Computers are beginning to learn to read between the lines of our tweets, Facebook updates, and email messages.

Humans are fairly sophisticated when it comes to understanding the complex meanings beneath spoken or written word. For example, we can tell that a statement like, “My car had a flat. Brilliant!” is sarcastic, and not actually brilliant.

And with the help of machine learning, computers are beginning to get better at the same, resulting in a new kind of analytics: sentiment analysis.

Sentiment analysis, also known as opinion mining, seeks to discover subjective opinion or sentiment from text, video, or audio data.

The basic aim of sentiment analysis is to determine the attitude of an individual or group regarding a particular topic or overall context. The sentiment or attitude may be a judgement, evaluation or emotional reaction.

For example, Expedia in Canada used sentiment analysis to determine that the music accompanying one of their commercials was receiving an overwhelmingly negative response online, and they were able to respond to that sentiment appropriately: by releasing a new version of the commercial in which the offending violin was abruptly smashed.

“What do you really think?”

Say you have a lot of text data from your customers originating from emails, surveys, social media posts, etc. There are several hundreds of thousands of words in the English language and while some are neutral others have a distinctly positive or negative vibe. This polarity of sentiment can therefore be applied to your customer text to establish what your customers as a stakeholder group really think of you.

There are number of software tools that can help you to measure text sentiment around your product or service. Twitter for example allows you to separate the positive tweets about your company, brand, product or service from the negative and neutral tweets so you can see how well you are doing in the Twitterverse.

Plus, you are no longer hampered by the scientific notion that what you observe, changes. People have long known that surveys and focus groups aren’t necessarily indicative of broader sentiment. The people who choose to respond to a survey may be the ones who have the most to complain about or the most to praise, but not the middle-of-the-road customers. People brought in for a focus group may alter their opinions based on what they think the company wants to hear.

With something like Twitter analysis, however, you’re getting the unfiltered opinions of potentially millions of users, not a dozen people sitting in a white room.

Sentiment analysis can help you gauge opinion, which can in turn guide strategy and help decision-making. In the current business landscape, it’s increasingly important that we know what our customers, competitors and employees think about the business, products and brand and sentiment analytics can help us do that — often relatively inexpensively.

More than market research

The technology is also being put to good use outside the marketing and sales arenas.

Researchers at the Microsoft Research Labs in Washington discovered that it was possible to predict which women were at risk of postnatal depression just by analysing their Twitter posts with text-based sentiment analysis.

The research focused on verbal cues that the mother would use weeks before giving birth. Those who struggle with motherhood tended to use words that hinted at an underlying anxiety and unhappiness. There was more negativity in the language used with an increase in words such as “disappointed”, “miserable”, “hate” as well as an increase in the use of “I” – indicating a disconnection from the “we” of impending parenthood.

Co-director of Microsoft Labs Eric Horvitz acknowledged that this type of information can be incredibly useful in reaching out and helping women at this vulnerable time and also to help break down the stigma around postnatal depression. It would be a relatively simple step, for example, for a welfare group to create an app that could run on a smart phone and alert pregnant women to the onset of potential postnatal depression and direct them to resources to help them cope.

Audio sentiment analytics is also being used to measure stress levels in call centres so that customer service representatives can measure how upset the caller is and intervene earlier before things escalate. People often talk into the receiver, even when they are on hold or listening to the soothing music, they can also make various sounds such as heavy sighing which can indicate the caller is getting increasingly frustrated.

Even Wimbledon began using sentiment analysis this year to help predict which headlines and news topics emerging from the tournament would most interest its fans and followers. Their systems could analyse existing Tweets, updates, and comments, and make predictive suggestions about what types of storeys fans would react most positively to.

Of course, sentiment analysis is not yet 100% accurate, and still needs a human’s watchful eye to ensure that the nuances of human speech are being fully understood by the computer.

In addition, it’s important to note that not all communications can be classified as positive, negative, or neutral; we’re just too complex for that. What we are seeing at the moment is that sentiment analytics is moving beyond a simple positive/negative scale and expanding into classifying a broader range of human emotions.

Similarly, organisations will become more comfortable with the idea of sentiment analytics, and begin using it in new and even more exciting ways.

Where to go from here If you would like to know more about analytics and customer measurement, cheque out my articles on:

• Measuring Customer Satisfaction With NPS: The Good, The Bad, And The Ugly
• Amazon: Using Big Data To Understand Customers
• Tesco: Measuring Customer Performance & Gaining Insights From Clubcard Data
• Disney: Using Big Data, The IoT And Machine Learning To Enhance The Customer Experience

Or browse the Big Data and Analytics and KPI & Metrics sections of this site to find more articles and many practical examples.