As mentioned during my intial data cleaning, here is the description that Spotify’s provides about this list of “inferences” it has made about me:
We draw certain inferences about your interests and preferences based on your usage of the Spotify service and using data obtained from our advertisers and other advertising partners. This includes a list of market segments with which you are currently associated. Depending on your settings, this data may be used to serve interest-based advertising to you within the Spotify service.
Only when looking through my settings on the Spotify website to download the data they have collected about me did I realize that I had linked my Facebook account 🤦🏻.
Spotify reports making 661 inferences about me.
# A tibble: 6 x 1
inference
<chr>
1 1P_Custom_Auto_BMW
2 1P_Custom_Discovery_Streamers
3 1P_Custom_Passionate_Curators
4 1P_Custom_Soothing_Sounds
5 1P_Custom_T-Mobile_Switchers
6 1P_Custom_iPhone_11_Users Most of the bottom 25 are labeled TEST. I do not know what this means, but will need to take the variance into account as I try to extract the key words and phrases.
# A tibble: 25 x 1
inference
<chr>
1 PMR TEST - 3P_Custom_Current Snuggle Buyers_04May2020_US
2 PMR TEST - 3P_Custom_Office Supplies Buyers_16April2020_US
3 PVHannahS2_US_AmazonPrimeVideo_Q2Q32020_Exposed
4 Recaptcha_scores_filtered
5 TEST_CUSTOM_FacialCleaning_US
6 [TEST] 3P_Custom_BeerLovers_US
7 [TEST] 3P_Custom_Parents of College Students_US
8 [TEST] 3P_Skin Care Buyers_US
9 [Test] - 3P_ TV_Sports TV Enthusiasts_US
10 [Test] - 3P_Custom_HHI $80K+_06Jan2020_CA
# … with 15 more rows
Spotify believes I am both a registered Republican and registered Democrat and hold strong right and left political views.
[1] "3P_Custom_NBC Viewers _US"
[2] "3P_Custom__CNN_US"
[3] "3P_Custom__MSNBC Viewers_US"
[4] "3P_Custom__MSNBC_ Political News Viewers_US"
[5] "3P_Politics - Active in Politics_US"
[6] "3P_Politics - Any Democrat_US"
[7] "3P_Politics - Any Left_US"
[8] "3P_Politics - Any Republican_US"
[9] "3P_Politics - Any Right_US"
[10] "3P_Politics - Interested in Politics_US"
[11] "3P_Politics - Political Donor Any_US"
[12] "3P_Politics - Registered Democrat_US"
[13] "3P_Politics - Registered Republican_US"
[14] "3P_Politics - Registered Voters_US"
[15] "3P_Politics - Strong Left_US"
[16] "3P_Politics - Strong Right_US"
[17] "[Test] - 3P_Custom__CNN_US"
[18] "[Test] - 3P_Custom__MSNBC Viewers_US" # A tibble: 1 x 1
inference
<chr>
1 3p_Anime_Manga_Enthusiast_Es# A tibble: 12 x 1
inference
<chr>
1 3P_Couple without Kids_DE
2 3P_Custom_Expectant Parents_US
3 3P_Custom_Parents of Boys 6-11_US
4 3P_Custom_Parents of Kids 5-13_28Apr2020_US
5 3P_Custom_Parents of Toddlers_30Nov2020_US
6 3P_Custom_Parents with Children Age 6-17_15March2020_US
7 3P_Custom_Parents with Children Ages 9-18_23Nov2020_US
8 3P_Custom_Parents with Kids 6-12_US
9 3P_Custom_Parents_12Nov2020_CA
10 3P_Family/Children's Shoppers_US
11 3P_Parents_US
12 [TEST] 3P_Custom_Parents of College Students_US Clean inferences.
Analyses:
- word cloud
- sentiment analysis?
- make graph with edge weight as word similarities
- distribution of size of components
- use centrality metrics to identify most important