?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=IMPLEMENTASI+PERAMALAN+SUHU+MINIMUM+MINGGUAN+MENGGUNAKAN+METODE+HYBRID+SEASONAL+AUTOREGRESSIVE+INTEGRATED+MOVING+AVERAGE+(SARIMA)-LONG+SHORT+TERM+MEMORY+(LSTM)++++++&rft.creator=ANINDYA+DAFA+%2C+SALSABILA&rft.subject=510+Matematika&rft.description=Indonesia++is++a++country++with++a++tropical++climate++that++has++two++seasons++in++one++year%2C+which+can+influence+the+weather+changes+in+a+region+over+time%2C+including+extreme+weather++changes.++Forecasting++is++a++statistical++method++that++can++be++used++to++predict+future++weather++changes.++The++SARIMA++method++is++a++forecasting++method++that++has+the++advantage++of++capturing++seasonal++patterns.++However%2C++SARIMA++is++difficult++to+use++for++processing++nonlinear++components.++LSTM++is++a++part++of++Deep++Learning++that+can+process+large+amounts+of+data+and+has+three+main+layers+capable+of+processing+fluctuating++nonlinear++components%2C++namely++the++input++layer%2C++hidden++layer%2C++and+output++layer.++Therefore%2C++the++hybrid++SARIMA-LSTM++method++is++expected++to+efficiently+process+data+with+both+linear+and+nonlinear+patterns%2C+including+seasonal+patterns.+The+hybrid+model+used+in+the+research+performs+well+in+the+data+splitting+scheme++of++80%25++for++training++and++20%25++for++testing%2C++with++an++MSE++value++of++0.1174%2C+RMSE+value+of+0.3426%2C+and+MAPE+value+of+0.0104%25.%0D%0AKeywords%3A+Hybrid+SARIMA-LSTM%2C+SARIMA%2C+LSTM%2C+Forecasting%2C+Deep+Learning%2C+Neural+Network%2C+Accuracy.%0D%0A+Indonesia++merupakan++negara++yang++mempunyai++iklim++tropis++yang++memiliki++2+musim++lam++1++tahun++yang++dapat++mempengaruhi++perubahan++cuaca++suatu++wilayah+setiap+waktunya%2C+seperti+perubahan+cuaca+ekstrim.++Peramalan+merupakan+metode+statistika++yang++dapat++digunakan++untuk++memprediksi++perubahan++cuaca++yang++akan+terjadi++di++masa++depan++++Metode++SARIMA++merupakan++metode++yang++dapat+digunakan++dalam++peramalan++dan++mempunyai++kelebihan++dalam++menangkap++pola+musiman.++++Akan++tetapi%2C++SARIMA++sulit++digunakan++untuk++memproses++komponen+nonlinear.+++LSTM+merupakan+bagian+dari++Deep+Learning++yang+dapat+memproses+data++berjumlah++besar++dan++mempunyai++3++layer++utama++yang++mampu++memproses+komponen++nonlinear++data++yang++berfluktuasi%2C++yaitu++input++layer%2C++hidden++layer++dan+output+layer.++Maka+dari+itu%2C+metode+hybrid+SARIMA-LSTM+diharapkan+mampu+untuk++memproses++data++dengan++pola++linear++dan++nonlinear++serta++seasonal++secara+optimal.++++Model++hybrid++pada++penelitian++yang++dilakukan++memiliki++performa++yang+baik++pada++skema++splitting++data++80%25++training++dan++20%25++testing++dengan++nilai++MSE+sebesar+0.1174%2C+RMSE+sebesar+0.3426+dan+MAPE+sebesar+0%2C0104+%25.+++++++Kata++Kunci%3A++Hybrid++SARIMA-LSTM%2C++SARIMA%2C++LSTM%2C++Peramalan%2C++Deep+Learning%2C+Neural+Network%2C+Akurasi.++&rft.publisher=FAKULTAS+MATEMATIKA+DAN+ILMU+PENGETAHUAN+ALAM+&rft.date=2023-07-14&rft.type=Skripsi&rft.type=NonPeerReviewed&rft.format=text&rft.identifier=http%3A%2F%2Fdigilib.unila.ac.id%2F74554%2F1%2FABSTRAK.pdf&rft.format=text&rft.identifier=http%3A%2F%2Fdigilib.unila.ac.id%2F74554%2F2%2FSKRIPSI%2520FULL.pdf&rft.format=text&rft.identifier=http%3A%2F%2Fdigilib.unila.ac.id%2F74554%2F3%2FSKRIPSI%2520TANPA%2520BAB%2520PEMBAHASAN.pdf&rft.identifier=++ANINDYA+DAFA+%2C+SALSABILA++(2023)+IMPLEMENTASI+PERAMALAN+SUHU+MINIMUM+MINGGUAN+MENGGUNAKAN+METODE+HYBRID+SEASONAL+AUTOREGRESSIVE+INTEGRATED+MOVING+AVERAGE+(SARIMA)-LONG+SHORT+TERM+MEMORY+(LSTM).++FAKULTAS+MATEMATIKA+DAN+ILMU+PENGETAHUAN+ALAM+%2C+UNIVERSITAS+LAMPUNG.+++++&rft.relation=http%3A%2F%2Fdigilib.unila.ac.id%2F74554%2F