By Ian Ransom, Contributor
Where talking to a machine was once like pulling teeth, natural language processing (NLP), the artificial intelligence (AI) that allows a computer to understand human language, has now made the conversation a lot more fulfilling.
Voice assistants like Siri and Alexa may fall short in the empathy stakes, but the algorithms that underpin them ensure they are dedicated and quick learners. In recent years, NLP has been transformed by the use of neural networks and deep learning, so that it is now possible to build automated systems that can interact with people more naturally than ever before.
YOU MAY ALSO LIKE THIS PODCAST: Not-Entirely-Human Resources
Forward-looking businesses have gotten the message, and are incorporating NLP into both customer-facing and internal operations. The market for NLP could reach more than $80 billion by 2026, with the technology set to unleash sweeping changes over industries ranging from manufacturing to healthcare.
NLP technologies work with both text and voice, powering more effective—and conversational—chatbots, and tools that can scan legal and regulatory documents. NLP is also being used in predictive applications that can act as a first warning of potential disease outbreaks, such as the coronavirus (COVID-19). While AI is the big buzzword, it is often NLP doing the heavy lifting and delivering many of the practical benefits in diverse areas, from customer service to compliance.
Freedom of Speech
One of NLP’s fastest emerging applications is voice interface technology for digital assistants in multiple languages. NLP identifies words and terminologies in speech, and works with AI to trigger a response. When integrated with the Internet of Things (IoT), the technology can literally open doors without human interaction. It is also enabling chatbots to become better customer service agents through automated coaching. This is evident in the recruitment industry where NLP-enabled chatbots are introducing organizations to prospective candidates and answering rudimentary queries.
YOU MAY ALSO LIKE: How to Eliminate Hiring Bias
Recruitment tech firm Pomato uses NLP in its resume-scanning software to identify the most promising candidates., while Utah-based firm Hirevue produces NLP-enabled tools that help companies evaluate video interviews of prospective hires.
“AI reduces the task of manually reviewing thousands of medical claims, and instead focuses our staff on performing warm human outreach, and thinking through complex problems together with our members.”
—Dr. Sanjay Basu, director of research and analytics, Collective Health
Insurance is another industry capitalizing on NLP, with established firms like Geico and insurtechs such as Collective Health developing chatbots to handle basic customer queries, allowing human agents to tackle more complex tasks. Collective Health, which manages employee health plans and benefits for companies, uses NLP-powered machine learning to assess risk and match members with tailored healthcare solutions.
“AI reduces the task of manually reviewing thousands of medical claims, and instead focuses our staff on performing warm human outreach, and thinking through complex problems together with our members,” says Dr. Sanjay Basu, Collective Health’s director of research and analytics.
Compliance Made Easy
Many sectors of business and government are crushed by a weight of documentation, much of it regulatory or legislative. The volume and pace of regulatory change, and the increasing regulatory burden are the two biggest challenges for compliance officers, according to Thomson Reuters’ 2019 Cost of Compliance report, which surveyed 900 financial services firms across the globe. However, NLP is providing relief, with tools able to slash compliance lead times and process huge volumes of data in seconds rather than hours.
Global insurance firms are enthusiastic adopters, including British insurance firm Lloyd’s and its consultancy Lloyd’s International Trading Advice (LITA), which produces regulatory intelligence for over 200 jurisdictions. Providing client advice on complex regulatory regimes was previously a manual process that could take staff days of combing through documents. NLP has automated much of the process and cut average turnaround times to less than an hour.
“The AI augments the role of employees,” says Craig Civil, head of data innovation, R&D, and analytics at Lloyd’s. “It makes their job far more satisfying because we automate 80 percent of the work, and the 20 percent [left] are truly interesting one-off questions that you do need an experienced team that can do the research and answer.”
Virus-Tracking Software
One of the more topical applications for NLP is its use in generating real-time alerts from unstructured information and alternative data sets. Apart from helping trading desks scan company filings and central bank releases, the technology is being employed to generate alerts for potentially high-risk events like health crises.
YOU MAY ALSO LIKE: Solving Complex Communication Problems with NLP & AI
Canadian tech firm BlueDot, which sells software that detects disease outbreaks, uses NLP and machine learning to wade through a trove of data ranging from digital media to government releases and air traffic information. Patterns and anomalies are presented to human experts, who then make an assessment on the validity of threats and alert clients including government agencies, airlines and health services.
BlueDot said its technology picked up on the COVID-19 outbreak when it emerged in Wuhan, China, and alerted clients more than a week before the World Health Organization.
“We didn’t necessarily know it would be of this size,” BlueDot founder and CEO Kamran Khan said in an interview with the University of Toronto. “But what we did know is that it had the ingredients.”
Researchers at Indiana University, the Regenstrief Institute in Indianapolis, and pharmaceutical giant Merck have used NLP to examine electronic medical records with the aim of identifying patients at risk of developing dementia. While many who develop Alzheimer’s or other dementia-related conditions are never diagnosed, the researchers have credited their NLP-enabled algorithm for being able to predict the onset of dementia within one to three years of diagnosis.
“This research is exciting because it potentially provides significant benefit to patients and their families,” says Patrick Monahan, a Regenstrief affiliate scientist.
Understanding the Unstructured
In the age of big data, organizations are awash with a tsunami of information from myriad sources. Humans can only spend so much time mining for the specks of gold that deliver actionable insights. NLP, however, offers support to do the legwork in processing huge volumes of data and allow humans to concentrate on adding value for their clients. The dialogue with voice assistants and chatbots may still be functional and lack the human touch, but the technology’s capacity for learning means NLP is set to dominate the digital conversation in years to come.
“The confidence [NLP] brings to the decision-making process is enabling innovation to happen at a more rapid pace than ever before,” says Kevin See, vice president, digital products at Lux Research, “but don’t think this means humans are no longer needed.”