Teburin Abubuwan Ciki[Boye][Nuna]
- 1. Titanic
- 2. Rarraba furannin Irish
- 3. Hasashen Farashin Gidan Gidan Boston
- 4. Gwajin ingancin ruwan inabi
- 5. Hasashen Kasuwar Hannu
- 6. Shawarar Fim
- 7. Load Cancantar Hasashen
- 8. Analysis na Sentiment ta amfani da bayanan Twitter
- 9. Hasashen tallace-tallace na gaba
- 10. Gano Labaran Karya
- 11. Hasashen Siyan Kuɗi
- 12. Abokin ciniki Churn Hasashen
- 13. Hasashen tallace-tallace na Wallmart
- 14. Uber Data Analysis
- 15. Binciken Covid-19
- Kammalawa
Koyon inji wani bincike ne mai sauƙi na yadda ake ilmantar da shirin kwamfuta ko algorithm don inganta hankali a kan takamaiman aikin da aka gabatar a babban matakin. Gano hoto, gano zamba, tsarin shawarwari, da sauran aikace-aikacen koyon injin sun riga sun tabbatar sun shahara.
Ayyukan ML suna sa aikin ɗan adam ya zama mai sauƙi da inganci, adana lokaci da tabbatar da sakamako mai inganci. Hatta Google, mashahurin ingin bincike a duniya, yana amfani da shi injin inji.
Daga nazarin tambayar mai amfani da canza sakamakon bisa ga sakamakon zuwa nuna batutuwa da tallace-tallace masu tasowa dangane da tambayar, akwai zaɓuɓɓuka iri-iri da ake da su.
Fasahar da ke da hankali da kuma gyara kanta ba ta da nisa a nan gaba.
Ɗaya daga cikin manyan hanyoyin da za a fara shine don samun hannu da tsara aikin. Don haka, mun tattara jerin manyan ayyukan koyan inji guda 15 don masu farawa don fara ku.
1. Titanic
Ana ɗaukar wannan a matsayin ɗayan ayyuka mafi girma kuma mafi daɗi ga duk mai sha'awar ƙarin koyo game da koyon injin. Kalubalen Titanic sanannen aikin koyon inji ne wanda kuma ke aiki a matsayin hanya mai kyau don sanin dandalin kimiyyar bayanai na Kaggle. Saitin bayanan Titanic yana kunshe ne da bayanai na gaske daga nutsewar jirgin da ba shi da lafiya.
Ya haɗa da cikakkun bayanai kamar shekarun mutum, matsayin zamantakewa, yanayin zamantakewa, jinsi, lambar gida, tashar tashi, kuma, mafi mahimmanci, ko sun tsira!
Dabarar maƙwabci na K-Neaest da yanke shawara mai rarraba bishiyar an ƙaddara don samar da kyakkyawan sakamako na wannan aikin. Idan kuna neman ƙalubalen karshen mako don inganta naku Iyawar Koyon Inji, wannan akan Kaggle na ku ne.
2. Rarraba furannin Irish
Masu farawa suna son aikin rarraba furen iris, kuma wuri ne mai kyau don farawa idan kun kasance sababbi ga koyon injin. Tsawon sepals da petals suna bambanta iris blooms daga sauran nau'in. Manufar wannan aikin ita ce raba furanni zuwa nau'ikan uku: Virginia, setosa, da Versicolor.
Don darussan rarrabuwa, aikin yana amfani da bayanan furen Iris, wanda ke taimaka wa xalibai don koyan tushen ma'amala da ƙima da bayanai. Saitin bayanan furen iris ƙarami ne wanda za'a iya adana shi cikin ƙwaƙwalwar ajiya ba tare da buƙatar sikeli ba.
3. Hasashen Farashin Gidan Gidan Boston
Wani sananne dataset for novices a cikin koyon inji shine bayanan Gidajen Boston. Manufarta ita ce yin hasashen ƙimar gida a yankuna daban-daban na Boston. Ya haɗa da ƙididdiga masu mahimmanci kamar shekaru, ƙimar harajin kadara, ƙimar laifi, har ma da kusanci ga wuraren aiki, waɗanda duk zasu iya shafar farashin gidaje.
