Open source equivalent to opencalais (Preferably PHP or Python?)

Is there an open source equivalent of op开发者_如何学运维encalais - preferably in PHP or Python?

---

Conceptually it's an interesting idea. It seems to be parsing basic text or HTML content, then wrapping entities in microformats

Why don't you just the web API for their service?

Implementing a solution yourself should be straightforward - it's just a case of lots of regexes and microformat templates.

---

It seems their main service is Named Entity Recognition with some extrapolations made on top of that. NER is still a research problem in Natural Language Processing. This means there's no one-size-fits-all algorithm, though commercial providers might try to convince you otherwise.

That said, Python's nltk does have some relevant features.

---

Stanford NER for the Named Entity Recognition and a Dependency parser would be some of the building blocks of making an equivalent to Opencalais.