بناء نموذج تنبؤ بمناطق انتشار كوفيد-19 باستخدام خوارزمية الذاكرة طويلة قصيرة المدى

الباحث

ابتسام عباس عثمان

المشرف

بان شريف مصطفى

الجامعة

جامعة الموصل

الكلية / القسم

كلية علوم الحاسوب والرياضيات / علوم الحاسوب

الشهادة

دبلوم عالي

السنة

2022

لغة الرسالة

عربي

الكلمات المفتاحية

كوفيد-19 ; التنبؤ ; الذاكرة طويلة قصيرة المدى ; التعلم العميق

The massive spread of COVID-19 has made it a new global pandemic. Predicting the extent of the virus' spread is critical to containing the threat, because it helps to take appropriate measures and decisions at the state level as well as at the personal level, where it is possible to avoid travel to the places of spread and take the necessary measures to limit the spread of the virus. In this research, a smart model was built to predict the extent of the spread of Covid-19 disease in the Iraqi governorates., COVID-19 data for Iraq's governorates was obtained from a website affiliated with the Iraqi Ministry of Health. The data was reconstructed according to a certain structure to be used in training the prediction model. The vector regression support algorithm (LinearSVR) and the gradient boosting algorithm were used to build the prediction model. The performance of the models used was measured using the average squared logarithmic error (Root Mean Squared Logarithmic Error) The value of the loss function for the first model was (0.49943), and the value of the loss function for the second model was (0.671821).The deep learning algorithm LSTM (Long Short-Term Memory) was adopted for its effective performance in predicting the direction of the recorded cases in the future. for its effective performance in predicting the direction of the recorded cases in the future. The results showed a high accuracy that outperformed the rest of the algorithms.

طلب نسخة من الملف

الملفات غير منشورة على الموقع. اترك بياناتك وسيتواصل معك موظفو قسم الشؤون الثقافية والفنون لإرسال النسخة مجاناً.

الاسم

رقم الهاتف

البريد الإلكتروني

الغرض من الطلب (اختياري)

إرسال الطلب