识别推文中提到的大学

Question

I am looking for a means of identifying UK University names mentioned in Tweet text.

I have a list of full University names, but the issue is dealing with shortened versions such as "aber uni" (Aberystwyth Uni), "staffs uni" (Staffordshire University) or "portsmouth" (University of Portsmouth).

I have looked down the route of Apache Stanbol and OpenNLP to attempt Named Entity Recognition, and although these will match for the full names I cannot seem to find a means of training them to identify variations of the names (or indeed lowercase versions of the name which are not identified).

Answer 1

收集大学列表（这很容易做到），并从Freebase刮取每所大学的名称列表： 使用网络查找相关名称的一种方法是什么？