Google Continues Efforts to Predict What Users Want

    September 17, 2010
    Chris Crum

Google is hellbent on predictions, it would seem. At Google I/O earlier this year, Google announced the Prediction API. More recently, it launched Google Instant, with the goal of giving you your results before you even finish typing the query. Today, Google is talking about some research its doing in trying to predict when searchers are frustrated with their results. 

 Google has also made some new additions to the Prediction API, which is designed to give developers access to Google’s algorithms to analyze historic data and predict likely future outcomes within their apps. 

 New additions include multi-category prediction, continuous output, mixed inputs, and combining continuous output with mixed inputs. 

Have Google Tell Your Fortune"Imagine you’re writing a news aggregator that suggests articles based on the kinds of stories the user has read before," says Travis Green of Google’s Prediction API Team. "Previously, using the Prediction API, each article could only be tagged with one label – the most pertinent one. For example, an article about a new truck might be labeled as ‘truck,’ but not ‘roomy’ or ‘quiet.’ Now articles can be tagged with all of those labels, with the labels ranked by pertinence, enabling your app to make better recommendations."
"You’d like to create a wine recommendation app," he continues.  "Matching a wine to personal preferences is a tricky task, dependent on many factors, including origin, grape, age, growing environment, and flavor presence. Previously, your app could only label wine as ‘good,’ ‘decent,’ ‘bad,’ or some other set of pre-defined values. Using the new continuous output option, your app can provide a fine-grained ranking of wines based on how well they fit the user’s preferences."

 That’s just a couple of examples of what can be done with the Prediction API now that it has these new features. Green provides a couple more here

 Google’s continued efforts of prediction are something to keep an eye on, as the company seems to be putting a great deal of emphasis on on them lately. It’s all about getting users information they want more quickly, which can’t be too bad, unless Google’s predictions are inaccurate. Maybe one day they’ll release an Android app that just reads our minds and delivers accordingly.