Inspiration
Everyday thousands of daily traders spend tens of hours pulling up information on current stock prices, history of the stock and expert recommendation of the stock. I wanted to automate this process, which might only save them a few seconds on each stock they search up, but in the grand scheme would help them save a lot of time and effort. This would especially helpful to people who trade as a hobby or a side job, as they do not have the time resources as a full-time trader.
What it does
Aisha is a bot made using Wit.ai which helps you to get up and running with basic information about stocks of a company. You can search stock prices, history of the share, majority holders of the share and advice on the share by industry experts and so on.
How I built it
Aisha is built using the python API for Wit.ai. Aisha also utilises the yfinance API to fetch data relating to the stock.
Challenges I ran into
The first challenge was to form a problem statement. Once I had that sorted out, figuring out python for wit was another challenge, which was actually quite simple.
Integrating with messenger was another challenge, but I overcame this by doing some research and finding some really helpful tutorials.
Accomplishments that I'm proud of
This is my first time working on a chatbot and it was really cool. Wit.ai is super easy and convenient to use. I am proud of figuring out how to get a basic chatbot up and running.
What I learned
This project was a huge learning curve for me. It gave me a lot of insight to me on how NLP works and it made me more curious to research further of NLP and important stuff that can be automated using chatbot. Sentiment extraction using wit seemed really cool to me, even if I didn't use it in my project.
What's next for Aisha - Bot to help everyday trader get stock prices
Aisha, in this form, is very basic and I would like to extend its functionality further. I think what I could do is to find a way to predict share price using Facebook Prophet. More functionality that can be added is to find the best performing stock and recommend it to the user and more helpful stuff like this.
Note
Please use NASDAQ name of stocks during testing, otherwise the program crashes.
Log in or sign up for Devpost to join the conversation.