You are new to HR Analytics, data science, machine learning and are looking for challenges to get experience with real-life examples.
In your second challenge, you will perform text analytics and make recommendations on how best to write job descriptions! How cool is that!
Text Analytics is a powerful tool, especially in the Human Resources Department, where a lot is based on texts. Text Analytics can help with several HR processes, for example:
- Semantic search within application forms
- Quick reading of employee appraisals
- Content extraction from text-based engagement surveys
In this challenge, we ask you to submit an R markdown or a Jupyter Notebook that analyses the text contained in 23 winning job descriptions provided by LinkedIn. It will be your recommendation on how best to write job descriptions! You are free to analyse whatever aspect you wish.
The job descriptions from LinkedIn represent proven practices and ready-made posts for connecting with candidates, so something definitely worth analysing in detail. On this occasion, you will be analysing job descriptions, but in real life, you might be analysing the characteristics of what makes applications forms successfully. Please click here if you want to know more about the origin of the dataset.
You are free to get inspiration from Julia Silge’s “Text Mining with R” book or quickly learn more about text mining.
How are we going to evaluate the submissions?
We will focus on the following key elements:
- the R Markdown file or the Jupyter Notebook is your final recommendation containing a reproducible analysis of the 23 job descriptions,
- your recommendation needs to be clear and easy to understand,
- appropriateness and appeal of visualisations,
- reusability of the code for creating similar analyses.
Public announcements
On 1st March 2019, the competition closed. The winner received its gift voucher. However, you can still download the dataset to have a try in analysing the job descriptions.