The past few weeks i have been experimenting with Named Entity Recognition tools. In particular, i tried out the opennlp tool suite and the name recognizer was pretty dismal. It really didnt recognise everything well and i wasnt sure if i should use the same in my research purposes. I guess i will either have to develop my own tools or use something else.
A shame since i did spend a little effort on trying to figure out how to use these tools and thereafter only to see that they dont perform as well as i expected it to. Perhaps i expected a lot for i do know that named entity recognition is not easy and of course there will always be an ambiguity in recognizing names.
Lets see how it works out. My adviser has asked me to read a few papers on the named entity recognition and i need to see if this will lead me to an overall idea on named entity recognition and also on how easy or difficult it is to get the kind of results that i am expecting.
Friday, February 15, 2008
Subscribe to:
Posts (Atom)