Salsabila, Tiara Amalina (2025) Predicting citation count growth in scopus documents using the decision tree algorithm / Tiara Amalina Salsabila</p>. Diploma thesis, Universitas Negeri Malang.
Full text not available from this repository.Abstract
p With the advancement of education in the digital era the number of academic publications continues to grow making citation growth analysis a crucial topic for evaluating the impact of scientific publications. This study aims to predict the growth in citation counts of Scopus documents using the Decision Tree algorithm over a shorter period than previous studies (citation growth within two months). The study applies the SMOTE technique to address class imbalance in the dataset. The dataset undergoes a process of attribute selection and development followed by clustering with K-Means to categorize citation growth into two label classes rendah (low) and tinggi (high). Initial results show that the Decision Tree model achieves a high accuracy of 98.34% but fails to predict the tinggi (high) class due to data imbalance. The use of SMOTE (Synthetic Minority Oversampling Technique) successfully improves the F1-score of the tinggi (high) class from 0.00% to 78.30%. However the overall model accuracy changed to 75.69% indicating a trade-off in performance when handling the minority class. This study also analyzes strategies to increase authors rsquo citation counts using a correlation matrix. The results suggest that publishing articles collaborating with more foreign institutions and ensuring publications are available in open access formats may contribute to higher citation growth. /p
| Item Type: | Thesis (Diploma) |
|---|---|
| Divisions: | Fakultas Teknik (FT) > Departemen Teknik Elektro (TE) > S1 Teknik Informatika |
| Depositing User: | library UM |
| Date Deposited: | 28 May 2025 04:29 |
| Last Modified: | 09 Sep 2025 03:00 |
| URI: | http://repository.um.ac.id/id/eprint/400281 |
Actions (login required)
 |
View Item |