Hi, welcome to the course Pattern Discovery in Data Mining. Here I'm giving you a general overview of this course. This course is one of six courses being offered by the University of Illinois at Urbana-Champaign, as a data mining specialization series. Data mining specialization consists of six courses. Data Visualization, Text Retrieval and Search Engines, Text Mining and Analytics, Pattern Discovery in Data Mining, Cluster Analysis in Data Mining, and finally, Data Mining Capstone. This final Data Mining Capstone is a project oriented courses using the knowledge you learned in the first five, six courses. So first you may wonder, what is pattern discovery? So I'd better first give you example. Considering massive shopping transaction data, pattern discovery may help you answer the following questions. What groups of items are frequently bought together? If a person buys diapers at night, what is possibility of this person buying beer as well? If a customer buys an iPhone 5 or iPhone 7, what other electronic products will the customer be most likely to buy in the next three months? So you probably can see pattern discovery is quite interesting. What is the value of pattern discovery? Pattern discovery may help you find the hidden and inherent data patterns in massive data. Pattern mining will play a unique and critical role in mining massive data. What roles does pattern discovery play in this particular data mining specialization series? In this course, you will learn scalable methods to find patterns from massive data. The patterns maybe the set of data items strongly correlated to each other. You will learn how to mine a large variety of patterns. You will also learn how to evaluate the patterns. And finally, you will find the pattern discovery may help classification, clustering, and many other data mining tasks. Pattern mining has very broad applications. We already see it may predict shopping transaction data. For example, for a customer who buys products A and B, what is the likelihood the customer will buy product C as well? It may predict a web click streams for example, based on the current data you may want to see which webpage is most likely to be clicked the next. It may help mining software bugs. Where is the likely bug in this program? It may identify objects or sub-structures in images, videos, and social media. It may help finding quality phrases, entities, and attributes in massive text data. It may help repeating DNA and protein sequences in genomes. It may help find hidden communities in massive social network. There are several major reference readings for this course. We will have a textbook, the textbook is written by me, Micheline Kamber and Jian Pei, published in year 2011, called Data Mining Concepts and Techniques. This is the third edition published by Morgan Kaufmann. The following three chapters are most related to the course. Chapter 1, Introduction. Chapter 6, Mining Frequent Patterns, Associations, and Correlations, Basic Concepts and Methods. Chapter 7, Advanced Pattern Mining. Other references will be listed at the end of each lecture video. The course has the following structures. We have Lesson 1: Pattern Discovery: Basic Concepts. Lesson 2: Efficient Pattern Mining Methods. These two lessons were from Module 1, which usually you can finish in one week. Then we have lesson 3, Pattern Evaluation. Lesson 4, Mining Diverse Frequent Patterns. These two lessons were from Module 2. We have lesson 5, Sequential Pattern Mining. Lesson 6, Pattern Mining Applications, Mining Spatiotemporal and Trajectory Patterns. These two lessons form Module 3. We have lesson 7, Pattern Mining Applications, Mining Quality Phrases from Text Data. And lesson 8, Advanced Topics in Pattern Discovery. These two lessons from Module 4. For the general information about this course, I am the instructor. I'm a Professor in Department of Computer Science, University of Illinois at Urbana-Champaign. We will have some teaching assistants who will help me editing this courses and in the meantime, giving some quizzes and program assignments. The course prerequisite is as long as you are familiar with basic data structures and algorithms, you will have no problem to take this course. The course will be assessed by the following three measures. We have in-video questions. At the end of the lesson, we have lesson quizzes. We also will be giving two programming assignments. We have one required programming assignment. The second one is an optional one, it is not required for passing the course. Hope you will enjoy this interesting course. Welcome. [MUSIC] [SOUND]