Saitin bayanan yana da sauƙi kuma ƙanƙanta, yana sauƙaƙa don gwaji tare da novice. Don gano abubuwan da ke tasiri farashin kadarorin a Boston, ana amfani da dabarun sake jujjuyawa akan sigogi daban-daban. Yana da kyakkyawan wuri don aiwatar da dabarun koma baya da tantance yadda suke aiki sosai.
4. Gwajin ingancin ruwan inabi
Wine wani abin sha ne wanda ba a saba gani ba wanda ke buƙatar shekaru na haifuwa. A sakamakon haka, tsohuwar kwalban giya yana da tsada da inganci. Zaɓin kwalaben ruwan inabi mai kyau yana buƙatar shekaru na ilimin ɗanɗano ruwan inabi, kuma yana iya zama tsari mai lalacewa.
Aikin gwajin ingancin ruwan inabi yana kimanta giya ta amfani da gwaje-gwajen kimiyyar sinadarai kamar matakin barasa, tsayayyen acidity, yawa, pH, da sauran dalilai. Aikin kuma yana ƙayyade ma'auni na ingancin ruwan inabi da kuma adadin. A sakamakon haka, siyan giya ya zama iska.
5. Hasashen Kasuwar Hannu
Wannan yunƙurin yana da ban sha'awa ko kuna aiki a fannin kuɗi ko a'a. Masana ilimi, kasuwanci, har ma a matsayin tushen samun kudin shiga na sakandare ana yin nazarin bayanan kasuwar hannun jari sosai. Ƙarfin masanin kimiyyar bayanai don yin nazari da gano bayanan jerin lokaci shima yana da mahimmanci. Bayanai daga kasuwar hannun jari wuri ne mai kyau don farawa.
Ma'anar wannan yunƙurin shine yin hasashen ƙimar haja ta gaba. Wannan ya dogara ne akan aikin kasuwa na yanzu da kuma ƙididdiga daga shekarun baya. Kaggle yana tattara bayanai akan ma'aunin NIFTY-50 tun daga 2000, kuma a halin yanzu ana sabunta shi kowane mako. Tun daga 1 ga Janairu, 2000, ya ƙunshi farashin hannun jari na ƙungiyoyi sama da 50.
6. Shawarar Fim
Na tabbata kun ji wannan jin bayan ganin fim mai kyau. Shin kun taɓa jin sha'awar karkatar da hankalin ku ta hanyar kallon fina-finai irin wannan?
Mun san cewa sabis na OTT kamar Netflix sun inganta tsarin shawarwarin su sosai. A matsayinka na ɗalibin koyan na'ura, za ka buƙaci fahimtar yadda irin waɗannan algorithms ke kaiwa abokan ciniki hari bisa abubuwan da suke so da bita.
Bayanan IMDB da aka saita akan Kaggle yana iya zama ɗaya daga cikin mafi cikakke, yana ba da damar samfuran shawarwarin da za a iya tantance su dangane da taken fim ɗin, ƙimar abokin ciniki, nau'in, da sauran dalilai. Hakanan kyakkyawar hanya ce don koyo game da Tacewar Abubuwan Abubuwan ciki da Injiniyan Fasaloli.
7. Load Hasashen Cancantar
Duniya tana tafe da lamuni. Babban tushen riba na bankuna yana fitowa ne daga ribar lamuni. Don haka su ne ainihin kasuwancin su.
Daidaikun mutane ko ƙungiyoyin daidaikun mutane za su iya faɗaɗa tattalin arziƙi ne kawai ta hanyar saka kuɗi a cikin kamfani cikin fatan ganin ya tashi da ƙima a nan gaba. Wani lokaci yana da mahimmanci a nemi lamuni don samun damar yin kasada da wannan yanayin har ma da shiga cikin wasu abubuwan jin daɗi na duniya.
Kafin a karɓi lamuni, bankuna yawanci suna da tsayayyen tsari don bi. Kamar yadda lamuni ke da mahimmanci ga rayuwar mutane da yawa, hasashen cancantar lamuni da wani ya nema zai yi matukar fa'ida, yana ba da damar ingantaccen tsari fiye da rancen da ake karɓa ko ƙi.
