Online Video Games based Mental Depression Prediction using Novel Recurrent Neural Network Over Convolutional Neural Network

Online Video Games based Mental Depression Prediction using Novel Recurrent Neural Network Over Convolutional Neural Network

연구 분야: Artificial Intelligence

학회: 2024 International Conference on Recent Innovation in Smart and Sustainable Technology (ICRISST)

The goal of this study is to detect online gaming addiction in young individuals using two types of neural networks: a novel recurrent neural network and a convolutional neural network. The study will compare the performance of these networks to determine their effectiveness in detecting online gaming disorder in youth. The collection of data and training the model has to be done. The number of groups chosen for this study are two and the machine learning algorithms used for this approach are Recurrent Neural Network and Convolution Neural Network Each group contains 10 samples and it totals 20 samples. For SPSS calculation, a G power value of 80% is used. The parameters considered are CI and alpha, which were determined as p= 0.000 (p<0.05). The analysis of online gaming mental disorder or disability has been done using novel recurrent neural network and convolutional neural network models. A greater proportion of 94% accuracy is attained by novel recurrent neural networks which is larger than 89%accuracy attained by Convolutional neural networks. In light of the acquired outcomes the Novel Recurrent Neural Network gives better accuracy (94%) when contrasted with convolutional neural networks which provide (89%) accuracy.