Applied Business Analytics in Production 生产中的应用商业分析 (MAV560)

MAV560 Applied Business Analytics in Production comprises an in-depth understanding of designing and deploying business analytics systems end-to-end. The course covers crucial aspects such as project scoping, data needs, modelling strategies, and deployment patterns and technologies. Students will learn strategies to address common production challenges, including establishing a model baseline, handling concept drift, and performing error analysis. A comprehensive framework will be adopted for developing, deploying, and continuously improving a productionised analytics application. While understanding analytics concepts is essential, building an effective career in business analytics requires experience in preparing projects for deployment. This course combines foundational analytics concepts with the skills and best practices of modern software development, enabling students to successfully deploy and maintain analytics systems in real-world environments. MAV560生产中的应用商业分析课程深入讲解了如何从头到尾设计和部署业务分析系统。课程涵盖关键方面，如项目范围界定、数据需求、建模策略以及部署模式和技术。学生将学习应对常见生产挑战的策略，包括建立模型基准、处理概念漂移和执行错误分析。课程采用综合框架，用于开发、部署和持续改进生产化的分析应用。尽管理解分析概念很重要，但在业务分析领域建立有效的职业生涯需要具备项目部署的实践经验。该课程将基础分析概念与现代软件开发的技能和最佳实践相结合，使学生能够在实际环境中成功部署和维护分析系统。

• Introduction to Business Analytics in Production 生产中的业务分析概述
• Project Scoping and Stakeholder Management 项目范围界定和利益相关者管理
• Data Needs and Collection Methods 数据需求和收集方法
• Data Cleaning and Transformation 数据清洗与转换
• Feature Engineering 特征工程
• Modelling Strategies 建模策略
• Evaluation Metrics 评估指标
• Deployment Patterns and Technologies 部署模式和技术
• Handling Concept Drift 处理概念漂移
• Error Analysis and Continuous Improvement 错误分析与持续改进
• Performance Monitoring in Production 生产环境中的性能监控
• Case Studies and Real-World Applications 案例研究和实际应用

• Evaluate the entire lifecycle of an analytics project from scoping to maintenance 评估从项目范围界定到维护的整个分析项目生命周期
• Critique essential techniques for data cleaning, transformation, and feature engineering 批判性地评价数据清洗、转换和特征工程的基本技术
• Assess various modelling strategies and evaluation metrics for effective model building and assessment 评估各种建模策略和评估指标，以实现有效的模型构建和评估
• Formulate deployment strategies and technologies for transitioning models to production environments 制定将模型转移到生产环境的部署策略和技术
• Design methods to address common production challenges like concept drift and error analysis 设计解决常见生产挑战（如概念漂移和错误分析）的方法
• Create objectives, identify stakeholders, and develop project roadmaps 制定目标，识别利益相关者，并开发项目路线图
• Construct and optimise machine learning models using Python 使用Python构建和优化机器学习模型
• Implement and monitor performance systems for models in production environments 在生产环境中实施和监控模型性能系统
• Appraise and refine error analysis and continuous improvement practices 评估和完善错误分析及持续改进实践