End of Unit 3 and 4 Design Project and Oral Presentation: Water Usage

End of Unit Design Project and Oral Presentation: Water Usage

Objective:

Students will apply their learning of the third and fourth units of the curriculum by completing an end of unit design project.

Materials:

  1. Computers

  2. IDS Unit 4 – Project and Oral Presentation (LMR_U4_Design Project)

End of Unit 4 Project and Oral Presentation: Water Usage

At the beginning of this unit, you explored a 2010 data set from the Los Angeles Department of Water and Power (DWP). You also created a Participatory Sensing campaign to investigate water usage around your community.

For this assignment, you will use both data sets to apply what you have learned in unit 4 and to answer the research question from the beginning of the unit:

How can we help city officials use Participatory Sensing to find out how water is being used around your neighborhood?

Your assignment is as follows:

  1. You and a partner will predict water usage for the moth of June using a subset of the dwp_2010 data set, which is called dwp_student.

    • Load the dwp_student data set.

    • Using this data, create two data sets: training and testing. Name these data sets student_train and student_test.

    • Create the best prediction model that you can based on your training data. Remember to set.seed(123) when creating your own training and testing data.

    • You’re building this model with data from July 2010 to May 2011. You will use your model to predict water usage for June 2011.

    • After you settle on a specific model, submit your model (code) to your teacher. If you created any new variables, submit the code you used to create them as well.

    • What do the variables included in your prediction model say about how Angelenos use water?

    • You will evaluate the prediction accuracy based on a separate set of data. Your teacher will give you another data set. Use this data set to evaluate your prediction. The pair with the smallest prediction error based on mean squared error (MSE), is the winner.

  2. Using your Participatory Sensing data, explain how water is being used in your neighborhood. Make sure you use evidence from your PS data analysis. Be sure to answer the research question and your statistical questions.

Create a 5-minute presentation comprising 4 to 5 slides that explains your model, the predicted value for June 2011 water consumption, and the findings using your campaign data. Be sure to include detailed explanation of how you and your partner decided to create your prediction model, and how it performed on the test data set your teacher provided in your presentation. Each person must participate in the presentation. In addition to the presentation, submit a 2-4 page, double-spaced summary of your analysis including plots/graphs.

Note to teacher about the testing data set: The data set you will provide for students to test their prediction models is called dwp_teacher. It is recommended that you provide the data set’s name upon students’ submission of the code for their prediction models.