A Recommender Model based on Bayesian Theory for Detection of COVID-19 on the Internet of Things Ecosystem

Subject Areas : Multimedia Processing, Communications Systems, Intelligent Systems

Zahra Ghorbankhani ¹ , Mani Zarei ² , Rahil Hosseini ³

1 - MSc, Computer Engineering, Shahr-e-Qods Branch, Islamic Azad University, Tehran, Iran
2 - Assistant Professor, Department of Computer Engineering, Shahr-e-Qods Branch, Islamic Azad University, Tehran, Iran
3 - Associate Professor, Department of Computer Engineering, Shahr-e-Qods Branch, Islamic Azad University, Tehran, Iran

Received: 2023-09-26 Accepted : 2024-11-05 Published : 2024-12-21

Keywords: IoT, Bayesian theory model, Data mining, COVID-19, Bayesian boosting,

Abstract :

Abstract

Introduction: Utilizing artificial intelligence (AI) techniques in healthcare with recommending techniques is highly effective in improving the smart health system, accelerating the control of mutated versions of viral diseases such as COVID-19, and reducing treatment costs. Therefore, using data mining techniques in the Internet of Things (IoT) ecosystem can be considered a reliable recommender system to identify people suffering from viral diseases such as COVID-19 in AI-assisted healthcare societies. Although the COVID-19 disease does not cause the death of humans at the moment, it still has a longer treatment period, and it is still catastrophic in elderly people or people with underlying diseases.

Method:  In this article, a recommender AI-assisted model is proposed based on Bayesian theory to detect COVID-19 in the IoT ecosystem. For this purpose, the factors affecting the coronavirus epidemic are discussed, and the data set of 50,000 patients is used to identify the disease of COVID-19 in different people. The data mining techniques used in this article include the Bayesian theory model, the proposed hybrid Bayesian boosting model, the nearest neighbor model, and the logistic regression model as machine learning methods.

Results: The Bayesian recommendation model in this research, can calculate the probability of contracting COVID-19 disease in new people with higher accuracy than the nearest neighbor and logistic regression models. By Referring to the performance evaluation results for the Bayesian theory model, Bayesian boosting proposed hybrid model, nearest neighbor model, and logistic regression model we conclude that the Bayesian belief model with 87% and its enhanced model with 98% have the highest accuracy.

Discussion: The results measured in the algorithms show that the Bayesian theory recommender model, considering five features related to the symptoms of COVID-19 in the training data, calculated the probability of  COVID-19 in new people with higher accuracy than the two nearest neighbor and logistic regression models. Considering that the focus of this research was on the accuracy of the proposed model, combining it with the decision tree model, which is named the Bayesian reinforcement model in this research, has been able to increase a significant percentage of accuracy. In addition, it compares the results by implementing two other models: nearest neighbor and logistic regression. Considering that the time index is one of the key components, the methods that have been introduced have been able to predict the probability of contracting COVID-19 in the fastest time.

 

References:

[1] P. Sunhare, R. R.Chowdhary, and M. K.Chattopadhyay. "Internet of things and data mining: An application oriented survey". Journal of King Saud University-Computer and Information Sciences, vol. 34, no. 6, pp.3569-3590. 2022
[2] M. Gholami, Sh. Damanabi, P. Rezaei Hachesu, and F. S. Ghyassi. "Evaluation of Nursing Information Systems using the HIS-Monitor Instrument: Nurses Perspectives". Iranian Journal of Medical Informatics, vol. 8, no. 1, 2019.
[3] E. M. Silva, and R. Jardim-Goncalves. "IoT ecosystems design: A multimethod, multicriteria assessment methodology". IEEE Internet of Things Journal, vol. 7, no. 10, pp.10150-10159. 2020.
[4] D. Kumar, S. K. Sood, and K. S. Rawat. "IoT-enabled technologies for controlling COVID-19 Spread: A scientometric analysis using CiteSpace". Internet of Things, vol. 23, pp.100863. 2023
[5] W. Lindquist, S. Helal, A. Khaled, and W. Hutchinson. "IoTility: architectural requirements for enabling health IoT ecosystems". IEEE Transactions on Emerging Topics in Computing, vol. 9, no. 3, pp.1206-1218. 2019.
[6] M. Ndiaye, S. S. Oyewobi, A. M. Abu-Mahfouz, G. P. Hancke, A. M. Kurien, and K. Djouani, K. "IoT in the wake of COVID-19: A survey on contributions, challenges and evolution". Ieee Access, 8, pp.186821-186839. 2020
[7] R. Sandhu, S. K. Sood, and G. Kaur" An intelligent system for predicting and preventing MERS-CoV infection outbreak". The Journal of supercomputing, vol. 72, no. 8, pp.3033-305. 2016.
[8] P. Verma. and S. K. Sood, 2018." Fog assisted-IoT enabled patient health monitoring in smart homes". IEEE Internet of Things Journal, vol. 5, no. 3, pp.1789-1796. 2018.
[9] P. Verma, and S. K. Sood. "A comprehensive framework for student stress monitoring in fog-cloud IoT environment: m-health Perspective". Medical & biological engineering & computing, vol. 57, no. 1, pp. 231-244. 2019.
[10] M. Bhatia and S.K. Sood. "Exploring temporal analytics in fog-cloud architecture for smart office healthcare". Mobile Networks and Applications, 24(4), pp.1392-1410. 2019.
[11] Z. Karimi, H. Haghihi. "The Opinion Mining of Digikala Reviews by Semi-Supervised Support Vector Machine" Intelligent Multimedia Processing and Communication Systems, vol. 1, no. 11, pp. 51-61, 2023.
[12] M. Bazergani, M. and Homayounpour, Z. "Providing A New Method to Discover the Closest Neighborhood in Recommendation Systems Based on Collaborative Filtering" Intelligent Multimedia Processing and Communication Systems, vol. 1, no. 1, pp. 55-64, 2020.
[13] G. Latif , A. Shankar, J. M. Alghaz, V. Kalyanasundaram, C. S. Boopathi, and M. Arfan Jaffar. "I-CARES: advancing health diagnosis and medication through IoT". Wireless Networks. vol. 26, no. 4, pp. 2375-2389. 2020.
[14] P. Verma, S. K. Sood. "Cloud-centric IoT based disease diagnosis healthcare framework". Journal of Parallel and Distributed Computing. vol. 116, pp. 27-38. 2018.
[15] J. Han, M. Kamber, and J. Pei. "Data Mining Concepts and Techniques", Third Edition, Elsevier Publisher, USA, 2012.
[16] G.F. Cooper, and E. Herskovits, "A Bayesian method for the induction of probabilistic networks from data". Machine learning, vol. 9, no. 4, pp.309-347. 1992.
[17] M. Esmaeili, Azad University publisher, "Concepts and techniques of data mining". 2018.
[18] Kaggle. (March 2020-November 2021). Covid-19 Dataset for year 2020 & 2021. https://www.kaggle.com/datasets/mykeysid10/covid19-dataset-for-year-2020/data?select=covid_data_2020-2021.csv