Skip to content

Using Decision Tree model,predict if a given review is Positive or Negative and trying various methods to improve accuracy.

Notifications You must be signed in to change notification settings

adityamatt/Sentiment-classification-using-decision-tree

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Author: Aditya Tiwari
Entry: 2016csb1029
Email [email protected] for any queries
##################################
The code is in python ,so make sure your system has python on it
OS:linux or linux-like

To run the code on Ubuntu 16.04 to reproduce the results
Go inside the folder Named "code"
Open Terminal in there
Enter: python main.py <EXP_NO>
where EXP_NO is from [2-5] for the respective experiments whose results are shown in Report
##################################
Experiment 2:

It asks you a input value of Information Thresh to be used to make the Decision Tree Then it finds the accuracy of it on 10 different test sets and returns the average of 10
Similarly it asks for a depth threshold and does tha same as above
##################################
Experiment 3:
I have saved the tree in files,Now the script asks for a choice for 0.5,1,5,10 percent Noise and simply Loads the Corresponding Noise Set and makes a tree using and and then
returns the accuracy of it using single test set
##################################
Experiment 4:

I have saved the Tree created initially and simply prune it greedily and print the result

##################################
Experiment 5:
It asks number of trees in the forest and starts creating it and prints the accuracy of forest after each insertion of tree
It also asks if you want to creat pruned tree or not in the forest


About

Using Decision Tree model,predict if a given review is Positive or Negative and trying various methods to improve accuracy.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages