Final Project: Job Recommender

Introduction: Kaggle conducted a survey to gather information on the state of data science and machine learning around the world. Glassdoor data science job dataset …. Online resume dataset ….

Purpose: In this project, we present a recommender system designed for job seeking and recruiting purpose. The proposed recommender system aims at leveraging the jobs and companies that are important for a target candidate and vice-versa. To meet this objective user profile such as job descriptions and candidate resumes are examined along with other user inputs. The recommendation approach is modeled on content-based analysis …. The dataset consisted of survey dataset of 23,859 respondents and job postings from Glassdoor and resumes from Post Resumes Free

Research Question: How reliable is the job description in the listings that a candidate who had applied to the position in predicting the strength of their preference for the skills in the resume ? How reliable are company ratings in predicting candidates preference towards the job?
Approach: We have applied the following process to organize our workflow:
  1. Business Understanding - TBDs
  2. Data Understanding - …
  3. Data Preparation - …
  4. Modeling - …
  5. Evaluation - …
  6. Integration and Deployment - …

Instructions

This project is a proof-of-concept(POC) with certain assumptions on the data. For this implementation, following are the guidelines …

Load the data

Filter the datasets

Join the dataframes

Summaruze the data

Data Modelling

In our system, each entity has a profile. We use the content- and interaction-based relations to connect entities.

Profile match: Bidirectional relation based on the hypothesis that if a candidate’s resume matches a job description, they are probably interested in each other.

…..

Calculate ratings based on … algorithm

Training and test data

Build Recommender

Connect the entities

Find similarities

Examine the relationships

Test the Recommender

Create the Shiny UI

Authenticate the user

Gather Inputs

Present visualizations

Conclusion

Evaluation: Our recommender predicted …

Conclusion: …

Q & A