巨量資料探勘與應用

臺北科技大學 114-2 學期課程，課號 357438。

課程大綱

本課程是研究所程度的進階課程，聚焦於巨量資料探勘與應用，涵蓋資料探勘理論及發展趨勢、技術與其在人工智慧及大數據領域中的創新應用。課程主要內容包括：資料科學的核心基礎(如資料倉儲與線上分析處理)、資料清理與前處理技術、頻繁樣式與關聯性分析、分類方法、群集分析及離群值偵測，並探討資料探勘中的前沿議題(如暗資料)。 本課程也將會深入介紹機器學習中常用的方法原理及分散式計算的基本架構，幫助學生掌握巨量資料處理技術。為加深學生的實作能力，課程安排了Python程式練習，結合資料科學模組（如NumPy、Pandas、Scikit-learn）進行資料探勘應用程式開發，並透過期中報告與期末專題的形式，培養學生分析資料、解決問題及團隊合作的能力。 透過本課程，預期同學將能夠學習到： 1. 理解巨量資料探勘的基本概念、技術與應用場景(如鐵道系統應用)。 2. 掌握資料清理、前處理、模型建立與評估等關鍵技能。 3. 使用Python與相關工具進行資料分析與應用開發。 4. 探索資料探勘在大數據與AI創新中的潛在發展方向，並培養研究與實踐能力。 This advanced graduate-level course focuses on big data mining and its applications, covering theories and developmental trends, as well as techniques and novel applications in the artificial intelligence and big data domains. The course covers fundamental topics like core foundations of data science (e.g., data warehousing and online analytical processing), data cleaning and preprocessing techniques, frequent patterns and association analysis, classification methods, cluster analysis, and anomaly detection. The course also covers cutting-edge topics in data mining, such as dark data. This course also delves deeply into the principles of commonly used machine learning methods and the fundamental architecture of distributed computing, providing students with critical skills for handling big data. To improve hands-on skills, the course includes Python programming exercises that incorporate data science modules (such as NumPy, Pandas, and Scikit-learn) for developing data mining applications. Mid-term reports and final course projects will encourage students to analyze data, solve problems, and develop teamwork skills. By finishing this course, students are expected to: 1) Understand the fundamental concepts, techniques, and use cases for big data mining (e.g., railway system applications). 2) Develop critical skills in data cleaning, preprocessing, model building, and evaluation. 3) Use Python and related tools for data analysis and application development. 4) Investigate potential developments in data mining for big data and AI innovations, while also developing research and practical skills.

課程進度

Week 1: Introduction to Data Mining Week 2-3: Python Development Environment Setup & Basic Programming Week 4-5: Python Data Science Modules Week 6: Data Type & Preprocessing Week 7-8: Classification Algorithms Week 9: Mid-Exam Week (Submit Final Project Proposal) Week 10: Clustering Algorithms Week 11: Association Analysis Week 12: Regression Evaluation & Dimensionality Reduction Week 13: Anomaly Detection Week 14: Artificial Neural Network Week 15: Distributed Computing Week 16: Special Lecture Week 17-18: Oral Presentation of Final Projects

評量方式

Assignments Practical group exercises Proposal of final projects Final course project Class attendance & involvement

教材

【遵守智慧財產權觀念，請使用正版教科書，不得使用非法影印教科書】 使用外文原文書：是 1. Introduction To Data Mining, 2/e. Pang-Ning Tan, Michael Steinbach, Anuj Karpatne & Vipin Kumar. Pearson FT Press. 2. Python Machine Learning By Example, 3/e. Yuxi (Hayden) Liu. Packt Publishing. 3. Data Mining and Machine Learning: Fundamental Concepts and Algorithms, 2/e. Mohammed J. Zaki & Wagner Meira, Jr. Cambridge University Press.