Claustro
An incipient examination over origins of cloisters
When working with categories of things, we want to know where do thoe belong.
For example, we have the business traveler that stays at the best hotels, the coach that travels in their own van, the family that is only exploring the most modest B&B. These are all different segments, different tastes, and they all belong to a specific segment that we can tease out of the data that comes out of some sufficiently motivated traveling agency.
The Cloisters exploration is going to be about how data presents enough characteristics to give an indication of what segments are relevant to our epxloration. We can get these segments based oin what our goals are, what is shown at a given moment, and from the instrument that we used to collect the data: all instruments have an underlying bias, or a idiosincratic limitation that allow the researcher to explain a point, because that is relevant to what is being considered crucial to the business at hand.
The procedure is simple: we identofy numeroc characteristics and create clusters based on those, and use those to predict how a market will fit into the exsiting clusters. Another approach is ti use data that uses categorical data and find another set of making clusters, and approach those to explain and understand the phenomema described by the categorical data.
In the first case we will use simply k-clusters, while in the second we will approach this using medoids. The relevant sources will be added later, so users can see for themsleves whereas the aproach fit their interests.
Since this analysis is geared towards business uses, the explanation is going to be a simple function, managing to extract salient features and highlight prominent facts, and limiting features that might ditrsact from urgent identification of characteristics that might be useful.
Exercise One
Reading the data from dataplants, using the states as categories with which to find poklants that might share characteristics about their habitat.
Also, the exercise uses technoiques to read from a file where there is minimal organization and covert it into something that can be used to study categorical data.
Ultimately, the goal is to use a fucntion to describe the readoing and cleaning.