Follow Datanami:
November 18, 2015

Watson App Gives Big Data for Christmas

Holiday shopping anxiety? There’s an app for that.

IBM launched its Watson-based “Trend” app this week designed to keep users posted on what’s selling this Christmas and why. Watson Trend was released in time for the holidays as an iPhone or desktop app.

The app seeks to create another consumer use case for the Watson cognitive technology platform that leverages natural language processing and machine learning to glean insights from heaps of unstructured data. In the case of Watson Trend, IBM hopes to offer consumers sounder advice on, say, what color tie to buy hubby, than traditional social media pointers designed to advise consumers on purchasing decisions.

“With this app we are trying out the idea that Watson’s insights can go beyond the usual gift guide to help users find and give more inspired gifts,” IBM declared.

The company said Watson Trend is designed to interpret data on holiday consumer sentiment on a daily basis and then report on the raw numbers, “not its opinion,” IBM stressed.

The “trendy” part comes from the apps’ ability to graph consumer data over several months. It then predicts whether a trending Christmas gift idea is a passing fad or a must-have item by predicting its staying power throughout the holidays.

If the Watson app manages to come up with the gift that keeps on giving, it also will inform harried holiday shoppers how and where to purchase the perfect present.

The platform works by gathering data on purchasing decisions gleaned from social networks ranging from blogs, forums, comments, ratings and reviews. Watson cognitive computing technology is used to interpret context and gauge consumer sentiment. The framework looks for information about what was purchased, where, consumer satisfaction with the transaction and what product features drove the purchase.

Then, the Watson app combines similar conversations to discover purchasing trends that are translated in to a daily “Trend Score.” Individual comments or reviews are grouped by topic, and similar topics constitute a “trend.” The “Trend Score” represents a grade assigned by the app based on metrics like the number of comments and replies.

The payoff, IBM asserts, is a 21-day forecast for each holiday shopping trend. Beyond forecasting, the company claims the Watson app also is able to define trends, or what it calls a trend’s “texture.” The texture for each trend is selected by the Watson app, reviewed by an IBM data analyst and then posted to the Watson Trend app.

U.S. retailers could certainly use the help: In October, the National Retail Federation forecast a 3.7 percent increase in holiday sales, slightly less than last year.

IBM is betting that Watson Trend will help shoppers come up with better holiday gift-giving ideas that don’t end up on the top shelf of the closet.

Or, you could just give a book….

Datanami