Final Plag Report

Document Information
Analyzed document 201_Deepti.docx (D132129220)
Submitted 2022-03-31T11:15:00.0000000
Submitted by Anil Gupta
Submitter email [email protected]
Similarity 89%
Analysis address [email protected]
Sources included in the report
URL: https://shodhgangotri.inflibnet.ac.in/jspui/bitstream/123456789/9436/1/deepti%20malhotra%
20synopsis.pdf 14
Fetched: 2022-03-31T11:15:28.2270000
1/15
Entire Document
COVID-19 PREDICTION MODEL: A LITERATURE SURVEY
Deepti Malhotra1, Gurinder Kaur Sodhi2
1Department of Electronics & Communication Engineering, Desh Bhagat University (Punjab) India 147301 2Department of
Electronics & Communication Engineering, Desh Bhagat University (Punjab) India 147301
Corresponding Author EmailID: [email protected]
Abstract: Believed to have been originated Chinese province Wuhan in December 2019, the coronavirus havesaid to cause
95 million cases with overall death rate of 2%of overall cases(as per Jan 2022). This fast spreading pandemic virus poses a
challenge at world level and proposes serious danger to people’s health as well as the economy. With time and regions
this virus has undergone several mutations resulting in rise of various other viruses, OMICRON being the latest.The most
common and widely faced threat in this disease was in the case of asymptomatic patients, the ones who showed no
symptoms and yet were carriers of corona virus. In recent times,many researchers have started exploring various methods
to predict the disease on the basis of medical parameters. Few of the commonly tools used are machine learning and
artificial intelligence. The present paper aims to compile the various models used by researchers in last year in predicting
COVID.
Keywords: covid19, prediction techniques, artificial intelligence, machine learning
1.Introduction
1.1
100% MATCHING BLOCK 1/14 https://shodhgangotri.inflibnet.ac.in/jspui/bi ...
Covid 2019 The coronavirus originated in Chinese province, Wuhan in December 2019.
The coronavirus disease 2019 is a virus of the coronavirus family. Studies suggest that COVID-19 encompasses
medicaltra its similar to SARS-CoV [1]. Major symptoms are fever and cough, while gastrointestinal signs are rare. The
absence of fever in novel coronavirus- infected people is more common than in patients diseased with the same virus,
that is, MERS corona virus (2%) and SARS corona virus (1%). Thus, non-febrile patients are likely to be exempted from a
monitoring device with the main attention on diagnosing fever. 1.2
Machinelearning The
healthcare sector is a huge sector requiring medical data whose collection and processing is required in real-time.
The emergence of AI has proved to be a technological breakthrough of the 21th century. It includes many applications
in areas ranging from weather forecasting, astrophysical investigation to self-directed systems. Machine learning based
prediction has gained hugeadmiration in the last few years. The course of learning is basically learning from past work
experience or observations, such as instances, or instructions, to see patterns in data. This process makes use of
examples to improve the decision-making ability of the system[2]. 1.3
2/15
Covid-19 Disease Prediction using MachineLearning The epidemiological data containing the number of people who
suffered with this disease previously and the total population is employed in the predictive models of corona virus 2019.
Different parameters namely latency period and probability of healing are utilized for predicting the infection trends. But,
these models are inefficient to reflect diverse socio-static and economic factors due to which the course of the virus is
affected.
All stages of this predictive framework are as;Data collection, Relevant dataset, Pre-processing, Feature Selection and
Machine learning Data Collection Relevant dataset Data Pre-processing Patient data Feature selection COVID SARS
ARDS Both (SARS,COVID Machine Learning Fig. 1.1.General Composition of COVID-19 prediction model
Following are the some commonly used Covid-19 outbreak prediction models. K-Nearest neighbor algorithm
Logistic Growth Model Long short term memory
2.
LITERATURE REVIEW Pratima Kumari, et.al (2020) intended ML (machine learning) method called ANN in order to
predict the Coronavirus Disease 2019 outbreak in India for the first time [8]. In addition, a mathematical curve fitting
system was adopted for determining the performance of the intended method. The development of the epidemic at
diverse transmission rates for analyzing the impact of preventive measures such as lockdown and social distancing. The
intended method was proved appropriate to predict the development rate of metrics of corona disease at lower MAPE
values and cumulative deceasedcases. Yasin Khan, et.al (2020) formulated a CNN algorithm to carry out a multi-class
classification of the chest X-ray images of Coronavirus Disease 2019 as infected patients,patients of pneumonia and
healthy persons [9]. Moreover, the Monte-Carlo simulation was utilized on the original data distributions for improving
the data so that the disease was predicted. Moreover, COVID-19 pandemic was forecasted using LR of the components
of GMM.
X-ray pneumonia images. The accuracy obtained from the formulated algorithm was computed 100% in training phase
and 96.66% in the testingphase. Choujun Zhan, et.al (2021) projected a ML (machine learning) mechanism in order to
forecast the COVID-19 on the basis of BLS (Broad Learning System) [10]. The significant attributes were extracted using
RF (Random Forest). Therefore, RF-Bagging-BLS (Random- forest-Bagging Broad Learning System) was constructed by
integrating the bagging strategy with BLS with the objective of predicting the COVID-19 pandemic. The projected
mechanism performed more effectively with regard to RMSE (relative mean square error), MAD and MAPE in
comparison with other models. Thus, the projected mechanism outperformed the
traditionalmodels.
