ISSN (Online): 2583 648X

IJIRID
International Journal of Ingenious Research, Invention and Development
Volume 1 | Issue 3 | April 2023
Scientific Journal Impact Factor (SJIF 2023): 3.647
DOI: 10.5281/zenodo.7970771
www.ijirid.in

An Automated College Prediction Model Using Machine Learning

Prof. R. L. Pardhi1, Yash Patne2, Meenal Shekokar3, Ajit Thakare4, Maitrideep Popatkar5, Shrutika Bijawe6
1Assistant Professor, Sipna College of Engineering and Technology, Amravati, Maharashtra, India
2, 3, 4, 5, 6UndergraduateStudent, Sipna College of Engineering and Technology, Amravati, Maharashtra, India

Abstract: The Maharashtra State Government administers the MHT-CET, also known as the
Common Entrance Test, every year. The Directorate of Technical Education oversees its operation.
This entrance exam focuses primarily on the degree programme offered by the following streams:
engineering, pharmacy, law, and management. In 2019, the test is administered in an online format
for the first time. The syllabus and scoring system for the computer-based test (CBT) were made
public by the State Common Entrance Test Cell, Maharashtra. In this project, we suggest creating
a machine learning model that will be able to forecast the college based on the MHT-CET score. We
utilized a dataset we made for this. Name, MHT-CET score, application college, branch, and
assigned college are all taken into consideration. Based on the outcomes of the college entrance
exams, the prediction results demonstrate that the support Nave Bayes Classifier can predict
college course performance with accuracy. Then, by providing performance indicators to the model,
we anticipate the college based on the score from prior years. Self-driving cars and chatbots are
just a couple of the new uses that machine learning (ML) has made possible. ML can process
enormous amounts of data to draw conclusions from the data. These insights can be used to inform
decisions. One application of machine learning is prediction. Predictive analytics uses data,
statistics, and machine learning algorithms to make predictions based on historical data. Beyond
understanding what has already happened, the goal is to better predict what will happen in the
future.

Keywords: College Prediction Process, Machine Learning, Dataset, Decision-Making, Naïve Bayes
Classifier, Predictions, Past Data, Predictive Analytics, Future.

I. INTRODUCTION
In recent years, the industry has experienced a sharp rise in competition. Problems brought on by increased
competitiveness include job loss and a strong need for new skills. Students should enroll and pursue their
education at the institution that best suits them in this competitive atmosphere. This will assist students in
developing their abilities to meet industry standards and guarantee appropriate placement. So, a student's
future is greatly influenced by their ability to get into the top colleges. Technology improvements and data
analysis have significantly changed the educational landscape in recent years. Academic performance is a staged
evaluation of the student's learning condition and a fundamental criterion for assessing the quality of the
teacher's instruction. Learning is the responsibility of every student.
More and more educational academics are beginning to concentrate on the analysis and mining of educational
data as a result of the rapid development of data science. In order to enhance educational results, researchers
have resorted to machine learning and other data-driven methodologies as digital data and tools for data
analysis become more widely available. The MHT-CET college prediction system is a platform that makes it easier
for students to obtain advice about which colleges to apply to.

Content from this work may be used under the term of the Creative Commons Attribution-Non-commercial (CC BY-NC) 4.0 licence. This license
allows refusers to distribute, remix, adapt, and build upon the material in any medium or format for non-
commercial purposes only, and only so long as attribution is given to the creator. Any further distribution of this
work must maintain attribution to the creators. 37
 ISSN (Online): 2583 648X

IJIRID
International Journal of Ingenious Research, Invention and Development
Volume 1 | Issue 3 | April 2023
Scientific Journal Impact Factor (SJIF 2023): 3.647
DOI: 10.5281/zenodo.7970771
www.ijirid.in

The method predicts the various courses provided by various colleges based on the results students received on
the MHT-CET exam. Both students and colleges can utilize the system easily thanks to its user-friendly layout.
Based on their MH-CET scores and rank, students can register on the portal and apply to numerous colleges and
courses. Based on the students' MHT-CET scores and rank, the system utilizes an algorithm to analyze their
applications automatically and determine which colleges they should attend. The MHT-CET-based college
prediction method is an effective, open, and equitable process that guarantees students are able to enroll in the
courses and schools of their choice.

