Question

1. Penyelidik-penyelidik dari Singapore University of Technology and Design (SUTD) telah membangunkan suatu sistem bagi pemantauan ramalan pandemik COVID-19 untuk banyak negara di dunia. Adakah anda bersetuju dengan kenyataan "Rangkaian neural buatan adalah sesuai untuk pemantauan ramalan pandemik COVID-19"? Sokong pandangan anda dengan berpandukan teori pengkomputeran lembut. Researchers from Singapore University of Technology and Design (SUTD) had developed a system for predictive monitoring of the COVID-19 pandemic for many countries around the world. Do you agree with the statement “Artificial Neural Network (ANN) is suitable for predictive monitoring of the COVID-19 pandemic”? Justify your view with regard to soft computing theory. (5 markah/marks)

Answer 1

ANN feasibility for predictive Monitoring of the COVID-19 pandemic

Artificial neural network is viewed as one of the most productive techniques in handling immense informational sets that can be broke down computationally to uncover designs, patterns, expectation, forecasting etc.It has proved to be extraordinarily successful in engineering as well as in medical applications. The current work utilizes artificial neural network-based curve fitting procedures in prediction and forecasting of the COVID-19 number of rising cases and death cases in India, USA, France, and UK, thinking about the dynamic patterns of China also, South Korea.

Three cases that are considered to examine the spread of COVID-19 pandemic are.,

(i)according to the current pattern of rising instances of various nations

(ii) determining of multi week catching up with the improvement slants according to China and South Korea, and

(iii)determining whenever followed up the dynamic drifts according to China and South Korea before seven days.

The outcomes have indicated that ANN can productively estimate the future instances of COVID-19 flare-up of any nation.

The artificial neural network (ANN) is an information processing framework comprising of a enormous number of simple and highly interconnected processing elements resembling a natural neural system. It has the capability of learning from an experimental test data or real world data set to portray the nonlinear and connection impacts with incredible success. ANN-based curve fitting technique is one of the widely utilized artificial intelligence techniques that are designed for the purpose of forecasting and prediction. It comprises of fundamentally three layers i.e., input layer, hidden layers, and output layer, the current work incorporates the quantity of days as input layer and the every day number of COVID-19 cases as output data for the network. In ANN curve fitting the training data followed the pattern of the given determined information and permits to make forecasts of how the information series will act later on. In the present study least square error method is utilized to build up the polynomial curve. During the training procedure, the network modifies its weights to limit the errors between the predicted and desired outputs. The ideal network is arrived upon a minimum error by randomly differing weights. The back propagation algorithm in ANN is utilized to train the network, and the network is used for estimating extrapolated data sets and the prediction is compared and contrasted with real data available at WHO official site. The ANN with neurons which are also called as nodes in each layer are interconnected with nodes of the subsequent and preceding layer with synaptic weights. In addition to this, a bias is also added to each neuron of the hidden and output layer.

Thanks!

Upvote if it helped!