deniederhut / weather_report

Unsupervised sentiment / emotion classification

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Title Author
Weather Report
Dillon Niederhut

DOI Build Status

Description

Weather Report is a collection of sentiment classifiers and location-based API readers. It is the engine that powers the Berkeley Mood Twitter Bot1

Installation

git clone https://github.com/deniederhut/weather_report.git
cd weather_report
python setup.py install

Use

See bots/jaco.py for an example of a CLI client. Running:

python jaco.py top_10_us data/counts.csv --classifiers count_dict polar_summary

Yields:

head counts.csv
city,year,month,mday,wday,hour,source,type,variable,value,n_items,n_terms
Pythopolis,2016,3,2,2,11,<class 'weather_report.classifiers.CountDict'>,count,angry,0,1,12
Pythopolis,2016,3,2,2,11,<class 'weather_report.classifiers.CountDict'>,count,sad,1,1,12
Pythopolis,2016,3,2,2,11,<class 'weather_report.classifiers.CountDict'>,count,loving,1,1,12
Pythopolis,2016,3,2,2,11,<class 'weather_report.classifiers.CountDict'>,count,afraid,0,1,12
Pythopolis,2016,3,2,2,11,<class 'weather_report.classifiers.CountDict'>,count,happy,1,1,12

About

Unsupervised sentiment / emotion classification

License:BSD 2-Clause "Simplified" License


Languages

Language:Python 96.2%Language:Shell 3.8%