Assignments in Data Mining I - 1DL360
Fall 2011
Contents
News
- 2011-09-04: Instruction is updated / TT
- 2011-08-31: Links updated /LM
- 2011-08-29: Page created /EZ
Assistants
- Assignment 1: Thanh
Truong
Office hours: Monday and Friday 13:15 -15:00
- Assignment 2: Lars
Melander, Sobhan
Badiozamany
Office hours: Lars Wed 13:15 - 15:00, Sobhan Tuesday 10:15 - 12:00
- Assignment 3: Andrej
Andrejev
Office hours: Tuesday and Thursday 13:15 - 15:00
Tools
Here follows links to tools used in the assignments and instructions to install and configure them.Working directory
- [Lab] indicates that the following instruction should be applied when logged in at the computer in the computer lab room at the university.
- [Home] indicates that the following instruction should be applied at home with your own computer.
- [Lab] At computer lab
rooms: H:\DataMining
Because data on H:\ is auto synchronized, you can access and work on this folder at any Windows computers in the campus. You also can access your files via Studentportalen - My Files
- [Home] Your own computer: C:\DataMining
Amos II
- Amos II tutorial [ slides + script + data ].
- Download Amos II for Windows and unzip it into DataMining directory. amos2.exe should be located in "H:\DataMining\amos2\bin\".
- Note: After you have installed Xemacs and updated init.el, the path to amos2 is set when Xemacs is started. init.el contains a code snippet that sets the PATH variable.
- Amos II User's Manual. The section about vectors is important for Data Mining.
Xemacs
- A good text editor that is customizable for Amos II.
- [Lab] It is pre-installed in all computers in labs.
- [Home] Download and install Xemacs [ Xemacs ].
- Customize Xemacs for Amos II
- Start Xemacs :
- [Lab] Start - All programs - Students - Xemacs
- [Home]
- Go to Options - Edit Init File . If the file does not exist, click Yes
- Overwrite the content of init.el file by [Lab] init_lab.el or [Home] init_home.el
- Save and Exit.
- Start Xemacs :
- It is often very practical to run programs under the xemacs shell
as it provides full editing and logging facilities for shell commands
and program interactions.
- Ctrl-x 2 splits the window into two windows(buffers) vertically, Ctrl-x 3 splits horizontally. One windows is for script, the other is interactive shell.
- Press Alt-x and type "shell" to start Xemacs shell.
- To copy code to the shell, place the cursor at the beginning of a function and press F2.
GNUPLOT (assignment 2 & 3)
- Amos II uses GNU Plot to plot graphs of 2 and 3 dimensions. GNU
Plot is installed in the PC Lab. Follow these steps to install GNU Plot
on your PC:
- GNU Plot for windows. Download and install gp426win32.zip.
- Add the GNU Plot bin
directory to your PATH:
- Right click "My Computer" (you should find it in the Start Menu), select Properties
- Choose the Advanced tab
- Click Environment Variables
- Under System variables, select Path, click Edit
- Add the path to the location of wgnuplot.exe. If GNU Plot was installed in C:\Program Files\gnuplot\bin, you will add ;C:\Program Files\gnuplot\bin to the Path variable value (notice the semi colon separator).
AmosMiner
- Download AmosMiner [ AmosMiner ] and unzip it into DataMining directory
- Start Xemacs's shell : Alt-x shell
If you have followed "Working directory", "Amos", and "Xemacs" 's instructions correctly, you will be at [Lab] H:\DataMining\AmosMiner or [Home] C:\DataMining\AmosMiner.
Otherwise, please go back and check the instructions again. Make a database image by running: install.cmd
- Run the system with : amosMiner.cmd
- Make your own scripts extending and modifying the system and load it into amosMiner : e.g. < 'a1.osql';
- Follow exercise specific instructions
Assignments
Examination
Important information about examination of assignments [ DOC | PDF ]Assignment 1: Classification using k-Nearest Neighbor
- Instructions [ PDF ]
- Tutorial slides [ PDF ] See also the lecture notes on Chapter 4.
- Lab (optional, but highly recommended):
- Answer form: [ DOC ][ PDF ] Answer the questions and bring the form to the examination.
- Examination (mandatory): Sep 20th 8:00-12.00 and Sep 21st 13:15-17:00 in P1345. Sign up outside P1346.
Assignment 2: k-Means and DBSCAN clustering
- Instructions [ PDF ]
- Tutorial slides [ PDF ] See also the lecture notes on Chapter 8 & 9 and the PCA talk.
- Lab (optional, but highly recommended): Sign up outside P1346.
- Answer form: [ PDF ] Answer the questions and bring the form to the examination.
- Examination (mandatory): The examination is scheduled for 26
September.
Location is P1345. Sign up sheets will be available outside P1346. - Related articles:
Assignment 3: Frequent itemset and association rule mining
- Instructions [ PDF ]
- Tutorial slides [ PDF ] See also the lecture notes on Chapter 6 & 7.
- Answer form: [ PDF ] Answer the questions and bring the form to the examination.
- Examination (mandatory): The examination is scheduled for October 5 and 6. Sign up outside
P1346.
--------------------------------------