[ Back to Toppage of Toma's website ]

fiscal 2016 : Data Mining Theory

Syllabus: [ Japanese | English ]
Contact: see also syllabus.
Slack: dmt-ie-univ-ryukyu.slack.com for annances, chattings, etc.
References
Lecture documents
- #1 (10/6) Guidance
- 10/13: cancelled (business trip)
- Part 1: Introduction to data mining
  - #2 (10/20): What is Data Mining? Some definitions, simple examples and applications.
  - #3 (10/27): Machine Learning and Statistics, aspects of personal information.
  - #4 (11/3): Instances and Attributes as Input, and Knowledge Representations as Output
  - #5 (11/10): The basic methods and algorithms 1
  - #6 (11/17): The basic methods and algorithms 2
  - #7 (11/24): Credibile experiment designs and evaluation ways
- 12/1: cancelled (day of entrance exam)
- Part 2: Discussion about applications
  - #8 (12/8): readings 1
  - #9 (12/15): readings 2
  - #10 (12/22): readings 3
- Cont. Part 2 or Part 3: Practice work
  - #11 (1/5)： readings 4 or Consideration about applications, group making
  - #12 (1/12)： readings 5 or exercise 1
  - #13 (1/19)： readings 6 or exercise 2
  - #14 (1/26)： readings 7 or exercise 3
  - #15 (2/2)： readings 8 or Final presentation about outcomes

#1 (10/6): Guidance

Syllabus: [ Japanese | English ]
Slack: dmt-ie-univ-ryukyu.slack.com for annances, chattings, etc.
- 10/13: cancelled
- 10/20: readings of Data Mining: Practical Machine Learning Tools and Techniques, 3rd Edition.
[Lecture documents]
- Guidance: dm-theory-2016-01.pdf
grouping
- 2 or 3 students for one group.
- 2 groups (presentations) for each week.
  max 45 minutes for one group. Each group has 30-35 minutes for presentation, 10 minutes for discussion.
assignments
- #2: (1) from chapter 1.1 to 1.2 (18 pages), (2) from chapter 1.3 to 1.7 (18 pages).
- #3: (1) from chapter 2.1 to 2.3 (13 pages), (2) from chapter 2.4 to 2.5 (10 pages).
- #4: (1) from chapter 3.1 to 3.4-Associatio rules (11 pages), (2) from chapter 3.4-Exceptions to 3.7 (11 pages)
- #5: (1) from chapter 4.1 to 4.2 (15 pages), (2) from chapter 4.3 to 4.4 (16 pages)
- #6: (1) from chapter 4.5 to 4.6 (15 pages), (2) from chapter 4.7 to 4.10 or 4.11 (15 pages)
- #7: (1) from chapter 5.1 to 5.6 (16 pages), (2) from chapter 5.7 to 5.10 or 5.11 (23 pages)
Google documents
Homework
- read and ready your assigned references to explain/introduce. You must make any presentation documents and uploat them to google documents until the previous day (10/19).
- e.g., PowerPoint (slides), commentary documents, demos, etc.

#2 (10/20): What is Data Mining? Some definitions, simple examples and applications.

(1) from chapter 1.1 to 1.2 (18 pages)
(2) from chapter 1.3 to 1.7 (18 pages)

References

Books
- Data Mining: Practical Machine Learning Tools and Techniques, 3rd Edition
Frameworks or Tools to use Machine Learning
- Weka 3: Data Mining Software in Java
- scikit-learn: Machine Learning in Python
- The R Project for Statistical Computing
- Apache Spark MLlib
- Caffe
- Chainer
- Google TensorFlow
- SkFlow for Tensorflow
- CNTK: Computational Network Toolkit by Microsoft
- DSSTNE: Deep Scalable Sparse Tensor Network Engine by Amazon
Others
- The 38 best tools for data visualization (Chart.js, Tableau,,,)