Leonardo
3/15
et. al(2021) introduced an ANN with the objective of predicting the number of covid confirmed cases and deaths and
also the future seven days for the time series of Brazil, Portugal and the United States [12]. The simulation results
depicted that the introduced method was capable to predict the confirmed cases and deaths from Coronavirus Disease
2019. The ANN utilized with a specific test set provided 50% higher MSE as compared to the Artificial Neural Network
with a random test set. The least MSE was acquired for all scenarios when the sigmoidal and linear activation functions
were integrated with the LMfunction. Durga Prasad Kavadi, et.al (2020) recommended a PDR-NML technique to predict
Coronavirus Disease 2019 which was spread all over the world [13]. The optimal metrics were investigated efficiently
using PPDLR model. Subsequently, the NML was adopted for the normalized attributes so as the accurate predictions
were achieved. The results indicated the superiority of the recommended technique over existing schemes in the Indian
population and it was presented as a promising tool to make predictions for other countries. Ankan Ghosh Dastider, et.al
(2021) investigated a CNN (convolutional neural network) algorithm in which the fusion of an autoencoder network and
separable convolutional branches was done with a modified DenseNet-201 network [14]. Hence, a vigorous, noise- free
classification system was constructed. The accuracy of investigated algorithm was evaluated using a five-fold cross-
validation system. The in-depth result analysis proved that when LSTM layers were utilized, the classification
performance was enhanced. This algorithm provided performed more effectively as compared to conventional
techniques. The extensive analysis confirmed the applicability of the investigated algorithm to detect the severity scores
of Coronavirus Disease 2019 from LUS images. Abhishek Dixit, et.al (2021) established a 3-step approach that employed
K-means clustering and extracted the attributes in the data pre-processing stage [15]. The subsequent stage focused on
optimizing the selected attributes. For this, a novel feature optimization technique was deployed on the basis of hybrid
DEA (differential evolution algorithm) and PSO. Thereafter, the SVM made the utilization of optimized attributes. The
empirical outcomes demonstrated that the established approach yielded the accuracy around 99.34% and their
robustness and adaptability of this approach was also proved in detecting the patients suffered from COVID-19. Elena
Casiraghi, et.al (2020) constructed a computerized mechanism whose major intend was to extract the most relevant
radiological and laboratory variables for improving the process of predicting the patient risk [16]. After that, an adequate
ML (machine learning) model was put forward. A robust and consistent variable was selected by integrating Boruta with
RF (Random Forest) in a 10-fold cross-validation system for achieving a variable importance estimate that was impartial
when the surrogates were available. The selection of significant variables was done for training the RF algorithm whose
robustness and efficacy was presented. The implementation of the developed system was easy and assisted the
emergency departments in predicting the rapid and accurate risk in COVID-19 patients. Anwar Jarndal, et.al (2020)
discussed that a model was formulated for estimating the number of deaths in the affected cases on the basis of
documented number of older diabetic and smoking cases [17]. The GPR algorithm was adopted for building a model. A
comparative analysis was done on this algorithm against ANN (Artificial Neural Network) model. A trustworthy data
gathered from WHO for distinct countries was utilized in this algorithm. The adopted algorithm was efficient to estimate
the number of deaths because of any arbitrary number of inputs. The effective measures were planned for mitigating
the number of deaths using this algorithm. Huan Zhao, et.al (2021) designed the BP (back propagation neural network)
algorithm for analyzing and predicting the growth rate of Coronavirus Disease 2019 [18]. Initially, the gathering of
epidemic data of Italy was done from WHO. Subsequently, R and MATALB were employed to analyze a number of
indicators. A computer language was executed to put together the results obtained from multiple indicators for
predicting the COVID-19 in Italy. The predictive outcomes indicated that the trend of Italy epidemic situation in the
study period was found good. But, the measures were required to prevent and control the corona disease. Eventually, a
comparison of fitting value of the designed algorithm was done with real value and the fitting degree was obtained
upto0.99. Nanning Zheng, et.al (2020) suggested a hybrid AI (artificial-intelligence) technique in order to forecast the
Coronavirus Disease 2019 [19]. At first, an ISI (improved susceptible-infected) model was presented for estimating
several infection rates so that the transmission laws and the growth trend were analyzed. Later, the effects of measures
taken to prevent and control the disease were taken in account for embedding the NLP (natural language processing)
and LSTM network into the presented model with the objective of developing the suggested hybrid technique to predict
