Rob Kleiman

View Original

Better Brews Come Delivered By Data Analytics

Cognitive and social context was the launchpad for these algorithmically alcoholic beverages

Using sophisticated data science, IBM and Havas helia created a beer that tastes of joy and optimism. How you may ask? They used a tool called Watson Personality Insights, which analyzes language to produce a personality profile. The technology uses linguistic analysis to find meaning words. In linguistics, semantic analysis is the process of taking syntactic structures and relating them to each other.

First, the team observed the sentiment and the emotions found in the social media shared on and around New Year’s eve by running a data set of New Year’s related messages and shares on social media and matched them with a wide range of emotional states. The partners leveraged this powerful tool to extract cognitive and social characteristics from input text such as email, SMS, tweets, forum posts, and more. Through their analysis, the team found that the top most shared emotions were love, joy, harmony, cheerfulness, optimism, resolution and excitement.

With this profile the team worked together to capture the mood of the nation during the New Year party season to create the world’s first beer crafted and based on human emotions.

It then takes that profile and can categorise each beer according to different human adjectives, such as “assertive,” “friendly” or “intelligent.” Then the IBM Watson team began to analyse 2,800 different beer recipes while giving the computer descriptions about the ingredients, recipes, tasting notes and beer reviews. This method helped to identify the perfect recipe.

The top 10 beers that matched the most shared New Year emotions found in the data were then identified and, through further analysis, all of these beer recipes were combined to find the most common ingredients.

Honey, the Nelson Sauvin hop variety and the Hallertauer hop emerged as the top three most common ingredients among the beers.

  • Honey denotes love and cheerfulness

  • Nelson Sauvin is for optimism, imagination and resolution

  • Hallertauer is for excitement and emotion

Each of these ingredients was used to create flagship data-powered New Year beer: 0101. For the complicated project — the team picked High Peak Brew Co, an independent microbrewery based in the UK’s Peak District, to head up the brewing project. They tapped this particular brewery because the company’s brews are unfined and unfiltered, like the content of the social activity they tested. They worked with them to get an exact taste that would match the data as closely as possible.

Helia works with data to uncover patterns in human behavior, mixing that with cultural understanding to inspire more creative ideas. Why does this matter? The service helps users to understand, connect to, and communicate with other people on a more personalized level. With this powerful tool we can derive consumers’ cognitive and social preferences just with the language they use. The service applies linguistic analytics and personality theory to infer attributes from a person’s unstructured text.

Through tools like this we can gain insight into how and why people think, act, and feel the way they do. This means companies and coders can use data and technology to interpret something abstract such as positive feelings and emotions and turn them into an experience to build upon.

Havas helia

Originally published at www.psfk.com on February 10, 2016.