Loading...
Image Image Image Image

Funsi (Funksion) Supervise Learning

Supervised Learning iha funksion principal nebee útil iha vida real:

1.Predisaun (Prediction) Previsa dadus foun baseia ba dadus historiku nebee iha label. Exemplu: Prediz prezu uma ka valor akasaun stoku.

2.Klasifikasaun (Classification) Mapeia dadus foun ba klase espesífiku. Exemplu: Klasifika email spam ka la spam.

3.Estimasaun / Previsaun Futuru (Estimation / Forecasting) Previsa valor ba tempu foun baseia ba dadus historiku. Exemplu: Previsaun vendas mensal ka demanda merkadu.

4.Suporta ba Desizaun (Decision Support) Ajuda toma desizaun baseia ba dadus. Exemplu: Sistema medis ajuda doktor husi decisao doensa pasiente.

kona-ba Supervise Learning

🔍 Oinsá mak Supervised Learning Servisu?.

1.Rekolle dadus ne’ebé iha label Ezemplu: Nota estudante (input) no status pasa/ka la pasa (output)

2.Hanorin model Makina aprende padraun husi dadus

3.Teste model Haree model nee loos to’o hira

4.Halo prediksaun Model pronto atu uza ba dadus foun

INFORMASAUN
KAKUTAK INTELEGENCYA

Robot

Arteficial Intelegencya

Tipo Supervise Learning

Regression (Regresaun)

1.Objektivu / Funsi: Prediz numeru kontínuu husi dadus input.

2.Exemplu Output: Prezu uma, salaríu, temperatura, stock produtus.

3.Algoritmu Komum: Linear Regression, Polynomial Regression, Decision Tree Regression.

4.Exemplu Kazu: Prediz prezu karu baseia ba tinan, marka, no kilometrajen.

Classification (Klasifikasaun)

1.Objektivu / Funsi: Prediz kategori/klase husi dadus input.

2.Exemplu Output: Spam / La Spam, Doensa A / B, Lulus / La Lulus

3.Algoritmu Komum: Logistic Regression, Decision Tree, Random Forest, SVM.

4.Exemplu Kazu: Prediz se email mak spam ka lae.

Image
Iha perguntas ruma? Kontak: +670 74641795
Machine Learning

Machine Learning (ML) mak bagian husi Artificial Intelligence (AI) nebee buat komputador aprende husi dadus.

Signifika: Komputador bele husi esperiensia nebee iha (data) apende regras no pola tanba prediksaun ka klasifika dadus foun. Iha ML(Machine Learning), komputador aprende sem tempu kodigu eksplisit.

Image

Benefísiu Aprendizajen Supervised Learning

1.Habilita kolesaun no prosesamentu dadus hosi dadus istóriku ka esperiénsia anteriór sira.

2.Ajuda rezolve problema oioin iha mundu reál, hanesan deteksaun saude.

3.Ajuda klasifika dadus formasaun nian ho loloos no espesífiku liu.

Explore More
Image

Objectivu Supervised Learning

Objectivu aprende Supervised Learning mak hodi aprende husi dadus ne’ebé iha label hodi prediksaun valor numeriku (regression) ka klase (classification), komprende forma ka problema iha dadus, no suporta toma desizaun ne’ebé informadu iha dadus foun..

Explore More
Image

Ezemplu Supervised Learning

Ezemplu supervised learning mak hanesan sistema detekta spam iha email, predisaun prezu uma, rekoñese imajen, diagnostiku doensa, no predisaun nota estudante, ne’ebé hotu aprende husi dadus ne’ebé iha label hodi prediksaun ka klasifika dadus foun..

Explore More
Sientista

Saida Mak Sira Koalia??

Iha siemtista barak mak koalia ona kona ba supervised learning no saida deit mak iha supervised learning inklui nia servisu no lalaok sistema

Sientista

Matenek Nain Neebe Mak Hamosu

_________________________________________________________________________

Frank Rosenblatt

(1957): Psikolojista iha Cornell Aeronautical Laboratory ne’ebé desenvolve Perceptron. Nee mak primeira “neuro-computer” ne’ebé sukses husi deskobre imajen.

Nia uza metoda supervised klasik: fó dadus iha label, no bainhira fali predisaun la diak, pesu internu-nia sei ajusta hodi aumenta akurásia.

Bernard Widrow

(1960)Pesquisador husi Stanford University ne’ebé kreia ADALINE (Adaptive Linear Neuron)

modelo supervised learning foin ona ne’ebé uza ba servisu real hanesan halo reduza ekos iha linha telefone.

Ted Hoff

(1960)Pesquisador husi Stanford University ne’ebé kreia ADALINE (Adaptive Linear Neuron)

modelo supervised learning foin ona ne’ebé uza ba servisu real hanesan halo reduza ekos iha linha telefone.