fiscal 2019 : 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 : Guidance
- Case 1: Introduction to data mining
- #2 : What is Data Mining? Some definitions, simple examples and applications.
- #3 : Machine Learning and Statistics, aspects of personal information.
- #4 : Instances and Attributes as Input, and Knowledge Representations as Output
- #5 : The basic methods and algorithms 1
- #6 : The basic methods and algorithms 2
- #7 : Credibile experiment designs and evaluation ways
- Case 2: Introduction to machine learning
- #2 : What is Machine Learning?
- #3 : Classification by perceptron learning
- #4 : Classification models in Scikit-learn
- #5 : Building good training sets and pre-processing
- #6 : Compressing data via dimensionality reduction
- #7 : Model evaluation and Hyperparameter tuning
- #8 : reserve day (e.g., adjustments for Part 2)
- Part 2: Discussion about applications
- #9 : readings 1
- #10 : readings 2
- #11 : readings 3
- #12 : readings 4
- #13 : readings 5
- #14 : readings 6
- #15 : readings 7
#1 : Guidance
- Syllabus: [ Japanese | English ]
- Slack: dmt-ie-univ-ryukyu.slack.com for annances, chattings, etc.
- Google drive (required @ie.u-ryukyu.ac.jp or authorized google account)
- [Lecture documents]
- Guidance: Data Mining Theory: Guidance
- 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: see assignment page on the google drive also.
- choose book1 or book2 by voting (majority).
- book1: ref1-DM-TOC.pdf (Data Mining: Practical Machine Learning Tools and Techniques, 3rd Edition)
- book2: ref2-ML-TOC.pdf (Python Machine Learning)
- make an assignment by randamly.
- choose book1 or book2 by voting (majority).
- 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.
- e.g., PowerPoint (slides), commentary documents, demos, etc.
References
- Books
- 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,,,)