A shunned science not long ago, Natural Language Processing is enabling humans to interact with machines in ways not possible only a few years ago. The new types of human-machine collaboration will enhance our abilities as humans and enable us to process more informed decisions in business, in health, in communication. Natural Language Processing is changing how we interact with machines…. and with each other.

As I watched a family re-run of “Lost” (yes, the first 25 chapters ending in 2005), I could not fail to notice a missing integral part of any 2020 citizen: mobile phones, pictures, internet and machine interaction. Only once, in a flashback, one character picks up a voice call from his sister. Even as the series closed in 2010, there is mobile data collection was in its infancy, and most technologies did not go beyond statistical processes and Bayesian probability. Move on 10 years from the island, and find a Natural Language Processing (NLP) maturing as hardware and software make it increasingly possible for computers to understand and process natural human language in written form, in speech, on the web, as audio or in a combination of all of them. All of us use Google Search, Alexa, Siri, or Google Assistant – and that is just the beginning. From 2017, language technologies have become presents. The advantage of Natural Language Processing is that it allows humans to make queries without first having to translate them into “computer language”.

NLP applications are being applied to make business, processes and consumer applications easier to use. Software developers are already incorporating it in more and more applications: embedded machine translation, anonymization, face and speech recognition, chatbots, summarization, sentiment analysis, text and document classification, predictive writing, market intelligence, and spell checking to name a few. The extensive use of all NLP technology not only affects how we interact with devices and these devices understand and process our voice, pictures or interpret texts in other languages for us. NLP is also affecting how we communicate with other humans.

NLP technology is particularly useful in data analytics, the process of structuring and analyzing unstructured data to provide insights to business leaders, researchers, and the public that will help them make better informed and more effective decisions. NLP can support data analytics efforts in a number of ways, e.g. solving major global problems and helping more people, including those who are not trained in data processing, use these systems.

 
Managing Big Data with NLP – the Covid-19 Example

With the help of NLP, users can analyze more data than ever before, even for critical processes such as medical research. This technology is especially important now as researchers try to find a vaccine for COVID-19. The Pangea Group is proud to help such efforts via the NLP Covid-19 initiative at ELRC.

In a recent article, the World Economic Forum (WEF) points out that NLP can help researchers fight COVID-19 by searching for key concepts through huge amounts of data that would be impossible for humans to filter and analyze. NLP can also bring summarization, or fast translation, such as in the above ELRC initiative. “Machines can find, evaluate and summarize tens of thousands of research papers on the new Covid-19, to which thousands are added every week …”. In addition, this technology can help track the spread of the virus by detecting new outbreaks.

According to the WEF article, as it is known by industry professionals, NLP can aid the research process when data analysts “[train] machines to analyze a user question in a sentence, then read the tens of thousands of scientific articles in the database, organize them, and generate answer snippets and abstracts or summaries” . For example, a doctor may ask, “Is COVID-19 worse in colder climates?” and the system will review the data and return relevant responses.

 
Enabling Professionals and Solving Problems

Our involvement in the machine translation and data search initiative is a small example to help with current health problems. But we can go further and see how NLP can be used together AI for business applications. Areas such as legal, banking, insurance, international documentation for commerce, and global challenges, such as clean energy, global hunger, improving education, and natural disasters. A Council Post appearing on Forbes, states that “huge companies like Google are setting their sights on flood prevention, utilizing AI to predetermine areas of risk and notify people in impacted areas”.

On the other hand, our searches are soon gong to be affected too, as NLP is going to affect search engines and provide more natural understanding , bringing more neutral results that do not depend so much on page format of keyword density. We, as users don’t need to understand SQL or Boolean search. We will simply notice and possibly modify search to our liking within a few years. Other search engines, perhaps multilingual, may appear. Asking the right questions may soon become essential for intelligence staff, C-level executives, researchers, and administrative staff.

A typical hard question we have been working on our chatbot systems has been “give me a list of pictures by Rembrandt not at the Rijksmuseum of Amsterdam”. Turn this into deeper NLP business queries to a BI system with a question such as, “give me the highest stock month this year as a percentage compared to last fiscal year in dollar value”. The system would convert each phrase to computer numeric information, then search for the data in the BI system, and return it to the user as an understandable sentence in a chosen language. This type of queries will allow company employees at any department to gain key insights and help them make informed decisions.

 
From 2020: Moving to a Data-Driven Culture

Business intelligence (BI) has traditionally required trained data professionals to correctly input queries and understand numbers and figures. NLP is now changing that dynamic, resulting in what we call “data democratization”. Nowadays, more people to have access to more data sets, something that previously was reserved to just those with the advanced skills needed to interpret it.

The more people within a company who know how to gather insights based on data, the more that company can benefit from a data-driven culture, which is one that relies on hard evidence rather than guesswork, observation, or theories to make decisions. Such a culture can be nurtured in any industry, including healthcare, manufacturing, finance, retail, or logistics.

For example, a retail marketing manager might want to determine the demographics of customers who spend the most per purchase and target those customers with special offers or loyalty rewards in any language or with machine generated language. With NLP, the commands needed to get this information can be executed by anyone in the business.

 
Our take

NLP has not fully matured nor given its best yet. There are numerous processes that can still be embedded in applications. A few BI and analytics vendors are offering NLP capabilities but they’re in the minority for now. To stay competitive, more applications will become embedded in more systems and offerings.

As it becomes more ubiquitous, NLP is going to enable humans to interact with more devices in ways not envisaged or not possible only our friends of Lost. However, this human-NLP enhanced device collaboration will allow customization and improvements in a wide variety of fields. In short, our conversations with devices are going to get more and more interesting, and continue to sound more and more human.