Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Re-focusing on 3 Twitter accounts Follow up to the first part published on Slideshare: https://www.slideshare.net/seinecle...
In the previous study, we found 3 clusters of interest for epidemiology specialists: See https://www.slideshare.net/seinec...
Using these 3 accounts as seeds, we run the method to find related accounts Because our interest here is to zoom in on epi...
Group 1 (270 members, 17% total) key terms: infectious disease epidemiologist (32), epidemiologist (23), epidemiology (17)...
Zoom on each key group The following slides zoom on one of the colored regions of the global picture
Group 1 decomposed in subgroups. 270 members. Key terms: health, infectious disease, epidemiologist sub-group 1-2 (87 memb...
Group 2 decomposed in subgroups. 405 members. Key terms: health, disease, coronavirus sub-group 2-1 (117 members, 29%) key...
Group 3 decomposed in subgroups. 490 members. Key terms: health, disease, global sub-group 3-3 (162 members, 33%) key term...
Group 4 decomposed in subgroups. 128 members. Key terms: china, beijing, reporter sub-group 4-2 (34 members, 27%) key term...
Group 5 decomposed in subgroups. 284 members. Key terms: health, reporter, science sub-group 5-4 (50 members, 18%) key ter...
Upcoming SlideShare
Loading in …5
×

Part 2: covid-19 on Twitter, with a focus on 3 new seed accounts

31 views

Published on

First part of the analysis chose to picture the global conversation on Twitter by picking 3 accounts in Italian, Spanish and English languages.
We identified several clusters gathering professional epidemiologists.
In this follow up, we run a new analysis where the starting points are 3 Twitter accounts of epidemiologists that were found in these clusters. The end goal is to identify many more epidemiologists.

Published in: Data & Analytics
no profile picture user

  • Be the first to comment

  • Be the first to like this

Part 2: covid-19 on Twitter, with a focus on 3 new seed accounts

  1. 1. Re-focusing on 3 Twitter accounts Follow up to the first part published on Slideshare: https://www.slideshare.net/seinecle/covid-actors-relations-and-topics-on-twitter-as-of-march-30-2020 published on April 1, 2020 by Clément Levallois, Associate Professor at emlyon business school Based on the methodology published in (free pdf download): Benabdelkrim, M., Levallois, C., Savinien, J., & Robardet, C. (2020). Opening Fields: A Methodological Contribution to the Identification of Heterogeneous Actors in Unbounded Relational Orders. M@n@gement, 23(1), 4-18. https://doi.org/10.37725/mgmt.v23.4245 Contact: levallois@em-lyon.com or @seinecle on Twitter
  2. 2. In the previous study, we found 3 clusters of interest for epidemiology specialists: See https://www.slideshare.net/seinecle/covid-actors-relations-and-topics-on-twitter-as-of-march-30-2020/12 -> We now pick the Twitter accounts that are most central to each cluster: @SCBriand, @JenniferNuzzo and @SCBriand
  3. 3. Using these 3 accounts as seeds, we run the method to find related accounts Because our interest here is to zoom in on epidemiologists. We could have chosen any other focus (journalists, healthcare companies, state organizations, etc.)
  4. 4. Group 1 (270 members, 17% total) key terms: infectious disease epidemiologist (32), epidemiologist (23), epidemiology (17), disease dynamic (15), virus (12), epitwitter (11), infectious disease epidemiology (11), mer (11), studying (8), oxford (8) key accounts: @richardneher, @PeterHotez, @profvrr Group 2 (405 members, 26% total) key terms: coronavirus (54), covid (44), covid coronavirus (19), real (19), disaster (19), coronavirus covid (15), update (13), report (11), disease control (11), assistant (11) key accounts: @trvrb, @angie_rasmussen, @VirusWhisperer Group 3 (490 members, 31% total) key terms: access (96), organization (64), peer reviewed (58), health system (47), society (45), online (42), european (36), privacy policy (35), oncology (32), nih (27) key accounts: @IDSAInfo, @ECDC_EU, @AmeshAA Group 4 (128 members, 8% total) key terms: beijing (156), china (66), chinese (64), china correspondent (15), scmp (11), shanghai (11), macro (8), asia (7), hong kong (6), freelance (5) key accounts: @ChuBailiang, @xinyanyu, @XijinHu Group 5 (284 members, 18% total) key terms: public (58), politico (52), covering health (36), statnew (32), health reporter (32), senior correspondent (27), policy reporter (23), send (23), reporter (15), washingtonpost (15) key accounts: @DrewQJoseph, @ashishkjha, @marynmck The result is a network of 1,579 Twitter accounts, decomposed in:
  5. 5. Zoom on each key group The following slides zoom on one of the colored regions of the global picture
  6. 6. Group 1 decomposed in subgroups. 270 members. Key terms: health, infectious disease, epidemiologist sub-group 1-2 (87 members, 32%) key terms: candidate (8), data (8), prof (6), writer (6), dean (5), epitwitter (5), disease ecologist (5), focusing (5), asst prof (5), policy (5) key accounts: @joshmich, @thelonevirologi, @PeterHotez sub-group 1-0 (38 members, 14%) key terms: microbiology (5), statistician (3), blogger (3), lumc leiden (3), occasional (3), gym (3), drug (3), • (3) key accounts: @EvolveDotZoo, @PaulSaxMD, @EpiEllie sub-group 1-1 (70 members, 26%) key terms: global (15), dynamic (8), imperial college (5), public health epidemiologist (5), london (5), health policy (5), infectious disease modelling (5), data scientist (5), control (5), gate (5) key accounts: @MMFill, @Caroline_OF_B, @SRileyIDD sub-group 1-3 (75 members, 28%) key terms: lab (11), biology (6), emerging infectious disease (5), preparedness (5), centre (5), molecular (5), diagnostic (5), immunity (5), dad (3), officer (3) key accounts: @richardneher, @KindrachukJason, @K_G_Andersen
  7. 7. Group 2 decomposed in subgroups. 405 members. Key terms: health, disease, coronavirus sub-group 2-1 (117 members, 29%) key terms: correspondent (15), host (8), american (8), staff (8), real (8), senior (6), modeling (5), biologist (5), obama (5), math (5) key accounts: @CT_Bergstrom, @JenniferNuzzo, @JHSPH_CHS sub-group 2-2 (171 members, 42%) key terms: reporter (27), interested (19), hiv (15), student (15), microbiology (15), doc (15), global (15), epidemiologist (14), university (12), emerging (11) key accounts: @trvrb, @angie_rasmussen, @VirusWhisperer sub-group 2-3 (16 members, 4%) key terms: york (5), city (3), event (3) key accounts: @TheWarMonitor, @NYCMayor, @IntelCrab sub-group 2-0 (101 members, 25%) key terms: coronavirus (33), coronavirus covid (15), latest (11), spread (11), update (11), covid (10), daily (8), source (6), covid coronavirus (6), dave robert (5) key accounts: @MRC_Outbreak, @BNODesk, @2019nCoVwatcher
  8. 8. Group 3 decomposed in subgroups. 490 members. Key terms: health, disease, global sub-group 3-3 (162 members, 33%) key terms: microbiology (83), infection (76), antibiotic (42), emerging (27), infectious disease (26), public (19), evolution (19), microbial (19), resistance (19), control (18) key accounts: @IDSAInfo, @ECDC_EU, @AmeshAA sub-group 3-2 (99 members, 20%) key terms: globalhealth (32), development (27), health system (27), global health (26), organization (21), health policy (15), advocate (15), country (8), london (8), connecting (8) key accounts: @WHOAFRO, @GHS, @francetim sub-group 3-1 (43 members, 9%) key terms: oncology (32), cancer (27), food safety (8), medical oncology (5), hematology (5), commission (5), clinical oncology (5), standard (3), specialty (3), peer reviewed article (3) key accounts: @OncJournal, @PennMedicine, @ASCO sub-group 3-5 (120 members, 24%) key terms: weekly (15), leading (11), doctor (8), trusted (8), journal published weekly (8), source (8), analysis (8), medical journal (8), food (8), surgery (8) key accounts: @statnews, @picardonhealth, @EricTopol sub-group 3-0 (62 members, 13%) key terms: massachusett (19), std (8), nation (8), privacy policy (7), boston (5), largest (5), cdcgov (5), safety (5), emergency (5), governor (5) key accounts: @CDCMMWR, @ASTHO, @NIOSH
  9. 9. Group 4 decomposed in subgroups. 128 members. Key terms: china, beijing, reporter sub-group 4-2 (34 members, 27%) key terms: scmp (8), bloomberg (8), national (5), junkie (3), chinese law (3), post (3), freelance (3), political (3), international (3), scmp new view (3) key accounts: @Chao_Deng, @LiYuan6, @QiZHAI sub-group 4-3 (35 members, 27%) key terms: foreign (5), chinese (5), living (3), year (3), focusing (3), foreign affair (3), consumer (3), article (3), çº½çº¦æ— ¶æš¥é© (3), york time reporter (3) key accounts: @S_Rabinovitch, @ChuBailiang, @hancocktom sub-group 4-1 (48 members, 38%) key terms: based (11), shanghai (11), reuter (8), bbc (5), researcher (5), view mine (5), policy (5), beijing (4), professor (4), editor (3) key accounts: @xinyanyu, @XijinHu, @TheJohnSudworth
  10. 10. Group 5 decomposed in subgroups. 284 members. Key terms: health, reporter, science sub-group 5-4 (50 members, 18%) key terms: nature (8), phd (8), science writer (8), science (5), natgeo (5), nerd (5), climate (5), space (5), energy (5), rocket (3) key accounts: @laurahelmuth, @ferrisjabr, @NewHorizonsBot sub-group 5-1 (42 members, 15%) key terms: surgeon (5), ft (5), ap (5), public (4), senior (4), atlantic (3), officer (3), senior health (3), nursing (3), john (3) key accounts: @emoryhealthsci, @jameshamblin, @marynmck sub-group 5-5 (21 members, 7%) key terms: emergency physician (8), ontario (5), foamed (5), physician (4), care (4), health quality ontario (3), father (3), ucsf (3), host (3), husband (3) key accounts: @grahamwalker, @DavidJuurlink, @skathire sub-group 5-2 (28 members, 10%) key terms: khnew (5), calhealthline (3), jersey (3), station (3), staff (3), journalist covering (3), hospital (3), health policy reporter (3), reporter covering health (3), cq (3) key accounts: @_melaevans, @jburns18, @IshaniG sub-group 5-3 (77 members, 27%) key terms: statnew (64), journal (23), health science (23), pharma (19), biotech (19), bloomberg (11), wall street journal (11), wsj (11), managing editor (8), reporter covering (8) key accounts: @DrewQJoseph, @meggophone, @GideonGil sub-group 5-0 (66 members, 23%) key terms: medicaid (15), kaiser (11), service (11), medicare (8), professor (8), center (7), policy (6), associate (6), native (5), vp (5) key accounts: @sangerkatz, @neel_shah, @ashishkjha

×