Fundamental concepts: The importance of constructing mining-friendly data representations; Representation of text for data mining.
Exemplary techniques: Bag of words representation; TFIDF calculation; N-grams; Stemming; Named entity extraction; Topic models.
Up to this point we’ve ignored or side-stepped an important stage of the data mining process: data preparation. The world does not always present us with data in the feature vector representation that most data mining methods take as input. Data are represented in ways natural to problems from which they were derived. If we want to apply the many data mining tools that we have at our disposal, we must either engineer the data representation to match the tools, or build new tools to match the data. Top-notch data scientists employ both of these strategies. It generally is simpler to first try to engineer the data to match existing tools, since they are well understood and numerous.
In this chapter, we will focus on one particular sort of data that has become extremely common as the Internet has become a ubiquitous channel of communication: text data. Examining text data allows us to illustrate many real complexities of data engineering, and also helps us to understand better a very important type of data. We will see in Chapter 14 that although in this chapter we focus exclusively on text data, the fundamental principles indeed generalize to other important sorts of data.
We’ve encountered ...