STATS 101-108 - Assignment 2 - Part A
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
Assignment 2 - Part A
This assignment is based on the content covered in Chapter 3 Prediction | Matapae. Before starting on this assignment you should have read the corresponding notes and completed the interactive exercises for this chapter. Lecture 3b covers many of the skills that you need to complete this assignment. For support in completing this assignment head along to the drop-in sessions. |
The process for submitting this assignment has changed since assignment 1. Refer to the assignment information video included with Part B for how to create an html Gle to submit via Canvas. |
Assignment information
|
Click this link to make a copy of a Google doc with a template for your answers.
DO NOT delete any of the template text. Type your answers and paste your images directly into this document.
See the assignment information video with assignment 2 part B for more information on using this template.
Introduction
|
|
|
|
Question One |
|
Review the article Study IDs visual characteristics that make people 'like' images on Instagram. Summarise FIVE things you learned from this article inyourownwords.
Write these as Gve bullet points, using the bullet point option within CANVAS quizzes. Each bullet point needs to be a full sentence. Question Two |
Head to unsplash.com and search for photos based on one word. You can choose whatever word you want, but make sure there are at least 60 photos listed when you search for this word.
For each photo, the website provides information about its popularity, such as views, likes, downloads, and other relevant information, like how many days the photo has been on the website.
Use the Popularity contest app to obtain images and related popularity data about 60 photos. Take a screenshot (snip) of your photos and copy this as an image to paste into your assignment.
The app will also create a training data set and testing data set using photos from the website based on your chosen search term. You will be given two links: The Grst link is to iNZight Lite with your training data. The second link is to a CSV Gle with your testing data. Copy both links.
Open your training data in iNZight Lite using the Grst link and create a plot of mean_likes_per_day and
mean_views_per_day . You will need to decide which variable to use as the explanatory variable and which variable to use as the response variable (both options are valid). Copy this plot as an image to paste into your assignment.
Describe the key features of the data visualised in your scatterplot. Include an interpretation of the rank correlation coeacient in your description.
Construct your answer using the following template which is already in the template document.
The one word used to search for photos:
An image based on a screenshot of your photos:
The link to your training data set through iNZight Lite: The link to your testing data set as a CSV:
The Grst scatterplot you created:
Your description of the key features of data visualised in the scatterplot:
Question Three
You now need to develop a prediction model using your training data and test your model using the testing data.
Using the training data from Question Two and iNZight Lite, add a linear model to the plot. Copy this plot as an image to paste into your assignment.
Copy the equation of the linear model from the summary tab.
Using the training data and the plot determine an appropriate amount of error for your prediction model. Refer to lecture 3b and the notes for how to do this.
Write the equation for your prediction model which includes the error term. Make sure you use the names of the variables in your model e.g. mean_likes_per_day and mean_views_per_day , being careful to be consistent with which variable you have decided is the explanatory variable, and which variable is the response variable.
Use the Prediction model testing app embedded below to generate a plot using your prediction model and your testing data. Copy the plot generated as an image to paste into your assignment.
Prediction model tester
Paste the link to your testing data CSV file:
Load data and enter prediction model
Use the plot generated to calculate the percentage of photos that
were correctly “captured” by the prediction intervals generated from your model.
Use the percentage correct calculation to write one sentence that
describes how well your model made predictions.
Construct your answer using the following template which is already in the template document.
The scatterplot you created with the linear model Gtted: The equation for the linear model:
The equation for your prediction model:
The plot generated from using your prediction model with the testing data:
The percentage of photos that were correctly “captured” by your prediction model:
One sentence describing how well your prediction model “worked”:
Copyright information. Remember that the content and delivery of content in this course are protected by copyright. Material belonging to others may have been used in this course and copied by and solely for the educational purposes of the University of Auckland under licence. You may copy the course content for the purposes of private study or research, but you may not upload onto any third party site, make a further copy or sell, alter or further reproduce or distribute any part of the course content to another person.For example, you are not allowed to download and share any materials from this course book on external websites such as Chegg, StuDoc, Course Hero, or other similar platforms. |
2023-03-27