Skripsi

KLASIFIKASI TALENT SCOUTING RENANG MENGGUNAKAN HYBRID DECISION TREE DAN ALGORITMA GENETIKA

---

Talent Scouting in sports is a systematic process of finding, identifying, and evaluating individuals who show potential for achievement from an early age, so that their talents can be developed through a tiered training program to achieve optimal performance. The use of classification methods is one effective solution to support this process. Classification is the main method in machine learning used to group data into certain classes based on patterns or characteristics. This study used a Talent Scouting swimming dataset of 100 data consisting of 13 features and 1 feature as a label, namely long distance and short distance as the classification target. The Decision Tree algorithm was chosen because the decision tree-based approach is able to classify data with complex criteria and is easy to interpret. However, Decision Tree has weaknesses such as a tendency to overfit and sensitivity to data changes. To overcome this, Genetic Algorithm is used as a feature selection method to identify the most relevant attributes so that the decision tree formation process becomes more optimal and efficient. The results of feature selection show one of the best subsets consisting of six features, namely age, height (cm), vertical jump (cm), agility run, shuttle run, and gender (PA/PI). The use of this feature subset is able to improve model performance, where the Decision Tree without producing accuracy of 60%, while the Decision Tree optimized using Genetic Algorithm with parameters Criterion = Entropy, Max Depth = 7, Min Samples Split = 5, Min Samples Leaf = 2, population 25, generation 10, crossover rate 0.9, and mutation rate 0.04 increased to 90%.

---

Availability

Detail Information

Series Title

-

Call Number

T1949882026

Publisher

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

Collation

xxiii, 61 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.310 7

Content Type

Text

Media Type

unmediated

Carrier Type

-

Edition

-

Subject(s)

Algoritma genetika
Prodi Teknik Informatika

Specific Detail Info

-

Statement of Responsibility

KA

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment