Walking Through Twitter: Sampling a Language-Based Follow Network
0 likes448 views
The document discusses the challenges and methodologies in analyzing social media follow networks, particularly on Twitter, focusing on the Australian context. It outlines research questions related to filter bubbles, social communities, and the spread of information, while detailing methods adapted for API restrictions. The analysis involves community detection and keyword extraction from sampled influential accounts, addressing the implications of changing data access on future research.
Walking Through Twitter: Sampling a Language-Based Follow Network
- 1. Walking Through Twitter Sampling a Language-Based Follow Network Felix Victor Münch1 , Ben Thies1 , Cornelius Puschmann1 , Axel Bruns2 1 Leibniz Institute for Media Research | Hans-Bredow-Institut (HBI), Germany 2 Digital Media Research Centre (DMRC), Queensland University of Technology (QUT), Australia AoIR Flashpoint Symposium, Urbino, 24. June 2019
- 2. Problem ‘I can’t get no population …’ ● follow/friend networks of Online Social Networks (OSNs) arguably main predictor for content exposure (despite sponsored content and algorithmic sorting of timelines) ● APIs too restrictive for collection of comprehensive follow networks
- 3. Opportunities Possible Research Questions ● “Are there filter bubbles/echo chambers?” ● “Are there social and/or topical communities/issue publics?” ● “Who can spread which content most efficiently?” ● “Can we predict the spread of (fake) news?” ● “Who let the bots out? And where?”
- 4. Background
- 5. The Australian Twittersphere Bruns, A., Moon, B., Münch, F. V., & Sadkowsky, T. (2017). The Australian Twittersphere in 2016: mapping the follower/followee network. Social Media + Society, 3(4). https://doi.org/10.1177/2056305117748162 Münch, F. V. (2019). Measuring the Networked Public – Exploring Network Science Methods for Large Scale Online Media Studies. Queensland University of Technology. https://doi.org/10.5204/thesis.eprints.125543
- 6. Method: Sample the influentials
- 7. Our adaptation of the ‘rank degree’ method Based on: Salamanos, N., Voudigari, E., & Yannakoudakis, E. J. (2017). Deterministic graph exploration for efficient graph sampling. Social Network Analysis and Mining, 7(1), 24. https://doi.org/10.1007/s13278-017-0441-6
- 8. Our adaptation of the ‘rank degree’ method Main adaptations (amongst others) mostly due to API restrictions: ● Undirected → Directed ● Fewer walkers (200) ● Walkers do not collapse when ending up on the same node ● Last 5000 friends only ● ‘Degree’ stays constant and is equal to follower count reported by Twitter API ● Only collect edges to accounts with German as their interface language (not possible anymore due to API changes ¯_(ツ)_/¯ … we work on a solution.)
- 12. Coverage Ranked distribution of the percentage of ‘friends’ of accounts in a random sample (n=1000, filtered for having at least 2 ‘friends’) found in our influencer sample (excl. nodes with in-degree 0) and a random sample of the same size (181k accounts)
- 13. Reach Ranked distribution of the percentage of accounts reached in a random sample (n=1000, filtered for having at least 2 ‘friends’) by accounts in our influencer sample (excl. nodes with in-degree 0) and by accounts in a random sample of the same size (181k accounts)
- 14. Test study: Community detection and keyword extraction
- 15. Community detection with infomap algorithm 3-core of our sample network; coloured by communities detected with the infomap community detection algorithm; node size represents Page Rank
- 16. Keyword extraction with chi-squared criterion # active accounts keywords top accounts tag short tag 4 2015 e3, stream, xd, e32019, nintendo, twitch, game, crossing, pc, zelda, animal, gameplay, cyberpunk2077, games, switch, xbox, trailer, cyberpunk, gaming, xboxe3, uff, nice, geil, keanu, pk, spiele, live, nen, lol, mega unge, dagibee, Gronkh, MelinaSophie, LeFloid, iBlali, Taddl, rewinside, HandIOfIBlood, PietSmiet YouTubers & Gaming Youtube & Games 3 1855 berlin, innen, spd, berliner, studie, unternehmen, diskutieren, cdu, fordern, themen, wichtiger, bundesregierung, deutschen, digitalisierung, thema, wurde, wurden, annefrank, klimaschutz, mehr, juni, zeigt, deutschland, interview, politik, brandenburg tazgezwitscher, Die_Gruenen, Tagesspiegel, c_lindner, gutjahr, dunjahayali, sigmargabriel, sixtus, HeikoMaas, spdde German politics German Politics 11 1414 frauenstreik, schweiz, schweizer, glarner, svp, bern, frauenstreik2019, zürich, kanton, grosse, basel, nationalrat, frauen NZZ, 20min, viktorgiacobbo, Blickch, tagesanzeiger, srfnews, MikeMuellerLate, watson_news, migros, srf3 Swiss politics / women's strike Swiss Politics 2 1044 saison, trainer, bundesliga, neuzugang, spieler, dfb, fc, gerest, em, wechselt, estland, wechsel, transfer, tor, sieg, mannschaft, wm, cup, liga DFB_Team, FCBayern, ToniKroos, MarioGoetze, esmuellert_, Podolski10, Manuel_Neuer, Bundesliga_DE, ZDFsport, JB17Official German football Soccer 17 767 övp, spö, fpö, wien, österreich, bierlein, oenr, österreichs, ibiza, wiener, strache, kickl, türkis, hofer, parlament, nationalrat, zib2, abdullah, mandat, wahlkampf, zentrum, bundeskanzlerin, heinz, österreicher, glyphosat, antrag, blau florianklenk, sebastiankurz, IngridThurnher, kesslermichael, HannoSettele, vanderbellen, Gawhary, HBrandstaetter, HHumorlos, michelreimon Austrian politics Austrian Politics 33 751 menschen, nazi, merz, deutschland, afd, wer, akk, cdu, spd, jahre, leben, wäre, grünen, obwohl, jemand, klar, nazis, ja, liebe, wähler, bitte, viele, immer, mann, wissen, wünsche, herr, seit ntvde, Muhterem_Aras, munich_startup, Literaturtest, _schwarzeKatze, bmzimmermann, ninjawarriorrtl, cesurmilusoy, MetalabVie, S_chill_ing NTV (news channel) NTV 6 720 album, keanu officiallyjoko, damitdasklaas, siggismallz, thisiscro, latenightberlin, gamescom, EtienneToGo, ralphruthe, TheRocketBeans entertainment Entertainment 20 563 saschalobo, StefanOsswald, CarstenRossiKR, ragnarh, cbgreenwood, breitenbach, BBDO, Ugugu, ring2, julianheck Digital Communication DigiCom 22 510 Kachelmann, schmitt_it, RomeoMicev, Perspektive360, paulespcforumde, naturkosmetix, TUIDeutschland, glockendoktor, nachtnebel, HaraldKlein ? ? 13 453 lt, bett, morgens, hast, schön, hasse, müde, essen, ja, möchte, manchmal, mama, bitte, frau, nachts, immer, bier, nein, kaffee, gern, gar, jemand, weiß, mal, geh, glaube, ach, scheiße, tweets, hund, kinder, hätte, abends KuttnerSarah, Regendelfin, vergraemer, DrWaumiau, HappySchnitzel, katjaberlin, ArminRohde, RenateBergmann, hashcrap, peterbreuer "Feuilleton" / authors / copy Writers Authors 23 429 10jahrejk, joko, klaas, mtv, canadiangp, hitzekindofmagic, fernsehen ProSieben, MattBannert, newshausen, haveonenme, Mone_Horan, about_riki_, delay1, sarahofer8, sweettweetangie, JoanBleicher Pro7 (Private TV broadcaster) Pro7 42 376 dessau, roßlau, afd, görlitz, islam, vergewaltigt, migranten, asylbewerber, gefährder, abgelehnter, sed, wippel, patrioten, greta, niger, hosni, stärkste, grüne, fridayforfuture, afrikaner, rosslau, linke, gretathunberg, asylbewerbe, verbrechen, altparteien, merkel, linksradikale, grünen, habeck, schweigt, zerreißen, wählt, mädchen, flüchtling, sachsen, vergewaltigung, maas, bürger, islamistischer, sexuell DonJoschi, AfD, MSF_austria, Alice_Weidel, SteinbachErika, Joerg_Meuthen, Beatrix_vStorch, GrumpyMerkel, krone_at hard rights / migration / refugees Hard Right 30 369 mydirtyhobby, femdom, boobs, latex, milf, mistress, cock, anal, tits, blowjob, busty, fetish, livecam, pornstar, mdh, sexy, findom, highheels, clip, manyvids, cam, cumshot, cum, webcam, pussy, goddess, dirty, bdsm, porn, hot, horny, booty, slave Erotik_Center, AnnyAuroraPorn, texas_patti, LenaNitro1, sandy226, Julietta_com, ModelRoxxyX, swo2212, LucyCatOfficial, SuziAnneVX Porn Porn
- 17. Tagged Community graph Community graph of communities in the 3-core of our sample with over 300 accounts, at least 80 active accounts during the examined time frame, and edges with a weight of at least 150; edge width represents weight; edge direction follows clockwise curvature; edges coloured by source node; node size represents the number of accounts in each community
- 18. Outlook ● Adaption to API changes ● Bootstrapping the seed pool ● Other language-based spheres ● Topical mining ● Other community detection methods ● Bot detection
- 19. Thanks! @FlxVctr, @BenAThies, @cbpuschmann, @snurb_dot_info
- 20. Interesting 🤔 Münch, F. V. (2019). Measuring the Networked Public – Exploring Network Science Methods for Large Scale Online Media Studies. Queensland University of Technology. https://doi.org/10.5204/thesis.eprints.125543