WriteRight

• 

  Analytics

• 

  Home page

Inspiration

Based on our personal experience and the number of international students at UC Davis, we realize there was a major need for more ESL education. Students especially struggle at the college writing level, and we believe our tool can help solve that.

What it does

Gives writing prompts for ESL speakers and performs analysis using machine learning techniques and NLP APIs, giving detailed feedback and charts that help users improve their writing skills.

How we built it

Node.js, C++ (for markov decision process), languagetool API (for text analysis), charts.js (for visuals), HTML/CSS

Challenges we ran into

We struggled with finding enough data to train our model to accurately determine whether word combinations were actually non-native sounding, as opposed to not in our training set. We also struggled with CSS, because it's CSS.

Accomplishments that we're proud of

We had relatively clean code for a hackathon, we were able to build out all the features we wanted and learned a lot about web technologies.

What we learned

We learned how to create servers and perform routing in node.js, POST & GET requests, how n-grams work, and lots of front-end web.

We also learned about the ESL community and the challenges they face.

What's next for WriteRight

Expansion to more advanced feedback metrics.

Built With

• api
• c++
• css3
• html5
• javascript
• machine-learning
• markov
• node.js