Uyezve komendari

Chikamu chino chakagadzirirwa kushandiswa sokunoratidza, pane kuti kuverenga somunhu rondedzero.

  • Nhanganyaya (Section 2.1)

Imwe mhando hwokutevera kuti haamo muchitsauko ichi ethnography. Kuti uwane pamusoro ethnography mune digitaalinen Spaces ona Boellstorff et al. (2012) , uye mamwe ari ethnography makawanda digitaalinen uye mumuviri Spaces ona Lane (2016) .

  • Big Data (Section 2.2)

Kana uri repurposing data, pane vaviri pfungwa mazano anogona kukubatsira kunzwisisa matambudziko zvinobvira kuti kusangana. Chokutanga, unogona edza kufungidzira yakanaka dataset nokuda dambudziko rako uye enzanisa kuti kuna dataset yauri kushandisa. Sei vane zvakafanana uye sei vari akasiyana? Kana hamuna unganidza mashoko ako pachako, pangangova kuva musiyano pakati zvaunoda uye zvaunofanira kuita. Asi, unofanira kusarudza kana kupesana aya zviduku kana zvikuru.

Chechipiri, yeuka kuti munhu akasikwa uye dzakaunganidza mashoko enyu nokuda kwechikonzero. Unofanira kuedza kunzwisisa pfungwa dzavo. mhando iyi zvisizvo-okugadzira kunogona kukubatsira kuziva matambudziko zvinobvira uye vakanga vaine kwavakarerekera ako repurposed mashoko.

Hapana tabvumirana chete yeshoko "mashoko guru", asi tsanangudzo dzakawanda dzinoita omerera 3 Vs: (semuenzaniso, vhoriyamu, siyana, uye velocity Japec et al. (2015) ). Pane kungotarira zvechadenga data, tsanangudzo yangu rinonyanya zvakawanda kuti nei mashoko akasikwa.

usanganidzo Kwangu hurumende administrative mashoko mukati muchikwata che guru mashoko ari zvishoma rechienzi. Vamwe vakaita iyi, inosanganisira Legewie (2015) , Connelly et al. (2016) , uye Einav and Levin (2014) . Kuti uwane nezvokukosha hurumende administrative mashoko yokutsvakurudza, ona Card et al. (2010) , Taskforce (2012) , uye Grusky, Smeeding, and Snipp (2015) .

Nokuti maonero administrative tsvakurudzo kubva mukati hurumende uwandu hurongwa, kunyanya US Census Bureau, ona Jarmin and O'Hara (2016) . Nokuti bhuku pakureba pakurapa administrative zvinyorwa tsvakurudzo Statistics Sweden, ona Wallgren and Wallgren (2007) .

Muchitsauko, ini zvishoma akaenzanisa bhurasho kuongorora yakadai General Social Survey (GSS) kuti munzanga vezvenhau Data bhuku akadai Twitter. Nokuti kwazvo uye nokungwarira kuenzanisa pakati chinyakare kuongorora uye magariro Vezvenhau data, ona Schober et al. (2016) .

  • Common unhu mashoko makuru (Section 2.3)

Ava 10 zvechadenga mashoko makuru dzave chinorondedzerwa nenzira dzakasiyana-siyana dzakasiyana maererano dzakasiyana-siyana nevanyori vakasiyana. Kunyora kuti kukanganiswa mafungiro angu panyaya idzi dzinosanganisira: Lazer et al. (2009) , Groves (2011) , Howison, Wiggins, and Crowston (2011) , boyd and Crawford (2012) , Taylor (2013) , Mayer-Schönberger and Cukier (2013) , Golder and Macy (2014) , Ruths and Pfeffer (2014) , Tufekci (2014) , Sampson and Small (2015) , Lewis (2015) , Lazer (2015) , Horton and Tambe (2015) , Japec et al. (2015) , uye Goldstone and Lupyan (2016) .

Kugara chitsauko chino, ndava akashandisa izwi digitaalinen zvisaririra, izvo ndinofunga achiri enyika. Chimwe yakakurumbira izwi nokuti digitaalinen zvisaririra zviri digitaalinen footprints (Golder and Macy 2014) , asi sezvo Hal Abelson, Ken Ledeen, uye Harry Lewis (2008) kutarawo, yakawanda zvakakodzera izwi zvimwe zviri digitaalinen Aparecen. Kana ukaronga footprints, imi muri kuziva zviri kuitika uye footprints vako havagoni kazhinji kurondwa kwauri pachako. Ndizvowo hakusi kwechokwadi nokuda kwenyu digitaalinen zvisaririra. Kutaura zvazviri, unenge kusiya zvisaririra zvose nguva pamusoro zvaunazvo ruzivo zvishomanana. Uye, kunyange zvazvo zvisarira izvi havana zita renyu pamusoro pavo, vanogona kazhinji kuva chokuita shure kwamuri. Nemamwe mashoko, ivo vava akafanana Aparecen: asingaoneki uye pachake chinozivisa.

guru

Kuti uwane pamusoro nei datasets makuru, anoshandura uwandu nemiedzo matambudziko, ona Lin, Lucas, and Shmueli (2013) uye McFarland and McFarland (2015) . nyaya idzi anofanira kutungamirira vatsvakurudzi kuti dzifunge hunobatsira kukosha pane nhamba kukosha.

Nguva dzose-mberi

Pamunenge muchikurukura nguva dzose-pamusoro data, zvakakosha kufunga kana uri kuenzanisa chaiyo vanhu chete pamusoro nguva kana uri kuenzanisa dzimwe kuchinja boka revanhu; ona Somuenzaniso, Diaz et al. (2016) .

Non-reactive

A bhuku chaiwo pane vasiri reactive matanho ari Webb et al. (1966) . Mienzaniso mubhuku kwerubwinyo rwakagara rwuripo-zuva iri digitaalinen zera, asi ivo vachiri richivhenekera. Nokuti mienzaniso yevanhu achichinja maitiro avo nokuti kuvapo vakawanda zvokuongorora, ona Penney (2016) uye Brayne (2014) .

sakwana

Kuti uwane pamusoro rekodhi linkage, ona Dunn (1946) uye Fellegi and Sunter (1969) (dzakaitika) uye Larsen and Winkler (2014) (ano). Similar aswedera yave yakatangawo kombiyuta zvesayenzi pasi mazita akadai date deduplication, Somuenzaniso kuzivikanwa, zita enzanirana, kutevedzera vasabatwa, uye kutevedzera rekodhi vasabatwa (Elmagarmid, Ipeirotis, and Verykios 2007) . Kunewo zvakavanzika kuchengetedza nzira kunyora linkage isingapwi zvinoda nokupfuuridzirwa pachake azivikanwe mashoko (Schnell 2013) . Facebook zvakavawo akakudziridza mberi kubatanidza zvinyorwa zvavo kuvhota maitiro; ichi chakaitwa kuti vaongorore chinhu kuedza kuti Ndichaudza kukuudza pamusoro muChitsauko 4 (Bond et al. 2012; Jones et al. 2013) .

Kuti uwane pamusoro kuvaka ukoshi, ona Shadish, Cook, and Campbell (2001) , Chapter 3.

sasvikika

Kuti uwane pamusoro AOL kutsvaka nerogi debacle, ona Ohm (2010) . Ini mazano pamusoro partnering mapoka nehurumende muChitsauko 4 apo ini tsanangura kuedza. Vanoverengeka vanyori vakaratidza nezvemhosva tsvakurudzo kuti rinovimba dzisingambosvikiki data, ona Huberman (2012) uye boyd and Crawford (2012) .

Imwe nzira yakanaka yokuti vanotsvakurudza kuyunivhesiti kuwana mashoko mukana ndiko kushanda pakambani ive Intern kana kushanyira mutsvakurudzi. Mukuwedzera kugonesa date kuwana, izvi ruchatibatsira nevatsvakurudzi kudzidza zvakawanda nezvekuti mashoko akasikwa, izvo zvinokosha kuongorora.

Non-mumiririri

Non-representativeness idambudziko guru kune vaongorori uye hurumende vanoda kuti mitsara pamusoro rose vemo. Ndiyo zvishoma nehanya makambani kuti vari yemanyorero nezvezvandaifanira vanoshandisa avo. Kuti uwane pamusoro sei Statistics Netherlands anoona nyaya asiri representativeness vemabhizimisi makuru data, ona Buelens et al. (2014) .

In Chapter 3, Ndichaenda kutsanangura zvakabudiswa uye uchimutarira zvizhinji kuchakurukurwa zvakadzama. Kunyange kana date vari isiri mumiririri, pasi zvimwe ezvinhu, vanogona kuva weighted kubudisa kufungidzira zvakanaka.

kukukurwa

System vasakukurwa zvakaoma chaizvo kuona kubva kunze. Zvisinei, MovieLens chirongwa (inokurukurwa zvakawanda muChitsauko 4) rave kumhanya kwemakore anopfuura 15 inopiwa tsvakurudzo boka zvechikoro. Saka, havana kunozivikanwa uye vakagoverana ruzivo pamusoro nzira kuti hurongwa yave rakavapo pamusoro nguva uye sei izvi kuti tinzwe ongororo (Harper and Konstan 2015) .

Vanoverengeka nyanzvi nezvezvandaifanira vasakukurwa muna Twitter: Liu, Kliman-Silver, and Mislove (2014) uye Tufekci (2014) .

Algorithmically nenyadzi

Ini kutanga kunzwa shoko rokuti "algorithmically nenyadzi" rakashandiswa Jon Kleinberg muhurukuro. Nepfungwa huru shure performativity ndechokuti dzimwe dzidziso sayenzi evanhu vari "injini kwete makamera" (Mackenzie 2008) . Kureva, chaizvoizvo kuumba nyika pane chete akaritora.

sviba

Masangano uwandu Governmental vanodana date rokutsvaira, uwandu mashoko fora. De Waal, Puts, and Daas (2014) tsanangura uwandu date fora mitoo dzakabudirira kuti ongororo Data uye kuongorora izvo papi ivo kushandiswawo guru mashoko zvinyorwa, uye Puts, Daas, and Waal (2015) rinotaura dzimwe chete pfungwa nokuda mamwemazwi vateereri.

Nokuti mienzaniso zvidzidzo nezvezvandaifanira spam muna Twitter, Clark et al. (2016) uye Chu et al. (2012) . Pakupedzisira, Subrahmanian et al. (2016) inorondedzera migumisiro DARPA Twitter Bot Challenge.

nzwisisa

Ohm (2015) akaongororazve tsvakurudzo kwamboita pfungwa inonyatsoziva mashoko onyengetera multi-chinhu bvunzo. The ina proposes zvinhu ndiwo: ndingangokuvara chakaipa; ndingangokuvara chakaipa; pamberi zvakavanzika ukama; uye kana ngozi vanoratidza majoritarian zvaimunetsa.

  • Tichiverenga zvinhu (Section 2.4.1)

Farber kuti ongororo nemuchovha muNew York kwaibva kudzidza ndamboudzwa Camerer et al. (1997) kuti vakashandisa nhatu dzakasiyana kurerukirwa mimwe bepa rwendo machira-paper siyana dzinoshandiswa vatyairi kunyora rwendo kutanga nguva, yokuguma, uye yokufamba. Izvi kudzidza vamba vakawana kuti vatyairi airatidzika kuva chipfuro earners: vaishanda zvishoma mazuva apo mubayiro avo vaiva yepamusoro.

Kossinets and Watts (2009) akanga rinotarisa kwakatangira homophily munzanga network. Ona Wimmer and Lewis (2010) kuti imwe nzira kune imwe dambudziko iro rinoshandisa mashoko kubva Facebook.

In unozotevera basa, Mambo uye vashandi Uyezve Akaongorora paIndaneti kusenza muna China (King, Pan, and Roberts 2014; King, Pan, and Roberts 2016) . Nokuti kufanana nzira kuyera paIndaneti kusenza muna China, ona Bamman, O'Connor, and Smith (2012) . Kuti uwane pamusoro uwandu nzira yakafanana munhu akashandiswa King, Pan, and Roberts (2013) kufungidzira naA mamiriyoni 11 mbiru, ona Hopkins and King (2010) . Kuti uwane pamusoro aitarisira kudzidza, ona James et al. (2013) (shoma hwokugadzira) uye Hastie, Tibshirani, and Friedman (2009) (more rwokugadzira).

  • Kufungidzira (Section 2.4.2)

Kufungidzira chakakura chikamu maindasitiri Data sayenzi (Mayer-Schönberger and Cukier 2013; Provost and Fawcett 2013) . Imwe mhando kufungidzira kuti anowanzonzi zvinoitwa pasocial vatsvakurudzi vari wevanotungamirira kufungidzira Somuenzaniso Raftery et al. (2012) .

Google flu Trends haasi chirongwa chokutanga kushandisa kutsvaka Data kuti nowcast yokuSpain hwakapararira. Kutaura zvazviri, vanotsvakurudza muUnited States (Polgreen et al. 2008; Ginsberg et al. 2009) uye Sweden (Hulth, Rydevik, and Linde 2009) vakaona kuti mamwe mashoko kutsvaka (semuenzaniso, "besa") vakafanotaura National pachena utano ongororo date risati akasunungurwa. Naizvozvo vazhinji, dzimwe zvirongwa vazhinji vakaedza kushandisa digitaalinen zvisaririra Data kuti chirwere zvokuongorora vasabatwa, ona Althouse et al. (2015) kuti kudzokorora.

Kuwedzera pakushandisa digitaalinen zvisaririra Data kufanotaura utano zvaitika, ndipo wakandibatsirawo hombe hwebasa achishandisa Twitter Data kufanotaura sarudzo zvaitika; nokuti nhoroondo ona Gayo-Avello (2011) , Gayo-Avello (2013) , Jungherr (2015) (Ch. 7), uye Huberty (2015) .

Kushandisa kutsvaka Data kuti vachifanotaura yokuSpain Huwandu uye kushandisa Twitter Data kufanotaura sarudzo dziri vaviri mienzaniso kushandisa digitaalinen zvisaririra vamwe mutsa kufanotaura nezvechiitiko munyika dzimwe mhando. Ikoko wakakura nhamba yezvidzidzo kuti ichi zvayo chivako. Table 2.5 rinosanganisira mimwe mienzaniso mishomanana.

Table 2.5: Kuzara mazita zvidzidzo vanoshandisa vamwe digitaalinen zvisaririra chokufungidzira vamwe chiitiko.
Digital zvisaririra gumisiro Citation
Twitter Bhokisi hofisi kwemari pamusoro mafirimu muna US Asur and Huberman (2010)
Search matanda Sales pamusoro mafirimu, nziyo, mabhuku, nemitambo yemavhidhiyo muna US Goel et al. (2010)
Twitter Dow Jones Industrial Avhareji (US pamusika wemari) Bollen, Mao, and Zeng (2011)
  • Approximating kuedza (Section 2.4.3)

Magazini PS Political Science aiva Hurukuro yokusiyirana guru data, causal inference, uye yakarongwa dzidziso, uye Clark and Golder (2015) muchidimbu mumwe mupiro. Magazini Proceedings National Academy of Sciences of United States of America aiva Hurukuro yokusiyirana causal inference uye mashoko makuru, uye Shiffrin (2016) muchidimbu mumwe mupiro.

Tichitevera dzinongoitika kuedza, Dunning (2012) chinotipa mukana wakanakisisa bhuku pakureba kurapwa. Kuti uwane pamusoro kushandisa Vietnam gwaro rotari sechinhu chokurapisa kuedza, ona Berinsky and Chatfield (2015) . Nokuti nzira muchina pakudzidza kuti vanoedza kuchinja vanoona dzinongoitika kuedza mukati guru mashoko manyuko, ona Jensen et al. (2008) uye Sharma, Hofman, and Watts (2015) .

Tichitevera enzanirana, ive netariro review, ona Stuart (2010) , uye kuti tariro ongororo ona Sekhon (2009) . Kuti uwane pamusoro chienderane sechinhu mhando kuchekerera, ona Ho et al. (2007) . Nokuti mabhuku anopa kwazvo kurapa pamusoro enzanirana, ona Rosenbaum (2002) , Rosenbaum (2009) , Morgan and Winship (2014) , uye Imbens and Rubin (2015) .