Pengembangan Prediksi Pertandingan Badminton Berbasis Supervised Learning Dengan Model Regresi

Titahena, Anastasya Syanne (2023) Pengembangan Prediksi Pertandingan Badminton Berbasis Supervised Learning Dengan Model Regresi. Bachelor thesis, Universitas Pelita Harapan.

	Text (Title) Title.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (129kB)
	Text (Abstract) Abstract.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (254kB)
	Text (ToC) ToC.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (764kB)
	Text (Chapter 1) Chapter1.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (683kB)
	Text (Chapter 2) Chapter2.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB)
	Text (Chapter 3) Chapter3.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB)
	Text (Chapter 4) Chapter4.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (905kB)
	Text (Chapter 5) Chapter5.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (351kB)
	Text (Bibliography) Bibliography.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (317kB)
	Text (Appendices) Appendices.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (8MB)

Abstract

Pendefinisian masalah dalam penelitian ini melakukan pengembangan prediksi pertandingan badminton berbasis supervised learning dengan menggunakan model regresi logistik dengan tujuan utamanya adalah untuk mengembangkan prediksi pertandingan badminton lalu melakukan implementasi tampilan prediksi menggunakan Streamlit [1]. Solusi yang diusulkan adalah menggunakan pendekatan regresi logistik dalam melakukan pengembangan hasil pertandingan badminton internasional lalu kemudian dilakukan evaluasi menggunakan metrik evaluasi yang sesuai kemudian diimplementasikan kedalam sebuah sistem dengan tujuan menghasilkan tampilan prediksi probabilitas kemenangan pada Streamlit yang paling akurat. Penelitian dilakukan dengan menggunakan metode Cross-Industry Standard Process for Data Mining (CRISP-DM) yang berisi teknik praproses data seperti, penanganan data kosong, korelasi, dalm-lain yang akan diterapkan untuk mempersiapkan data set . Model yang telah diuji dengan data set akan dievaluasi menggunakan metrik yang relevan. Hasil akhir yang diharapkan adalah sebuah aplikasi berbasis Streamlit yang mampu memprediksi probabilitas kemenangan tim dalam pertandingan badminton internasional dengan tingkat akurasi yang tinggi berdasarkan metode regresi logistik [2], [3]. Hasil pengujian menunjukkan bahwa model regresi logitik menghasilkan nilai akurasi sebesar 0.53, yang berarti model dapat mengklasifikasikan dengan benar 53% dari keseluruhan data. Kata Kunci: Badminton, Supervised learning , Regresi Logistik, CRISP-DM, Metrik Evaluasi, Streamlit / The problem definition of this research is to develop a prediction model for badminton matches using supervised learning with logistik regression model, and to implement the prediction interface using Streamlit. The main objective is to develop a predictive model for international badminton matches and evaluate its performance using appropriate evaluation metrics [1]. The proposed solution involves using logistic regression approach to develop the outcome prediction for international badminton matches, followed by evaluation using relevant evaluation metrics. The developed model will then be implemented into a system to provide an accurate prediction interface on Streamlit. The research will be conducted using the Cross-Industry Standard Process for Data Mining (CRISP-DM) methodology, which includes data preprocessing techniques such as handling missing data and exploring correlations, among others, to prepare the data set. The model will be trained and tested using the data set, and evaluated using relevant metrics. The expected outcome is a Streamlit-based application capable of predicting the probability of winning for teams in international badminton matches with high accuracy using logistic regression method [2], [3]. The testing results show that the logistic regression model achieved an accuracy of 0.53, indicating that the model can correctly classify 53% of the total data. Keywords: Badminton, Supervised learning , Logistic Regression, CRISP-DM, Evaluation Metrik, Streamlit

Actions (login required)

View Item