8. Analysis na Sentiment ta amfani da bayanan Twitter
Godiya ga hanyoyin sadarwar kafofin watsa labarun kamar Twitter, Facebook, da Reddit, ra'ayoyi masu ban sha'awa da abubuwan da suka faru sun sami sauƙi sosai. Ana amfani da wannan bayanin don kawar da ra'ayi akan abubuwan da suka faru, mutane, wasanni, da sauran batutuwa. Ana amfani da dabarun ilmantarwa na inji mai alaƙa da ra'ayi a cikin saitunan daban-daban, gami da kamfen na siyasa da kimanta samfuran Amazon.
Wannan aikin zai yi kyau a cikin fayil ɗin ku! Don gano motsin rai da bincike na tushen al'amari, ana iya amfani da dabaru kamar na'urori masu goyan baya, koma baya, da algorithms rarrabuwa (neman gaskiya da ra'ayi).
9. Hasashen tallace-tallace na gaba
Babban kasuwancin B2C da 'yan kasuwa suna son sanin nawa kowane samfur a cikin kayan aikin su zai sayar. Hasashen tallace-tallace yana taimaka wa masu kasuwanci wajen tantance abubuwan da ake buƙata sosai. Madaidaicin hasashen tallace-tallace zai rage yawan almubazzaranci yayin da kuma ke tantance tasirin ƙara kan kasafin kuɗi na gaba.
Dillalai irin su Walmart, IKEA, Babban Kwando, da Babban Bazaar suna amfani da hasashen tallace-tallace don kimanta buƙatar samfur. Dole ne ku saba da dabaru daban-daban na tsabtace danyen bayanai don gina irin waɗannan ayyukan ML. Har ila yau, ana buƙatar kyakkyawar fahimtar nazarin koma baya, musamman sauƙi mai sauƙi na layi, ana buƙatar.
Don irin waɗannan ayyuka, kuna buƙatar ɗaukar dakunan karatu kamar Dora, Scrubadub, Pandas, NumPy, da sauransu.
10. Gano Labaran Karya
Wani yunƙurin koyon injinan yankan-baki ne wanda ke nufin yaran makaranta. Labaran karya suna yaduwa kamar wutar daji, kamar yadda muka sani. Ana samun komai akan kafofin watsa labarun, daga haɗa mutane zuwa karanta labaran yau da kullun.
Sakamakon haka, gano labaran karya ya daɗa wahala a kwanakin nan. Yawancin manyan hanyoyin sadarwar zamantakewa, irin su Facebook da Twitter, sun riga sun sami algorithms a wurin don gano labaran bogi a cikin aikawa da ciyarwa.
Don gano labaran karya, wannan nau'in aikin ML yana buƙatar cikakken fahimtar hanyoyin NLP da yawa da algorithms rarrabuwa (PassiveAggressiveClassifier ko Naive Bayes classifier).
11. Hasashen Siyan Kuɗi
Abokan ciniki suna ƙara yin tunanin siye ta kan layi lokacin da coronavirus ya kai hari a duniya a cikin 2020. Sakamakon haka, an tilasta wa cibiyoyin siyayya su canza kasuwancin su akan layi.
Abokan ciniki, a gefe guda, har yanzu suna neman manyan tayi, kamar yadda suke a cikin shaguna, kuma suna ƙara farautar kuɗaɗen ceto. Akwai ma gidajen yanar gizon da aka sadaukar don ƙirƙirar takardun shaida ga irin waɗannan abokan ciniki. Kuna iya koyo game da haƙar ma'adinan bayanai a cikin koyan na'ura, samar da ginshiƙan mashaya, ginshiƙan kek, da lissafin lissafi don hango bayanai, da fasalin aikin injiniya tare da wannan aikin.
Don samar da tsinkaya, zaku iya kuma duba hanyoyin ƙirƙira bayanai don sarrafa ƙimar NA da kamanceceniyar masu canji.
12. Hasashen Abokin Ciniki
Abokan ciniki sune mafi mahimmancin kadari na kamfani, kuma kiyaye su yana da mahimmanci ga kowane kasuwancin da ke da niyyar haɓaka kudaden shiga da gina alaƙa mai ma'ana ta dogon lokaci tare da su.
Bugu da ƙari, farashin samun sabon abokin ciniki ya ninka sau biyar fiye da farashin ci gaba da kasancewa. Abokin ciniki Churn/Attrition sanannen matsalar kasuwanci ce wacce abokan ciniki ko masu biyan kuɗi suka daina kasuwanci tare da sabis ko kamfani.
Ba za su ƙara zama abokin ciniki mai biyan kuɗi ba. Ana tsammanin abokin ciniki ya mutu idan ya kasance wani adadin lokaci tun lokacin da abokin ciniki ya yi hulɗa da kamfanin. Gano ko abokin ciniki zai ɓata, da kuma ba da hanzarin ba da bayanan da suka dace da nufin riƙe abokin ciniki, suna da mahimmanci don rage jinkirin.
Ƙwaƙwalwarmu ba ta da ikon hango canjin abokin ciniki ga miliyoyin abokan ciniki; anan ne inda koyon inji zai iya taimakawa.
13. Hasashen tallace-tallace na Wallmart
Ɗaya daga cikin fitattun aikace-aikacen koyon injin shine hasashen tallace-tallace, wanda ya haɗa da gano halayen da ke tasiri tallace-tallacen samfur da kuma tsammanin girman tallace-tallace na gaba.
The Walmart dataset, wanda ya ƙunshi bayanan tallace-tallace daga wurare 45, ana amfani da shi a cikin wannan binciken koyon injin. Ana haɗa tallace-tallace a kowane shago, ta nau'in, kowane mako a cikin tsarin bayanai. Manufar wannan aikin koyo na inji shine hasashen tallace-tallace ga kowane sashe a cikin kowane shago ta yadda za su iya yin ingantacciyar hanyar inganta tashoshi bayanai da yanke shawarar tsara ƙira.
Yin aiki tare da bayanan Walmart yana da wahala tunda ya ƙunshi abubuwan da aka zaɓa waɗanda ke da tasiri akan tallace-tallace kuma yakamata a yi la'akari da su.
14. Uber Data Analysis
Idan ya zo ga aiwatarwa da haɗa koyan inji da zurfafa ilmantarwa a cikin aikace-aikacen su, mashahurin sabis ɗin raba keke bai yi nisa ba. Kowace shekara, tana aiwatar da biliyoyin tafiye-tafiye, yana ba masu ababen hawa damar yin balaguro a kowane lokaci dare ko rana.
Saboda yana da babban tushe na abokin ciniki, yana buƙatar sabis na abokin ciniki na musamman don magance korafe-korafen mabukaci da sauri.
Uber yana da tarin tarin miliyoyin abubuwan da za su iya amfani da su don tantancewa da nuna tafiye-tafiyen abokin ciniki don buɗe haske da haɓaka ƙwarewar abokin ciniki.
15. Binciken Covid-19
COVID-19 ya mamaye duniya a yau, kuma ba kawai a cikin ma'anar annoba ba. Yayin da kwararrun likitocin ke mai da hankali kan samar da ingantattun alluran rigakafi da rigakafi a duniya, masana kimiyyar bayanai ba su yi nisa a baya ba.
Sabbin kararraki, kididdigar ayyukan yau da kullun, mace-mace, da kididdigar gwaji duk ana bayyana jama'a. Ana yin hasashen kowace rana dangane da barkewar cutar SARS a karnin da ya gabata. Don wannan, zaku iya amfani da bincike na koma baya da goyan bayan ƙirar tsinkayar tushen injin vector.
Kammalawa
Don taƙaitawa, mun tattauna wasu manyan ayyukan ML waɗanda za su taimaka muku wajen gwada shirye-shiryen Koyon Inji tare da fahimtar ra'ayoyinsa da aiwatarwa. Sanin yadda ake haɗa Ilimin Injin na iya taimaka muku ci gaba a cikin sana'ar ku yayin da fasahar ke mamaye kowane masana'antu.
Yayin koyan Injin Learning, muna ba da shawarar ku aiwatar da ra'ayoyin ku kuma ku rubuta duk algorithms ɗin ku. Rubuta algorithms yayin koyo yana da mahimmanci fiye da aiwatar da aiki, kuma yana ba ku fa'ida wajen fahimtar batutuwan yadda ya kamata.
Leave a Reply