Result in diagram

This dataset contains project management data extracted from Jira, used in the research “Predictive Analysis Using Machine Learning with CRISP-DM and Random Forest for Software Project Delivery Delay Mitigation in an HRIS SaaS Company.” The dataset includes historical epic-level project records from 2024, covering attributes such as project identifiers, issue identifiers, status and workflow information, time and effort metrics, team and resource attributes, and quality indicators. Data preprocessing involved handling missing values, outlier treatment, data type conversion, and feature engineering (including time-based metrics, effort ratios, aggregated project-level features, and textual features). The dataset is intended for use in predictive analytics research, specifically focusing on project delay prediction using machine learning models. It supports experimentation with techniques such as Random Forest, XGBoost, and other classification algorithms, and can be used to reproduce or extend the results presented in the associated research paper.本数据集包含从Jira提取的项目管理数据，用于题为《基于CRISP-DM（跨行业数据挖掘标准流程）与随机森林（Random Forest）结合机器学习的预测分析，以缓解HRIS SaaS企业软件项目交付延迟》的研究工作。 该数据集涵盖2024年的历史史诗级项目记录，包含项目标识符、问题标识符、状态与工作流信息、时间与工作量度量指标、团队与资源属性以及质量指标等属性。数据预处理环节包括缺失值处理、异常值修正、数据类型转换及特征工程（涵盖基于时间的度量指标、投入比率、聚合型项目级特征与文本特征）。 本数据集面向预测分析研究，核心聚焦于依托机器学习模型实现项目延迟预测。其支持随机森林、XGBoost及其他分类算法相关技术的实验验证，可用于复现或拓展关联研究论文中呈现的研究成果。