You are here:  Home  >  Data Education  >  BI / Data Science News, Articles, & Education  >  Current Article

Look to Semantic Tech — Not Psychic Readings — To Predict Outcomes

By   /  July 20, 2011  /  No Comments

On the way from Saplo – that’s the company whose tradeshow trademark is the wearing of shocking green suits by CEO Mattias Tyrberg and his co-founders – is a Prediction API for its text analytics platform. The vendor already provides through its API access to services for entity and topic tagging, related and similar articles, sentiment analysis and contextual recognition upon which developers can build applications.

The Prediction API, due around summer’s end, seeks to predict outcomes from text, as Tyrberg describes it. That is, it assesses how a company name or any other word has been described in text and  finds a correlation between that and expected outcomes, such as sales volumes.

It works by having the user submit historic text and historic data points, from which the technology analyzes the relationship between the meaning of the text and the data that the user wants to have predicted (it also will return data of how good it believes it can predict the outcome, Tyrberg says). After that, the user submits new text data to Saplo for a new time period, and based on that text Saplo returns a prediction of the next outcome.

“Think of it like BI,” says Tyrberg. “You might be able to predict new numbers based on previous numbers, but a lot of information that is available is in written text, and we can find the correlations between the meaning of that text and numerical data.”

It’s about predicting the meaning of how this company or that product needs to be described in order to have higher sales, he says, as an example. Tyrberg expects that organizations could build “really cool apps” for a variety of predicted outcome requirements using this.

Saplo’s contextual recognition API also was a recent add, providing a platform from which developers can create ways of filtering information for news sites or social media feeds. For example, based on the meaning it derives of a handful of articles a user likes on Facebook that are grouped together, the API can be used to reveal any relationships between that group of content and new articles. The goal is to determine whether the new articles are interesting for that particular user or not.

“It’s a way of not using topics or ontologies for understanding the meaning and what you are interested in,” says Tyrberg, who adds that he has seen it perform better at recommending content than other recommendation engines that do use the topics approach. “We don’t define topics but relate information, you could say, say in a meaningful way to find other articles that have similar meanings to your interest.” Contextual recognition also has a role to play in optimizing online advertising, he says, by trying to understand other content consumed by those who have clicked on a particular ad to optimize its placement.

About a month ago Saplo also signed a deal with the news paper that is the largest one in the Nordic region, Aftonbladet, with reportedly more than 25 million visitors per week. It will soon start using Saplo’s technolgoy for tagging, related articles and topics.

Saplo also counts among users of its technology Monster.com, for reports regardingcharacteristics companies are looking for in employees. In June the report was based on the analysis of more than 8,000 job ads and shows that the characteristic of being accurate is the most important.









About the author

Jennifer Zaino is a New York-based freelance writer specializing in business and technology journalism. She has been an executive editor at leading technology publications, including InformationWeek, where she spearheaded an award-winning news section, and Network Computing, where she helped develop online content strategies including review exclusives and analyst reports. Her freelance credentials include being a regular contributor of original content to The Semantic Web Blog; acting as a contributing writer to RFID Journal; and serving as executive editor at the Smart Architect Smart Enterprise Exchange group. Her work also has appeared in publications and on web sites including EdTech (K-12 and Higher Ed), Ingram Micro Channel Advisor, The CMO Site, and Federal Computer Week.

You might also like...

Metadata Management vs. Master Data Management

Read More →