[SOUND] Hi welcome to the course, Cluster Analysis in Data Mining. Before starting this course, I'm going to give you a general course overview. First, what is cluster analysis? Actually when you fly over a city, you can easily identify fields, forests, commercial area, and residential areas based on their features, without anybody's explicit training or labeling. This is the power of cluster analysis. In this course, I'm going to systematically introduce you the concepts and methods of cluster analysis. And help answering the following question. What are the different proximity measures for effective clustering? Can we cluster a massive number of data points efficiently? Can we find clusters of arbitrary shape and at multilevels of granularity? How can we judge the quality of the clusters discovered by our system? Cluster analysis bring a lot of value to data mining. What is the value of cluster analysis? Cluster analysis may help you partition massive data into groups based on its features. Cluster analysis also will help subsequent data mining processes, such a pattern discovery, classification and outlier analysis. What roles does cluster analysis play in the data mining specialization? You will learn various scalable methods to find clusters from massive data. You will learn how to mine different kinds of clusters effectively. You'll also learn how to evaluate the quality of the clusters you'll find. And you will see cluster analysis will help with classification, outlier analysis and other data mining tasks. Cluster analysis has its broad applications. For example, for data summarization, compression, and reduction, like image processing or vector quantization, you will need cluster analysis. For collaborative filtering, recommendation systems, or customer segmentation, you will find like-minded users or similar products by cluster analysis. For dynamic trend detection, you will find clustering stream data and detecting trends and patterns will be very effective. For multimedia analysis, biological data analysis, and social network analysis you'll find you may have effecting measures to group audio/video clips and finding clusters of gene protein sequences. Cluster analysis is a key intermediate step for many other data mining tasks. For example, you want to generate a compact summary of data for classification, pattern discovery, hypothesis generation and testing your neat cluster masses. You wana find outlier detection. Actually, outliers are those far away from any cluster. In this course, the major reference of the readings is my own textbook published in 2011, which is, Data Mining: Concepts and Techniques, the 3rd edition, published by Morgan Kaufmann. In this book, we will only used two chapters. The first is Chapter 2. [INAUDIBLE] just cover section 2.4: Measuring Data Similarity and dissimilarity. And then Chapter 4 is a major source related to this textbook. The textbook of Chapter 4 is a major source related to this seminar, this lecture, called Cluster Analysis: Basic Concepts and Methods. Other references will be listed at the end of each lecture video. For the course structure, we will have six lessons. Lesson 1 is, Cluster Analysis: An Introduction. Lesson 2 is, Similarity Measures for Cluster Analysis. These two lessons from the first module, Module 1. Then, Lesson 3: Partitioning-Based Clustering Methods. An, Lesson 4(Part I): Hierarchical Clustering Methods (I), from the contents of Module 2. Then the Part II of Lesson 2, Hierarchical Clustering Methods (II). And, Lesson 5: Density-Based and Grid-Based Clustering Methods, form the contents of Module 3. Finally, Lesson 6: Clustering Validation, forms the material for Module 4. For general information about this course, I'm Jiawei Han. I'm a professor in the Department of Computer Science, University of Illinois at Urbana-Champaign. I'm the instructor. We have some teaching assistants on the web to help you. The course prerequisites. Just as long as you are familiar with basic data structure and algorithms, you like to work hard, you're quite okay. For course assessments, we have in-video questions to help you understand the course materials. Then we need you to pass some minimum requirements for lesson quizzes and two programming assignments. Thank you. Hope you will enjoy this course. Thank you. [MUSIC]
