Case study 2 - predicting topics of hotel reviews data
Our second case study will take a look at hotel reviews data and attempt to cluster the reviews into topics. We will be employing a latent semantic analysis (LSA), which is a name given to the process of applying a PCA on sparse text document—term matrices. It is done to find latent structures in text for the purpose of classification and clustering.
Applications of text clustering
Text clustering is the act of assigning different topics to pieces of text for the purpose of understanding what documents are about. Imagine a large hotel chain that gets thousands of reviews a week from around the world. Employees of the hotel would like to know what people are saying in order to have a better idea of what they are doing well and what can be improved.
Of course, the limiting factor here is the ability for humans to read all of these texts quickly and correctly. We can train machines to identify the types of things that people are talking about...
