Sentiment analysis of students' Facebook comments toward university announcements

Students' opinions are among the critical indicators to evaluate the university teaching process. However, due to the absence of an official online system in most universities that provides a mechanism for obtaining students' opinions on several university announcements, most students use various social networks to express their feelings and provide their opinions toward these announcements. We present, through this paper, sentiment analysis of Facebook comments written in the Moroccan Arabic dialect. These comments reflect the opinions of students about university announcements during the COVID-19 pandemic, especially those related to teaching mode and ex-am planning. Then, the comments collected were cleaned, preprocessed, and manually classified into four categories, namely positive, neutral, negative, and bipolar. Further, data dimensionality reduction is applied using TF-IDF and Chi-square test. Finally, we evaluated the performance of three standard classifiers, i.e., Naïve Bayesian (NB), Support Vector Machines (SVM), and Random Forests (RF) using k-fold cross-validation. The results showed that the SVM-based classifier performs as well as the RF-based classifier regarding the classification's accuracy and F1-score, while the NB-based classifier lags behind them. © 2022 IEEE.

Analysis of Sentiment Tendency Based on Major Public Health Events

The outbreak of the COVID-19 has seriously affected the lives of the public. On the network platform, there has been a lot of controversy about this issue, which caused panic among the people. Accurate Sentiment propensity analysis of user statements on various platforms can better guide public opinion and avoid unnecessary panic. This paper classifies data based on the Naive Bayesian algorithm because the traditional Naive Bayesian algorithm does not consider that the feature weights of the same feature word in different classes are different when classifying. Under the strong hypothesis of independence between features, the same feature word has the same importance, which will reduce the accuracy of the classifier. Therefore, this paper uses an improved TF-IDF algorithm to weight it and performs classification experiments. The experimental results show that this model can achieve better performance in text classification. © 2022 IEEE.

COVID-19 detection based on combined domain features

The computed tomography (CT) scan delivers more detailed information and higher judgment accuracy than a chest X-ray, which has a wide range of uses in diagnosing and decision-making to aid medical professionals. This paper proposed a method to detect COVID-19 from CT scan images using the combination of spatial domain and transform domain features. Using the lung segmentation step, the CT image is first processed and segmented, and then various domain features are extracted. From these domain features, the highest combined domain features (CDF) are obtained. Finally, the detection task is completed using random forest (RF) and Naive Bayesian (NB) classifiers. The proposed method is tested using a dataset of CT scan images, and the results are compared to several current techniques. The results showed that our method based on CDF outperforms previous methods, with an overall accuracy of nearly 98%. As can be shown, CDF is the best domain feature to apply for detecting COVID -19. © 2022 Institute of Advanced Engineering and Science. All rights reserved.

Machine Learning Applications in Forecasting of COVID-19 Based on Patients' Individual Symptoms

Predicting the COVID-19 outbreak has been studied by many researchers in recent years. Many machine learning models have been used for the prediction of the transmission in a country or region, but few studies aim to predict whether an individual has been infected by COVID-19. However, due to the gravity of this global pandemic, prediction at an individual level is critical. The objective of this paper is to predict if an individual has COVID-19 based on the symptoms and features. The prediction results can help the government better allocate the medical resources during this pandemic. Data of this study was taken on June 18th from the Israeli Ministry of Health on COVID-19. The purpose of this study is to compare and analyze different models, which are Support Vector Machine (SVM), Logistic Regression (LR), Naive Bayesian (NB), Decision Tree (DT), Random Forest (RF) and Neural Network (NN). © 2021 ACM.

INTELLIGENT LOW-COST PRELIMINARY IDENTIFICATION OF COVID-19 USING IOT

: The World Health Organization(WHO)has proclaimed a worldwide health emergency of international concern due to the coronavirus (COVID-19)disease outbreak. This viral outbreak has caused more than 2,863,225 deaths in the world. It has spread over into all areas of the globe. Excessive national and international action is being taken to stop the outbreak. The WHO suggested taking the necessary steps and measures to reduce the risk of the disease or importation.WHO's suggested measures are not to contact the infected person and do not touch the frequently used areas. People are observing these suggestions, but it is still spreading. The process of vaccination around the world has started. Coronavirus disease can be avoided or stopped, with the instant widespread of internet technologies. Current Internet of Things (IoT) developments on coronary virus protection is discussed in this paper from a fever control point of view on airports, religious sites, borders, events, etc. The design of the technique developed in this paper is a very low-cost remote temperature monitoring system model IoT Naïve Bayesian (INB) which measures body temperature by the sensor with infrared rays, processes and learns intelligently with Naïve Bayesian System and sends the data to a cloud system without any human intervention. It is extremely useful in preventing the epidemic on airports, religious sites, border crossings, and activities, among other places.

A multi-task CNN learning model for taxonomic assignment of human viruses.

BACKGROUND: Taxonomic assignment is a key step in the identification of human viral pathogens. Current tools for taxonomic assignment from sequencing reads based on alignment or alignment-free k-mer approaches may not perform optimally in cases where the sequences diverge significantly from the reference sequences. Furthermore, many tools may not incorporate the genomic coverage of assigned reads as part of overall likelihood of a correct taxonomic assignment for a sample. RESULTS: In this paper, we describe the development of a pipeline that incorporates a multi-task learning model based on convolutional neural network (MT-CNN) and a Bayesian ranking approach to identify and rank the most likely human virus from sequence reads. For taxonomic assignment of reads, the MT-CNN model outperformed Kraken 2, Centrifuge, and Bowtie 2 on reads generated from simulated divergent HIV-1 genomes and was more sensitive in identifying SARS as the closest relation in four RNA sequencing datasets for SARS-CoV-2 virus. For genomic region assignment of assigned reads, the MT-CNN model performed competitively compared with Bowtie 2 and the region assignments were used for estimation of genomic coverage that was incorporated into a naïve Bayesian network together with the proportion of taxonomic assignments to rank the likelihood of candidate human viruses from sequence data. CONCLUSIONS: We have developed a pipeline that combines a novel MT-CNN model that is able to identify viruses with divergent sequences together with assignment of the genomic region, with a Bayesian approach to ranking of taxonomic assignments by taking into account both the number of assigned reads and genomic coverage. The pipeline is available at GitHub via https://github.com/MaHaoran627/CNN_Virus .

Intrusion detection by machine learning for multimedia platform.

The multimedia service company, Netflix, increased the number of new subscribers during the Coronavirus pandemic age. Intrusion detection systems for multimedia platforms can prevent the platform from network attacks. An intelligent intrusion detection system is proposed for the security IP Multimedia Subsystem (IMS) based on machine learning technology. For increasing the accuracy of the classifiers, it is vital to select the critical features to construct the intrusion detection system. Two-class classifiers, including the Decision Tree, Support Vector Machine, and Naive Bayesian, are selected to evaluate intrusion detection accuracy. According to the three classifiers' accuracy values, the most critical features are selected based on the features' ranking orders. Six critical features are selected:Service, dst_host_same_srv_rate, Flag, Protocol Type, Dst_host_rerror_rate, and Count. Numerical comparison with state_of_the_art shows that critical features improve intrusion detection accuracy, which can be better than the deep learning method.