Textalytics believes that when it comes to tuning in on Internet banter, different types of businesses have different needs. To that end, the Madrid-based text mining specialist this week launched a family of industry-specific semantic APIs. The company claims its RESTful APIs are plug-and-play, easy to use, and multilingual.
Text mining is man’s search for meaning across the ocean of unstructured data on the Internet. To computers, which prefer data that is neatly arranged in columns and rows, unstructured data is nothing but mumbo jumbo and gobbledygook. Yet, most forms of human communication (documents, emails, social media conversations, and news) exist as unstructured data. To remedy this, businesses rely on data mining tools to quickly sort through huge quantities of data and pull out relevant information. These tools allow companies to do things like better target their advertising and improve communications.
A new company launched by Daedalus, Textalytics has two products available now for developers wanting to integrate data mining into their apps. The Texalytics Core API filters words based on syntax (noun, verb, article) to extract the key words of a document. The platform examines text patterns and identifies entities (a process known as named entity recognition) in order to differentiate between people and places. It also works on text in English, Spanish, French and Italian.
A second product, the Media Analysis API, is aimed at media firms. This product pinpoints buying signals in social conversations and identifies whether customers are happy or unhappy about you and your products. It also picks up context and meaning in conversations, allowing firms to target their ads. (Shoes, did you say shoes? We have just the thing.) Security surveillance is another use for the product.
A third product, a Semantic Publishing API is in the works but not quite ready for primetime. When released, this product will help publishers create more valuable content based on what is trending in the news.
Textalytics claims its semantic APIs are easier to use and implement than similar ones on the market. You don’t need to know the difference between named entity recognition and concept extraction, when to use statistical or rule-based classification, or how to process the language of social networks. The company handles all those complicated matters for you.
If you are still not clear on how data mining works, the Textalytics has a demo that does a nice job illustrating the process.