The company has just published a new API that allows developers to automatically tag text, identify related and similar text elements and recommend the new. Use of the API is free for developers to a number of method calls. After that pricing is based on the total number of calls. There is a model of sharing special recipes for start-ups.
The technology is built around the idea of a context. A context is created from several text elements that are related in some way. Three different sports together articles could be used to create a context for the sport. Once the context is created, it can be compared to new articles to predict how relevant they are, in this case to the sport. The text elements are used to build a context and more closely they linked the best results. CEO Mattias Tyrberg explained that the five elements of blog article length text are generally enough to create a context that can give good recommendations, based on relevance.
Saplo context can even be used to create a personal context setting your interests. A sample application called Saplo flow analysis articles distributed on Twitter to recommend new articles in your Twitter stream that you may have missed, and even gives you estimates how much time for reading each article.
Saplo technology can also automatically tag text elements. Currently, it is mainly performed manually. When I finished this article I set the post tags. Finally, this type of text analysis can be used to feel sophisticated analysis. Not only can determine if a text element is positive or negative about a product or an organization but also, for example, how easy environment to an organization is seen way.
Tyrberg sees all kinds of applications for the technology but particular potential for contextual advertising spots and personal filter. Contextual ads are based on the meaning instead of key words. If an advertiser wants to an announcement is displayed next to an article on cars, but a news site is not an automobile section, Saplo can automatically find articles related to the car and place announced it. Tyrberg predicts that Customization via the analysis of text will be the next great innovation in the search. Instead show the same results search for everyone, the prediction engine can display each individual unique results based on their interests.
Competitors include OpenAmplify, Lexalytics OpenCalais and perhaps even from Google. Saplo technology is language independent Tyrberg telling is its main advantage over the competition. The new language learning process is fully automated. To start a new language marking, training set with the sample tag is required and to start a sense analysis training together with examples of positive and negative. Technology then generalized training sets and learns later feedback provided by users.
Saplo is based in Sweden has 10 employees and has received $500,000 in funding for Angels and private investors. CEO Mattias Tyrberg won the prize of "to young entrepreneurs of the year" of the Sweden in 2010.
Tags: analysis of feeling, marking, analysis of text
Enterprises: Lexalytics OpenAmplify, Opencalais Saplo
Copyright 2010 VentureBeat. All rights reserved.VentureBeat is an independent technology blog. "Read more".
No comments:
Post a Comment