Happy Halloween to those who celebrate! We posted the solution JSON file to Cloud Quest 19. Check it out and let us know what you think! Send suggestions to academy@alteryx.com or leave a comment below!
For more detailed instructions on how to import and export Designer Cloud workflow files, check out the pinned article Cloud Quest Submission Process Update.
In this week’s Cloud Quest, you have been given a small sample of reviews submitted for Halloween costumes purchased on Amazon during the years 2010–2020. Use the dataset to answer the following questions:
Bonus: One of the exciting capabilities of modern large language models (LLMs) is their ability to quickly parse complex text. While not necessary for this quest, try using an LLM to extract simplified costume names from the product name column such as "Top Gun Flight Suit," or "Train Engineer."
Your results will vary depending on the prompt and LLM vendor used. For this example, we copied a summarized list of product names into Claude 3.5 and used the following prompt:
“Attached is a list of Halloween costume descriptions from a dataset. Attempt to isolate JUST the name of the costume. Remove any information about size, gender, or brand. Brand names are typically uppercase. Keep any descriptive text relating to a costume’s material such as ‘leather’ and if it is a specific article of clothing such as ‘dress.’”
A combination of the RegEx, Sort, DateTime, Summarize, Join, Filter, and Multi-Row Formula tools should solve your problem, but not necessarily in this sequence.
If you find yourself struggling with any of the tasks, feel free to explore these interactive lessons in the Alteryx Academy for guidance:
Once you have completed your quest, go back to your Analytics Cloud library.
Data Source: https://www.kaggle.com/datasets/thomaskonstantin/popular-halloween-costumes-amazon-reviews/code
The "% YoY Change" values in the solution for Task 3 are all 100 higher than they should be. Or maybe I don't understand that metric.
It is good and fun one.
I think the Costume name the 2nd answer is coming out from LLM, which dont have access now.
So I am using the long one in the input data instead.
Same as everyone else, I didn't use LLM so costume names differ. FUn challenge though Happy Halloween !
Done!