Tech Glimpse

Take a glimpse at your next project!

Comment

Inspiration

In today's era of continuous innovation newer technologies keep emerging. However, students find it difficult to select the appropriate field they would like to pursue. Our project recommends best sources for a beginner to start their journey into upcoming technologies like Web Dev , Julia, AI , Virtual reality etc.

About

Technology is a booming sector, with huge job opportunities for many. If you're trying to enter this field, it can be intimidating figuring out which side you should learn first. With projects being the best way to learn things, we're using a logistic regression classification model in order to recommend to the user what project would suit them well, and with what projects they should start off with. In order to use this web application, answer the questions as best as you can, and check out a recommended project.

How we built it

In order to build the classification model, we had to gather a dataset. We sent a survey and had multiple responses. With these responses, we had trained a logistic regression model to predict what the user should learn. However, the logistic regression model was not able to make it to the final project. A Google cloud trigger function receives the input from website form, processes it and stores the output in Google firestore. For the Frontend, we built a website using REACT. We used Github search API to give the desired project recommendation

Challenges we ran into

Unable to find a dataset for our needs, we went ahead and created our own with a survey. However, given the time constraints, we were unable to get as many responses as we would've liked to have. As a result, our classification model is not as accurate as we would've liked it to be.

Google cloud fuctions were not returning JSON but HTTP response . So we had to figure out an alternative and decided to go for firestore trigger instead of http trigger.

Reading data from request and converting it to numpy array was also a challenge to figure out.

Understanding firestore structure and adding json response to sub collection.

Accomplishments that we're proud of

Deploying Google cloud functions Making a catchy website

What we learned

Usage of google cloud functions NodeJS ReactJS Creating and preparing a dataset Logistic Regression

Built With

Share this project:

Updates