News Mining
Analysing the top trends in news from news headlines from ABC news Australia.
The data contains news headlines from ABC news Australia.
Tools used:-
- Pycharm (Community Edition)
- Numpy
- Panda
- Matplotlib
- sklearn
Algorithm used:-
I have used the k-means cluster algorithm to calculate the most common words to find out what the news headlines were about.
Observation:-
Using the tf-idf approach, the following words were the most common amongst the various headlines:-
polic, say, plan, investig, probe, council, govt, offic, search, hunt, death, attack, car, drug, arrest, water, miss, seek, shoot, minist, driver, union, need, crash, hous charg, man, murder, face, assault, drug, polic, child, sex, woman, court, teen, death, stab, drop, alleg, rape, men, attack, guilti, bail, shoot, sydney, fatal, driver new, zealand, law, year, plan, open, council, polic, home, hospit, centr, deal, set, hope, australia, appoint, look, announc, chief, govt, south, minist, say, mayor, welcom win, council, govt, australia, report, kill, court, warn, urg, fund, australian, nsw, water, open, interview, crash, death, chang, face, claim, qld, wa, year, day, sydney man, jail, court, polic, die, murder, arrest, miss, stab, accus, kill, guilti, car, crash, attack, face, death, assault, search, plead, shoot, shot, sydney, dead, bash
Link to the data file:-
https://www.dropbox.com/sh/52bv1sy64dn1mo9/AAACUscB7n1-PmweirjjpU7aa?dl=0