II. LITERATURE REVIEW

A brief overview of existing work in various papers, which have referred for implementation.

Mishra, Sushruta, Soumya Sahoo, Brojo Kishore Mishra, and Santosh Satapathy, "A quality based automated
admission system for educational domain" [1]. The authors of this work conducted research from a university
perspective to forecast the possibility that a student will enroll in the university after inquiring about courses
there. They utilised the K-Means algorithm to group pupils based on feedback, family income, family occupation,
parent education, parent motivation, etc. to determine whether or not the student would enroll in college. The
pupils were divided into clusters based on how similar their traits were, and decisions were made. The model's
goal was to increase the number of students enrolled in the university.

Pushpa, S. K., T. N. Manjunath, T. V. Mrunal, Amartya Singh, and C. Suhas, "Class result prediction using machine
learning" [2]. In this study, machine learning is used to predict a class's outcome. Before taking the final test, a
student's performance from the previous semester is taken into account, along with the results of internal exams
from the current semester, to determine whether the student will pass or fail. To calculate the outcome, the
author used SVM, Naive Bayes, Random Forest Classifier, and Gradient Boosting. A variety of learning algorithms
are combined in the ensemble learning algorithm known as "boost" to improve prediction performance.

Jamison, Joseph, "Applying machine learning to predict Davidson College's admissions yield" [3]. In this study,
machine learning techniques were used to estimate the yield of college admission. The rate at which students
who have been admitted by the institution actually enroll in the course is known as the yield rate. In order to
develop the model, many machines learning techniques, including Random Forest, Logistic Regression, and SVM,
were utilised. The models were compared based on their performance and accuracy, and Random Forest
performed better than the other models with 86% accuracy. Also underlined were the elements that have been
shown to be crucial in foretelling successful applications.

2021 JETIR, “Machine Learning Model for Prediction of Post Graduate Admissions”, July 2021, Volume 8, Issue 7
[4]. This essay aims to assist students in narrowing down universities based on their personal profiles. The
anticipated results offer them a good indication of their prospects of admission to a particular university.
Students who are now preparing for their higher education should also benefit from this analysis. In the current
system, students must speak with consultants using their MH-CET scores in order to receive advice and reach a
decision. In the second step, information was gathered from the internet and universities were chosen using an
online search procedure. A student's entire career may be destroyed if they choose the incorrect path. This
programme uses machine learning techniques to forecast college names and anticipate changes to predictions.
The student completes the MH-CET form by providing information about his schooling and test results. The
application then uses trained machine learning models to check the data and makes a prediction. The system

Content from this work may be used under the term of the Creative Commons Attribution-Non-commercial (CC BY-NC) 4.0 licence. This license
allows refusers to distribute, remix, adapt, and build upon the material in any medium or format for non-
commercial purposes only, and only so long as attribution is given to the creator. Any further distribution of this
work must maintain attribution to the creators. 38
 ISSN (Online): 2583 648X

IJIRID
International Journal of Ingenious Research, Invention and Development
Volume 1 | Issue 3 | April 2023
Scientific Journal Impact Factor (SJIF 2023): 3.647
DOI: 10.5281/zenodo.7970771
www.ijirid.in

then shows the student's prospects of admission to a university based on his test results, educational
background, and employment history, if any.

Mohana Bangale, Shubham Bavane, Akshay Gunjal, Rohit Dandhare, Sudhir D. Salunkhe, “A Survey on Placement
prediction system using machine learning” [5]. To overcome the issues with the current manual system in use,
"College placement Prediction using Machine Learning" is developed in this work. The difficulties that the
current system faces are supported by this programme, which can sometimes even lessen them. Additionally,
this system is created to meet a business's desire for efficient and seamless functioning. To fill their roles, the
majority of businesses are concentrating on university recruitment. Before students finish their studies, the
corporations find talented and skilled professionals. The best technique to work with the correct resource at the
right moment to land good firms at the start of their careers is through this method.

III. IMPLEMENTATION
Naïve Bayes Classifier
The Nave Bayes algorithm is a supervised learning method for classification issues that is based on the Bayes
theorem. It is mostly employed in text categorization with a large training set. The Naive Bayes Classifier is one
of the most straightforward and efficient classification algorithms available today. It aids in the development of
quick machine learning models capable of making accurate predictions. Being a probabilistic classifier, it makes
predictions based on the likelihood that an object will occur. Spam filtration, Sentimental analysis, and article
classification are a few examples of Naive Bayes algorithms that are frequently used.

The Naïve Bayes algorithm is comprised of two words Naïve and Bayes, which can be described as:
Naïve: It is called Naïve because it assumes that the occurrence of a certain feature is independent of the
occurrence of other features. Such as if the fruit is identified on the bases of color, shape, and taste, then red,
spherical, and sweet fruit is recognized as an apple. Hence each feature individually contributes to identify that
it is an apple without depending on each other.
Bayes: It is called Bayes because it depends on the principle of Bayes' Theorem.

Bayes Theorem:
Bayes' theorem is also known as Bayes' Rule or Bayes' law, which is used to determine the probability of a
hypothesis with prior knowledge. It depends on the conditional probability.

The formula for Bayes' theorem is given as:

Where,
P(A|B) is Posterior probability: Probability of hypothesis A on the observed event B.
P(B|A) is Likelihood probability: Probability of the evidence given that the probability of a hypothesis is true.
P(A) is Prior Probability: Probability of hypothesis before observing the evidence.
P(B) is Marginal Probability: Probability of Evidence.

Content from this work may be used under the term of the Creative Commons Attribution-Non-commercial (CC BY-NC) 4.0 licence. This license
allows refusers to distribute, remix, adapt, and build upon the material in any medium or format for non-
commercial purposes only, and only so long as attribution is given to the creator. Any further distribution of this
work must maintain attribution to the creators. 39
 ISSN (Online): 2583 648X

IJIRID
International Journal of Ingenious Research, Invention and Development
Volume 1 | Issue 3 | April 2023
Scientific Journal Impact Factor (SJIF 2023): 3.647
DOI: 10.5281/zenodo.7970771
www.ijirid.in

Naïve Bayes and Text Classification:

1. The Naïve Bayes algorithm is one of the most popular and simple machine learning classification
algorithms.
2. It based on Bayes theorem for calculating probabilities and conditional probabilities.
3. It works on Bayes theorem of probability to predict the class of unknown dataset.

IV. TECHNOLOGY USED

Programming Language: NodeJS
I. Libraries Used:
• Scikit-learn
• Pandas
• TensorFlow
• NumPy
• Flask
• Plotly.js
• Sequelize
• Express

II. Algorithms/ Tools/ Techniques/ Classes:

• Naïve Bayes Classifier
• Bayes Theorem

Content from this work may be used under the term of the Creative Commons Attribution-Non-commercial (CC BY-NC) 4.0 licence. This license
allows refusers to distribute, remix, adapt, and build upon the material in any medium or format for non-
commercial purposes only, and only so long as attribution is given to the creator. Any further distribution of this
work must maintain attribution to the creators. 40
 ISSN (Online): 2583 648X

IJIRID
International Journal of Ingenious Research, Invention and Development
Volume 1 | Issue 3 | April 2023
Scientific Journal Impact Factor (SJIF 2023): 3.647
DOI: 10.5281/zenodo.7970771
www.ijirid.in

IV. DATA FLOW DIAGRAM

IV. RESULT
Dataset: We built the dataset with parameters of name, MHT-CET scores, applied colleges (minimum two),
branch, and allotted college in Excel and created a csv file. Then we convert the csv file into JSON and train it,
which will help predict certain college courses for students. We built around 500 rows into the dataset.

Algorithm: Naïve Bayes Classifier is used. Here, input is given to the predictor.

Content from this work may be used under the term of the Creative Commons Attribution-Non-commercial (CC BY-NC) 4.0 licence. This license
allows refusers to distribute, remix, adapt, and build upon the material in any medium or format for non-
commercial purposes only, and only so long as attribution is given to the creator. Any further distribution of this
work must maintain attribution to the creators. 41
 ISSN (Online): 2583 648X

IJIRID
International Journal of Ingenious Research, Invention and Development
Volume 1 | Issue 3 | April 2023
Scientific Journal Impact Factor (SJIF 2023): 3.647
DOI: 10.5281/zenodo.7970771
www.ijirid.in

Here, Naïve Bayes Classifier is used for prediction.

Login / Signup Page: Here are the login and signup pages for basic security.

Content from this work may be used under the term of the Creative Commons Attribution-Non-commercial (CC BY-NC) 4.0 licence. This license
allows refusers to distribute, remix, adapt, and build upon the material in any medium or format for non-
commercial purposes only, and only so long as attribution is given to the creator. Any further distribution of this
work must maintain attribution to the creators. 42
 ISSN (Online): 2583 648X

IJIRID
International Journal of Ingenious Research, Invention and Development
Volume 1 | Issue 3 | April 2023
Scientific Journal Impact Factor (SJIF 2023): 3.647
DOI: 10.5281/zenodo.7970771
www.ijirid.in

Students to Enter Details: All the students have to enter the given parameters for the prediction.

After details are entered:

Output of Model:

V. CONCLUSION
In conclusion, the MHT-CET-based college prediction system has the potential to completely transform the
application process for colleges and universities. Based on a dataset of student admissions from the previous
year, the system may forecast a student's college choice. Overall, the MH-CET marks-based college prediction
system has the potential to be advantageous to the entire educational ecosystem, including students, colleges,
universities, and the government. The predicted college may occasionally be incorrect due to poor accuracy
now, which is a drawback. The method can offer a more effective, transparent, and fair procedure with proper
preparation, execution, and monitoring, thereby assisting students in understanding prediction.

Content from this work may be used under the term of the Creative Commons Attribution-Non-commercial (CC BY-NC) 4.0 licence. This license
allows refusers to distribute, remix, adapt, and build upon the material in any medium or format for non-
commercial purposes only, and only so long as attribution is given to the creator. Any further distribution of this
work must maintain attribution to the creators. 43
 ISSN (Online): 2583 648X

IJIRID
International Journal of Ingenious Research, Invention and Development
Volume 1 | Issue 3 | April 2023
Scientific Journal Impact Factor (SJIF 2023): 3.647
DOI: 10.5281/zenodo.7970771
www.ijirid.in

REFERENCES
[1] Mishra, Sushruta, Soumya Sahoo, Brojo Kishore Mishra, and Santosh Satapathy. "A quality based
automated admission system for educational domain." In 2016 international conference on signal
processing, communication, power and embedded system (SCOPES), pp. 221-223. IEEE, (2016).
[2] Pushpa, S. K., T. N. Manjunath, T. V. Mrunal, Amartya Singh, and C. Suhas. "Class result prediction using
machine learning." In 2017 International Conference on Smart Technologies for Smart Nation
(SmartTechCon), pp. 1208-1212. IEEE, (2017).
[3] Jamison, Joseph. "Applying machine learning to predict Davidson College's admissions yield." In
Proceedings of the 2017 ACM SIGCSE Technical Symposium on Computer Science Education, pp. 765-766.
(2017).
[4] “Machine Learning Model for Prediction of Post Graduate Admissions” 2021 JETIR July 2021, Volume 8,
Issue 7 www.jetir.org (ISSN-2349-5162).
[5] Mohana Bangale, Shubham Bavane, Akshay Gunjal, Rohit Dandhare, Sudhir D. Salunkhe, “A Survey on
Placement prediction system using machine learning” (2023).

Content from this work may be used under the term of the Creative Commons Attribution-Non-commercial (CC BY-NC) 4.0 licence. This license
allows refusers to distribute, remix, adapt, and build upon the material in any medium or format for non-
commercial purposes only, and only so long as attribution is given to the creator. Any further distribution of this
work must maintain attribution to the creators. 44