/Predicting-Kickstarter-Campaign-Outcomes-Using-NLP-Feature-Engineering

Turning raw kickstarter text data => Campaign predictions using SpaCy, Scikit-learn, SQLAlchemy, SQLite3 & XGBoost Classifier (feat eng = Bag-of-Words, Tfdvectorizer)

Primary LanguageJupyter Notebook

alt text

🕵️‍♀️ Classifying Kickstarter Campaigns Utilizing NLP Feature Engineering Techniques 📚

👋 Hi!

My name is Mikiko Bazeley and this is my second capstone project: 🔬 Classifying Kickstarter Campaigns Utilizing NLP Feature Engineering Techniques. 📖

From Oct 2018 to April 2019 I completed a number of projects, including this, as part of the Springboard Data Science Track. 🧠

For this project I incorporated NLP feature engineering techniques (Bag-of-Words, N-Grams and TFID-vectorizer) & the SpaCy 👩🏻‍🚀 package to use both quantitative & text data to predict outcomes of Kickstarter campaigns 💡.

To find out more about this project, check out the attached presentation below!

☑️ Jupyter notebook for project: LINK

☑️ Final write up: LINK

☑️ Slides presentation: LINK

For more information about my Springboard work: 📝 All of the documentation, code, and notes can be found here, as well as links to other resources I found helpful for successfully completing the program.

💬 For questions or comments, please feel free to reach out on LinkedIn.

⚠️ If you find my repo useful, let me know OR ☕ consider buying me a coffee! https://www.buymeacoffee.com/mmbazel ☕.


alt text


alt text


alt text


alt text


alt text


alt text


alt text

alt text


alt text


alt text

alt text


alt text