Our final discussion in basic text processing is segmenting out sentences from text.
We use a decision tree to solve this question. But it's doesn't enough, we should use more sophisticated decision tree features to gain the classifier. For example, u can get the probablity of one word end of sentences, such ".The".
Actually, the building of decision trees is possible only for every simple features about with six or seven rules. But it's very hard to do for numeric feature. So we can use other classifiers such as linguistic regression or SVMs or neural nets, we can put features into those kinds of classifier.
Well, it's just introducting. We will talk later.