Natural Language Processing – How it compliments SAS analytics platforms?

Natural Language Processing helps machines understand how human minds communicate and help deliver the correct information, as a result, playing a huge role in SAS analytics platforms.

Earlier this year, SAS announced that it was deploying several AI-based advancements to its SAS Viya platform, including natural language processing (NLP). NLP is not new to SAS (it is used in SAS Text Analytics), but the inclusion of the AI-based tech into Viya expands the functionality of that particular platform. The news is a welcome development, but, while experts in the analytics industry (including myself) were delighted to hear the news, others expressed confusion. Therefore, I am going to explain what exactly NLP is, and why we should be excited about its expansion into other SAS analytics platforms.

What is Natural Language Processing? (NLP)

Humans communicate on multiple mediums. We just use language (English or Mandarin), tone and select our words carefully to convey sentiment. By contrast, when machines communicate, it is with a series of 0s and 1s to produce logical action. As you can imagine, machines cannot understand human language and vice versa. That is where natural language processing (NLP) comes into play. Drawing from several disciplines, like computer science and computational linguistics, NLP is used to help machines understand and manipulate human language. As you can imagine, the AI-based tech plays a huge role in SAS analytics platforms

While it is not brand new, NLP has received a lot of attention in the past few years, due to increasing interest in big data, growing computing power, more sophisticated algorithms and growth in machine-human interaction. Amazon Assistant, Alexa, is an excellent example of NLP in action. When you give a voiced command, Alexa not only executes the command but also stores away data for the future. For example, if you give Alexa a command to play your favourite song, the assistant will play the song, but also note your preferences for the future.

Better SAS analytics platforms through NLP

NLP is valuable to analytics platforms because it breaks down the language into short pieces, and explores their connection to create meaning and value.

Analyse large volumes of text

NLP allows SAS analytics platforms to break down and analyse large text volumes. Without NLP, it is unlikely that analytics can assess text accurately. SAS analytics platforms break down text to identify the sentiment behind the words, i.e. the most important parts. An excellent example is when an international hotel chain used analytics to assess thousands of user reviews to better understand the source of discontent. The hotel then used the information to address the core problems behind their services.

Analyse unstructured content

Of course, we no longer just produce structured (text) data. We also produce unstructured content like video chat. Not only are we dealing with different mediums, but the rules of syntax and grammar change accordingly. If you were to consider regional dialect, slang and a person’s tendency to stutter or mumble, then processing the language becomes even more complicated. If analytics platforms are going to make any sense of unstructured content, then NLP is necessary. NLP clarifies ambiguity and adds a numerical structure to the language so that applications, like speech recognition and text analytics, can process the language easily.

Different approaches to assessing language

Humans communicate in different ways. Therefore, it makes sense for SAS analytics platforms to have different approaches to assessing and breaking down language. These different approaches are all made possible, thanks to NLP. Functions for analysing language include parsing, stemming, language detection, tokenisation, parsing and identifying semantic relationships.

With NLP capable of so much, SAS analytics platforms can perform several functions like content categorisation (duplication detection, search indexing and content alerts), topic discovery and modeling (capturing meaning and themes in text), sentiment analysis (identifying the mood or subjective opinions expressed in a large body of text) and machine translation (automatically translates text or speech to another language).

In short, NLP makes it easier to process text to derive meaningful value.

Key takeaways

NLP is an advanced branch of AI that enables SAS analytics platforms to break down human language into a format that machines can understand. The technology boosts the capability of analytics platforms by performing several tasks like search indexing, content alerts and sentiment analysis. These functions breakdown human communication for different functions, which can be used by organisations for revenue-generating operations. Hence, the reason why SAS has incorporated NLP into the SAS Viya platform.

If you are interested in using an upgraded SAS Viya, you will need a team of SAS experts to install the platform. Visit our SAS Viya installation page to get more details.