Social Media recruiting with NLP

Challenge

Online recruiting through social media platforms allows companies to access a large pool of candidates today. However, the process of identifying promising profiles on relevant platforms is time-consuming and often results in a low response rate. As a professional provider of social media recruiting, our client faced this situation daily.

Approach

Since the suitability of an online profile depends on the text of the respective job posting, a word embedding combined with a neural network from the field of natural language processing (NLP) was first developed to relate the requirements of the profiles to those of the job posting. Building on this, machine learning models can evaluate all available online profiles regarding their relevance to a job posting and their likelihood of response. The data obtained was also used to automatically enhance the text of existing job postings with requirements such as knowledge of specific programming languages, increasing the attractiveness of a posting for candidates.

Result

The implemented recommendation engine supports our client's recruiting process by suggesting promising profiles and optimizing existing job postings. This significantly increased the response rate and reduced manual effort.