Topic Modeling

Topic Modeling using Latent Dirichlet Allocation and Non Negative Matrix Factorization

Required:

Python 3.5
Sklearn

Execution:

python topicmodeling.py

Dataset

data file (20 Newsgroup Dataset)

NMF Algorithm Output ( 5 Topic ):

Topic: 0

people don just think like god know time say good

Topic: 1

thanks windows does edu use know mail file card program

Topic: 2

game team year games season players play hockey win league

Topic: 3

key chip encryption clipper keys government use escrow public algorithm

Topic: 4

drive scsi hard drives disk ide controller floppy cd bus

LDA Algorithm Output ( 5 Topic ):

Topic: 0

people don think just god like know say time did

Topic: 1

thanks use windows know does like drive mail card using

Topic: 2

mike radio tv hear jim article station com city chicago

Topic: 3

car game year team games good new like 00 sale

Topic: 4

bike edu soon deleted david stuff dod insurance road just

Screenshots