This project takes solar installer Tweets to better understand and classify their marketing strategies. The analysis utilizes tf-idf weighting, k-means clustering, and topic modeling/LDA to determine key topics solar installer marketing. As tweets have a time reference, we look for trends in marketing over time across both the entire industry and among individual companies.

• scrape_teets/ python script to get tweets from solar installer twitter accounts
• Solar Marketing Analysis.ipynb jupyter notebook with scripts and outputs for text processing, k-means clustering and topic modeling
• data/ contains data and script for analysis
• analysis/ future analysis to be integrated into online analysis

• Import tweets from the top US residential solar installers to an SQL database
• Manually tag tweet to better understand and classify tweets.
• Apply Natural Language Processing (NLP) tools to process the text data for each company's description: word tokenization, TFIDF
• Condense terms using Non-negative matrix factorization (NMF) for topic modeling
• Assign TFIDF term scores and NMF topic scores to each business and investor (average scores across all companies invested in)
• Use cosine distance as a metric to measure similarity between solar installers

Scripts were written in Python 2.7. You'll need the following modules:

matplotlib >= 1.5.1  
nltk >= 3.1
numpy >= 1.10.1  
pandas >= 0.17.1  
python-dateutil >= 2.4.2
scipy >= 0.16.0
seaborn >= 0.6.0
sklearn >= 0.17
spacy >= 0.100
statsmodels >= 0.6.1
pattern >= 2.6
vaderSentiment 

To install modules, run:

$ pip install <module>

Thanks to:

• Twitter REST API
• List of Top 100 solar installers