Mo'Mentum

First draft of the demo for my insight project

Transcript

1. Consulting Proposal September 4, 20XX Mo’Mentum Helping you optimize your momentum for social change Peter Winslow Insight Data Science Fellow
2. Mo’Mentum Mission Statement: Predict the likelihood and time scale for petition success Petition Mo’Mentum Probability of success Time scale to reach signature goal
3. Data Source Change.org sitemap Petition urls Petition id’s Petition text and metadata Change.org api Over ~ 40,000 Petitions
4. Sentiment Feature Engineering Text data Metadata Stopwords, Lemmatization Grammatical Structure Word/Sentence Count Signatures Time Petition topic Petition target Month of petition submission Signature Accumulation Rate = Timestamps Signature count Wanted: A metric for petition momentum
5. Algorithms: Classification Random Forest Classifier (Scikit-Learn) Predict success/failure of petition 18 features after selection Reasons for choosing: • Lots of complication yet resistant to overfitting Challenges: • Class imbalance in the data Validation: Train-Test-evaluation split with 5-fold CV
6. Algorithms: Regression GradientBoostingRegressor (Scikit-Learn) Predict signature accumulation rate 17 features after selection Reasons for choosing: • Many features, highly non-linear, can return predicted “quantiles” Challenges: • The right evaluation metric? Validation: Train-Test-evaluation split with 5-fold CV
7. About me Peter Winslow The Professional PhD + 1 Postdoc in theoretical High Energy Physics and Cosmology. Specific interests: building models to explain the existence of matter in the Universe + searching for these models in next-generation super colliders! Peter Winslow The New Father! Kiana Winslow, born Nov. 29th 2016