About the Practice Problem: HR Analytics Challenge category

Use this category for the discussion related to contest: Practice Problem: HR Analytics Challenge. Feel free to share your approach and ask your questions here.

For more information, visit:

Welcome to Data Science, Analytics and Big Data discussions

Reagarding the test file of HR analytics(https://datahack.analyticsvidhya.com/contest/wns-analytics-hackathon-2018-1/)

I am not able to download test file , appreciate if anyone can help

Hi Team, Can anyone explain me about the target variable values 0 and 1. 1 means promoted and 0 not promoted or vice versa.

Promoted -1, Not Promoted - 0

Thank you Ankit.

Hi Team, Could you please elaborate about previous_year_rating?

How it is calculated? Is it cumulative of other parameters like KPIs_met >80%/awards_won?/avg_training_score from last year?

Thank you!

Previous year rating is basically the overall rating for the employee during the last financial year. It has nothing to do with current year’s KPIs met / awards won.

I am working on HR Analytics Challenge ,Could you please help how to generate a submission file in R as i have created the model using Tree ,but how to map this model to submission form means mapping employee id to is prompted or not.

Any link present in Analytics Vidhya would be greatly appreciated.


I did not understand one thing…what is the difference between solution file and code file here

1 Like


Kindly Help.


I m not able to read file and getting this error
XLRDError: Unsupported format, or corrupt file: Expected BOF record; found b’employee’

You can read a csv file by Pandas as a DataFrame object with this code:
import pandas as pd
data_df = pd.read_csv(‘file.csv’)

You can read a csv file by Pandas as a DataFrame object with this code:
import pandas as pd
data_df = pd.read_csv(‘file.csv’)

Hi team,

Could you please upload a solution in R?

I was doing the data analysis and was surprised to the see the below fact

folks who have not Met the KPIs are being promoted and vice versa …It is interesting that meeting criterias or key performance indicator is a basic criteria for any organization and when it comes to promotions ,I thought it will be 100% true …

Since this is not a real time data ,I suppose we might find more interesting facts like this often …

Guys i need help i have is_promoted column “Y” present in training data set but same column is not present in testing data set how can i predict for is_promoted" Y" in test data set

1 Like


I tried making a SVM model for this HR Analytics, though I did the sampling on Train data to try and model was working fine, but when I am trying to predict on the original Test file I am getting following error, can someone advice.
Error in eval(predvars, data, env) : object ‘departmentFinance’ not found

I need solution for this competition. from where i can get the solution for this competition please help me.

Hi Vijay,

0- Not Promoted
1- Promoted
and we can observe the promoted percentage is very less compared to the not promoted.

Thank you,

© Copyright 2013-2019 Analytics Vidhya