en-GB

EN

  • EN
  • DE
  • NL
£
en-GB

Home / Our blog / Driving Business Growth with Automation and Google’s NLP

Does your business have a customer service department? Does your website have a chatbot or an online form? Do you handle any other kind of data, like call centre recordings? If the answer to any of these questions is yes, chances are you’re sitting on huge business potential.

These days, many organisations collect more data than they know what to do with. Take the example of a chatbot on your website. Every day you receive a few to a few hundred messages with questions, comments and requests. An employee might be able to follow-up on a few dozens a day, but not hundreds, and your company will inevitably fall behind. This means losing out on potential leads or solving important problems. On top of that, when handling the content of these forms, employees are unlikely to look for patterns, build models and make predictions. Another missed opportunity.

So, what can you do?

Automate.

You need to find a way to automatically and quickly process all the input you receive from chatbots and, ideally, you’d want to automate your responses too. This is exactly where Natural Language Processing (NLP) can help. NLP is the combination of Machine Learning and linguistics, and it has become one of the most heavily researched subjects in the field of Artificial Intelligence recently.

Gain valuable insights with Google

Google has a Natural Language API that can help you unlock insights from unstructured text using Machine Learning. This API is an easy-to-use interface linked to a set of powerful NLP models which have been pre-trained by Google to perform various tasks.

Because they are pre-trained on enormously large datasets, these models are ready to use. With a simple API call, you can instantly make predictions. This is especially valuable in situations where little labelled data is available.

The Natural Language API comprises four different services:

  • Syntax Analysis
  • Sentiment Analysis
  • Entity Analysis
  • Content classification

These can have a multitude of uses. A good example is the work we did with LUMC. During this collaboration, we developed a fit-for-purpose application that records, transcribes and analyses doctor-patient interviews automatically.

Combining Natural Language with Speech-to-Text, we were able to develop a speech recognition application that extracts insights from doctor-patient interviews. This way, doctors don’t need to transcribe these interactions and can now use their valuable time caring for patients.

Increased flexibility with Google AutoML Natural Language

If the Natural Language API is not flexible enough for your needs, then AutoML Natural Language might be better for you. AutoML is a new Google Cloud Service that enables you to create customised Machine Learning models, trained on your own dataset to fit a specific task.

The AutoML service requires a bit more effort from your side, mainly because you’ll need to provide a dataset to train the model. However, the training and evaluation of the models are completely automated and no Machine Learning knowledge is required. The whole process can be done without writing any code, just by using the Google Cloud console. And, of course, if you want to automate these steps, there is support for all common programming languages.

You can use AutoML Natural Language to extract information from a range of content, such as collections of articles, chat data, audio recordings or scanned PDFs, etc. These are some of the products you can use to create custom Machine Learning models:

  • AutoML Text Classification: This model analyses a document and returns a list of content categories that apply to the text found in the document. An excellent example of how this is being used is media companies. They’re using AutoML Text Classification to gain insights into the thousands of pieces of content they publish daily, analysing how it resonates with their audiences.
  • AutoML Sentiment Analysis: This model inspects a document and identifies the prevailing emotional opinion within it, especially to determine a writer's attitude as positive, negative, or neutral. This can be used to analyse customer complaints and prioritise urgent requests. This way, companies can ensure that more pressing issues can be solved faster, improving their customer experience.
  • AutoML Entity Extraction: An entity extraction model inspects a document for known entities referenced in the document and labels those entities in the text. This is a useful model for obtaining customer insights. It can help you find and label common fields within documents such as emails, chat, audio recordings and social media. Then, using sentiment analysis, you can understand your customer’s opinion on certain products better and tailor your strategy around that information.

Google’s NLP API and AutoML can be applied to a wide variety of datasets to quickly solve a wide variety of business challenges. To find out more about how we can help you outsmart your ambitions, get in touch. Our enthusiastic experts are waiting to hear from you.

Similar Stories