Gender Prediction of Indian and Hungarian Students Towards ICT and Mobile Technology for the Real-Time
Chaman Verma1, Zoltán Illés2, Veronika Stoffová3

1Chaman Verma, Faculty of Informatics, Eötvös Loránd University, Budapest, Hungary.

2Zoltán Illés, Faculty of Education, Trnava University, Trnava, Slovakia.

3Veronika Stoffová, Faculty of Education, Trnava University, Trnava, Slovakia.

Manuscript received on 09 July 2019 | Revised Manuscript received on 21 July 2019 | Manuscript Published on 23 August 2019 | PP: 1260-1264 | Volume-8 Issue-9S3 August 2019 | Retrieval Number: I32780789S319/2019©BEIESP | DOI: 10.35940/ijitee.I3278.0789S319

Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (

Abstract: The present paper focused on the prediction of the university student’s gender towards Information Communication Technology (ICT) and Mobile Technology (MT) in Indian and Hungary.In this paper, four experiments were performed on dataset using three popular classifiersnamed Support Vector Machine (SVM), Artificial Neural Network (ANN) and Random Forest (RF) with three numerous testing technique such as K-fold Cross Validation (KCV), Hold Out (HO) and Leave One Out (LOO).Three different applications named Explorer, Experimenter and KnowledgeFlow (KF) of Weka 3.9.1 are used for predictive modeling. The class balancing has been also applied using Synthetic Minority Over-Sampling (SMOTE) to enhance the prediction accuracy of each algorithm. Further, a significant difference among classifier’s accuracies has also been tested using T-test at the 0.05 confidence level. Also, CPU user time has been calculated to train each model to justify to present real-time prediction of gender towards ICT and MT.The results of the study inferred that the CPU time is significantly differed in between RF (0.18 Seconds), SVM (0.06 seconds) and ANN (4.40 seconds).Also, the RF classifier (89.4%) outperformed others with LOO method in terms of accuracy.The authors recommended these predictive models to be deployed as an online prediction for the gender of the student towards ICT and MT at both universities to track technological activities.

Keywords: Gender Prediction, Machine Learning, LOO, Prediction Accuracy, HO, KCDV, SMOTE.
Scope of the Article: Machine Learning