Predicting Assignment Submissions in a Multiclass Classification Problem
Journal Title: TEM JOURNAL - Year 2015, Vol 4, Issue 3
Abstract
Predicting student failure is an important task that can empower educators to counteract the factors that affect student performance. In this paper, a part of the bigger problem of predicting student failure is addressed: predicting the students that do not complete their assignment tasks. For solving this problem, real data collected by our university’s educational platform was used. Because the problem consisted of predicting one of three possible classes (multi-class classification), the appropriate algorithms and methods were selected. Several experiments were carried out to find the best approach for this prediction problem and the used data set. An approach of time segmentation is proposed in order to facilitate the prediction from early on. Methods that address the problems of high dimensionality and imbalanced data were also evaluated. The outcome of each approach is shown and compared in order to select the best performing classification algorithm for the problem at hand.
Authors and Affiliations
Bogdan Drăgulescu, Marian Bucos, Radu Vasiu
Keywords
Related Articles
- EP ID EP163823
- DOI -
- Views 176
- Downloads 0
System for Detection of Network Threats based on Classifiers
In this paper we present a system that automatically detects and profiles threats on a real network. The realised Threat Detection System (TDS)is based on Snort software and it allows the security experts to evalua...
Self-Tuning Control of Linear Systems Followed by Deadzones
The aim of the present paper is to increase the efficiency of self-tuning generalized minimum variance (GMV) control of linear time-invariant (LTI) systems followed by deadzone nonlinearities. An approach, based on...
Drying Kinetics of Sunflower Seeds using Pulsed UHF Energy Intake
In this paper the drying kinetics of sunflower seeds using pulsed UHF energy intake are given. It is shown and argued that the application of pulsed internal heat source made possible the reduction of heat treatmen...
An Image Compression Method for Fleck Board-3
The recent availability of inexpensive hardware, such as CMOS and CCD cameras, has enabled the new research field of wireless multimedia sensor networks. This is a network of interconnected devices, capable of retrieving...
Study of Power Loss Reduction in SEPR Converters for Induction Heating through Implementation of SiC Based Semiconductor Switches
This paper presents a power loss analysis for a Single Ended Parallel Resonance (SEPR) Converter used for induction heating. The analysis includes a comparison of the losses in the electronic switch when the circui...