We've brought several new API services online in the last couple of weeks.
As mentioned before API "services" are essentially classes of question which are suitable for machines and where the support has been added onto our platform to answer them reliably and usefully. A key advantage of our technology is that building knowledge services on top of what is there already can be done very rapidly compared with existing methods, and once there they can be accessed through the same interface and account just by changing the query. If you would like to help us test this technology, please sign up here.
Most of these services have an online demo so they can be experimented with before obtaining an account. The new services are:
Public Holidays. Given a geographical location and a date we can say whether it is likely to be a trading day or not. Possible uses include timing emails so that emails reach the recipient when they are likely to be at work (possibly combined with our Local Time API service so as to arrive during the middle of the day); avoiding holidays when scheduling phone conversations or online meetings; pricing loans (or double checking assumptions in existing pricing systems)....
Location to Language. This takes a geographical location and returns the written languages that are likely to be understood by people resident at that location. It can also return other languages that are comprehensible to native users of those languages. Possible uses include automatically selecting an appropriate translation of a document or website; automatic selection of media that will be understood by the user etc. The geographical location could be derived from the users' IP address or come from another source.
Name to gender. Where possible this converts a name into a probable gender from knowledge of whether first names and titles are exclusively female or male. For example, "Jenny Smith" can be identified as female; "Mr John MacDonald" will be identified as male. This has possible uses for steering the behaviour of automatic systems based on the gender of the user when all it has is a name; deriving demographic information from customer data; selecting the right pronoun when referring to a user when all that is known is the name; pre-selecting an appropriate gender in a sign-up form when a name has been entered...
Email to name. Where possible, this derives a probable first name from an arbitrary email address. e.g. "John" can be extracted from johnsmith564@hotmail.com but it will say that no name can be extracted from editor8822@yahoo.com. The main use is making automatic emails more friendly when all you have is an email address but in combination with the Name to gender service it can also be used to get a gender from an email address and obtain demographic information.
More API services will be coming online in the New Year.
Comments