ºÝºÝߣ

ºÝºÝߣShare a Scribd company logo

Trend Makers and Trend Spotters in a Mobile Application

?
?
Daniele Quercia
Daniele Quercia

WHO creates trends in a mobile sharing app? accidentals or influentials? Answer: influentials DO exist, yet they are not few but many! http://profzero.org/publications/trend13sha.pdf

1 of 16
Downloaded 13 times
Trend Makers and Trend Spotters in a Mobile Application Xiaolan Sha? Daniele Quercia? Pietro Michiardi? Matteo Dell¡¯Amico? ?EURECOM ?Yahoo! Research Barcelona
Who create trends?
Two-Step Flow Mass Media In?uentials Normal
444 Accidental Influentials JOURNAL OF CON FIGURE 2 of these assumptions is demonstrabl clearly correct either¡ªthe empirical ev SCHEMATIC OF NETWORK MODEL OF INFLUENCE inconclusive. Thus we will also presen variations of the basic model that relax and the randomness assumptions. Another advantage of formally de? work, even with such a simple mod de?ne more precisely what we mea Previous empirical work has address should be considered in?uential, b mains elusive (Weimann 1991). Clas of Coleman et al. (1957) and Merton individuals who directly in?uence m of their peers should be considered cent market research studies have co ber may be as high as 14 (Burson-M studies, by contrast, de?ne in?uent terms: Keller and Berry (2003), fo ?uentials as scoring in the top 10% ership test, while Coulter et al. (2002 treat the top 32% as in?uentials. Here we follow the latter approac ential as an individual in the top q% tribution p (n). From a theoretical pers value of q that we specify is necessa we have already argued that dichoto tween opinion leaders and followers a 1 in?uence can only ?ow from opinion leaders to fol- derived nor empirically supported. O lowers, in ?gure 2, it can ?ow in either direction. Second, ever, is not to defend any particular d in ?gure 2 in?uence can propagate for many steps, [D. Watts,but to examine the claim that in?uen P. Dodds JSTOR 2007] whereas in ?gure 1 it can propagate only two. We note, reasonable, self-consistent manner¡ª however, that, in both cases, ?gure 2 is consistent with of diffusion processes. From this pers available empirical evidence¡ªarguably more so than ?g- de?nition has the advantage (over d ure 1. Numerous studies, including that of Katz and La- absolute numbers) that it can be app
Context
Dataset Users technology-savvy, design-conscious Pictures technology, lifestyle, music, design and fashion 9,316 users uploaded 6,395 pictures and submitted 13,893 votes.
Identification Trends A simple burst detection method Spotters/Makers Spotter Score: how many, early, popular of the trends Maker Score: how often Typical Users All active users (>=2 votes/uploads) who is not spotter or a maker. 140 Makers; 671 Spotters; 1,705 Typical Users
Characterizations Features Activity Content Network Geographical Hypotheses [Kolmogorov-Smirnov tests] Spotters/Makers vs. Typical Users Spotters vs. Makers
Results Spotters/Makers vs. Typical users More active, more popular Spotters vs. Makers More votes, less uploads,wider spectrum of interests
Prediction Features £û Activity Content Trend Spotters Every User Social Network Trend Makers User Space Geography User Space
