Skip to content

ericchiang/weatherTweets

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

33 Commits

data

data

 
 

test

test

 
 

.gitignore

.gitignore

 
 

README.md

README.md

 
 

clean.sh

clean.sh

 
 

crossvalidate.py

crossvalidate.py

 
 

featextract.py

featextract.py

 
 

hmm.py

hmm.py

 
 

naivebayes.py

naivebayes.py

 
 

predict.py

predict.py

 
 

utils.py

utils.py

 
 

Repository files navigation

weatherTweets

UPDATE

Please refer to my "cloudy-tweets" repo for the improved solution.

Machine Learning solution for Kaggle.com competition "Partlmy Sunny with a Chance of Hashtags"

The challenge:
Given a training set of tweets, can we extract weather information?
The Machine Learning:
Use of a customized Bag-of-Words Naive Bayes algorithm. Modifing similar algorithms for spam filters, the feature space of a given tweet is simply the set of words which comprise it, similar to an email. However, for five of the target fields in-class confidence intervals (between 0 and 1) are offered as input, and required output rather than a discrete classification (such as ham or spam).

About

Machine Learning challenge from Kaggle.com.

Resources

Readme
Activity

Stars

1 star

Watchers

3 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages