Reference no: EM132018257
Project -
During this week you will have chance to work with what you have learned on a new set of data.
Step 1: Please choose one of the following sets of data and build either a classification or a regression model.
Classification Problem: student-por.csv. This is just like the student data we have been working with, but for portuguese class grades instead of math class grades. You may already have this file from when you downloaded the student data for this course.
Regression Problem: Download the concrete compressive strength data. The goal with this dataset is to use the 8 features of the concrete to predict the concrete compressive strength. This is similar to how we built a regression model that used car features to predict car mpg.
Step 2: Complete a "description" of the data. This is a high level description of the dataset, just like we did in class.
Step 3: Complete an "exploration" of the data. This is an indepth exploration of the data, using plots. Just like we did in class.
Step 4: Build two models on the data. For example, if you chose the classification dataset, build a decision tree and logistic regression model. If you chose the regression dataset, build a linear regression and a neural network or a decision tree regression model.
Step 5: Evaluate the performance of the models using the metrics and diagnostics we used in class.
Step 6: Describe and interpret the model, what features did the model use select via regularization or splitting?
(I realize we did not cover methods to determine which features are used in the neural network after applying regularization, so you can skip this if you build a neural network). What does the model tell us about the data? For example: Are students more likely to fail if the drink on the weekend? Do increase amounts of fly ash in concrete increase its compressive strength? (I am not asking you to answer these specific questions, these are just examples of the kinds of questions you will want to answer when describing and interpreting the model.
Attachment:- Assignment File.rar