NATURAL LANGUAGE PROCESSING (NLP) APPLICATIONS FOR HUMAN RESOURCES
Brainster, R.N. Macedonia
Human Resources teams have diverse and important areas to focus on. From hiring to managing payroll, retaining talent, and ensuring competitive benefits, acting as a culture steward is one of the most important roles. Monitoring the health and well-being of company culture is critical to ensuring happy, productive and engaged employees. The value of a surveillance culture is only amplified now that so many people are working from home and grappling with the new challenges of work-life balance. One of the most popular ways to understand and monitor culture is employee engagement surveys. Open-ended questions are a valuable way to get an idea of what employees are thinking. The ability to provide feedback through open-ended questions is a great way to learn first-hand about issues, suggestions, and concerns relevant to current employees. Even with the resources available to read and analyze all comments, reaching consistent and unbiased results is a challenge for any organization. Advances in data science are enabling Human Resource professionals to address these challenges and discover timely, actionable insights. In particular, the use of natural language processing (NLP) is increasingly applied to this challenge.
Keywords: Human Recourses, natural language processing, employee, culture, data
Human Recourses is the perfect candidate to adopt natural language processing -based technologies because Human Recourses is people-centered and communication-based. As a result, Human Recourses business processes generate large amounts of data in natural language.
Natural language has great potential as a source of valuable insight. But until recently, it was rarely analyzed or used in decision making. This is because the process takes too long. Sometimes even reading and analyzing thousands of lines of text is impossible. Natural language processing technology automatically processes and analyzes textual content. They provide valuable insights and transform this “raw data” into structured and valuable information. Most companies strive to improve work efficiency, reduce costs, and improve the employee experience. Also rapidly incorporating AI, machine learning, and natural language processing into our strategies. In recent years, chat bots have started to become part of the digital transformation agenda. This impacts key HR areas such as recruitment, onboarding, training, career development and benefits.
In the following research we will use the method of critical literature to give us better knowledge on the topic of this research. Based on the theoretical and empirical data we will try to prove the contribution and effect that Natural Language Processing can have on Human Recourses. We will discuss three ways NLP can integration can transform HR. To be more precise:
- Natural Language Processing in the Recruitment Process
- Natural Language Processing as a Tool for Employee Engagement
- Other application of Natural Language Processing to Human Recourses
Natural language has great potential as a source of valuable insight. But until recently, it was rarely analyzed or used in decision making. This is because the process takes too long. Sometimes even reading and analyzing thousands of lines of text is impossible. Natural language processing technology automatically processes and analyzes textual content. They provide valuable insights and transform this “raw data” into structured and valuable information. Most companies strive to improve work efficiency, reduce costs, and improve the employee experience but also rapidly incorporating AI, machine learning, and natural language processing into our strategies. In recent years, chat bots have started to become part of the digital transformation agenda. This impacts key Human Recourses areas such as recruitment, onboarding, training, career development and benefits.
- Natural Language Processing in the Recruitment Process
Being a recruitment professional, is understandable how challenging it can be to sift through multiple resumes. And as a result, most recruiters tend to focus on keywords when screening resumes. This practice is in effect, one of the most inefficient and inaccurate methods to select a candidate.
The bigger the organization, the harder it is for Human Recourses to dedicate time and resources to making employees and candidates feel valued. NLP technology enables HR professionals and recruiters to communicate and maintain relationships with employees and candidates at a scalable level. Many positions require candidates to complete multiple interviews. NLP can be integrated early in the interview process to better understand and identify the best fit from the candidate pool. NLP-enhanced automated interviews help recruiter’s blind spots and behaviors they might otherwise miss. By analyzing a candidate’s word choice, speech patterns, and facial expressions, NLP tools can assess a candidate’s fit for a company and provide insights that recruiters may not be aware of.
According to research, making a poor hiring decision based on unconscious prejudices can cost a company up to 75% of that person’s annual income. Therefore, it’s critical to hire the appropriate personnel. (Applied.com– research published by Joe Cacavale)
NLP provides a data-driven approach to resume screening, which not only gives more time per candidate but also helps make better hiring decisions. NLP also enable to rank and classify candidate profiles, identify personal traits, and eliminate human biases.
It reduces time-to-hire by automating pre-qualification via chat bots and automated interviews, also improving the candidate experience.
There are studies lying at the basis of this statement, namely, the staff fluctuation is reduced by 30% if the recruitment and selection process use instruments which can really identify the values of the potential employees and the type of organizational culture they would fit (Mădălina Bălan, Organizational Culture – Strategic Advantage or Decorative Element).
In general NLP can contribute during the recruitment in the following ways:
- To recognize the skills, professional background and experience presented in resumes, job profiles or positions to benefit from relevant matches.
- To find the best profile using a robust semantic search engine that can pre-select the best prospects/employees including vacancies and positions.
- To customize candidate or job search with advanced filters and multi-criteria searches.
- To Automate profile entry when collecting internal profiles or updating a single HR application or existing HRIS software to manage the entire recruitment and internal transfer process.
- Natural Language Processing as a Tool for Employee Engagement
Companies spend hundreds of millions of dollars on employee engagement programs, yet scores on engagement surveys remain shockingly low. A staggering 87% of employees worldwide are not engaged. Employee engagement is sometimes referred to as the holy grail for today’s business leader, and for good reason. According to Gallup, a highly engaged workforce means the difference between a company that outperforms its competitors and one that fails to grow. It is a matter of life or death for a business (Naz Beheshti, “Our Approach to Employee Engagement is Not Working,” Forbes, Sep 30, 2018)
After implementing great customer listening programs, companies are increasingly turning to listening to their employees. To retain talent, maximize employee engagement, unite them around a common project and improve operational processes, giving employees a voice at all levels of the company is key necessity. These are systems that aim to collect Employee Feedback, either through a periodic satisfaction measure or through a system that assesses employees’ emotions at certain critical points in their career. Therefore, Semantic Analysis tools used in the field of Customer Relations can be used to analyze Employee Feedback, prioritize them and turn them into strategic lessons.
There are many uses for NLP technology, but highly relevant to HR departments, NLP can be used to perform sentiment analysis and topic analysis.
- Sentiment analysis uses NLP technology to understand the emotion (positive, negative, or neutral) behind text responses. When applied to employee surveys, sentiment analysis helps assess employee satisfaction with an organization’s overall performance and with specific organizational decisions. Combined with information about where employees sit and what role they have within the organization, use this analysis to identify top performing teams. This is key information for deciding how to focus employee retention efforts.
- Topic analysis automatically assigns topics to text data. Some companies are already using this form of his NLP to quickly sift through customer feedback to understand and respond to customer pain points. Similarly, topic analysis can be applied to employee survey results to understand employee feedback and improve employee experience. Problem analysis can be combined with the subject matter expertise of HR leaders to understand how employees are responding to ongoing HR initiatives and whether they are aware of those initiatives. Alternatively, topic analysis can be applied unattended to identify the most frequently occurring topics such as: Workplace bias, recession concerns, or commitment to the organization’s mission.
Whether applied in a wide-ranging or narrow way, whether supervised or unsupervised, sentiment and topic analysis can be used to answer key HR questions.
The benefits of NLP analysis of employee engagement data are numerous. NLP analysis and automated data ingestion will save time. Rather than months of manual analysis, initial design of an NLP algorithm would take weeks to create. This will allow HR departments to quickly analyze survey results and adjust their work and policies accordingly. A one-time creation of an NLP algorithm can be used year after year, proffering continued ROI. Subsequent use of an already-created algorithm on future surveys would take mere hours (How NLP Analysis of Workforce Engagement Surveys Can Boost HR Strategy – Ariel Silbert)
- Employee Social Media Analytics with Natural Language Processing
NLP is a powerful “listening” technique that HR teams can use to evaluate employees’ social media content to reveal interests, identify employee potential and talents, and identify employees behavioral trends.
According to a research, during the job search, most candidates discover potential employers through company content and media, such as blogs, news and videos. Employers can use social media analytics through these tools to engage and identify potential employers during candidate screening, learn about their interests, and get the ability to drive retention with insights driven by social media analytics.
NLP in social media analytics can also be a useful addition to employee advocacy program, allowing to demonstrate the program’s return on investment. It can also be of great help in assessing general candidate behavior in recruitment data. NLP makes it possible to monitor employee social behavior, interests, and perception to identify opportunities or issues inside the company. In 2019, more than 473,400 tweets and more than 100 million messages have been sent every minute globally. NLP algorithms make it possible to analyze this amount of data in a short amount of time and bring many details information to the business owners about their customers’ and employees’ attitudes on social media. The process can start with generic text analytics (sentiment analysis), continue with advanced insights (via computational linguistics models) and can even include potential semi-automation (NLP and Text Analytics for HR, Why Is It Important? – ComeMit Article 2020)
These days the amount of text-data at organizations is increasing rapidly. The HR department is one of the main resources in gathering and analyzing these data and using them to get better decisions for the company.
These are just a few examples of natural language processing applications for human resources. We can also discuss automatic translation tools that make it easier for international teams to understand each other, and social media monitoring tools that companies can use to monitor employer branding and identify relevant profiles. Whenever a task involves reading and understanding written or spoken text, AI has the potential to help humans, especially with newer deep learning technologies. HR is inherently human-centric and communication-based, making it a prime candidate for adoption of NLP-based technologies. Therefore, HR business processes generate large amounts of natural language data. It also allows Human Recourses to have more intelligence and influence within the organization. However, deciding what is “right” and what really matters is a human prerogative. In the recruitment and placement process, the role of natural language processing (NLP) is to free up time for meaningful person-to-person contact. Here, the recruitment process is streamlined, valuable insights are revealed, and attendees are engaged. NLP protects against information overload and lack of focus, helping to turn the highly tactile hiring process into a fun virtual tour. Natural language processing is a branch of artificial intelligence that is still a fairly new idea in the HR industry.
- Naz Beheshti, “Our Approach to Employee Engagement is Not Working,” Forbes
- Jacob Morgan “Why the Millions We Spend on Employee Engagement Buy Us So Little,” Harvard Business Review
- Adam Rogers “How Unified Employee-Feedback Tools are Revolutionizing HR”
- Raja Sengupta “How Natural Language Processing can Revolutionize Human Resources” AIHR Blog & Academy.
- Cyrus Sanati “How big data can take the pain out of performance reviews,” Fortune
- Dave Zielinski “Artificial Intelligence and Employee Feedback”, Society for Human Resource Management
- Frank Partnoy “What Your Boss Could Learn by Reading the Whole Company’s Emails”