73rd Annual Golden Globes Twitter Analysis Boying Gong, Jianglong - - PowerPoint PPT Presentation

73rd Annual Golden Globes Twitter Analysis

Boying Gong, Jianglong Huang, Peter Sujan, Shamindra Shrotriya, Tomofumi Ogawa February 8, 2016

Data Sources and Limitations

◮ Metadata - Golden Globe Nominees

◮ 87 people nominees and 35 movie nominees ◮ Manually collected/ annotated list of all nominees ◮ Twitter Screen Names ◮ Gender Flag ◮ Film/ TV Show Flag ◮ Age of Nominee/ Release Date

◮ Timelines

◮ Typically searched for top 3200 tweets from API ◮ Based on most recent tweets since Dec 10 2015 ◮ NLP processing performed e.g. removing stopwords etc

Quick summary of tweet data collected

Key themes of our data exploration

◮ Twitter Influence and Temporal Patterns - Jianglong ◮ Social Popularity of Winners and Nominees - Peter ◮ Sentiment Analysis - Boying ◮ Pre-Post-During Golden Globe Analysis - Tomo

When Do They Tweet?

1Mon 2Tue 3Wed 4Thu 5Fri 6Sat 7Sun 5 10 15 20

Hour of the day Day of the week

500 1000

count

All Tweets Heatmap

1Mon 2Tue 3Wed 4Thu 5Fri 6Sat 7Sun 5 10 15 20

Hour of the day Day of the week

250 500 750 1000

count

Official Accounts Tweets Heatmap Heatmap For Tweet Density

When Do Celebrities Tweet? By Gender

1Mon 2Tue 3Wed 4Thu 5Fri 6Sat 7Sun 5 10 15 20

Hour of the day Day of the week

100 200 300

count

Female Tweets Heatmap

1Mon 2Tue 3Wed 4Thu 5Fri 6Sat 7Sun 5 10 15 20

Hour of the day Day of the week

50 100 150

count

Male Tweets Heatmap Heatmap For Tweet Density

Tweet “POWER” VS Twitter “TENURE”

0.0 2.5 5.0 7.5 10.0 1000 2000 3000

Tweet Tenure Tweet Power

factor(factor)

Old and Powerful Old and Weak Young and Powerful Young and Weak

Scatter Plot for Tweet Power VS Tweet Tenure

Profile of “Weak and Old” VS “Young and Powerful”

female tv winner 0.0 0.2 0.4 0.6 0.8

Proportion in Each Group Categories of Interest

as.factor(group)

Old and Weak Young and Powerful

Profile Plot

Follower/Following Behavior Exhibits Distinct Groups

Leonardo DiCaprio Steve Carell Mark Ruffalo Amy Schumer Anomalisa The Fencer

• Mr. Robot

Veep Wolf Hall Taraji P. Henson Aziz Ansari Rachel Bloom Queen Latifah

Active Popular Power

Not shown: Lady Gaga: 55029737 , 131197 Game of Thrones: NA , NA

1000 2000 3000 4000 0.0e+00 5.0e+06 1.0e+07 1.5e+07

Number of followers Number followed Result

L W

Nominee Following vs. Follower Counts

Movie/TV Accounts Appear Most Active

Active Other Popular Power 10 20 10 20 F FILM/SHOW M F FILM/SHOW M

Type count WINNER

L W

Breakdown of User Type by Style of Twitter Use

Following Similarity Distribution

Very dissimilar Similarity with self = 1 Similar pairs

1000 2000 0.00 0.25 0.50 0.75 1.00

Similarity Frequency

Distribution of Following Similarity

Similarity Measures Match Real-life Connections

Regina King, Viola Davis Taraji P. Henson, Idris Elba Queen Latifah, Idris Elba Fargo, American Horror Story: Hotel Spy, Spy Inside Out, The Good Dinosaur Taraji P. Henson, Viola Davis Felicity Huffman, American Crime Flesh and Bone, Sarah Hay Taraji P. Henson, Queen Latifah Rachel Bloom, Adam McKay Rachel Bloom, Amy Schumer Adam McKay, Amy Schumer Queen Latifah, Viola Davis Steve Carell, The Big Short 0.0 0.1 0.2 0.3 0.4

Similarity Pair

Following Similarity: Most Similar Pairs

Popularity Among All Users = Popularity Among Peers

2 4 6 8 10 12 14 16

log of retweet count log of mention count

Mention Counts Grouped by Celebrities

Adam McKay Alan Cumming Christian Slater Daniel Pemberton Emma Donoghue Ennio Morricone Felicity Huffman Joanne Froggatt JudithLight Patrick Stewart Rami Malek Rob Lowe Robin Wright Sylvester Stallone Gina Rodriguez

• Mr. Bob Odenkirk

Queen Latifah Rachel Bloom Steve Carell Taraji P. Henson Tobias Menzies Tomlin and Wagner Idris Elba Jane Seymour Fonda Leonardo DiCaprio Liev Schreiber Sarah Hay Bryan Cranston Mark Ruffalo Regina King Uzo Aduba Carter Burwell Jamie Lee Curtis Julia Louis−Dreyfus Aziz Ansari ryuichi sakamoto Viola Davis Kirsten Dunst patrick wilson Lady Gaga Brie Larson Melissa McCarthy Jeffrey Tambor Amy Schumer Caitriona Balfe 2 4

logMentionCount Name Win

L W

Top Domains of External Links

Percentage of External Links Used by Gender

Male PercentageM Female PercentageF youtube 0.91 instagram 6.01 instagram 0.55 whattheflicka 1.56 whosay 0.40 youtube 0.39 apple 0.28 facebook 0.28 facebook 0.15 twimg 0.22 usanetwork 0.14 latina 0.14 ifc 0.14 ew 0.13 hollywoodreporter 0.12 variety 0.13 twimg 0.11 theguardian 0.13 ew 0.10 yahoo 0.10

Males tweeted more Post-Globes than Pre-Globes

1000 2000 3000 −20 −10 10 20

lag (days) retweet count WINNER_FLAG

black blue L W

lag vs retweet count

5 10 15 −20 −10 10 20

lag (days) tweet count WINNER_FLAG

black blue L W

lag vs tweet count

2000 4000 6000 8000 −20 −10 10 20

lag (days) favorite count WINNER_FLAG

black blue L W

lag vs favorite count

60 80 100 120 140 −20 −10 10 20

lag (days) tweet length WINNER_FLAG

black blue L W

lag vs tweet length

Male Winners vs Male Losers (mean)

Females winners were less favorited Post-Globes!

1000 2000 3000 4000 −20 −10 10 20

lag (days) retweet count WINNER_FLAG

black blue L W

lag vs retweet count

10 20 30 −20 −10 10 20

lag (days) tweet count WINNER_FLAG

black blue L W

lag vs tweet count

1000 2000 3000 4000 −20 −10 10 20

lag (days) favorite count WINNER_FLAG

black blue L W

lag vs favorite count

60 80 100 120 −20 −10 10 20

lag (days) tweet length WINNER_FLAG

black blue L W

lag vs tweet length

Female Winners vs Female Losers (mean)

Actors are using Twitter for activism post-Globes!

Conclusion and Next Steps

◮ Nominee Analysis shows distinct behaviour patterns when

summarised by gender, age, temporal components

◮ Next Steps:

◮ Do the analysis for Golden Globes 2015, 2014, 2013 ◮ Look at nominee influence via external data e.g. box office ◮ Download large amount of historical follower analysis ◮ Analysis of twitter users the nominees follow