Amherst, Massachusetts (PRWEB) October 31, 2013
Semantria, leading cloud-based text and sentiment analysis engine, is thrilled to announce its new API 3.5, which includes an impressive line up of new features designed to bring even more actionable insight to articles, tweets, posts, or any text content.
The latest version of the Semantria API 3.5 will support text and sentiment analysis in Mandarin. This is Semantria’s sixth fully supported language, in addition to English, French, Spanish, German, and Portuguese. Unlike other providers, Semantria relies upon native language packs and technologies developed by native language speakers in each language. Using native language packs delivers more accurate sentiment and the ability to catch linguistic nuances that are often lost when performing sentiment analysis using machine translation. The addition of Mandarin to its arsenal gives Semantria the opportunity to enter into the quickly growing Chinese market with an easily accessible and top quality Chinese text and sentiment analysis web service.
With the release of its API 3.5 Semantria has also updated its categorization functionality. Semantria now supports auto-categorization, which will classify content into over 4,000 automatically generated categories based on Wikipedia’s taxonomy. Auto-categorization will not require any curating of sample keywords, and is the first service to use the context of text to easily distinguish between an article about windows, versus one about Microsoft Windows. Users can continue to build custom categories according to their unique needs, but the auto-categorization feature allows for more diverse and easy-to-use categorization right out-of-the-box.
In order to facilitate the analysis of web content, the API 3.5 introduces an HTML processing option. Built with clients’ ease of use in mind, Semantria is now able to clean HTML code to extract the meaningful text and ignore the html markups, thus improving output quality and eliminating the need for an HTML parser.
The new API 3.5 brings users new functionality that allows them to delve further into their content and extract even more information than ever before. Some additional new features include:
- Enhanced discovery processing with which users can extract the important keywords and named entities from a collection of texts, allowing for more detailed analysis;
- Normalization of entity mentions for more precise disambiguation of content; and
- Regular Expressions for named entity extraction, which will allow for more granular control when extracting custom named entities.
“We’re very excited that the new API 3.5 will introduce Chinese language support so that companies around the world will be able to extract insight out of content generated by the world’s fastest growing market,” says Haseeb Tahir, VP Client Services. “In addition, we’ll be able to offer new and existing clients even more functionality. Many Semantria users are analyzing thousands of documents at a time and with the new auto-categorization and discovery analysis features we have dramatically simplified the classification process and increased the amount of valuable information, which can be extracted.”
Users interested in trying the new API 3.5 features can access the API with 1 of 7 SDK wrappers and a free trial of 10,000 transactions by registering at semantria.com/trial.
Founded in 2011, Semantria is a software-as-a-service and services company with a specialization in cloud-based text and sentiment analysis. Semantria is a joint venture, co-founded by the leading text analytics technology provider Lexalytics (Boston, MA), amongst other investors.