garimasarangal / non-fraudulent-job-recommender

Non Fraudulent Job Recommender

Our system filters out the fraudulent job postings before recommending the jobs based on the details provided by the user.

Note- Both the jupyter notebooks uploaded to GitHub contains the corresponding output with it.

Phase one - EDA and Fraudulent job Classifier

Kindly execute using https://colab.research.google.com/ and upload the necessary files as mentioned in the Input Data Files Section.

Code file - Phase_One_EDA.ipynb
Input data files - fake_job_postings.csv
Output - Preprocessed_data.csv
Description of code file - We performed EDA on our original dataset (fake_job_postings.csv) and the hypothesis mentioned in phase one of the Report are experimented. Also, we performed pre-processing on dataset. The output of this notebook is a dataset (Preprocessed_data.csv)

Creating a Job Recommender

Kindly execute using https://colab.research.google.com/ and upload the necessary files as mentioned in the Input Data Files Section. Code file - Filter_and_Recommender.ipynb
Input data files -

• fake_job_postings.csv
• Preprocessed_data.csv
• User_Resume.docx

Output - Recommended_jobs.csv
Description of code file - We have evaluated different ML models - Logistic Regression, K-Nearest Neighbors, Support Vector Classifier, Random Forest Classifier and Decision Tree Classifier using AUC-ROC score. Created fruadulent job filter using Decision Tree Classifier. Also, we preprocessed user resume and developed recommender system using cosine similarity.