Summary

In this chapter, we learned about the motivations behind converting human language in the form of text and speech into mathematical structures such as scalars, vectors, matrices, and tensors. This helps machine learning algorithms to execute mathematical functions on them, detect patterns in language, and gain a sort of understanding of the meaning of the text. We also saw the different types of vector representation techniques, such as simple integer encoding, character-level encoding, one-hot encoding, and word encoding.

In the next chapter, we will look at the area of sentiment analysis, which is the automated understanding of tone or sentiment in text sources. Sentiment analysis uses some of the vector representation techniques that we saw in this chapter.