ISL ENGINEERING COLLEGE
Approved by AICTE, recognized by the Govt. of TS, and affiliated to
OU Bandlaguda,Hyderabad – 500 005
Department of AI&DS
A REVIEW ON DATA MINING AND MACHINE
LEARNING METHODS FOR STUDENT SCHOLARSHIP
PREDICTION
Presented by :
Mohammed Safiullah Hussain(160521747001)
Under the Guidance of: Syed Abdul Qadeer(160521747005)
Mr. Mazhar(Head of Mohammed Mustafa Ahmed(160521747024)
Deoartment of
AI&DS)
CONTENT
• ABSTRACT • RESULT AND
ANALYSIS
• INTRODUCTION
• CONCLUSION
• LITERATURE SURVEY
• EXISTING SYSTEM
• PROPSED SYSTEM
• UML DIAGRAMS
• DATA FLOW
DIAGRAM
• CLASS DIAGRAM
ABSTRACT
• This presentation consists of literature survey to prediction of scholarship by using
Machine Learning and Data Mining technique. Along with this it contains a small
description of ML/DM which are used by the researchers. It also describes data sets as
very important in ML/DM methods. The different types of ML/DM algorithms are
addressed by using all this technique. The algorithms which give more accuracy results
in detection of continuity of every student’s scholarship such as Naïve Bayes, Decision
Tree and k-NN. Finally, the proposed model will provide a list of candidates, who
deserve to have a scholarship and also discussion has been made on accuracy of each
techniques which was used to get a result.
INTRODUCTION
• This presentation explores various Machine Learning and Data Mining techniques used
for predicting scholarship outcomes. It discusses the advantages and disadvantages of
these methods and identifies the most effective approach. Scholarships offer financial
support for students' education, boosting their confidence and enabling them to pursue
their goals. Different types of scholarships exist, such as merit-based and career-specific
ones, with eligibility criteria based on factors like GPA, performance, and income. The
paper focuses on using ML/DM methods to automatically predict eligible students
using constraints like percentage, GPA, marks, income, and communication skills.
LITERATURE
• SURVEY
Helping Students Get More than Their Money from an Engineering
Scholarship Program
• Characteristics of engineering faculty engaged in the scholarship of teaching
and learning.
• Data mining: Prediction for performance improvement of graduate students
using classification
• Artificial intelligence-based scholarship and credit pre-assessment system
• A comparative analysis of techniques for predicting academic performance
EXISTING SYSTEM
1.Scholarships provide financial assistance to students for their education and are
typically offered by government or non-government organizations, recognizing
students' achievements and boosting their confidence to pursue their goals.
2.Common types of scholarships include merit-based, student-specific, and
career-specific scholarships, with different eligibility criteria and schemes
provided to students.
3.The existing system utilizes the Random Forest algorithm for scholarship
prediction but faces limitations in terms of the number of parameters used for
prediction, suggesting a potential area for improvement.
DISADVANATGES OF EXISTING SYSTEM
• In the existing process we have implemented only machine learning
algorithms.
• In machine learning we implemented regression techniques only.
• For this concept we will get less accuracy and more complexity with
regression techniques.
.
PROPOSED
SYSTEM
• The proposed system integrates ML and DL algorithms for effective scholarship
prediction, utilizing various ML algorithms and training data to predict unknown
factors.
• Scholarships are vital for students' financial support and personal motivation, with
communication between teachers, industry professionals, and students playing a
key role. The proposed system helps upper-class engineering students develop
skills, knowledge, and leadership qualities.
• Scholarships have a significant impact on students' leadership skills, future
paths, and university experiences. The proposed system streamlines scholarship
processes by generating predictions using fewer parameters.
ADVANTAGES OF PROPOSED SYSTEM
• We have implemented both ML and DM algorithms in this method to
perform better accuracy.
• Naviebyesgives good accuracy, flexibility .
• Using ML and DM techniques the prediction of scholarship
much easier and effective also.
Modelling process
Data Flow Diagram
•Use case
Diagram (User)
Use case
Diagram (Server)
CLASS DIAGRAM
SEQUENCE DIAGRAM:
RESULT AND
ANALYSIS Home Page
USER REGISTER:
Admin Login Form
Activate users:
User Login Form
View Dataset
Naïve bayes results:
CONCLUSION
• By using ML and DM techniques the prediction of scholarship much easier and
effective also. While predicting the data we need to look upon many parameters.
ML and DM techniques unable to work without training data.
• For collecting of datasets makes more difficult. We cannot be able to collect the
best algorithm suitable for prediction. Because prediction is based on training data
values: By using the ML and DM methods, the instructions can easily track on
the scholarships and the problems faced by the recipients in their studies.
THANK YOU
