**EE448 Big Data Mining 2018 **

**Weinan Zhang, Assistant Professor**

John Hopcroft Center for Computer Science

SEIEE

Shanghai Jiao Tong University

Big data driven techniques have been revolutionizing various aspects of our daily life. Big data means not only big volume but also high dimension and diversity. How to collect, represent, process and compute so as to successfully mine valuable patterns and acquire benefit from the big data is a fundamental challenge to both academia and industry.

This course provides a comprehensive introduction of the fundamental problems and methodologies of big data mining. The organization of the course would be application oriented, which helps SEIEE students get familar with various data mining tasks and basic solutions. Via lectures, hands-on courseworks and poster presentations, the students are expected to acquire the basic theory, algorithms, and some practice experience of big data mining techniques. It would also help students find their interested research topics, which could benefit their further graduate study and industrial practice.

## Notice

The first quiz will be taken on Apr. 19, 2018.

The first course work (text classification) is launched on Kaggle.

## Course Works

link |
Course work 1: Text Classification
To create a classification model for recommending editor selected articles. Mar. 22, 2018 - May 5, 2018. |

link |
Course work 2: Link Prediction
Top-N prediction for a target node give a start node and a link type. Apr. 21, 2018 - June 8, 2018. |

## Slides

Lecture 1: Introduction to Big Data Mining
Basic concepts, history and some examples of data mining. Mar. 1, 2018 |

Lecture 2: Know Your Data
Data representation, visualization and proximity measures. Mar. 8, 2018 |

Lecture 3: Fundamental Data Mining Algorithms
Frequent patterns, association rules, Apriori, FPGrowth, KNN. Mar. 15, 2018 |

Lecture 4: Supervised Learning (Part I)
Intro to machine learning, linear regression and logisitic regression. Mar. 22, 2018 |

Lecture 5: Supervised Learning (Part II)
Support Vector Machines, Neural Networks. Mar. 29 and Apr. 12, 2018 |

Lecture 6: Supervised Learning (Part III)
Tree models, Ensemble Methods Apr. 19 and 26, 2018 |

Lecture 7: Unsupervised Learning
K-means Clustering, PCA, Mixture Gaussian, EM Methods May 3, 2018 |

Lecture 8: Search Engines
Information retrieval, inverted index, retrieval model, relevance model May 10, 2018 |

Lecture 9: Learning to Rank
Ranking problem, pairwise/listwise ranking, LambdaRank May 17, 2018 |

Lecture 10: Recommender Systems
Information filtering, collaborative filering, matrix factorization May 24, 2018 |

## Related Readings

- UIUC CS512 Data Mining: Principles and Algorithms. by Prof. Jiawei Han.
- Stanford CS229 Machine Learning course by Prof. Andrew Ng.
- Top 10 algorithms in data mining by Xindong Wu et al. 2008.
- Matrix Cookbook: fundamentals of matrix calculations.

## Teaching Assistants

link |
Yuchen Yan, IEEE Honored Class 2015 student, Computer Science
Research on data mining, knowledge graph, network analysis Email: xyxpzer [at] sjtu.edu.cn |

link |
Ruijie Wang, Computer Science 2015 student
Research on data mining, natural language processing Email: wjerry5 [at] sjtu.edu.cn |

link |
Jialu Wang, IEEE Honored Class 2015 student, Computer Science
Research on data mining, deep learning, reinforcement learning Email: faldict [at] sjtu.edu.cn |

## News

**Mar. 22, 2018**

First course work is launched. The deadline is May 5.

**Mar. 1, 2018**

First lecture is provided.

**Jan. 9, 2018**

Web site created!