coronavirus disease. The outcomes obtained in experimentation depicted that the suggested technique was capable of
mitigating the errors of the predictive results. Safa Bahri, et.al (2020) intended a LSTM (Long Short-Term Memory)-based
predictive framework to predict the future cumulative number of recovered cases of CoronaVirus Disease 2019 seven
days earlier, in India, USA and Italy [20]. This framework was suitable for the dynamic of evolution of recovered cases of
the observed Coronavirus Disease 2019 and predicting the disease accurately with a total error within 1.46 to 2.65%. The
experimental results revealed the effectiveness of intended framework which provided a minimal error below 3%. This
4/15
model was proved as an effective tool for practitioners for monitoring the occurrence of corona disease all over
theworld. Teddy Mantoro, et.al (2020) introduced two techniques namely SVR and SIR system for predicting COVID-19
[21]. The disease was predicted in the best-case as well as worst-case scenario. The initial scenario included the current
daily case as a maximum case and another country's maximum case were contained in the latter scenario utilized. The
SVR algorithm provided diverse end of epidemic. The SIR model was utilized to validate the similar end of epidemic on
January 2021 in both cases. However, a rapid increase of infectious people out of 450,000 people was found in initial
scenario to 5,500,000 people in latter scenario. The prediction obtained using the introduced approach offered an
insight for the policy maker to deal with the coronavirus diseasepandemic. B. Prabha, et.al (2021) constructed a novel
ML (machine learning) method known as HDDP (Hybrid Disease Detection Principle) which was developed using two
traditional ML model namely CNN (Convolutional Neural Network) and the AdaBoost Classifier [22]. The ML factors
were considered for the CT image of lung that was pre-processed, its attributes were extracted and the classification
was done. The constructed method had potential for handling the CT images on the basis of these effective image
processing strategies. Hence, the COVID-19 disease was predicted and a report was sent to the respective user with
properaccuracy ratio. This approach focused on predicting the disease effectively and assisted the doctors and patients
successfully. Wencheng Zou, et.al (2020) devised the ML (machine learning) techniques namely LR (linear regression)
and NN (neuron networks) to predict the number of positive cases of Coronavirus Disease 2019 [23]. The LR algorithm
was deployed for gathering the data of a state so that the corona disease was predicted. The number of infections was
predicted using the data taken from Georgia and Massachusetts. The dataset was split into3 consecutive time periods
and diverse models were trained to be suitable for each corresponding data for comparing MSE (mean square error)
values. Moreover, the general viewpoint despite of three time periods depicted the superiority of single variable linear
regression (LR) over accurately in comparison with the traditional technique with regard
toaccuracy.
References [1] Yifan Yang, Wenwu Yu, Duxin Chen, “Prediction of COVID-19 spread via LSTM and the deterministic SEIR
model”, 2020, 39th Chinese Control Conference(CCC) [2] Zainab Abbas AbdulhusseinAlwaeli, Abdullahi Abdu Ibrahim,
“Predicting Covid-19 Trajectory Using Machine Learning”, 2020, 4th International Symposium on Multidisciplinary
Studies and Innovative Technologies(ISMSIT) [3] Suraj Bodapati, Harika Bandarupally, M Trupthi, “COVID-19 Time Series
Forecasting of Daily Cases, Deaths Caused and Recovered Cases using Long Short Term Memory Networks”, 2020, IEEE
5th International Conference on Computing Communication and Automation(ICCCA) [4] Hanqing Chao, Xi Fang,
Pingkun Yan, “Integrative analysis for COVID-19 patient outcome prediction”, 2020, Medical ImageAnalysis [5] Danish
Rafiq, Suhail Ahmad Suhail, Mohammad Abid Bazaz, “Evaluation and prediction of COVID-19 in India: A case study of
worst hit states”, 2020, Chaos, Solitons &Fractals [6] Farah Shahid, AneelaZameer, Muhammad Muneeb, “Predictions for
COVID-19 with deep learning models of LSTM, GRU and Bi-LSTM”, 2020, Chaos, Solitons &Fractals [7] Anwar Jarndal,
Saddam Husain, Omar Zaatar, Talal Al Gumaei, Amar Hamadeh, “GPR and ANN based Prediction Models for COVID-19
Death Cases”, 2020, International Conference on Communications, Computing, Cybersecurity, and Informatics(CCCI)
[8] Pratima Kumari, Durga Toshniwal, “Real-time estimation of COVID-19 cases using machine learning and
mathematical models - The case of India”, 2020, IEEE 15th International Conference on Industrial and Information
Systems(ICIIS) [9] Yasin Khan, Pritam Khan, Sudhir Kumar, Jawar Singh, Rajesh M. Hegde, “Detection and Spread
Prediction of COVID-19 from Chest X-ray Images using Convolutional Neural Network-Gaussian Mixture Model”, 2020,
IEEE 17th India Council International Conference (INDICON) [10] Choujun Zhan, Yufan Zheng, Haijun Zhang, Quansi
Wen, “Random-Forest-Bagging Broad Learning System with Applications for COVID-19 Pandemic”, 2021, IEEE Internet
of Things Journal [11] Yifan Yang, Wenwu Yu, Duxin Chen, “Prediction of COVID-19 spread via LSTM and the
deterministic SEIR model”, 2020, 39th Chinese Control Conference(CCC) [12] Leonardo Sestrem de Oliveira, Sarah
Beatriz Gruetzmacher, João Paulo Teixeira, “COVID-19 Time Series Prediction”, 2021, Procedia ComputerScience [13]
Durga Prasad Kavadi, Rizwan Patan, Amir H. Gandomi, “Partial derivative Nonlinear Global Pandemic Machine Learning
prediction of COVID 19”, 2020, Chaos, Solitons & Fractals [14] Ankan Ghosh Dastider, Farhan Sadik, Shaikh Anowarul
Fattah, “An integrated autoencoder-based hybrid CNN-LSTM model for COVID-19 severity prediction from lung
ultrasound”, 2021, Computers in Biology andMedicine [15] Abhishek Dixit, Ashish Mani, Rohit Bansal, “CoV2-Detect-Net:
Design of COVID-19
predictionmodelbasedonhybridDE-PSOwithSVMusingChestX-rayimages”,2021,
5/15
Information Sciences [16] Elena Casiraghi, Dario Malchiodi, Gabriella Trucco, Marco Frasca, Luca Cappelletti, Tommaso
Fontana,“
Explainable Machine Learning for Early Assessment of COVID-19 Risk Prediction in Emergency Departments”, 2020,
IEEEAccess [17] Anwar Jarndal, Saddam Husain, Omar Zaatar, Talal Al Gumaei, Amar Hamadeh, “GPR and ANN based
Prediction Models for COVID-19 Death Cases”, 2020, International Conference on Communications, Computing,
Cybersecurity, and Informatics(CCCI) [18] Huan Zhao, Yichi Li, Shenglan Chu, Suling Zhao, Chenglin Liu, “A COVID-19
Prediction Optimization Algorithm Based on Real-time Neural Network Training—Taking Italy as an Example”, 2021, IEEE
Asia-Pacific Conference on Image Processing, Electronics and Computers(IPEC) [19] Nanning Zheng, Shaoyi Du, Jianji
Wang, He Zhang, Wenting Cui, Zijian Kang, Tao Yang, Bin Lou, Yuting Chi, Hong Long, Mei Ma, Qi Yuan, Shupei Zhang,
Dong Zhang,
FengYe,JingminXin,“PredictingCOVID-19inChinaUsingHybridAIModel”,2020,
IEEE Transactions on Cybernetics [20] Safa Bahri, MoetezKdayem, NesrineZoghlami, “Deep Learning for COVID-19
prediction”, 2020 4th International Conference on Advanced Systems and Emergent Technologies(IC_ASET) [21] Teddy
Mantoro, RahmadyaTriasHandayanto, Media Anugerah Ayu, Jelita Asian, “Prediction of COVID-19 Spreading Using
Support Vector Regression and Susceptible Infectious Recovered Model”, 2020, 6th International Conference on
Computing Engineering and Design(ICCED) [22] B. Prabha, Sandeep Kaur, HarikumarPallathadka, “Intelligent predictions
of Covid
diseasebasedonlungCTimagesusingmachinelearningstrategy”,2021,MaterialsToday:
Proceedings [23] Wencheng Zou, “The COVID-19 Pandemic Prediction in the US Based on Machine Learning”, 2020,
International Conference on Public Health and Data Science(ICPHDS)
National Conference on Advanced Research in Engineering Science and Technology (NCAREST 2022), Feb 24-25, 2022 at
Desh Bhagat University, Mandi Gobindgarh, Punjab, INDIA.
6/15
Hit and source - focused comparison, Side by Side
Submitted text As student entered the text in the submitted document.
Matching text As the text appears in the source.
1/14 SUBMITTED TEXT 12 WORDS 100% MATCHING TEXT 12 WORDS
Covid 2019 The coronavirus originated in Chinese COVID-19) The coronavirus originated in Chinese
province, Wuhan in December 2019. province, Wuhan in December 2019.
https://shodhgangotri.inflibnet.ac.in/jspui/bitstream/123456789/9436/1/deepti%20malhotra%20synops ...
The coronavirus disease 2019 is a virus of the coronavirus The coronavirus disease 2019 is a virus of the coronavirus
family. Studies suggest that COVID-19 encompasses family. Studies suggest that COVID-19 encompasses
medicaltra its similar to SARS-CoV [1]. Major symptoms medicaltra its similar to SARS-CoV [1]. Major symptoms
are fever and cough, while gastrointestinal signs are rare. are fever and cough, while gastrointestinal signs are rare.
The absence of fever in novel coronavirus- infected The absence of fever in novel coronavirus- infected
people is more common than in patients diseased with people is more common than in patients diseased with
the same virus, that is, MERS corona virus (2%) and SARS the same virus, that is, MERS corona virus (2%) and SARS
corona virus (1%). Thus, non-febrile patients are likely to corona virus (1%). Thus, non-febrile patients are likely to
be exempted from a monitoring device with the main be exempted from a monitoring device with the main
attention on diagnosing fever. 1.2 attention on diagnosing fever.
The emergence of AI has proved to be a technological The emergence of AI has proved to be a technological
breakthrough of the 21th century. It includes many breakthrough of the 21th century. It includes many
applications in areas ranging from weather forecasting, applications in areas ranging from weather forecasting,
astrophysical investigation to self-directed systems. astrophysical investigation to self-directed systems.
Machine learning based prediction has gained Machine learning based prediction has gained huge 2
hugeadmiration in the last few years. The course of admiration in the last few years. The course of learning is
learning is basically learning from past work experience or basically learning from past work experience or
observations, such as instances, or instructions, to see observations, such as instances, or instructions, to see
patterns in data. This process makes use of examples to patterns in data. This process makes use of examples to
improve the decision-making ability of the system[2]. 1.3 improve the decision-making ability of the system.
7/15
Covid-19 Disease Prediction using MachineLearning The Covid-19 Disease Prediction using Machine Learning The
epidemiological data containing the number of people epidemiological data containing the number of people
who suffered with this disease previously and the total who suffered with this disease previously and the total
population is employed in the predictive models of population is employed in the predictive models of
corona virus 2019. Different parameters namely latency corona virus 2019. Different parameters namely latency
period and probability of healing are utilized for predicting period and probability of healing are utilized for predicting
the infection trends. But, these models are inefficient to the infection trends. But, these models are inefficient to
reflect diverse socio-static and economic factors due to reflect diverse socio-static and economic factors due to
which the course of the virus is affected. which the course of the virus is affected.
Machine learning Data Collection Relevant dataset Data Machine Learning Data Collection Relevant dataset Data
Pre-processing Patient data Feature selection COVID Pre-processing Patient data Feature selection represents
SARS ARDS Both (SARS,COVID Machine Learning Fig. the COVID-19 prediction model on data
1.1.General Composition of COVID-19 prediction model Fig. 1.2: General Composition of COVID-19 prediction
model
Following are the some commonly used Covid-19 Following are the some commonly used Covid-19
outbreak prediction models. K-Nearest neighbor outbreak prediction models. 5 i. K-Nearest Neighbours
algorithm Algorithm:
8/15
LITERATURE REVIEW Pratima Kumari, et.al (2020) Literature Review Pratima Kumari, et.al (2020) intended
intended ML (machine learning) method called ANN in ML (machine learning) method called ANN in order to
order to predict the Coronavirus Disease 2019 outbreak in predict the Coronavirus Disease 2019 outbreak in India for
India for the first time [8]. In addition, a mathematical the first time [8]. In addition, a mathematical curve fitting
curve fitting system was adopted for determining the system was adopted for determining the performance of
performance of the intended method. The development the intended method. The development of the epidemic
of the epidemic at diverse transmission rates for analyzing at diverse transmission rates for analyzing the impact of
the impact of preventive measures such as lockdown and preventive measures such as lockdown and social
social distancing. The intended method was proved distancing. The intended method was proved appropriate
appropriate to predict the development rate of metrics of to predict the development rate of metrics of corona
corona disease at lower MAPE values and cumulative disease at lower MAPE values and cumulative
deceasedcases. Yasin Khan, et.al (2020) formulated a CNN deceased Yasin Khan, et.al (2020) formulated a
algorithm to carry out a multi-class classification of the CNN algorithm to carry out a multi-class classification of
chest X-ray images of Coronavirus Disease 2019 as the chest X-ray images of Coronavirus Disease 2019 as
infected patients,patients of pneumonia and healthy infected patients, patients of pneumonia and healthy
persons [9]. Moreover, the Monte-Carlo simulation was persons [9]. Moreover, the Monte-Carlo simulation was
utilized on the original data distributions for improving the utilized on the original data distributions for improving the
data so that the disease was predicted. Moreover, COVID- data so that the disease was predicted. Moreover, COVID-
19 pandemic was forecasted using LR of the components 19 pandemic was forecasted using LR of the components
of GMM. of GMM.
X-ray pneumonia images. The accuracy obtained from X-ray pneumonia images. The accuracy obtained from
the formulated algorithm was computed 100% in training the formulated algorithm was computed 100% in training
phase and 96.66% in the testingphase. Choujun Zhan, et.al phase and 96.66% in the testing Choujun Zhan,
(2021) projected a ML (machine learning) mechanism in et.al (2021) projected a ML (machine learning) mechanism
order to forecast the COVID-19 on the basis of BLS (Broad in order to forecast the COVID-19 on the basis of BLS
Learning System) [10]. The significant attributes were (Broad Learning System) [10]. The significant attributes
extracted using RF (Random Forest). Therefore, RF- were extracted using RF (Random Forest). Therefore, RF-
Bagging-BLS (Random- forest-Bagging Broad Learning Bagging-BLS (Random- forest-Bagging Broad Learning
System) was constructed by integrating the bagging System) was constructed by integrating the bagging
strategy with BLS with the objective of predicting the strategy with BLS with the objective of predicting the
COVID-19 pandemic. The projected mechanism COVID-19 pandemic. The projected mechanism
performed more effectively with regard to RMSE (relative performed more effectively with regard to RMSE (relative
mean square error), MAD and MAPE in comparison with mean square error), MAD and MAPE in comparison with
other models. Thus, the projected mechanism other models. Thus, the projected mechanism
outperformed the outperformed the
9/15
et. al(2021) introduced an ANN with the objective of et.al (2021) introduced an ANN with the objective of
predicting the number of covid confirmed cases and predicting the number of Coronavirus concerning
deaths and also the future seven days for the time series confirmed cases and deaths and also the future seven
of Brazil, Portugal and the United States [12]. The days for the time series of Brazil, Portugal and the United
simulation results depicted that the introduced method States [12]. The simulation results depicted that the
was capable to predict the confirmed cases and deaths introduced method was capable to predict the confirmed
from Coronavirus Disease 2019. The ANN utilized with a cases and deaths from Coronavirus Disease 2019. The
specific test set provided 50% higher MSE as compared to ANN utilized with a specific test set provided 50% higher
the Artificial Neural Network with a random test set. The MSE as compared to the Artificial Neural Network with a
least MSE was acquired for all scenarios when the random test set. The least MSE was acquired for all
sigmoidal and linear activation functions were integrated scenarios when the sigmoidal and linear activation
with the LMfunction. Durga Prasad Kavadi, et.al (2020) functions were integrated with the LM Durga Prasad
recommended a PDR-NML technique to predict Kavadi, et.al (2020) recommended a PDR-NML technique
Coronavirus Disease 2019 which was spread all over the to predict Coronavirus Disease 2019 which was spread all
world [13]. The optimal metrics were investigated over the world [13]. The optimal metrics were investigated
efficiently using PPDLR model. Subsequently, the NML efficiently using PPDLR model. Subsequently, the NML
was adopted for the normalized attributes so as the was adopted for the normalized attributes so as the
accurate predictions were achieved. The results indicated accurate predictions were achieved. The results indicated
the superiority of the recommended technique over the superiority of the recommended technique over
existing schemes in the Indian population and it was existing schemes in the Indian population and it was
presented as a promising tool to make predictions for presented as a promising tool to make predictions for
other countries. Ankan Ghosh Dastider, et.al (2021) other countries. Ankan Ghosh Dastider, et.al (2021)
investigated a CNN (convolutional neural network) investigated a CNN (convolutional neural network)
algorithm in which the fusion of an autoencoder network algorithm in which the fusion of an autoencoder network
and separable convolutional branches was done with a and separable convolutional branches was done with a
modified DenseNet-201 network [14]. Hence, a vigorous, modified DenseNet-201 network [14]. Hence, a vigorous,
noise- free classification system was constructed. The noise- free classification system was constructed. The
accuracy of investigated algorithm was evaluated using a accuracy of investigated algorithm was evaluated using a
five-fold cross-validation system. The in-depth result five-fold cross-validation system. The in-depth result
analysis proved that when LSTM layers were utilized, the analysis proved that when LSTM layers were utilized, the
classification performance was enhanced. This algorithm classification performance was enhanced. This algorithm
provided performed more effectively as compared to provided performed more effectively as compared to
conventional techniques. The extensive analysis conventional techniques. The extensive analysis
confirmed the applicability of the investigated algorithm confirmed the applicability of the investigated algorithm
to detect the severity scores of Coronavirus Disease 2019 to detect the severity scores of Coronavirus Disease 2019
from LUS images. Abhishek Dixit, et.al (2021) established a from LUS images. Abhishek Dixit, et.al (2021) established a
3-step approach that employed K-means clustering and 3-step approach that employed K-means clustering and
extracted the attributes in the data pre-processing stage extracted the attributes in the data pre-processing stage
[15]. The subsequent stage focused on optimizing the [15]. The subsequent stage focused on optimizing the
selected attributes. For this, a novel feature optimization selected attributes. For this, a novel feature optimization
technique was deployed on the basis of hybrid DEA technique was deployed on the basis of hybrid DEA
(differential evolution algorithm) and PSO. Thereafter, the (differential evolution algorithm) and PSO. Thereafter, the
SVM made the utilization of optimized attributes. The SVM made the utilization of optimized attributes. The
empirical outcomes demonstrated that the established empirical outcomes 10 demonstrated that the established
approach yielded the accuracy around 99.34% and their approach yielded the accuracy around 99.34% and their
robustness and adaptability of this approach was also robustness and adaptability of this approach was also
proved in detecting the patients suffered from COVID-19. proved in detecting the patients suffered from COVID-19.
Elena Casiraghi, et.al (2020) constructed a computerized Elena Casiraghi, et.al (2020) constructed a computerized
mechanism whose major intend was to extract the most mechanism whose major intend was to extract the most
relevant radiological and laboratory variables for relevant radiological and laboratory variables for
improving the process of predicting the patient risk [16]. improving the process of predicting the patient risk [16].
After that, an adequate ML (machine learning) model was After that, an adequate ML (machine learning) model was
10/15
put forward. A robust and consistent variable was selected put forward. A robust and consistent variable was selected
by integrating Boruta with RF (Random Forest) in a 10-fold by integrating Boruta with RF (Random Forest) in a 10-fold
cross-validation system for achieving a variable cross-validation system for achieving a variable
importance estimate that was impartial when the importance estimate that was impartial when the
surrogates were available. The selection of significant surrogates were available. The selection of significant
variables was done for training the RF algorithm whose variables was done for training the RF algorithm whose
robustness and efficacy was presented. The robustness and efficacy was presented. The
implementation of the developed system was easy and implementation of the developed system was easy and
assisted the emergency departments in predicting the assisted the emergency departments in predicting the
rapid and accurate risk in COVID-19 patients. Anwar rapid and accurate risk in COVID-19 patients. Anwar
Jarndal, et.al (2020) discussed that a model was Jarndal, et.al (2020) discussed that a model was
formulated for estimating the number of deaths in the formulated for estimating the number of deaths in the
affected cases on the basis of documented number of affected cases on the basis of documented number of
older diabetic and smoking cases [17]. The GPR algorithm older diabetic and smoking cases [17]. The GPR algorithm
was adopted for building a model. A comparative analysis was adopted for building a model. A comparative analysis
was done on this algorithm against ANN (Artificial Neural was done on this algorithm against ANN (Artificial Neural
Network) model. A trustworthy data gathered from WHO Network) model. A trustworthy data gathered from WHO
for distinct countries was utilized in this algorithm. The for distinct countries was utilized in this algorithm. The
adopted algorithm was efficient to estimate the number adopted algorithm was efficient to estimate the number
of deaths because of any arbitrary number of inputs. The of deaths because of any arbitrary number of inputs. The
effective measures were planned for mitigating the effective measures were planned for mitigating the
number of deaths using this algorithm. Huan Zhao, et.al number of deaths using this algorithm. Huan Zhao, et.al
(2021) designed the BP (back propagation neural network) (2021) designed the BP (back propagation neural network)
algorithm for analyzing and predicting the growth rate of algorithm for analyzing and predicting the growth rate of
Coronavirus Disease 2019 [18]. Initially, the gathering of Coronavirus Disease 2019 [18]. Initially, the gathering of
epidemic data of Italy was done from WHO. Subsequently, epidemic data of Italy was done from WHO. Subsequently,
R and MATALB were employed to analyze a number of R and MATALB were employed to analyze a number of
indicators. A computer language was executed to put indicators. A computer language was executed to put
together the results obtained from multiple indicators for together the results obtained from multiple indicators for
predicting the COVID-19 in Italy. The predictive outcomes predicting the COVID-19 in Italy. The predictive outcomes
indicated that the trend of Italy epidemic situation in the indicated that the trend of Italy epidemic situation in the
study period was found good. But, the measures were study period was found good. But, the measures were
required to prevent and control the corona disease. required to prevent and control the corona disease.
Eventually, a comparison of fitting value of the designed Eventually, a comparison of fitting value of the designed
algorithm was done with real value and the fitting degree algorithm was done with real value and the fitting degree
was obtained upto0.99. Nanning Zheng, et.al (2020) was obtained upto 0.99. Nanning Zheng, et.al (2020)
suggested a hybrid AI (artificial-intelligence) technique in suggested a hybrid AI (artificial-intelligence) technique in
order to forecast the Coronavirus Disease 2019 [19]. At order to forecast the Coronavirus Disease 2019 [19]. At
first, an ISI (improved susceptible-infected) model was first, an ISI (improved susceptible- 11 infected) model was
presented for estimating several infection rates so that the presented for estimating several infection rates so that the
transmission laws and the growth trend were analyzed. transmission laws and the growth trend were analyzed.
Later, the effects of measures taken to prevent and Later, the effects of measures taken to prevent and
control the disease were taken in account for embedding control the disease were taken in account for embedding
the NLP (natural language processing) and LSTM network the NLP (natural language processing) and LSTM network
into the presented model with the objective of developing into the presented model with the objective of developing
the suggested hybrid technique to predict coronavirus the suggested hybrid technique to predict coronavirus
disease. The outcomes obtained in experimentation disease. The outcomes obtained in experimentation
depicted that the suggested technique was capable of depicted that the suggested technique was capable of
mitigating the errors of the predictive results. Safa Bahri, mitigating the errors of the predictive results. Safa Bahri,
et.al (2020) intended a LSTM (Long Short-Term Memory)- et.al (2020) intended a LSTM (Long Short-Term Memory)-
based predictive framework to predict the future based predictive framework to predict the future
cumulative number of recovered cases of CoronaVirus cumulative number of recovered cases of CoronaVirus
Disease 2019 seven days earlier, in India, USA and Italy Disease 2019 seven days earlier, in India, USA and Italy
[20]. This framework was suitable for the dynamic of [20]. This framework was suitable for the dynamic of
evolution of recovered cases of the observed Coronavirus evolution of recovered cases of the observed Coronavirus
11/15
Disease 2019 and predicting the disease accurately with a Disease 2019 and predicting the disease accurately with a
total error within 1.46 to 2.65%. The experimental results total error within 1.46 to 2.65%. The experimental results
revealed the effectiveness of intended framework which revealed the effectiveness of intended framework which
provided a minimal error below 3%. This model was provided a minimal error below 3%. This model was
proved as an effective tool for practitioners for monitoring proved as an effective tool for practitioners for monitoring
the occurrence of corona disease all over theworld. Teddy the occurrence of corona disease all over the Teddy
Mantoro, et.al (2020) introduced two techniques namely Mantoro, et.al (2020) introduced two techniques namely
SVR and SIR system for predicting COVID-19 [21]. The SVR and SIR system for predicting COVID-19 [21]. The
disease was predicted in the best-case as well as worst- disease was predicted in the best-case as well as worst-
case scenario. The initial scenario included the current case scenario. The initial scenario included the current
daily case as a maximum case and another country's daily case as a maximum case and another country's
maximum case were contained in the latter scenario maximum case were contained in the latter scenario
utilized. The SVR algorithm provided diverse end of utilized. The SVR algorithm provided diverse end of
epidemic. The SIR model was utilized to validate the epidemic. The SIR model was utilized to validate the
similar end of epidemic on January 2021 in both cases. similar end of epidemic on January 2021 in both cases.
However, a rapid increase of infectious people out of However, a rapid increase of infectious people out of
450,000 people was found in initial scenario to 5,500,000 450,000 people was found in initial scenario to 5,500,000
people in latter scenario. The prediction obtained using people in latter scenario. The prediction obtained using
the introduced approach offered an insight for the policy the introduced approach offered an insight for the policy
maker to deal with the coronavirus diseasepandemic. B. maker to deal with the coronavirus disease B.
