Abstract: Nowadays electronic gadgets play an important role in students' life as a source of learning. The Dependency of services provided by electronic gadgets has reached a large scale. Electronic gadgets like smart phones have a major impact on people in their day-to-day life [1]. Among all, students are the important one, as they rely on electronic gadgets for their academic activities. The major impact is that it can affect the students’ mental and physical health. Students are getting addicted to these electronic gadgets as it becomes inevitable [3]. This project work uses machine learning techniques to demonstrate how gadgets affect students' daily lives. Many parameters used to find association among use of gadgets and student academic performance.
The parameters include how many electronic devices they use and how long they use them for, whether the usage of electronic gadgets shows any improvement in their academic performance. This system applies unsupervised machine learning (ML) techniques to discover which significant attributes that a successful learner often demonstrated in an academic course. Our project main goal is to find the correlation between the use of gadgets and the student academic performance. Many research works are there related to this topic, all works purely concentrated on building static models using machine algorithms but there is no real time application useful for educational sector[4][5][6]. In our New Proposed system we focus on this issue and our system major objective is to predict the student academic ups and downs based on the use of electronic gadgets. Efficient ML algorithms will be used either Apriori algorithm or ECLAT algorithm will be used. System build as real time cum browser based application useful for college. We use efficient tools such as VISUAL STUDIO and SQL SERVER, using these Microsoft technologies we can build attractive and impressive GUI based applications.
Keywords: Data Science, Machine Learning, Educational Sector, Supervised Learning, Student Data, Training Datasets. Naïve Bayes.
|
DOI:
10.17148/IJARCCE.2024.13816
