How to detect named entities

Question

I have a list of named entities. I would like to extract listed named entities from a certain text file with some Java libraries.

What I want to do are;

• "Currencies" should be detected, even if the list includes only "Currency". Conversely, "OECD country" should be detected, even if the list includes only "OECD countries."
• "recommendation system" should be detected if the list includes only "recommender system" and vice versa.
• "economic-buyer theory" should be detected, if the list includes only "economic buyer theory" and vice versa.

I tried GATE tools, but it did not work well. How can I do that? If you have any suggestions, please let me know.

Answer 1

You can use Illinois-NER but you need to annotate sample training data with your desired annotation and train the model again, since your desired are labels are different from the current labels: http://cogcomp.cs.illinois.edu/page/software_view/4