How often would you flush the tube, and how much water would you deliver with each flush?

A company that designs tractors wants to analyze thousands of customer messages to discover opportunities for improving their products.  They do not have predefined “opportunity” categories but want to uncover patterns in the data.  Which technique should they use?

Untutored in philosophy, Alexander the Great excelled as an empire builder. 

How much free water will this provide? 

BONUS (+5 points): In lecture 5, we discussed the different types of word embeddings, Frequency based and Prediction based. Which of the following fall under the Prediction Based Embeddings? Select all that apply.

In LDA, what are topics?

What is the main output of LDA?

In LDA, the number of topics (K) is automatically determined by the model and does not need to be manually specified.

Which of the following is the primary output of a “bag of words” model?

Which machine learning task is most likely unsupervised?