Business Problem: Can demographic changes in population be utilized in lieu of polling for the purpose of predictive electoral analysis? Convert this business problem into Research question and complete milestone 2. Write the Introduction In this milestone, you will write the introduction to your final paper. Specifically, the introduction should explain and discuss your research, including the final research question, dataset, and the chosen machine learning model. It should give the appropriate context of the analysis for your reader, and explicitly state your final research question as well as your hypothesis and the machine learning model you've selected.

A good introduction not only gives background material on the machine learning method and dataset, but it also dives into how data analysis principles specifically apply to the situation: why did you decide on that particular machine learning model for this specific dataset and research question? In particular, now that you have selected your dataset and formulated a research question, you should explore and characterize the dataset by doing an Exploratory Data Analysis. This data appraisal will inform your introduction, which will provide the context for your data analysis. Finally, it should consider the metrics you will utilize to evaluate the results of your analysis. Deliverable For milestone 2, please ensure you have the following sections at a minimum: Introduction: Describe the purpose, type, intended populations, and uses of the analysis report to establish an appropriate context for the data analysis plan. Appraise the data within the context of the problem to be solved and industry standards. How will you use the data? For example, expound upon the limitations of the data set in the context of your needs. Please do especially ensure you clearly state your final, possibly revised, research question in a separate sub-section Exploratory Data Analysis: Characterize the data set. For example, what is the purpose such data are generally used for? You should do the data preparation for analysis and compute and present the summary statistics and the Exploratory Data Analysis (EDA) for the dataset. Measurable Metrics and Utilities: Explain the utilities that you will be using and how the data supports that choice. The utilities, in this case, are the different machine learning algorithms you are considering, along with any particular software implementations you are thinking of using for it. You should also think of measurable metrics to evaluate the results of your analysis, as well. Please ensure you have an explicit discussion of which evaluation metrics you will utilize for each model and why. Tips Please see the following tutorials on doing Exploratory Data Analysis (EDA); pleae note, these are in IPython format so you'll need to run it in a Jupyter notebook or Google Co-Lab: tutorials_2-index-eda.html One of the main things you want to identify is those variables that have a high variability as the initial assumption is that the higher the variance, the more entropy, or information, that feature might have if it's correlated with the target variable. In addition, you should check for correlations between the predictor variables (multicollinearity) as well as with the response/target variable. You'll likely also want to determine which features seem most informative at this point. Additional References Notes from my previous undergraduate Data Analytics Course on doing Exploratory Data Analysis: