Addis Ababa Science and Technology University

College of Electrical and Mechanical Engineering

Department of computer engineering

Thesis proposal
Title: - Investigation and Prediction of Students’ Performance Using Deep Neural Network

By

Belayneh Bariga Abetu

Principal Advisor

Dr. Asrat .M

1
Table of Contents
1. Introduction................................................................................................................................1

1.2 Background study......................................................................................................................1

2. Literature review.........................................................................................................................2

3. Statement of the Problem............................................................................................................3

4. Objectives of the study ................................................................................................................3

4.1 General objective:- ....................................................................................................................3

4.2 Specific objective:- ....................................................................................................................3
5. Research Question ......................................................................................................................4

6. Methodology ..............................................................................................................................4

7. Scope of the study.......................................................................................................................5

8. Limitation of the study.................................................................................................................5

9. Significance of the study ..............................................................................................................5

10. Contribution ...............................................................................................................................5

11. Financial plan..............................................................................................................................6

12. Time plan....................................................................................................................................6

13. References..................................................................................................................................7

i
1. Introduction
Machine learning approaches have been extensively used to model student behavior and
performance in computer programming activities. An Artificial Neural Network (ANN) is an
arithmetical model that is motivated by the organization and/or functional feature of biological
neural networks. A neural network contains an interrelated set of artificial neurons, and it
processes information using a connectionist form to computation. As a general rule an ANN is an
adaptive system that adjusts its structure based on external or internal information that runs
through the network during the learning process. For happening from short time the neural
networks are non-linear numerical data modeling tools. It is usually used to model intricate
relationships among inputs and outputs or to uncover patterns in data. ANN has been applied in
numerous applications with considerable attainment [1]. However, Deep Learning techniques
aim to learn attribute hierarchies with attribute from higher levels that is formed by the
combining of other low features. This includes learning multiple methods for higher and deeper
architectures. DNN is a class of multiple NN models. It has Model with input layers, arbitrary
number of hidden layers and an output layer. The layers are made up of neurons which share
similarities to human brain neurons.

1.2 Background study

The quality of students enrolled into any university influences the research and training level
within the university, and furthermore, has an overall effect on the advances of the country itself,
as these students may become an influence in the affairs of the country in all sorts of the
economy. AASTU University, students get admitted to the Department of Mechanical and
Electrical Engineering in the Faculty of Engineering Technology after they successfully passed
the scientific branch of the high school with at least to enjoy some amount of percent, students
are required to study their freshman year without getting a major in engineering yet. Once the
student finished the freshman year, he or she can major in Computer Engineering, Power
Engineering and Communication Engineering or Control Engineering or Mechanical
engineering. A student can major in one of the Engineering major if he or she satisfy specific
requirements such as High school score, number of credits finished, pass some subjects in the
freshman year, such as Math, English, statics drawing etc. And so, We will be used the

1
 classification power of a neural network to predict the potential students with problems in
continuing their education and, It should be noted that this module of predicting student
performance will help in identifying which student possibly will succeed in study Engineering
programs. Hence this study takes an Engineering approach to make the process a student
selecting an Engineering major more effective and efficient. Specifically the study seeks to
explore the possibility of using a Deep Neural Network model to predict the performance of a
student before he/she starts his/her Studying in the next year in Engineering studies.

2. Literature review
A number of studies addressed the analysis of educational data to get useful information that
affect learning quality. Neural networks were used by many researchers for predictions of
students’ results.

In [3], discuss and applied a hybrid technique on K-Means Clustering and Artificial Neural
Network for students who are pursuing higher education while adopting a new foreign language
as a means of instruction and communication. Firstly, Neural Network was used to predict the
student's performance and then fitting them in a particular cluster which was form using the K -
Means algorithm. In [4], proposed a prediction model for students’ performance based on data
mining methods with some few features called student’s behavioral features. The model was
evaluated in three different classifiers; Naïve Bayesian, Artificial Neural Network and Decision
tree. Random Forest, Bagging and Boosting were used as ensemble methods to improve the
classifier’s performance. In [5], discuss to an Artificial Neural Network (ANN) classifier to
predict the performance of students in Mathematics. From their experiments they found that the
modified spectral Perry trained artificial neural network performs better classification compared
to other classifiers. [6] In their study predicted the performance of students in Mathematics using
an Artificial Neural Network (ANN) classifier. They found that the modified spectral Perry
trained artificial neural network performs better classification compared to other classifiers in
this context. [7] used Artificial Neural Network and linear regression models to predict student
performance after access to higher education. Data received from the Tshwane University of
Technology was utilized for the study. The total Average Point Scores (APS) students obtained
in grade 12 was employed as input variable. The results indicated a better agreement between
ANN model prediction and observed values compared to those in the linear regression.

2
 The good results of applying neural networks in classification problems lead us to use them for
predicting students’ results in higher education. But so, in this study by using a deep neural
network to describe and investigate and to predict the performance of the AASTU in the required
data set of the student.

3. Statement of the Problem

In the current situation educational systems, student performance prediction is getting worsen
day by day. Predicting or Evaluating student performance in advance can help students and their
teacher to keep track of progress of a student. Many institutes have adopted continuous
evaluation system today. Such systems are beneficial to the students in improving performance
of a student. The purpose of continuous evaluation system is to help regular students. So, a
significant problem in higher education is the poor results of students after admission. Many
students leave universities from a variety of reasons one of the reason as poor background
knowledge in the field of study, very low grades and the incapacity of passing an examination,
lack of financial resources and the adaptation of the environment. Therefore, evaluating or
predicting students’ results is an important problem for the management of the universities who
want to avoid the phenomenon of early school leaving.