Predictors Follower Geo Span Upload Diversity Daily Uploads Vote Diversity Daily Votes Wandering #Followers #Followees Life Time Age Life Time 0.21 Activity Daily Uploads 0.02 -0.12 Daily Votes 0.05 -0.09 0.47 ? Upload Diversity 0.02 0.09 0.40 ? 0.08 Content Vote Diversity 0.04 0.08 0.22 0.08 0.42 ? Wandering 0.004 0.13 0.16 0.11 0.06 0.05 Geographical Follower Geo Span 0.05 0.12 0.16 0.10 0.12 0.11 0.23 #Followers 0.03 0.23 0.37 ? 0.14 0.22 0.16 0.44 0.16 #Followees 0.05 0.17 0.52 ? 0.31 ? 0.29 ? 0.22 0.56 ? 0.21 0.64 ? Network Network Clustering 0.03 0.13 0.22 0.04 0.24 0.23 -0.001 0.27 ? 0.08 0.22 Spotter Score 0.07 0.18 0.03 0.01 0.05 0.10 0.04 0.07 0.13 0.11 0.15 Maker Score 0.07 0.10 0.06 0.01 0.07 0.06 0.02 0.12 0.12 0.09 0.10 Table 5. Pearson Correlation coef?cients between each pair of predictors. Coef?cients greater than ¡À0.25 with statistical signi?cant level < 0.05 are marked with a ?. Practical Implications CONCLUSION The ability of identifying trend spotters and trend makers has A community is an emergent system. It forms from the ac- implications in designing recommender systems, marketing tions of its members who are reacting to each other¡¯s behav- campaigns, new products, privacy tools, and user interfaces. ior. Here we have studied a speci?c community of individuals who are passionate about sharing pictures of items (mainly Recommender Systems. Every user has different interests fashion and design items) using a mobile phone application. and tastes and, as such, might well bene?t from personalized This community has a speci?c culture in which a set of habits, suggestions of content. These suggestions are automatically attitudes and beliefs guide how its members behave. In it, we produced by so-called ¡°recommender systems¡±. Typically,
Prediction True positive rate 0.8 S-logistic 0.4 S-svm M-logistic M-svm 0.0 0.0 0.2 0.4 0.6 0.8 1.0 False positive rate
y trend spot- Age 2e-04 0.001 preliminary Life Time 0.006 * 0.001 * ers opens up Successful Spotters/Makers Daily Votes (Daily Uploads) 0.007 * 0.16 * ferences be- Vote Diversity (Upload Diversity) 0.38 * 0.14 * Wandering -6e-15 -7e-15 #Followers 2e-05 0.009 * Network Clustering 0.08 0.28 * no previous spotters and l hypotheses (b) Linear Regression er that trend Features log(Score) tend to vote Spotters Makers compared to Age 0.36 * 0.01 (H3.1), vote Life Time 0.19 * 0.0001 Daily Votes (Daily Uploads) 0.16 - -1.03 * ote more di- Vote Diversity (Upload Diversity) 7.28 * - -1.09 * ). After run- Wandering -2.1e-13 -1.4e-15 nd that trend #Followers -0.06 0.01 * ers who, by Network Clustering 2.75 - -0.64 * oth H3.1 and R2 0.15 0.65 ote, we ?nd Adjusted R2 0.14 0.64 oad and vote s vote items kers act in a Table 3. Coef?cients of the linear regression. A correlation coef?cient d in [20, 18] within 2 standard errors is considered statistically signi?cant. We high- light and mark them with *. ality content. items in the le they vote of followees, daily uploads, daily votes, and content diver- de spectrum
Summary Successful Spotters Early adopters who vote items from various categories. Successful Makers Users who upload items belonging to speci?c categories, tend to be followed by users from different social clusters.
Conclusions Who Create Trends? Regular individual with speci?c interests connected with early adopters with diverse interests.
Questions?

