Quantum Modelling - IBM HR Attrition Dataset

This is the dataset developed using data Pavansubhash. (2017). IBM HR Analytics Employee Attrition & Performance. https://www.kaggle.com/datasets/pavansubhasht/ibm-hr-analytics-attrition-dataset/data

The data has been preprocessed as clean, noisy and imbalanced and machine learning has been attempted to apply and test classical models such as RandomForest, SVM, LogisticRegression, GradientBoosting, NNN, NaiveBayes, DecisionTree, and also Quantum Models such as QuantumKernel, QSVC, VQC_shallow, VQC_deep. The models were compared using metrics such as Accuracy, Precision, Recall, F1, Time and a Confusion Matrix was developed.

The code used for preprocessing, classical and quantum modelling has been provided,
Software and environment the code was operated details has been added
The Clean, Noisy and Imbalanced datasets were provided
Final Results were also added

本数据集基于Pavansubhash（2017）发布的《IBM人力资源分析：员工离职与绩效》（IBM HR Analytics Employee Attrition & Performance）数据开发，数据源链接为：https://www.kaggle.com/datasets/pavansubhasht/ibm-hr-analytics-attrition-dataset/data。本数据集已被预处理为清洁数据集、含噪数据集与不平衡数据集三类；研究团队尝试应用并测试了经典机器学习模型与量子机器学习模型，其中经典模型包括随机森林（RandomForest）、支持向量机（SVM）、逻辑回归（LogisticRegression）、梯度提升（GradientBoosting）、NNN、朴素贝叶斯（NaiveBayes）、决策树（DecisionTree），量子模型则涵盖量子核（QuantumKernel）、量子支持向量分类器（QSVC）、浅层变分量子分类器（VQC_shallow）与深层变分量子分类器（VQC_deep）。本次实验采用准确率（Accuracy）、精确率（Precision）、召回率（Recall）、F1值、运行耗时作为模型性能对比指标，并生成了混淆矩阵（Confusion Matrix）。本项目已提供数据预处理、经典机器学习建模与量子机器学习建模的代码，并补充了代码运行所需的软件与环境配置细节。已提供清洁、含噪与不平衡三类数据集。同时补充了最终实验结果。