4. Objectives of the study

4.1 General objective:-

 Studying and predicting the performance of AAST student using deep neural network. And to
determine students who would be expected do well in the Faculty of Electrical and Mechanical
Engineering Technology in AAST University.

4.2 Specific objective:-

 To examine the use of Deep Neural Network in predicting and evaluating students’ academic
performance.
 To examine the mode of operation of Deep Neural Network.
 To study and develop the deep neural network model from the data sets aside for training.
 To predict academic performance of the student using the test data set aside.

3
5. Research Question
In order to address the purpose of this thesis or research, the study will be try to finding to
answer the following questions:-

1, how can the deep neural network will be modelled from the AAST student data? And
2, how can a deep neural network be used for prediction of AAST student academic
performance?

6. Methodology
By looking deeply through the literature and ask for the experience of human experts on student
performance, a number of factors that are considered to have an effect on the performance of an
Engineering student will be outlined. These factors will cautiously to study and synchronize into
a convenient number appropriate for computer coding within the environment of the DNN
modeling. These factors to classify as input variables. The output variables are to visible form to
some likely levels of performance of a student in terms of the university of Addis Ababa
Science and Technology University grading system.
The methods it will be used in this study are:-
Data Collection:- In this study, will be use the data of four academic batches of the university
of Addis Ababa science and technology, Ethiopia, which is necessary to altogether of
undergraduate students enroll in the academic batches of at least 7 years data.
Dataset:-The data used in this study will be provided or finding by the Addis Ababa Science and
Technology University.
Document analysis:-analyzed recorded documents from AASTU registrar office and
administration office.
Model preparation: - Design the block diagram of the system.
Algorithm design: design the algorithm for this study.
Simulation & testing:-The designed model should be simulated by the aid of software.
Testing: In testing the algorithm will be fed with new records from the test set, for which no
information about the classes is available.
Evaluation: Evaluation will be done to determine how well the classification algorithm is
performing.

4
 Documentation: finally the results find from the simulation and testing would be documented.

7. Scope of the study

This study will cover the mode of operation of a Deep Neural Network and how it can be used to
predict student academic performance.

8. Limitation of the study

The limitation in this study are insufficient information sourcing for the relevant materials,
literature or information and in the process of data collection (internet) and also time constraints.

9. Significance of the study

This study will educate on the design and investigation of deep Artificial Neural network. It will
also educate on how Deep Neural Network can be used in predicting students’ academic
performance. This study will also serve as a resource base to other scholars and also to
identifying some suitable factors that affect a student performance, and so to model an a Deep
Neural Network that can be used to predict a student performance based on some predetermine
data for a given student.

10. Contribution
Findings of the study will contribute to the field of computer engineering in the following way:
1, Provide a process to design and create a prediction model of a deep neural network that
predicts academic performance for the AASTU electrical and mechanical engineering student.
And.

2, we will be proposed a feedback model to calculate the student’s knowledge for particular
course domain and provide feedback if the student needs to put more effort in that course based
on the predicting GPA.

5
11. Financial plan
The table below shows the equipment’s or material that is need for this studies and their costs.

Material Cost in birr

Data collection and material resources 17,000
8GB RAM and 4Intel cores computer(optional)
Software 800
Hard disc ,flash memory 6000
Printing, laminating and duplications 700
Total cost 24,500

12. Time plan

This is the proposed time schedule to finish the study. And it is the interests of me to work on
this schedule.

6
13. References

[1] P. K. a. S. Chitnis, "Handwritten Devanagari Character Recognition using Artificial Neural

Network," Journal of Artificial Intelligence, vol. 4, no. 1, 2011.

[2] e. a. Livieris, "Predicting students' performance using artificial neural networks 8th
PanHellenic Conference with International Participation Information and Communication
Technologies," in PanHellenic Conference , 2012.

[3] C. K. M. a. Z. W. Moucary, "Improving student’s performance using data clustering and

neural networks in foreign- language based higher education," The Research Bulletin of
Jordan ACM, vol. 2, no. 3, pp. 27-34, 2011.

[4] E. H. T. a. A. I. Amrieh, "Mining Educational Data to Predict Student’s academic

Performance using Ensemble Methods," International Journal of Database Theory and
Application, vol. 9, no. 8, pp. 119-136, 2016.

[5] e. a. Livieris, "Predicting Students’ Performance using Artificial Neural Networks," in 8th
PanHellenic Conference with International Participation Information and Communication
Technologies, 2012,312-328.

[6] a. A. O. G. Kanakana1, "Predicting student performance in Engineering Education using an

artificial neural network at Tshwane university of technology," in ISEM Proceedings, (2011)
September21 -23, Stellenbosch, South Africa. , 2011.

[7] B. a. D. Prabu, "Students' Performance Prediction Using Deep Neural Network,"

International Journal of Applied Engineering Research ISSN 0973-4562 , vol. 13, no. 2, pp.
1171-1176, 2018.

[8] A. a. A. Benyettou, "Using Probabilistic Neural Networks for Handwritten Digit

Recognition," Journal of Artificial Intelligence, vol. 4, no. 4, 2011.