profiles, everyday activities, study routines, etc. An ensemble learning layer combines these models and evaluates student behaviour characteristics. The suggested model classifies input data with over 90% accuracy, 85% precision, and 89% recall, which is greater than typical behaviour analysis models. The "Student Life" collection includes user information, sensing data, educational information, survey data, and educational maintenance allowance (EMA) data from over 10000 students of varied ages and qualifications. The suggested augmented ensemble model is scalable and deployable because it achieves consistent performance for each entity. The suggested model beats most current models in terms of precision, recall, accuracy, and other performance criteria.

Keywords

Student, behaviour, clinical, machine learning, augmented, ensemble

References

  1. [1]   Tang, Hua et al. “Predicting green consumption behaviors of students using efficient firefly grey wolf-assisted K-nearest neighbor classifiers”. IEEE access: practical innovations, open solutions 8 (2020): 35546–35562. Web.

    [2]   Vhaduri, Sudip, Sayanton Vhaduri Dibbo, en Yugyeong Kim. “Deriving college students? phone call patterns to improve student life”. IEEE access: practical innovations, open solutions 9 (2021): 96453–96465. Web.

    [3]   Saba, Tanzila et al. “Categorizing the students? activities for automated exam proctoring using proposed deep L2-GraftNet CNN network and ASO based feature selection approach”. IEEE access: practical innovations, open solutions 9 (2021): 47639– 47656. Web.

    [4]   Alotaibi, Norah Basheer. “Cyber bullying and the expected consequences on the students? academic achievement”. IEEE access: practical innovations, open solutions 7 (2019): 153417–153431. Web.