Chapter 8. Text Visualization
Machine learning is often associated with the automation of decision making, but in practice, the process of constructing a predictive model generally requires a human in the loop. While computers are good at fast, accurate numerical computation, humans are instinctively and instantly able to identify patterns. The bridge between these two necessary skill sets lies in visualization—the precise and accurate rendering of data by a computer in visual terms and the immediate assignation of meaning to that data by humans.
In Chapters 5 and 6 we examined several practical examples of applied machine learning models. Yet in the execution of these examples, we observed that the integration of machine learning is often not as straightforward as merely fitting a model. For one thing, the first model is rarely optimal, meaning that an iterative process of model fitting, evaluation, and tuning is frequently necessary.
Moreover, the evaluation, steering, and presentation of results from applied text analytics is significantly less straightforward than with numeric data. What is the best way to find the most informative features when features can be words, word fragments, or phrases? How do we know which classification model is best suited to our corpus? How can we know when we have selected the best value for k in a k-means clustering model?
It is these types of questions, coupled with our need to iterate toward an optimal, deployable solution as efficiently as ...