Training Arc

Inspiration

Many students struggle to translate math word problems to their respective equations that the students are more familiar with. Furthermore, with online learning, it is hard to give personalized feedback to the students to help familiarize them with the proper steps to take to solve word problems. Because of this, we wanted to develop a web application that could assess a student's abilities to solve a variety of word problems, and then provide step-by-step solutions as to how to solve word problems that they may be having trouble with.

What it does

Our project takes a word problem submitted by the user as input and runs it through the Wolfram Alpha Full Result API to generate a JSON with the solution and step-by-step process if available. The website then takes the JSON and parses it, printing the results and the step-by-step solution to the page. We also have a separate page with a diagnostic quiz so that the user can assess themselves by solving tagged word problems and seeing what their score is, as well as what percentage of each type of problem they got correct.

How we built it

We used a combination of ReactJS, NodeJS, and Python to create this web application. We ran API calls through Python and had NodeJS interact with the Python script to send the information to the React component.

Challenges we ran into

The two major challenges that we ran into were connecting the Python script to the backend and connecting the backend to the front end, and deploying the website onto heroku. This was the first time all four of us have done web development, so most of our time was spent looking up proper documentation and code that would work for our purposes. Furthermore, it was the first time we used heroku, so debugging our code to work with it took extra time.

Accomplishments that we're proud of

We are proud of the fact that we were able to make a full-stack website, even though it is all of our first times developing a website from scratch.

What we learned

We learned how to use the Wolfram Alpha API, as well as other technologies such as Flask and heroku. We also learned how to connect the backend server with the frontend client, making our website more dynamic.

What's next for Word Problem Solver

We hope to expand the functionality of our word problem solver to be able accept a larger variety of questions. Furthermore, we would like to generate word problem questions and tag them using an online word problem API to use for the quiz questions. We would like to make an automatic topic detection for all kinds of word problems to tag each of the randomly generated word problems.

Built With

Submitted to

SD Hacks 2021
- Winner Wolfram Alpha Prize

Created by

I worked on the frontend of our project. I focused on making sure the style was consistent throughout our webpage. This was my first time working with Node.js and I learned a lot.

Dorothy Chou
I worked on the front-end, and helped connect the backend to the front end. I mainly focused on parsing the data from backend so that it could be used on our webpage. I learned a lot about full-stack development through this project.

Shirley Qi
I connected the backend with the frontend using Flask, and deployed our site with Heroku. This was my first time working with these technologies, so I'm glad that our project is functional. :)

Ryan Niu
I worked on the python script for the Wolfram Alpha API and developed the code to extract topics from the step-by-step solutions. I also worked with Shirley to develop the diagnostic quiz code.

Matthew Cho