In this chapter, we have looked at the basic concepts of ML, as well as the various classification algorithms that are used in the NLP domain. In NLP, we mostly use classification algorithms, as compared to linear regression. We have seen some really cool examples such as spam filtering, sentiment analysis, and so on. We also revisited the POS tagger example to provide you with better understanding. We looked at unsupervised ML algorithms and important concepts such as bias-variance trade-off, underfitting, overfitting, evaluation matrix, and so on. We also understood features selection and dimensionality reduction. We touched on hybrid ML approaches and post-processing as well. So, in this chapter, we have mostly understood how to develop and fine-tune NLP applications.
In the next chapter, we will see a new era of machine learning--deep learning. We will explore the...
