课程大纲

COURSE SYLLABUS

课程代码/名称

Course Code/Title

金融数据挖掘

Financial Data Mining

课程性质

Compulsory/Elective

选修课 Elective Courses

课程学分/学时

Course Credit/Hours

3/48

授课语言

Teaching Language

中英双语 English & Chinese

授课教师

Instructor(s)

陈琨 Chen Kun

先修要求

Pre-requisites

无 No

教学目标

Course Objectives

此课程的目的是讲授数据分析以及数据挖掘的基本过程、模型和工具，及其在金融中的应用。此课

程将培养学生软件包（如 Excel 和 weka 软件）、编程（如 Python、JAVA）的实用技巧以及通过分

析最新的文献来讨论如何在技术领域分析和解决金融数据问题。通过本课程的学习，学生可学习金

融科技技术，为从事创新金融行业奠定基础。

The course aims to teach students the pro

cess, models, and tools for data analytics and data mining in

finance. The course will teach students the practical skills to employ software packages (such as Excel

and weka) , programming（such as python or JAVA） and apply necessary extensions to analy

tic

framework and tackle financial data analysis problems. The course will equip students the basic skills

social network analysis and models to pursue further study in the Fintech domain.

教学方法

Teaching Methods

理论课形式主要为课堂讨论研究文献及方法，并辅以研究方法实际操作，共

学时。

The lecture is mainly for the discussion of research articles and methods, and the practical operation of the

research methods is supplemented. There are total 48 hours of lectures.

教学内容

Course Contents

理论 lecture

Section 1

Data type, data quality, data preprocessing

This section explains how to describe data objects from attributes and metrics,

explains the detection and correction of data quality problems, and the ideas

and methods for preprocessing data.

Section 2

Classification: Basic Concepts, Decision Trees

This section mainly describes the basic concepts of classification and

introduces the general approach to solving classification problems, and

explains the working principle of the decision tree classification method and

the method of establishing the decision tree.

Section 3

Classification: Model Evaluation

This section mainly introduces some commonly methods for evaluating the

performance of a classifier, such as retention methods, random subsampling,

and cross-validation.

Section 4

Classification: Alternative Techniques

This section mainly explains some alternative classification techniques: Rule-

Based Classifier

、

Nearest-Neighbour classifiers

、

Bayesian Classifiers

、

Artificial Neural Network (ANN)

、

Support Vector Machine (SVM).

Section 5

Lab-Text Mining and Classification

This section mainly explains the use of different classification algorithms in

open sourced software to analyze the textual data, as well as in this chapter,

students will learn the features of different classification algorithms.

Section 6

Classification: Literature

Select reading Fintech related literature, especially about classification.

Section 7

Association Analysis: Apriori

This section mainly explains the effective techniques for generating frequent

itemsets and rules in the association rule mining algorithm, focusing on the

frequent itemsets and rules generation of the Apriori algorithm.

Section 8

Association Analysis: FP-Growth Algorithm

This section mainly introduces other algorithms that generate frequent

itemsets—the FP growth algorithm.

Section 9

Association Analysis: Evaluation

This section mainly discusses the evaluation metrics about association analysis.

Section 10

Lab—Basket Analysis

This chapter mainly explains the analysis of dataset using the association rules

in open sourced software.

Section 11

Association Analysis: Literature

Select reading Fintech

related literature, especially about association

analysis.

Section 12

Cluster Analysis: Basic Concepts and Algorithms

This section mainly introduces the K-means, condensed hierarchical clustering

algorithm, the basic principles of the DBSCAN algorithm, and the advantages

and disadvantages of each algorithm.

Section 13

Cluster Analysis: Cluster Evaluation

This section mainly introduces methods for evaluating clusters generated by

clustering algorithms.

Section 14

Lab—Clustering

This section explains how to use K-means to process the real data using

Python.

Section 15

Cluster Analysis: Literature

This section mainly discusses the current financial application of cluster

analysis technology in practice cases through reading literature.

Section 16

Project Demonstration and Financial Review

10.

课程考核

Course Assessment

20%

平时作业

+ 40%

期末报告

I + 40%

期末报告

20% Assignments + 40% Final Report

I +

40% Final Report

11.

教材及其它参考资料

Textbook and Supplementary Readings

Pang-Ning Tan, Michael Steinbach, Vipin Kumar, Introduction to Data Mining, posts & telecom press.