Int. J. Adv. Sci. Eng. Vol.3 No.2 325-329 (2016) 325 ISSN 2349 5359
Vivek Anand et al
International Journal of Advanced Science and Engineering www.mahendrapublications.com
*Corresponding Author: neela.madheswari@gmail.com
Received: 20.09.2016 Accepted: 20.11.2016 Published on: 30.11.2016
ABSTRACT: Machine learning is a subfield of computer science that evolved from the study of pattern recognition and
computational learning theory in artificial intelligence. The focus of this paper is based on the student’s result, which is a data
science work and applying the concept of machine learning. Nowadays at the end of every year or semester in the schools or
colleges, the teachers have to find out all the students in their classes who are eligible for writing exams. Every time they have
to analyze a lot of data and find who is eligible and who is not. To overcome this difficulty, this work proposed on students
result prediction model in the form of web service and applying machine learning concept, find the possibility of students
about their overall performance so that they can predict the eligible students for writing the exams.
KEYWORDS: machine learning, web service, result prediction, accuracy, precision, recall
© 2015 mahendrapublications.com, All rights reserved
Students Results Prediction using Machine Learning Techniques
Vivek Anand, Saurav Kumar, A. Neela Madheswari
*
Department of Computer Science and Engineering, Mahendra Engineering College, Namakkal-637503, India
I. INTRODUCTION
Machine learning is one of the fastest growing areas of
computer science with far-reaching applications. The term
machine learning refers to the automated detection of
meaningful patterns in data. It is also widely used in
scientific applications such as Bioinformatics, medicine and
astronomy [1].
The machine learning algorithms used for Human Activity
Recognition (HAR) are: 1) Decision tree, 2) Boosted Decision
Tree (AdaBoost), 3) Random Forest, and 4) Support Vector
machine. The comparison of these four algorithms are
analysed in terms of classification accuracy percentage. It is
one of the applications of machine learning techniques [2].
Machine learning technology has been successfully applied
in many information retrieval systems, both experimental
and fielded. The instant availability of enormous amounts of
textual information on the internet and in digital libraries
has provoked a new interest in software agents that acts on
behalf of users, sifting through what is there to identify
documents that may be relevant to the user’s individual
needs. The application of machine learning techniques to
information retrieval is very common [3].
Machine learning is a branch of artificial intelligence that
employs a variety of statistical, probabilistic and
optimization techniques that allows computers to learn from
past examples and to detect hard-to-discern patterns from
large, noisy or complex datasets. This capability is well-
suited for medical applications, especially those that depend
on complex proteomic and genomic measurements. As a
result, machine learning is frequently used in cancer
diagnosis and detection. Machine learning is applied to
cancer prognosis and prediction. A broad survey of the
different types of machine learning methods are used, the
types of data are integrated and the performance of those
methods in cancer prediction and prognosis is conducted.
A number of trends are noted, including a growing
dependence on protein biomarkers and microarray data, a
strong bias towards applications in prostate and breast
cancer, and a heavy reliance on older technologies such as
artificial neural networks (ANN) instead of more recent
developed or more easily interpretable machine learning
methods [4].
The application of machine learning is applied to an
important environmental problem such as detection of oil
spills from radar images of the sea surface. Only about 10%
of oil spills originates from natural sources such as leakage
from sea beds. Much more prevalent is pollution caused
intentionally by ships that want to dispose cheaply of oil
residues in their tanks. Radar images from satellites such as
RADARSAT and ERS1 provide an opportunity for monitoring
coastal waters day and night, regardless of weather
conditions. Oil slicks are less reflective of radar than the
average ocean surface, so they appear dark in an image. An
oil slick’s shape and size vary in time depending on weather
and sea conditions. A spill usually starts out as one or two
slicks that later break up into several small slicks. Oil spill
detection requires a highly trained human operator to assess
each region in each image. It was therefore essential that the
system be readily customizable to each user’s particular
needs and circumstances. This requirement motivates the
use of machine learning. The system will be customized by
training on examples of spills and nonspills provided by the
user, and by allowing the user to control the tradeoff
between false positives and false negatives [5]. Though
machine learning techniques are applied for various fields,
this paper demonstrates the application of machine learning
technique for students’ performance prediction.
II. RELATED WORK
There are different approaches for students result
prediction. Data mining is a computational method of
processing data which is successfully applied in many areas
that aim to obtain useful knowledge from the data. Data
mining technique is classified with two types of algorithms
namely: 1) unsupervised algorithms and 2) supervised
algorithms. Classification models in supervised algorithms
are: 1) Naive Bayes (NB) algorithm, 2) Multilayer Perceptron
(MLP), and 3) J48. Tests were conducted using the following
metrics: name of the attribute, Merit, Merit deviation, Rank,
Rank and deviation. Comparison of the three algorithms are