Self-Esteem Classification

This dataset was used in the study entitled “SMOTE Effectiveness and various Machine Learning Algorithms to Predict Self-Esteem Levels of Indonesian Student.” Data were collected through a questionnaire distributed to students aged 16–30 years, containing 19 features covering social, emotional, and demographic aspects related to self-esteem levels. A total of 47 responses were obtained, with 64% indicating high self-esteem and 36% indicating low self-esteem. Features include variables such as social relations, psychological well-being, social support, social media usage, emotional regulation, and others. The dataset was used to develop classification models using Naïve Bayes, Decision Tree, Random Forest, Logistic Regression, and Support Vector Machine (SVM) algorithms, and evaluated with preprocessing techniques such as SMOTE and min-max normalization. This dataset is suitable for research in psychology, education, mental health, and machine learning, particularly in studies related to psychological prediction using tabular data.

Keywords:
self-esteem, mental health, machine learning, SMOTE, normalization, survey dataset, psychological prediction, Indonesian students

Additional Information:
Sample size: 47
Number of features: 19
Data format: Tabular (CSV/Excel)
Questionnaire language: Indonesian
Measurement scale: Likert 1–5 and categorical data

Citation Suggestion:
Anshori, M., Siwi Pradini, R., & Teja Kusuma, W. (2025). SMOTE Effectiveness and various Machine Learning Algorithms to Predict Self-Esteem Levels of Indonesian Student. Engineering, MAthematics and Computer Science Journal (EMACS), 7(2), 175–182. https://doi.org/10.21512/emacsjournal.v7i2.13521

本数据集应用于题为《合成少数类过采样技术（SMOTE）有效性与多种机器学习算法预测印尼学生自尊水平》的研究。研究通过向16至30岁的印尼学生发放问卷收集数据，共包含19项特征，覆盖与自尊水平相关的社交、情感及人口统计学维度。最终回收有效问卷47份，其中64%的受访者自尊水平较高，36%较低。特征变量涵盖社交关系、心理幸福感、社会支持、社交媒体使用、情绪调节等多个维度。本数据集被用于构建基于朴素贝叶斯、决策树、随机森林、逻辑回归及支持向量机（SVM）的分类模型，并通过SMOTE与最小-最大归一化等预处理技术对模型进行评估。本数据集适用于心理学、教育学、心理健康及机器学习领域的研究，尤其适用于基于表格数据开展的心理预测相关研究。

关键词：自尊、心理健康、机器学习、SMOTE、归一化、问卷数据集、心理预测、印尼学生

补充信息：
样本量：47
特征数量：19
数据格式：表格型（CSV/Excel格式）
问卷语言：印尼语
测量量表：李克特1-5级量表与分类数据

引用建议：
安肖里（Anshori, M.）、西维·普拉迪尼（Siwi Pradini, R.）与特贾·库苏马（Teja Kusuma, W.）(2025). 《合成少数类过采样技术（SMOTE）有效性与多种机器学习算法预测印尼学生自尊水平》. 工程、数学与计算机科学期刊（EMACS）, 7(2), 175–182. https://doi.org/10.21512/emacsjournal.v7i2.13521