Skripsi

REKOMENDASI PEMILIHAN SEKOLAH MENENGAH ATAS BERBASIS DOMISILI DENGAN MENGGUNAKAN METODE STACKING LEARNING

---

School selection at the senior high school level is an important decision amid intense competition in the student admission process and multiple influencing factors such as academic performance, domicile, and school capacity. This study aims to develop a recommendation system based on stacking ensemble learning that combines Random Forest and XGBoost as base learners, with Logistic Regression as the meta-learner to predict student admission probability more accurately and reliably. The system is also equipped with SHAP (SHapley Additive exPlanations) to provide transparent interpretation of the recommendation results. The dataset consists of 6,120 student–school pairs, split into 70% training, 15% validation, and 15% testing data. Initial results show an accuracy of 83% with MRR@5 = 0.5308 and NDCG@5 = 0.6334. After incorporating school features and student–school interaction features, performance improves significantly, reaching 99.67% accuracy with better recommendation quality (MRR@5 = 0.5733 and NDCG@5 = 0.6739), indicating that the proposed system is more effective in helping students prioritize schools with the highest admission probability.

---

Availability

Detail Information

Series Title

-

Call Number

T1959772026

Publisher

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

Collation

xv, 140 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.310 7

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Sistem Rekomendasi
Prodi Teknik Informatika

Specific Detail Info

-

Statement of Responsibility

TUTI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment