Reminder System Study Period And Student Graduation Predictions Using Multilayer Percepstrons Algorithms And Support Vector Machines In The Telegram Application

Feb 27, 2025 by ADMIN 161 views

**Student Remembrance System and Student Graduation Prediction Using Multilayer Perceptron Algorithm and Support Vector Machine in Telegram Applications**

Introduction

The time graduation rate of students at an undergraduate level is a challenge that is often faced by study programs at the university every year. This problem has a negative effect on the accreditation assessment of the study program itself. To overcome this problem, this research develops a study system and graduation prediction that utilizes the Multilayer Perceptron (MLP) algorithm and the Support Vector Machine (SVM) in the form of chatbots in the Telegram application.

Background and Problem Statement

The study program at the university faces a significant challenge in terms of student graduation rates. The existing system relies heavily on manual tracking and prediction, which can lead to inaccuracies and inefficiencies. Moreover, the current system does not provide students with timely and accurate information about their graduation status, which can cause unnecessary stress and anxiety.

Methodology

The dataset used in this study is the Academic Data of Students of Computer Science and Information Technology Studies Program at the University of North Sumatra, especially from semester 1 to semester 4, and semester 5 to 6. The main purpose of this system is to predict the status of student graduation in semester 5, 6, or 7 using the two algorithms.

However, during data analysis, it was found that the existing dataset had an unbalanced class. To overcome this problem, the synthetic minority over-sampling technique (SMOTE) with edited nearest neighbor (ENN) is used. The application of this technique has succeeded in increasing prediction accuracy, which reached 98.28% for student data semester 5, 99.41% for semester 6, and 99.63% for semester 7.

System Analysis and Benefits

The implementation of the reminder system and graduation predictions offers several significant benefits. First, with a reminder, students can be more organized and easier to plan their studies. This will certainly reduce the possibility of being late in completing studies and increasing time efficiency.

Second, accurate graduation predictions will give students insight earlier about their academic status, so they can take the steps needed if it is estimated that they do not pass on time. For example, they can try harder in courses that are considered difficult or seek additional assistance, both from lecturers and classmates.

In addition, this system also provides opportunities for universities to improve the quality of their academic services. By analyzing the data and predictions obtained, the university can better understand the pattern of success or difficulties faced by students, so as to make improvements to the study programs and teaching strategies that are applied.

Implementation and Results

The system was implemented using the Telegram application, which provides a user-friendly interface for students to access the reminder system and graduation predictions. The system uses the MLP and SVM algorithms to predict the graduation status of students.

The results of the study show that the system is able to predict the graduation status of students with high accuracy. The accuracy of the system reached 98.28% for student data semester 5, 99.41% for semester 6, and 99.63% for semester 7.

Conclusion

The development of the study period of study period and student graduation predictions using MLP and SVM in this telegram application is an innovative step that can help students to achieve academic success more easily and directed. The use of modern technology such as chatbots in telegram not only provides easy access to information, but also creates more dynamic interactions between students and educational institutions.

Future Work

The study suggests several areas for future research. First, the system can be improved by incorporating more features, such as student behavior and academic performance. Second, the system can be expanded to include more universities and study programs. Finally, the system can be used to predict other academic outcomes, such as student retention and graduation rates.

Recommendations

Based on the results of the study, the following recommendations are made:

The university should implement the reminder system and graduation predictions in the Telegram application to provide students with timely and accurate information about their graduation status.
The university should use the data and predictions obtained from the system to improve the quality of their academic services.
The university should consider incorporating more features, such as student behavior and academic performance, into the system to improve its accuracy and effectiveness.

Limitations

The study has several limitations. First, the dataset used in the study is limited to the Academic Data of Students of Computer Science and Information Technology Studies Program at the University of North Sumatra. Second, the system relies heavily on the accuracy of the data and predictions obtained from the MLP and SVM algorithms. Finally, the system may not be able to predict the graduation status of students who have not completed their studies.

Conclusion

In conclusion, the study demonstrates the effectiveness of using MLP and SVM algorithms in predicting the graduation status of students. The system provides students with timely and accurate information about their graduation status, which can help them to achieve academic success more easily and directed. The study suggests several areas for future research and provides recommendations for the implementation of the system in the university.

Frequently Asked Questions

Q: What is the main purpose of the study?

A: The main purpose of the study is to develop a study system and graduation prediction that utilizes the Multilayer Perceptron (MLP) algorithm and the Support Vector Machine (SVM) in the form of chatbots in the Telegram application to predict the status of student graduation in semester 5, 6, or 7.

Q: What dataset was used in the study?

A: The dataset used in the study is the Academic Data of Students of Computer Science and Information Technology Studies Program at the University of North Sumatra, especially from semester 1 to semester 4, and semester 5 to 6.

Q: What technique was used to overcome the unbalanced class problem?

A: The synthetic minority over-sampling technique (SMOTE) with edited nearest neighbor (ENN) was used to overcome the unbalanced class problem.

Q: What are the benefits of the reminder system and graduation predictions?

A: The benefits of the reminder system and graduation predictions include:

Students can be more organized and easier to plan their studies.
Accurate graduation predictions will give students insight earlier about their academic status, so they can take the steps needed if it is estimated that they do not pass on time.
The system provides opportunities for universities to improve the quality of their academic services.

Q: How accurate is the system in predicting graduation status?

A: The accuracy of the system reached 98.28% for student data semester 5, 99.41% for semester 6, and 99.63% for semester 7.

Q: What are the limitations of the study?

A: The study has several limitations, including:

The dataset used in the study is limited to the Academic Data of Students of Computer Science and Information Technology Studies Program at the University of North Sumatra.
The system relies heavily on the accuracy of the data and predictions obtained from the MLP and SVM algorithms.
The system may not be able to predict the graduation status of students who have not completed their studies.

Q: What are the recommendations for the implementation of the system?

A: The recommendations for the implementation of the system include:

The university should implement the reminder system and graduation predictions in the Telegram application to provide students with timely and accurate information about their graduation status.
The university should use the data and predictions obtained from the system to improve the quality of their academic services.
The university should consider incorporating more features, such as student behavior and academic performance, into the system to improve its accuracy and effectiveness.

Q: What are the future research directions?

A: The future research directions include:

Incorporating more features, such as student behavior and academic performance, into the system to improve its accuracy and effectiveness.
Expanding the system to include more universities and study programs.
Using the system to predict other academic outcomes, such as student retention and graduation rates.

Q: What are the implications of the study?

A: The implications of the study include:

The development of the study period of study period and student graduation predictions using MLP and SVM in this telegram application is an innovative step that can help students to achieve academic success more easily and directed.
The use of modern technology such as chatbots in telegram not only provides easy access to information, but also creates more dynamic interactions between students and educational institutions.