More Related Content

Trend Makers and Trend Spotters in a Mobile Application

  • 1. Trend Makers and Trend Spotters in a Mobile Application Xiaolan Sha? Daniele Quercia? Pietro Michiardi? Matteo Dell¡¯Amico? ?EURECOM ?Yahoo! Research Barcelona
  • 3. Two-Step Flow Mass Media In?uentials Normal
  • 4. 444 Accidental Influentials JOURNAL OF CON FIGURE 2 of these assumptions is demonstrabl clearly correct either¡ªthe empirical ev SCHEMATIC OF NETWORK MODEL OF INFLUENCE inconclusive. Thus we will also presen variations of the basic model that relax and the randomness assumptions. Another advantage of formally de? work, even with such a simple mod de?ne more precisely what we mea Previous empirical work has address should be considered in?uential, b mains elusive (Weimann 1991). Clas of Coleman et al. (1957) and Merton individuals who directly in?uence m of their peers should be considered cent market research studies have co ber may be as high as 14 (Burson-M studies, by contrast, de?ne in?uent terms: Keller and Berry (2003), fo ?uentials as scoring in the top 10% ership test, while Coulter et al. (2002 treat the top 32% as in?uentials. Here we follow the latter approac ential as an individual in the top q% tribution p (n). From a theoretical pers value of q that we specify is necessa we have already argued that dichoto tween opinion leaders and followers a 1 in?uence can only ?ow from opinion leaders to fol- derived nor empirically supported. O lowers, in ?gure 2, it can ?ow in either direction. Second, ever, is not to defend any particular d in ?gure 2 in?uence can propagate for many steps, [D. Watts,but to examine the claim that in?uen P. Dodds JSTOR 2007] whereas in ?gure 1 it can propagate only two. We note, reasonable, self-consistent manner¡ª however, that, in both cases, ?gure 2 is consistent with of diffusion processes. From this pers available empirical evidence¡ªarguably more so than ?g- de?nition has the advantage (over d ure 1. Numerous studies, including that of Katz and La- absolute numbers) that it can be app
  • 6. Dataset Users technology-savvy, design-conscious Pictures technology, lifestyle, music, design and fashion 9,316 users uploaded 6,395 pictures and submitted 13,893 votes.
  • 7. Identification Trends A simple burst detection method Spotters/Makers Spotter Score: how many, early, popular of the trends Maker Score: how often Typical Users All active users (>=2 votes/uploads) who is not spotter or a maker. 140 Makers; 671 Spotters; 1,705 Typical Users
  • 8. Characterizations Features Activity Content Network Geographical Hypotheses [Kolmogorov-Smirnov tests] Spotters/Makers vs. Typical Users Spotters vs. Makers
  • 9. Results Spotters/Makers vs. Typical users More active, more popular Spotters vs. Makers More votes, less uploads,wider spectrum of interests
  • 10. Prediction Features £û Activity Content Trend Spotters Every User Social Network Trend Makers User Space Geography User Space
  • 11. Predictors Follower Geo Span Upload Diversity Daily Uploads Vote Diversity Daily Votes Wandering #Followers #Followees Life Time Age Life Time 0.21 Activity Daily Uploads 0.02 -0.12 Daily Votes 0.05 -0.09 0.47 ? Upload Diversity 0.02 0.09 0.40 ? 0.08 Content Vote Diversity 0.04 0.08 0.22 0.08 0.42 ? Wandering 0.004 0.13 0.16 0.11 0.06 0.05 Geographical Follower Geo Span 0.05 0.12 0.16 0.10 0.12 0.11 0.23 #Followers 0.03 0.23 0.37 ? 0.14 0.22 0.16 0.44 0.16 #Followees 0.05 0.17 0.52 ? 0.31 ? 0.29 ? 0.22 0.56 ? 0.21 0.64 ? Network Network Clustering 0.03 0.13 0.22 0.04 0.24 0.23 -0.001 0.27 ? 0.08 0.22 Spotter Score 0.07 0.18 0.03 0.01 0.05 0.10 0.04 0.07 0.13 0.11 0.15 Maker Score 0.07 0.10 0.06 0.01 0.07 0.06 0.02 0.12 0.12 0.09 0.10 Table 5. Pearson Correlation coef?cients between each pair of predictors. Coef?cients greater than ¡À0.25 with statistical signi?cant level < 0.05 are marked with a ?. Practical Implications CONCLUSION The ability of identifying trend spotters and trend makers has A community is an emergent system. It forms from the ac- implications in designing recommender systems, marketing tions of its members who are reacting to each other¡¯s behav- campaigns, new products, privacy tools, and user interfaces. ior. Here we have studied a speci?c community of individuals who are passionate about sharing pictures of items (mainly Recommender Systems. Every user has different interests fashion and design items) using a mobile phone application. and tastes and, as such, might well bene?t from personalized This community has a speci?c culture in which a set of habits, suggestions of content. These suggestions are automatically attitudes and beliefs guide how its members behave. In it, we produced by so-called ¡°recommender systems¡±. Typically,
  • 12. Prediction True positive rate 0.8 S-logistic 0.4 S-svm M-logistic M-svm 0.0 0.0 0.2 0.4 0.6 0.8 1.0 False positive rate
  • 13. y trend spot- Age 2e-04 0.001 preliminary Life Time 0.006 * 0.001 * ers opens up Successful Spotters/Makers Daily Votes (Daily Uploads) 0.007 * 0.16 * ferences be- Vote Diversity (Upload Diversity) 0.38 * 0.14 * Wandering -6e-15 -7e-15 #Followers 2e-05 0.009 * Network Clustering 0.08 0.28 * no previous spotters and l hypotheses (b) Linear Regression er that trend Features log(Score) tend to vote Spotters Makers compared to Age 0.36 * 0.01 (H3.1), vote Life Time 0.19 * 0.0001 Daily Votes (Daily Uploads) 0.16 - -1.03 * ote more di- Vote Diversity (Upload Diversity) 7.28 * - -1.09 * ). After run- Wandering -2.1e-13 -1.4e-15 nd that trend #Followers -0.06 0.01 * ers who, by Network Clustering 2.75 - -0.64 * oth H3.1 and R2 0.15 0.65 ote, we ?nd Adjusted R2 0.14 0.64 oad and vote s vote items kers act in a Table 3. Coef?cients of the linear regression. A correlation coef?cient d in [20, 18] within 2 standard errors is considered statistically signi?cant. We high- light and mark them with *. ality content. items in the le they vote of followees, daily uploads, daily votes, and content diver- de spectrum
  • 14. Summary Successful Spotters Early adopters who vote items from various categories. Successful Makers Users who upload items belonging to speci?c categories, tend to be followed by users from different social clusters.
  • 15. Conclusions Who Create Trends? Regular individual with speci?c interests connected with early adopters with diverse interests.