Good morning - I'm looking to get a little advice or direction on a project
I'm working on.

I'd like to parse the titles and descriptions of podcasts, looking to pull
out names and places.

For example, if we look at the current values in the BBC Global News
podcast, I'd like to identify values such as "Lebanon" and "Trump" and
"China", "Michael Brown".

http://www.bbc.co.uk/programmes/p02nq0gn/episodes/downloads.rss

What would be my best approach for doing this? I saw from this message

http://mail-archives.apache.org/mod_mbox/opennlp-users/201507.mbox/%3CCA%2BV%3DWqhbcmQ7%2BJVMyekAhde3kA%3DrJLJ3CzbwjS1P_yHWgAeM1w%40mail.gmail.com%3E

that the "default" models aren't necessarily the best to use. What else
should I be looking to? Are there any tools to help with the training?

Any advice is appreciated, thank you!

-Clay Mitchell

Reply via email to