Hmine

This is a port of the Hmine algorithm in Python.

Requirements:

Some sample datasets are given:

1. Market_Basket_Optimisation.txt
2. Random_Test.txt
3. Chess_data.txt

Any .txt dataset with items separated by delimiter (space) " " can be used.

The Project_Main.py runs both the H_mine and Apriori algorithms for all three datasets and outputs the result into files.

To Run the program :

1. Open Terminal(Mac) or Cmd(Windows).

2. Change your directory to the same directory where the program is stored.

   Command: cd /path/to/program

3. Enter The Command:

python Project_Main.py  

4. After Running, the Program will read all the 3 datasets one by one and run H-mine and Apriori with arbitrary value of minimum_support given in the file.

5. The output file is generated indicating the name of the dataset and the algorithm used. Confirmation of storage is given in the terminal with a prompt.

NOTE: Only 5 files are generated since the Apriori algorithm is not triggered for chess.txt.

Instruction: To Run algorithm H_Mine_Algo.py or Apriori_Algo.py Invidually.

NOTE: Python interactive shell is used to run the runHmine() or runApriori().

To Run the module :

1. Open Terminal(Mac) or Cmd(Windows).

2. Change your directory to the same directory where the program is stored.

   Command: cd /path/to/program Command: "python" (to enter python interactive shell) NOTE: use "pyhton3" for Bluenose

3. Enter The following Command in python interactive shell:

   import Apriori_Algo

   import H_Mine_Algo

   Apriori_Algo() and H_Mine_Algo() takes 3 parameters:

   1. Input_file_name.txt
   2. Output_file_name.txt
   3. Relative support (Valid range 0<x<1)

   For example,

   Apriori_Algo.runApriori("Random_Test.txt","Random_Test_Out.txt",0.1)

   H_Mine_Algo.runHmine("Random_Test.txt","Random_Test_Out.txt",0.1)

4. After Running, the program will read the input file and run the corresponding algorithm with the given minimum_support.

5. The output file is generated with the given name. If completed successfully, the Program will print a completion message.

CONTACT INFO

Email: aman.jaiswal.dal.ca