Natural Language Processing to transform the recruitment process
Through the course of the COVID-19 pandemic, the centrality of novel technologies (in tasks ranging from advancing remote work procedures to contract tracing for possible patients) has been reiterated continuously. Businesses today have genuinely realised the need to invest heavily in their digital strategies in order to ensure that they remain competitive in a digital-heavy global environment over the long run. An area where businesses have greatly benefitted from the influx of artificial intelligence technologies is in the recruitment of new talent – especially in using Natural Language Processing (NLP) to deal with issues such as a large talent pool and poor screening options.
Phase-out the Old
Most available screening systems being used around the world suffer from drawbacks in the way that they search and filter for possible candidates. Current systems rarely (successfully) allow the screening of candidates based on unique combinations of skills or other factors with adequate flexibility. The problem becomes even more accentuated in job openings with a very large and diverse group of possible candidates. Certain job profiles end up with hundreds, or even thousands, of applicants, thereby adding to the stress of hiring managers and recruiters in reviewing so many profiles: causing them to resort only to simple keyword searches.
This causes the biggest challenge for any HR team: a personalised, high-touch approach to recruitment becomes almost impossible to pull off. Furthermore, this may also lead to the introduction of several biases into the hiring process as well, with several worthy candidates ending up overlooked without even being given a chance.
Despite the ‘humanness’ of HR, technology is quickly being adopted as the central means to empower teams, improve the hiring process and engage the workforce. Natural Language Processing systems now form the core component of any human-to-computer interaction and is used almost every step of the way: right from the screening of the first resume to the final exit interview.
Phase-in the ‘Ideal Candidate’
Recent advances in text mining have gone a long way in making the hiring process much more descriptive and holistic. Instead of using keywords, the search for the next employee can be made a conversational process using NLP technology. AI-based analytics company Phase does exactly this: ‘reducing time-to-hire and finding niche candidates via Text Mining and NLP’. An AI-driven approach optimises processes for speed, automation, privacy and data security allowing for greater sophistication and robustness in the final analyses.
Using this technology, Phase recently hired a ‘uniquely qualified data science candidate’ out of a pool of thousands for global toy company Maison Battat in only 26 days. Instead of asking for keywords, Phase asked recruiters for a freeform description of the kind of candidate they were looking for. This essentially transforms the search process on its head. Instead of the conventional means of searching for keywords, such as, ‘Python’, ‘SQL’, ‘engineer’ or ‘toys’, the candidate to be hired was described as “A data scientist who has with experience with toys, education, or children’s products.”
According to Phase: “The algorithm knows that a ‘data scientist’ is a person that is likely to know languages like ‘Python’, ‘SQL’, or others and takes on roles such as an ‘analyst’ or ‘engineer’. By teaching the algorithm to seek out candidates who have experience with ‘toys’, ‘education’ and ‘children’s products’ we can find people with relevant experience in related areas like ‘gaming’ or ‘youth development’.” This process not only makes the search for an ‘ideal’ candidate easier, but also much more well-rounded.
The goal is to provide the algorithm with an idea of the type of candidate being searched for rather than limit it to specific keywords. This will allow the system to peruse the entire résumé and make connections between several aspects to find the strongest available candidates. “For instance, a candidate might outline an interest in ‘children’s products’ in one part of their resume, but not include this in their core skillsets elsewhere. This semantic approach tracks the entire resume and scores the themes that come up rather than just individual skills or keyword flags.” It also means recruiters will save time by automating searches while generating a ‘broader diversity of qualified candidates’ than ever before.