Aparna Ananthasubramaniam
@AparnaAnanth729
Followers
375
Following
779
Media
18
Statuses
170
PhD student @UMSI @UMSocialWork computational social sciences. Director @HHCollab. Researcher @MAFPajarito. Previously BS/MS @Stanford. she/her/hers
Ann Arbor, MI
Joined September 2019
RT @BrendanNyhan: New job ad: Assistant Professor of Quantitative Social Science, Dartmouth College . Please share with your networks. I am….
0
49
0
RT @AndersGiovanni: What actually happens on social media when users get access to integrated AI?🤖.In our new preprint, we present a contro….
0
2
0
@david__jurgens @DanielMRomero @TheWebConf Check out our paper here: And a prior paper where we developed the agent-based model and applied it to understand geographic diffusion:
nature.com
npj Complexity - Networks and identity drive the spatial diffusion of linguistic innovation in urban and rural areas
0
0
1
@david__jurgens @DanielMRomero @TheWebConf Our work suggests that modeling cultural production requires explicitly incorporating the role of multiple social factors in the process of diffusion -- including the important heterogeneity in the roles network and identity play in cultural production. 9/9.
0
0
0
@david__jurgens @DanielMRomero @TheWebConf Modeling hashtag diffusion using network+identity effects is especially important when initial adopters have high racial and demographic similarity, and for hashtags related to certain topics including sports, news, and film. 8/9
0
0
0
@david__jurgens @DanielMRomero @TheWebConf However, network effects alone best modeled specific properties related to a hashtag's popularity while identity effects alone best modeled the hashtag's adopter composition. 7/9
0
0
0
@david__jurgens @DanielMRomero @TheWebConf Aggregating over the performance on all 10 properties, we find that the diffusion of hashtags is best modeled using a combination of network + identity effects. 6/9
0
0
0
@david__jurgens @DanielMRomero @TheWebConf We got 1,337 innovative hashtags on Twitter and seeded each model with the hashtag's initial adopters and Twitter mention network. We analyze how well each model's *synthetic* cascades match the *actual* cascade on 10 properties of popularity, growth, and adopter composition. 5/9
0
0
0
@david__jurgens @DanielMRomero @TheWebConf Empirically, separating network and identity effects is virtually impossible, since they're closely linked. We use an agent-based model of network + identity effects in diffusion. Then we create network-only and identity-only counterfactual models to separate their effects. 4/9.
0
0
0
@david__jurgens @DanielMRomero @TheWebConf Our paper explains hashtag diffusion using the interaction between network and identity effects (e.g., geographic identity). The network exposes users to a hashtag. A user's identity helps determine whether they adopt the hashtag and, therefore, also shapes future exposures. 3/9
0
0
0
@david__jurgens @DanielMRomero @TheWebConf The diffusion of hashtags is often modeled as a network-only process, using only the social network to simulate the spread of cultural innovation. However, hashtag diffusion is actually the result of multiple interacting social factors. 2/9.
0
0
0
How did #IGotAThingFor become a thing?. Louise Zhu @david__jurgens @DanielMRomero and I explored the roles of networks and identity in the adoption of hashtags in our new @TheWebConf paper (Poster 01, Thu 5pm)!.. #www2025 #thewebconf2025 1/9.
9
4
4
RT @alexolegimas: Are you PhD student interested in AI/ML & economics? . Apply to the Machine Learning in Economics Summer Institute at Chi….
0
81
0
RT @jblumenstock: Great opportunity to join the team at Berkeley! . I'm recruiting a postdoc to work at the intersection of CS and applied….
0
16
0
RT @radamihalcea: 🌟I will be hosting a workshop for first-time NSF proposal writers, with strategies & tips for writing a successful first….
docs.google.com
Join an interactive Zoom workshop with Professor Rada Mihalcea from the University of Michigan, designed to guide first-time NSF grant writers through the key steps for crafting a competitive...
0
21
0
RT @NLIHC: HUD released today its 2024 Point-In-Time (PIT) Count, a snapshot of the number of people experiencing homelessness on a given n….
0
10
0
RT @dggoldst: Microsoft's Computational Social Science group may have the opportunity to hire one researcher. Senior: 0-3 yrs post PhD.htt….
0
47
0
RT @umsi: The COVID-19 pandemic reshaped mental health worldwide. In a groundbreaking study, @UMich researchers, including UMSI's @AparnaAn….
si.umich.edu
Narratives behind numbers point to need for better coordination, support by employers, communities, public health leaders in future crises, especially mental health
0
1
0
RT @EpLogMedia: Multiple debates have been held on relationships and mental health. We have tried to decipher how relationships and strong….
0
2
0
RT @agrima_seth: Honored and humbled to awarded the 2024 Google PhD Fellow in #NLProc. Thankful to my advisor @kentarotoyama, my mentors &….
0
4
0