Machine Learning in Customer Service
Machine Learning is going to play a big part in customer service. This is going to bring benefits to both consumers as well as support providers. Consumers will benefit from faster and better support while organizations will be able to leverage data and technology to streamline their operations providing better service at a lower cost. They will also be able to improve employee satisfaction by allowing their support personnel to work on value added tasks rather than any tedious, boring and manual tasks – tasks that machines can handle easily.
What does it mean to bring Machine Learning into customer service? In this 3 part series we will break this topic down to answer this question. In this first part we take a look at the landscape of technologies that are driving disruption. In the second part we will discuss how chatbots powered by Machine Learning form the “front line of defence” tackling most of the common questions. Finally we will review what techniques can assist humans by providing contextual information in real time so that they can answer questions better and faster.
Some of the key Machine Learning techniques that are at the heart of this customer support revolution include Natural Language Processing (NLP), Natural Language Generation (NLG), Language Translation, Speech Analysis etc.
NLP is is the ability of a computer program to understand human language as it is spoken or written. As simple as that statement sounds it’s actually very complex to make a computer program understand what humans take for granted. For humans, understanding language is so natural we usually don’t even have to make a conscious effort to do it. Without consciously thinking about it, humans correct grammar mistakes, resolve ambiguities, and infer meaning that isn’t explicitly stated. NLP has been an area of research for academicians for several decades. Recent advancements in deep learning along with the ability to process large amounts of data has taken accuracy of computers to understand intents of a given text to new heights of accuracy. One application of NLP is in the customer service industry where large amounts of text (via existing cases) can be analyzed.
Natural Language Generation on the other hand is the ability for a computer to generate text. The goal of NLG systems is to figure out how to best communicate what a system knows. The trick is figuring out exactly what the system is to say and how it should say it. In the past there were simple rules based systems that could generate text based on a finite list of set conditions. Such systems have limited functionality and are unable to learn and scale as an enterprise expands their product or support offerings. Google Home, Amazon Alexa and others are a few examples of systems that leverage NLG to a certain extent to make everyday tasks easy and efficient for consumers. The enterprise world however is catching up, with the hope that eventually intelligent conversational interfaces can facilitate engagement across employees and customers. Per a recent article in the Harvard Business Review, “Bots that Can Talk Will Help Us Get More Value from Analytics”: Conversations with systems that have access to data about our world will allow us to understand the status of our jobs, our businesses, our health, our homes, our families, our devices, and our neighborhoods — all through the power of Advanced NLG. It will be the difference between getting a report and having a conversation. The information is the same but the interaction will be more natural.
Other technologies that have seen huge advancements are in Speech recognition and Language translation. For instance Google’s Speech and Language Translation API’s allow a developer to easily integrate these features into their apps. These APIs are highly accurate making the end consumer experience enjoyable.
Similarly, Salesforce Einstein Language – Custom Intent and Community Sentiment APIs offer NLP capabilities to quickly understand the meaning in a body of text and its sentiment.
In this article we touched upon some key Machine Learning techniques that are going to be at the heart of changing customer service in the coming years. In the next two parts we’ll dive deeper into some of these areas. We will discuss tools in more detail and what an organization should keep in mind before embarking on such a transformational journey.