Prediction and Classification Of Vector-Borne and Communicable Diseases through Artificial Neural Networks

Karn, Shivam; Sangole, Shubham; Gawde, Abhishek; Joshi, Jyoti

doi:10.1109/iccs45141.2019.9065500

Cited by 4 publications

(1 citation statement)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Algoritma backpropagation mempropagasi kesalahan melalui jaringan saraf dan pembaruan bobot terjadi [17], memiliki fungsi latihan dan fungsi transfer. Fungsi-fungsi ini memiliki banyak teknik serta jalur memecahkan masalah kompleks [18].…”

Section: Pendahuluanunclassified

Analisa Dini Gangguan Disleksia Anak Sekolah dengan Metode Backpropagation

Yanti,

Setiawan,

Defit

2023

JEPIN

View full text Add to dashboard Cite

Disleksia sering disalah artikan sebagai kebodohan atau kemalasan pada anak. Gejala disleksia dikenal dengan gangguan belajar yang meliputi mengenal huruf, mengeja, membaca, dan menulis. Meskipun gejala disleksia tidak terlihat dengan jelas, kondisi ini dapat berdampak pada perkembangan pola belajar anak. Tujuan penelitian adalah untuk mengidentifikasi gejala disleksia sedini mungkin agar tidak mengganggu perkembangan belajar pada anak. Selain itu, penelitian juga bertujuan untuk mengevaluasi keakuratan teknik yang digunakan. Analisa menggunakan metode jaringan syaraf tiruan dengan teknik backpropagation dengan memberikan nilai bobot, sehingga dapat memberikan nilai input dengan benar. Penelitian menggunakan 150 dataset, 40 variabel input dan 40 lapisan tersembunyi. Keluaran yang diharapkan mencakup disleksia atau non-disleksia. Hasil implementasi dan pengujian untuk data latih dan data uji terbaik adalah 90:10. Dengan nilai epoch maksimum 5000 dan nilai error target 0,001. Metode backpropagation dapat memberikan hasil akurasi terbaik 100% pada learning rate 0,5. Sehingga metode backpropagation dapat dengan baik mendeteksi gangguan disleksia pada anak sejak dini.

show abstract

Section: Pendahuluanunclassified

Analisa Dini Gangguan Disleksia Anak Sekolah dengan Metode Backpropagation

Yanti,

Setiawan,

Defit

2023

JEPIN

View full text Add to dashboard Cite

show abstract

A Large-Scale IoT-Based Scheme for Real-Time Prediction of Infectious Disease Symptoms

Said

2023

Mobile Netw Appl

View full text Add to dashboard Cite

In recent decades, many infectious diseases have appeared that have negatively affected life in general and people in particular, causing many economic and human losses. Recently, many attempts have emerged to confront these diseases using computer-based technology for diagnosis, prediction, and data analysis using various techniques, the most important of which is deep learning. Previous research relied primarily on a set of images taken from the patient’s body while he was in a healthcare facility, and this is the main weakness of these studies. Not all people go to a doctor or hospital when they feel the symptoms of a disease. Hence, people moving in crowded places without knowing their health status can contribute to spreading infectious diseases quickly, and this is the issue that should be confronted. Therefore, this paper presents a people-monitoring scheme, which is based on the internet of things (IoT) technology, to predict infectious disease symptoms through people’s behavior as well as through a wireless body area network (WBAN). This scheme can predict the spread of disease by tracking the movements of infected persons. Additionally, a simple methodology for processing the data extracted from the monitoring process across a range of different computing centers is introduced. Moreover, to ensure the monitoring scheme operates in real-time, it was necessary to provide a powerful coverage model for its objects. Also, a simple COVID-19 case study is presented. Finally, the performance of the prediction model is measured using images, sounds and videos files. Furthermore, the performance of the data computing and coverage methodologies is measured using an intensive simulation environment for the IoT that was constructed using NS3 package. The results showed that the proposed scheme is able to predict the symptoms of disease and its spread with accepted level of accuracy. In addition, using a mixture of coverage tools and computing techniques is recommended.

show abstract