No image available for this title

Skripsi

Comparison of Random Forest and Decision Tree Performance in Evaluating the Quality of Knowledge Representation in Concept Maps / Roudhotulloh Nazakhan

Abstrak
Concept maps are structured visual tools that depict the relationships among key concepts and ideas. They are widely employed in educational settings to facilitate students rsquo comprehension of subject matter. Despite their usefulness the evaluation of concept map quality remains largely manual resulting in inefficiencies and inconsistent outcomes across assessors. This study aims to compare two algorithm methods Random Forest and Decision Tree for automatically evaluating the quality of concept maps. The data used includes 691 propositions from 30 concept maps on the topic of Relational Database which have been previously scored by experts on a scale from 0 to 3. This research follows the CRISP-DM process starting with text preprocessing (such as case folding tokenization stopword removal and stemming) then converting the text into numerical features using the TF-IDF method and finally training the Random Forest and Decision Tree models with optimized parameters. For evaluation performance metrics such as accuracy precision recall F1-score and Kappa value were used. The results showed that Random Forest performed better with an accuracy of 81.2% and a Kappa value of 0.869 compared to Decision Tree which had an accuracy of 79.8% and a Kappa value of 0.852. These findings suggest that machine learning models particularly Random Forest have the potential to be used as an initial approach in supporting automatic efficient and consistent assessment of concept map quality in educational settings

Informasi Detail
DDC
SKRIPSI DIGITAL
Prodi
Universitas Negeri Malang. Program Studi Teknik Informatika, .
Deskripsi Fisik
x, 38 hlm. : ilus.
Bahasa
No Reg
4308/RS/25
Edisi
Skripsi (Sarjana)--Universitas Negeri Malang. 2025
Subjek
1. PETA KONSEP - ALGORITMA RANDOM FOREST
2. CONCEPT MAP - RANDOM FOREST ALGORITHM
Pembimbing
1. Dr.eng Didik Dwi Prasetya, S.t., M.t.
Lampiran Berkas