1.0.0

What's Changed

• Created package structure, added datasets by @CraigAdlam in #1
• Added interactive app_radar_chart.py standalone dash code by @CraigAdlam in #2
• Re-added datasets, added interactive country map by @CraigAdlam in #3
• Updated country_map.py and tab3_content by @CraigAdlam in #4
• Changed country.py, tab3_content to be sized by proportion, not fixed by @CraigAdlam in #5
• Added reflection-milestone2.md to doc folder by @CraigAdlam in #6

New Contributors

• @CraigAdlam made their first contribution in #1
• @kulaphongj made their first contribution by pushing directly to the main branch.
• @Nijat27 made their first contribution by pushing directly to the main branch.

Full Changelog: https://github.com/CraigAdlam/spotify_dashboard/commits/1.0.0

Spotify Dashboard - Reflection

Tab 1 – Discover Music Taste

What has been done:

• Genres, artists, track names, filters
• Statistics table of music, including min, mean, and max, are shown in the table based on the filtering
• Genre proportion based on the filtering
• Radar chart of scaled values based on the filtering
• If the filters are not selected, the values of the table and charts will be calculated based on all data

Limitations:

• Slow loading (track names list of the filter contains about 80,000)

What needs to be improved:

• Remove data/make the dataset callable when needed (stored variable)
• Make search area smaller/equal to the chart area beside
• Use Card element header instead of paragraph

Tab 2 – Find New One?

What has been done:

• Created sliders for each music feature on the left pane which interacts with the plots on the right pane (Feature Bar Plot, Feature Bar Plot, and the Song Table)
• When the sliders are used to select feature ranges, Feature Bar Plot will be updated with the average musical feature proportions, showing the trends
• When the sliders are used to select feature ranges, Feature Bar Plot will be updated to reflect the musical features showing fine-tuned visualizations
• When the sliders are used to select feature ranges, Song Table gets updated with the top 10 songs based on the feature ranges selected

What needs to be improved:

• Rename ambiguous table columns
• Add mark ticks/slider position
• Potential rearranging of the plot positions for maximum impact
• Change background colors/Card colors to match the Spotify green theme
• Add a checkbox to implement selecting rows based on popularity

Tab 3 – Explore Globally

What has been done:

• Created interactive map where users can select specific countries to display the top 10 songs based on popularity (e.g., top 10 songs rated 100, 50 etc.)
• Created 2 horizontal bar charts that display the top 10 songs and artist globally
• Created main popularity slider to control the output that is presented
  • Included for both global top 10 songs and top 10 artists horizontal bar charts
  • Also controls the output for the top 10 songs when specific country selected
• Incorporated additional package (pycountry) that converts ISO-alpha3 country codes (required for map display) into full country names for interpretability

What needs to be improved:

• Change background colors/Card colors to match the Spotify green theme
• Add second index to slider (so there are 2) so a range of popularity can be selected
• Limit table to popularity or include 3/4 metrics (mainly covered in first 2 tabs)
  • If so, potentially add an additional plot or summary statistics in the bottom right to further describe popularity insights

General Concerns

• Data loading issues
• Flexible sizing or fixed (if can make scrollable, fixed might be better)
• When table exceeds background area, how to make background conform new width