back

Goal

         In this analysis, we want to determine how many times each artist has appeared in the Billboard Top 50, and whether the number of repeated artists in the charts has changed over the years.


Creating the Subdataset:

• Parsed Dates and Calculated Points
        We converted all dates into years only, extracting the year from the original “YYYY-MM-DD” format. Ranking positions were transformed into a point system (Position 1 = 100 points, Position 2 = 99, etc.).  


• Cleaned Performer Strings
        Performer names were cleaned by removing unnecessary characters, fixing whitespace, resolving unmatched symbols, correcting typos, and discarding incomplete or meaningless entries.  


• Applied the “Lloyd Price Rule”
        A custom mapping preserved unusual artist names. Collaborations were split using regex keywords (feat, ft, with, and, x, /, +, vs). Special cases like Lloyd Price were handled by summing solo and collaboration counts.  


• Expanded Performer Rows and Aggregated Artist Activity by Year
        Each performer received their own row after splitting collaborations. Appearances were aggregated by artist and year.  


• Selected the Top 50 Artists per Year
        Only the top 50 artists per year were retained to ensure readability and enable visualization.  

Final Structure

Year	The year
Artist	The name of the artist
Count	Number of appearances in the rankings per year
Rank	Total rank points accumulated per year