First Meeting 11 28 2018 - abpwrs/thee-flying-chicken GitHub Wiki
What do we want to do????
A stock market tracking algorithm
- data analysis
- web dev
- possible deep learning
Web Scraping
- Scripy or JS library What do we do with scraped data?
- Twitter Sentiment analysis is standard
- Scraped market-related data and see if there is a correlation to stock price
General WebDev -- based on technology
- Javascript --> MEAN
- Python --> Flask Django
Stock modeling with language processing WebApp
- Scrape text data to do analysis (NLP --> natural language processing, Frequency-based analysis)
- web-based interface
- Acess to current stock prices and trends
- Sentiment analysis on companies (text field to enter the companies name)
- Track sentiment over time
- Machine Learning/Data Analysis playground for stocks --> (percent certainty of increase or decrease)
- Paper Trading -- no real money
What will we need to learn (General)
-- web scraping (getting data) -- web dev (present data/ have cool features) -- sentiment analysis/NLP/NLU
What will we actually need to learn though
Languages
- Javascript for web dev
- Python for data analysis
- Py/JS for web scraping (scrapy/JSJSJSJSJSJS)
Spesific Libraries
- SCRAPY/JSJSJS for scraping
- MongoDB -- Database for webdev
- Node/Express -- backend
- Angular/React -- frontend
- sklearn -- python machine learning library
- pandas -- data manipulation
- numpy/scipy -- efficient data storage and access
- tensorflow/keras -- deep learning
- NLTK -- natural language tool kit
General
- Cloud service to host (GCP, Heroku, AWS)
How do we approach this
We are going to start with web scraping as a group The future plan is to work through area by area together
Action Items
AP --> Throw tutorials at JT DR and DC
JT --> scrapy -- add a file to this repo
DR --> scrapy -- add a file to this repo
DC --> scrapy -- add a file to this repo