Top Menu

Semantria and Diffbot partner up for Net text parsing and processing

One of the biggest problems companies have with the Internet is how to find text, the right text, online. Sure, you can use search engines but search engines can’t read whether a review written about your company is positive or not. Even if the text is positive, there is no way you can tell, from a simple search how positive the text is. This takes a human pair of eyeballs. Moreover, if you are looking to look over thousands of reviews scattered through dozens of sites, you have your work cut out for you. You can outsource but you will still face daunting quality control and volume issues. In short, it can get very pricey very quickly. Enter text parsing companies like Semantria, Alchemy API, and others. These companies have technology that parse text to figure out patterns. You can then craft your online marketing and branding presentation (or damage control) based on the text.

Semantria and other such companies have a problem though-digging up enough text. Retrieving the information is the first step and this can get very labor-intensive. Even if they were to automate, there is the issue of where to start and where to end. Thanks to a partnership with diffbot, Semantria taps into powerful technology that helps scour tons of online text in a fast and meaningful way without retrieving code junk. As a result, the text retrieved is clean and ready for parsing by Semantria’s technology. This is a powerful partnership because corporate users need this information to quickly size up what others think about their product or brand. Moreover, it can help put out fires before they grow to unmanageable proportions.

The secret to Semantria’s technology is its ability to tell the context of online messages. This is what trips up many other automated parsing systems. Even outsourced text parsing can stumble here if the outsourced analysts are based in another country and have very static interpretations of English words.
Diffbot’s big innovation is that it can retrieve clean text en masse and bundle them under positive and negative sentiment. Semantria works on the context.

, ,