Skripsi

KLASIFIKASI TALENT SCOUTING RENANG MENGGUNAKAN HYBRID LOGISTIC REGRESSION DAN ALGORITMA GENETIKA

---

Talent identification (Talent Scouting) in swimming is a crucial stage in long-term athlete development; however, selection processes that rely on subjective judgment may lead to bias. This study aims to develop a more objective classification model for identifying potential swimming athletes using a Machine Learning approach. The dataset consists of 100 records with 13 variables, including anthropometric, biomotor, and target potential variables. The method applied is Logistic Regression optimized using Genetic Algorithm for feature selection. Logistic Regression without feature selection achieved 80% accuracy, while the hybrid approach improved accuracy to 90%. The best feature subset was [1 1 0 0 0 0 1 1 1 1 0 1] with a fitness value of 0.7, obtained using C = 10, liblinear solver, population size of 35, 30 generations, a crossover rate of 0.5, and a mutation rate of 0.04. This fitness value was obtained from one particular run and may vary in subsequent runs due to the stochastic nature of the Genetic Algorithm. Overall, feature selection optimization significantly improved classification performance in swimming talent identification.

---

Availability

Detail Information

Series Title

-

Call Number

T1931432026

Publisher

Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2026

Collation

xiii, 130 hlm.; ilus.; tab.; 29 cm.

Language

Indonesia

ISBN/ISSN

-

Classification

797.207

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Informatika
Talent Scouting--Atlit Renang

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment