2.3.7 Drifting

Tsarin yawan yawan jama'a, amfani drift, da tsarin drift tsarin yana da wuya a yi amfani da manyan bayanan bayanan don yin nazarin al'amura na dogon lokaci.

Ɗaya daga cikin manyan abubuwan da aka samo asali daga manyan mahimman bayanai shi ne sun tattara bayanai a tsawon lokaci. Masana kimiyyar zamantakewa suna kiran irin wannan bayanai na tsawon lokaci . Kuma, a bayyane, bayanan sirri suna da matukar muhimmanci ga nazarin canji. Don dogara da sauya canji, duk da haka, tsarin da kanta ya zama dole ya zama barga. A cikin maganar masanin ilimin zamantakewa Otis Dudley Duncan, "idan kuna son gyara canji, kada ku canza ma'auni" (Fischer 2011) .

Abin takaici, yawancin tsarin bayanai-musamman tsarin kasuwanci-suna canzawa duk lokacin, wani tsari da zan kira drift . Musamman ma, waɗannan tsarin sun canza cikin hanyoyi uku: yawan tsararren yanki (canji a wanda ke amfani da su), gyaran hali (canji a yadda mutane suke amfani da su), da kuma tsarin tsarin (canza a cikin tsarin kanta). Hanyoyi guda uku na drift na nufin cewa kowane tsari a cikin babban mahimman bayanai zai iya haifar da wani canji mai mahimmanci a duniya, ko kuma zai iya haifar da wani nau'i na sauƙi.

Maganin farko na drift-population drift-yana haifar da canje-canje a wanda ke amfani da tsarin, kuma waɗannan canje-canje na iya faruwa a cikin gajeren lokaci da kuma tsawon lokaci. Alal misali, a lokacin zaben shugaban kasa na Amurka na shekara ta 2012, yawancin tweets game da siyasa da mata suka rubuta daga yau zuwa rana (Diaz et al. 2016) . Sabili da haka, abin da zai iya zama kamar canji a yanayi na Twitter-aya zai iya kasancewa canji ne kawai wanda ke magana a kowane lokaci. Bugu da ƙari, ga waɗannan haɓakaccen gajeren lokaci, akwai wasu lokuttan da suka dace da wasu kungiyoyin jama'a na yin amfani da su da barin Twitter.

Bugu da ƙari, canje-canje a cikin wanda ke amfani da tsarin, akwai kuma canje-canje a yadda aka yi amfani da tsarin, wanda na kira tarbiyyar hali. Alal misali, a lokacin zanga-zangar Gezi a shekarar 2013, masu zanga-zanga sun sake yin amfani da hashtags yayin da zanga-zangar ta samo asali. Ga yadda Zeynep Tufekci (2014) bayyana fasalin hali, wadda ta iya gane saboda tana kallon halin Twitter da mutum:

"Abin da ya faru shi ne cewa da zarar zanga-zangar ya zama mahimmanci, mutane masu yawa ... sun dakatar da yin amfani da hashtags sai dai su jawo hankali ga wani sabon abu ... Yayin da zanga-zangar suka ci gaba, har ma da tsanani, hashtags ya mutu. Tambayoyi sun bayyana dalilai biyu na wannan. Na farko, da zarar kowa ya san wannan batu, hashtag ya kasance mai ban mamaki da kuma raguwa a kan hanyar Twitter. Na biyu, ana ganin abubuwan da ake amfani da shi kawai don amfani da hankali ga wani abu, ba don magana game da shi ba. "

Saboda haka, masu bincike da suka nazarin boren da nazarin tweets da zanga-zanga da alaka Hashtags za a yi karkatacciyar ji abin da aka faruwa saboda wannan halin yin gantali. Alal misali, sun yi imani da cewa tattaunawa da zanga-zangar rage tun kafin shi a zahiri rage.

Na uku nau'i na drift shi ne tsarin tafiyar dashi. A wannan yanayin, mutane ba su canza ko halin su canza, amma tsarin kanta yana canza. Alal misali, a tsawon lokaci Facebook ya ƙãra iyaka akan tsawon ɗaukakawar halin. Saboda haka, duk wani nazarin binciken da aka yi na tsawon lokaci zai kasance mai sauƙi ga abubuwa masu yawa da suka haifar da wannan canji. Drift tsarin yana da alaka da matsala da ake kira algorithmic confounding, wanda zan rufe a sashi na 2.3.8.

Don ƙare, yawancin manyan kafofin bayanan bayanai suna drifting saboda canje-canje ga wanda yake amfani da su, yadda ake amfani da su, da kuma yadda tsarin ke aiki. Wadannan sauye-sauye wasu lokuta mahimman tambayoyin bincike ne, amma waɗannan canje-canje sunyi tasiri ga iyawar manyan bayanan bayanan don yin la'akari da canje-canjen lokaci mai tsawo.