A Symptom-Based Machine Learning Model for Malaria Diagnosis in Nigeria
Küçük Resim Yok
Tarih
2021
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Ieee
Erişim Hakkı
info:eu-repo/semantics/closedAccess
Özet
Malaria, with around 200 million cases worldwide, tends to kill more people than war and crises. With efforts to reduce mortality rates being futile, an inadequate malaria diagnosis is one of the barriers to a successful reduction in mortality. Machine learning methods were used to classify the stages of malaria in patients to improve diagnosis. To predict the stages of malaria, this research used knowledge of an algorithm of machine learning for a predictive model. A 77% accurate decision algorithm was developed using the symptoms of patients to identify their malaria stages. This research also discovered that malaria does not kill only children (between 0-5 years), in contrast to what has been pointed out in many research studies. This study shows that older women are more likely to experience severe stages of malaria. Therefore, adequate care should be considered for these women once they show some of the significant symptoms as described in the model. This approach applies to everyone with the symptoms set out in the model. This system will provide a preliminary test before conducting a confirmatory diagnosis in the laboratories.
Açıklama
9th International Symposium on Digital Forensics and Security (ISDFS) -- JUN 28-29, 2021 -- Fırat Univ, Elazig, TURKEY
Anahtar Kelimeler
Malaria, Sokoto, Decision Tree, Machine Learning
Kaynak
9th International Symposium on Digital Forensics And Security (Isdfs'21)
WoS Q Değeri
N/A
Scopus Q Değeri
N/A
