课程: Complete Guide to NLP with R
今天就开通帐号，24,700 门业界名师课程任您挑！
Lemmatization
- [Instructor] We just recently spoke about stemming. There is a related concept called lemmatization which accomplishes much the same task but does it in a different way and has different results. Let's take a quick look at lemmatization. In line four, I bring in the text stream package which provides stemming and lemmatization tools. In line seven, I'll define some sample texts that we'll play around with. Now, with lemmatization, what happens is that words are compared against a dictionary and that dictionary provides alternative words to the meaning of the first one. For example, ball, orb, globe, and sphere all mean essentially the same thing. You may want those to be counted as one word with four instances rather than four words with one instances. If you create or use the proper dictionary with lemmatization, lemmatization will accomplish that for you. In this case, I'm using the standard Hunspell dictionary to…
随堂练习，边学边练
下载课堂讲义。学练结合，紧跟进度，轻松巩固知识。
下载课程离线学习
下载课程到移动设备，轻松离线学习。在领英学习 iOS 或安卓版 APP 下载课程