Tal is an ML-powered utility for recruiters and hiring managers
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
.vscode
config
images
public
routes
server
views
.DS_Store
.gitignore
LICENSE
Procfile
README.md
app.js
final.md
milestone2.md
milestone3.md
milestone4.md
milestone6.md
package-lock.json
package.json
proposal.md
test_data.csv
train_data.csv

README.md

Tal

A machine learning powered tool to help recruiters.

Group Members + Contributions

Nikhil Bose: Developed and contributed to the UI + Design, the recommender, JS to transfer information and backend/database, video

Ribhu Lahiri: Developed the ML algorithm for the candidate recommender, built the REST API and designed the data schema of the database

Alvin Vanegas: Developed the UI and primarily wrote HTML and Javascript code to link the web pages together

Source Code Files

  • index.html: landing page of our webpage, contains front end code of our login page.
  • index.js: javascript code for our login page, handles user input for entering out web page and sends it to the database.
  • profile.html: front-end code for the user's profile which contains past models they have created
  • profile.js: javascript code for the user's profile, sends the model's name to database.
  • input.hmtl: front-end code for the input page, contains the necessary inputs to create the model.
  • input.js: javascript code that handles the user's inputs and sends it to the database and to the script to run the recommendor.
  • output.html: front-end code to display the data visualizations and the results of the recommendor base on the users input.
  • output.js: contains the javascript code to pull the necessary data from the database to create the data visualizations and to output the results of the model.
  • model (in server/model/): contains the data schema for the required data models for student, recruiters and groups.
  • controllers (in server/controller/: contains the queries to serve data from the database through a REST API
  • perceptron: a simple 2-layer perceptron, with an input layer of 10 units. Trained on given company data for binary classification of potential hires
  • test_data.csv - test data file
  • train_data.csv - train data file

Demo Video

https://youtu.be/S3X9qhw-OTY