?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=IMPLEMENTASI+MODEL+HYBRID+CONVOLUTIONAL+NEURAL+NETWORK%0D%0ADAN+LONG+SHORT-TERM+MEMORY+UNTUK+IDENTIFIKASI%0D%0AKERUSAKAN+DAUN+KEDELAI+AKIBAT+SERANGAN%0D%0ALARVA+LEPIDOPTERA+DAN+DIABROTICA%0D%0A&rft.creator=NANDA+%2C+EVITARINA&rft.subject=500+ilmu+pengetahuan+alam+dan+matematika&rft.subject=510+Matematika&rft.description=The+identification+of+soybean+leaf+damage+is+an+interesting+case+to+be+studied+in%0D%0Aconnection+with+the+increase+in+the+demand+for+soybean+commodities+in+Indonesia.%0D%0ASoybean+production+and+yields+are+affected+by+pest+attacks+that+cause+loss+of%0D%0Aagricultural+yields+each+year.+One+of+these+pests+is+insects.+Caterpillar+(insect+larvae+of%0D%0Athe+order+Lepidoptera)+and+Diabrotica+speciosa+insects+are+types+of+pests+that+often%0D%0Aattack+the+soybean+leaves%2C+damaging+the+quality+and+quantity+of+soybean+production.%0D%0ATherefore%2C+this+study+aims+to+assist+farmers+in+an+effort+to+provide+the+right+treatment%0D%0Afor+soybean+leaf+damage.+This+identification+requires+good+machine+learning%0D%0Aalgorithms+such+as+deep+learning.+Deep+learning+is+recommended+for+identification+or%0D%0Aclassification+because+it+can+learn+features+from+images+of+soybean+leaves+damage%0D%0Aautomatically.+In+this+research%2C+the+deep+learning+model+used+to+identify+soybean%0D%0Aleaves+damage+is+a+combination+of+Convolutional+Neural+Network+(CNN)+and+Long%0D%0AShort-Term+Memory+(LSTM)+where+CNN+can+solve+the+problem+of+spatial%0D%0Acharacteristics+and+LSTM+can+understand+the+temporal+context+of+image+data.+The%0D%0Amodel+is+developed+through+image+processing+stages+such+as+data+augmentation+using%0D%0Athe+roboflow+platform.+The+dataset+used+comes+from+the+Mendeley+Data+platform+with%0D%0Aa+total+dataset+of+6410.+The+combined+CNN-LSTM+hybrid+model+performed+well+in%0D%0Aidentifying+soybean+leaves+damage+with+an+accuracy+of+93%25%2C+precision+of+93%25%2C+recall%0D%0Aof+93%25%2C+and+f1-score+of+93%25.%0D%0A%0D%0AKeywords+%3A+Soybean%2C+Caterpillar%2C+Diabrotica+Speciosa%2C+Deep+Learning%2C+CNN%2C+LSTM%2C%0D%0ACNN-LSTM%0D%0A%0D%0AIdentifikasi+kerusakan+daun+kedelai+menjadi+kasus+yang+menarik+untuk+diteliti%0D%0Asehubungan+dengan+terjadinya+peningkatan+kebutuhan+komoditas+kedelai+di+Indonesia.%0D%0AProduksi+dan+hasil+panen+kedelai+dipengaruhi+oleh+serangan+hama+yang+menyebabkan%0D%0Ahilangnya+hasil+pertanian+tiap+tahun.+Salah+satu+hama+tersebut+adalah+serangga.%0D%0ACaterpillar+(larva+serangga+Ordo+Lepidoptera)+dan+serangga+Diabrotica+speciosa%0D%0Aadalah+jenis+hama+yang+sering+menyerang+daun+kedelai+sehingga+merusak+kualitas+dan%0D%0Akuantitas+produksi+kedelai.+Oleh+karena+itu%2C+penelitian+ini+bertujuan+untuk+membantu%0D%0Apetani+dalam+upaya+memberi+penanganan+yang+tepat+pada+kerusakan+daun+kedelai.%0D%0AIdentifikasi+ini+membutuhkan+algoritma+pembelajaran+mesin+yang+baik+seperti+deep%0D%0Alearning.+Deep+learning+disarankan+untuk+melakukan+identifikasi+atau+klasifikasi%0D%0Akarena+dapat+mempelajari+fitur+dari+gambar+kerusakan+daun+kedelai+secara+otomatis.%0D%0APada+penelitian+ini+model+deep+learning+yang+digunakan+untuk+mengidentifikasi%0D%0Akerusakan+daun+kedelai+adalah+penggabungan+Convolutional+Neural+Network+(CNN)%0D%0Adan+Long+Short-Term+Memory+(LSTM)+di+mana+CNN+dapat+memecahkan+masalah%0D%0Akarakteristik+spasial+dan+LSTM+dapat+memahami+konteks+temporal+pada+data+gambar.%0D%0AModel+tersebut+dikembangkan+melalui+tahap+pemrosesan+gambar+seperti+augmentasi%0D%0Adata+menggunakan+platform+roboflow.+Dataset+yang+digunakan+berasal+dari+platform%0D%0AMendeley+Data+dengan+jumlah+dataset+sebanyak+6410.+Model+hybrid+CNN-LSTM%0D%0Ayang+digabungkan+berkinerja+baik+dalam+mengidentifikasi+kerusakan+daun+kedelai%0D%0Adengan+capain+akurasi+sebesar+93%25%2C+presisi+sebesar+93%25%2C+recall+sebesar+93%25%2C+dan+f1-%0D%0Ascore+sebesar+93%25.%0D%0A%0D%0AKata+Kunci+%3A+Kedelai%2C+Caterpillar%2C+Diabrotica+Speciosa%2C+Deep+Learning%2C+CNN%2C%0D%0ALSTM%2C+CNN-LSTM%0D%0A&rft.publisher=FAKULTAS+MATEMATIKA+DAN+ILMU+PENGETAHUAN+ALAM&rft.date=2024-08-28&rft.type=Skripsi&rft.type=NonPeerReviewed&rft.format=text&rft.identifier=http%3A%2F%2Fdigilib.unila.ac.id%2F82956%2F1%2FABSTRAK%2520-%2520Nanda%2520Evitarina.pdf&rft.format=text&rft.identifier=http%3A%2F%2Fdigilib.unila.ac.id%2F82956%2F2%2FSKRIPSI%2520FULL%2520-%2520Nanda%2520Evitarina.pdf&rft.format=text&rft.identifier=http%3A%2F%2Fdigilib.unila.ac.id%2F82956%2F3%2FSKRIPSI%2520FULL%2520TANPA%2520PEMBAHASAN%2520-%2520Nanda%2520Evitarina.pdf&rft.identifier=++NANDA+%2C+EVITARINA++(2024)+IMPLEMENTASI+MODEL+HYBRID+CONVOLUTIONAL+NEURAL+NETWORK+DAN+LONG+SHORT-TERM+MEMORY+UNTUK+IDENTIFIKASI+KERUSAKAN+DAUN+KEDELAI+AKIBAT+SERANGAN+LARVA+LEPIDOPTERA+DAN+DIABROTICA.++FAKULTAS+MATEMATIKA+DAN+ILMU+PENGETAHUAN+ALAM%2C+UNIVERSITAS+LAMPUNG.+++++&rft.relation=http%3A%2F%2Fdigilib.unila.ac.id%2F82956%2F