Jump to navigation Jump to search
- Examining Trends in a Performance Sport
- Data set: WCA Database
- What can we discover about how people improve in a field over time? To explore this I looked through the World Cubing Association database. A database with tens of thousands of Rubik's cube solves for thousands of people.
-Identifying and downloading the target data set
- The WCA Dataset was easily downloaded as a set of SQL inserts. The file can be downloaded from here.
-Data cleaning and pre-processing
- The issue was that the .sql file was in MS-SQL or OracleSQL, so some mass modifications to the file had to be made. Primarily it was with changing smallint(n) to int, and `tablename` without the `.
-Load the data into your Postgres instance
- It took a few times to get everything from the script all working, but the script was successfully run on my directory on BigFe.
- Develop queries to explore your ideas in the data
- People with more than 100 3x3 averages of 5.
SELECT personname, count(average) FROM results
WHERE eventid = '333' GROUP BY personname HAVING count(average) > 100
ORDER BY count(average)
- Develop and document the model function you are exploring in the data
- Develop a visualization to show the model/patterns in the data
- Node: as3
- Path to storage space: /scratch/big-data/tristan
- The visualization(s)
- The story