The Epic Treasure Hunters are trying to determine if there is anything that they can do to improve their chances of winning treasure map portals. Winning means getting to the 5th floor of the dungeon. Dungeons are accessed via portals. If you haven’t already, view the video for a better understanding of the data.
You will be working with three different sets of data for this project:
1. Pre-Patch Training and Pre-Patch Scoring data
2. Post-Patch Training and Post-Patch Scoring data
3. Treasure map reviews
Complete the following steps to meet the requirements of this project. Use this document as a template, placing screen shots where indicated.
Part 1: Linear and Logistic Regression
1. In RapidMiner, set up the training and scoring sets for the Post-Patch data.
2. Run a Logistic Regression to determine if the attributes that capture data BEFORE a portal are good predictors of winning/losing the portal. You will need Location, MemberJob, #Members, #MobGroups, ChestLoot, and Outcome
3. Paste your coefficient table here
4. How many predicted wins are in the results?
5. Run a Logistic Regression to determine if the attributes that capture data about the portal experience are good predictors of winning/losing the portal. You will need the remainder of the attributes to predict Outcome excluding Feather (if a feather is 1 then it is for sure a win)
6. Paste your coefficient table here
7. How many wins are in the results?
8. What commonalities can you see across the records that are predicted to win?
9. Of those predicted to win, how many actually reached floor 5?
10. Repeat the run in Step 5, but use a Linear Regression to predict Floor instead of Outcome. Paste your coefficient table here.
11. What is the highest predicted floor?
12. Repeat the run in Step 2, but use a Linear Regression to predict Floor instead of Outcome. Paste your coefficient table here.
13. What is the highest floor predicted?
14. Now let’s use the Pre-Patch data. Swap your data sets and run Step 2 again using the Logistic regression to predict Outcome. Paste your coefficient table here.
15. How many predicted wins are in the results?
16. Repeat the run in Step 5. Paste your coefficient table here.
17. How many predicted wins are in the results?
18. What commonalities can you see across the records that are predicted to win?
19. Of those predicted to win, how many actually reached floor 5?
20. Repeat the run in Step 5, but use a Linear Regression to predict Floor instead of Outcome. Paste your coefficient table here.
21. What is the highest predicted floor?
22. Repeat the run in Step 2, but use a Linear Regression to predict Floor instead of Outcome. Paste your coefficient table here.
23. What is the highest predicted floor?
Part 2: Decision Trees
1. Using the Pre-Patch data run a Decision Tree that predicts Outcome. Include all attributes EXCEPT Floor and Feather. Paste an image of your Decision Tree here.
2. How many wins are predicted in the Decision Tree?
3. Swap your data sets and run a Decision Tree on the Post-Patch data. Paste an image of your Decision Tree here.
4. How many wins are predicted in the Decision Tree results?
Part 3: Text Mining
1. Read the document Treasure Map Reviews into RapidMiner.
2. Apply all operators you practiced with in Chapter 12 including the Sort. Filter Examples to those with a Total >= 3.
3. Paste an image of your main Process window here
4. Paste an image of your subprocess window here
5. How many rows are in your ExampleSet in your results?
6. What word shows up the most and how many times? Maps,
7. Paste a screen shot of the first 10 rows of your ExampleSet in your results here
Part 4: Storytelling with data
Now you must create a report for your audience. The report must include:
1. Title page (make it fun/interesting)
2. Introduction paragraph outlining what the audience will be reading through in the report. No more than 3-4 sentences.
3. Four visualizations.
a. These may be any type of visualization covered in the storytelling with data text.
b. Avoid the chart types noted in the text (like pie charts)
c. Do not paste images of your coefficient tables or results tables as visualizations – CREATE visualizations based on your findings
d. If you need to run additional processes to examine the data in order to tell the story you can. You are not bound to what you ran for Parts 1-3 of the project for information for your visualizations
e. Include a description for each visualization that clearly and concisely explains it.
4. Conclusion: the story matters – your audience is composed of gamers. Ultimately your report needs to answer this question:
a. What can we do to improve our chances of winning treasure map portals?
b. Explicitly answer the question which must be supported by your previous visualizations
5. Spelling and grammar matter and points will be deducted for errors in each
CS 340 Milestone One Guidelines and Rubric Overview: For this assignment, you will implement the fundamental operations of create, read, update,
Retail Transaction Programming Project Project Requirements: Develop a program to emulate a purchase transaction at a retail store. This
7COM1028 Secure Systems Programming Referral Coursework: Secure
Create a GUI program that:Accepts the following from a user:Item NameItem QuantityItem PriceAllows the user to create a file to store the sales receip
CS 340 Final Project Guidelines and Rubric Overview The final project will encompass developing a web service using a software stack and impleme