Prabha, et.al (2021) constructed a novel ML (machine Prabha, et.al (2021) constructed a novel ML (machine
learning) method known as HDDP (Hybrid Disease learning) method known as HDDP (Hybrid Disease
Detection Principle) which was developed using two Detection Principle) which was developed using two
traditional ML model namely CNN (Convolutional Neural traditional ML model namely CNN (Convolutional Neural
Network) and the AdaBoost Classifier [22]. The ML factors Network) and the AdaBoost Classifier [22]. The ML factors
were considered for the CT image of lung that was pre- were considered for the CT image of lung that was pre-
processed, its attributes were extracted and the processed, its attributes were extracted and the
classification was done. The constructed method had classification was done. The constructed method had
potential for handling the CT images on the basis of these potential for handling the CT images on the basis of these
effective image processing strategies. Hence, the COVID- effective image processing strategies. Hence, the COVID-
19 disease was predicted and a report was sent to the 19 disease was predicted and a report was sent to the
respective user with properaccuracy ratio. This approach respective user with proper 12 accuracy ratio. This
focused on predicting the disease effectively and assisted approach focused on predicting the disease effectively
the doctors and patients successfully. Wencheng Zou, and assisted the doctors and patients successfully.
et.al (2020) devised the ML (machine learning) techniques Wencheng Zou, et.al (2020) devised the ML (machine
namely LR (linear regression) and NN (neuron networks) learning) techniques namely LR (linear regression) and NN
to predict the number of positive cases of Coronavirus (neuron networks) to predict the number of positive cases
Disease 2019 [23]. The LR algorithm was deployed for of Coronavirus Disease 2019 [23]. The LR algorithm was
gathering the data of a state so that the corona disease deployed for gathering the data of a state so that the
was predicted. The number of infections was predicted corona disease was predicted. The number of infections
using the data taken from Georgia and Massachusetts. was predicted using the data taken from Georgia and
The dataset was split into3 consecutive time periods and Massachusetts. The dataset was split into3 consecutive
diverse models were trained to be suitable for each time periods and diverse models were trained to be
corresponding data for comparing MSE (mean square suitable for each corresponding data for comparing MSE
error) values. Moreover, the general viewpoint despite of (mean square error) values. Moreover, the general
three time periods depicted the superiority of single viewpoint despite of three time periods depicted the
variable linear regression (LR) over accurately in superiority of single variable linear regression (LR) over
comparison with the traditional technique with regard accurately in comparison with the traditional technique
with regard
12/15
References [1] Yifan Yang, Wenwu Yu, Duxin Chen, References [1] Yifan Yang, Wenwu Yu, Duxin Chen,
“Prediction of COVID-19 spread via LSTM and the “Prediction of COVID-19 spread via LSTM and the
deterministic SEIR model”, 2020, 39th Chinese Control deterministic SEIR model”, 2020, 39th Chinese Control
Conference(CCC) [2] Zainab Abbas AbdulhusseinAlwaeli, Conference (CCC) [2] Zainab Abbas AbdulhusseinAlwaeli,
Abdullahi Abdu Ibrahim, “Predicting Covid-19 Trajectory Abdullahi Abdu Ibrahim, “Predicting Covid-19 Trajectory
Using Machine Learning”, 2020, 4th International Using Machine Learning”, 2020, 4th International
Symposium on Multidisciplinary Studies and Innovative Symposium on Multidisciplinary Studies and Innovative
Technologies(ISMSIT) [3] Suraj Bodapati, Harika Technologies (ISMSIT) [3] Suraj Bodapati, Harika
Bandarupally, M Trupthi, “COVID-19 Time Series Bandarupally, M Trupthi, “COVID-19 Time Series
Forecasting of Daily Cases, Deaths Caused and Recovered Forecasting of Daily Cases, Deaths Caused and Recovered
Cases using Long Short Term Memory Networks”, 2020, Cases using Long Short Term Memory Networks”, 2020,
IEEE 5th International Conference on Computing IEEE 5th International Conference on Computing
Communication and Automation(ICCCA) [4] Hanqing Communication and Automation (ICCCA) [4] Hanqing
Chao, Xi Fang, Pingkun Yan, “Integrative analysis for Chao, Xi Fang, Pingkun Yan, “Integrative analysis for
COVID-19 patient outcome prediction”, 2020, Medical COVID-19 patient outcome prediction”, 2020, Medical
ImageAnalysis [5] Danish Rafiq, Suhail Ahmad Suhail, Image Danish Rafiq, Suhail Ahmad Suhail,
Mohammad Abid Bazaz, “Evaluation and prediction of Mohammad Abid Bazaz, “Evaluation and prediction of
COVID-19 in India: A case study of worst hit states”, 2020, COVID-19 in India: A case study of worst hit states”, 2020,
Chaos, Solitons &Fractals [6] Farah Shahid, AneelaZameer, Chaos, Solitons & Fractals [6] Farah Shahid, AneelaZameer,
Muhammad Muneeb, “Predictions for COVID-19 with Muhammad Muneeb, “Predictions for COVID-19 with
deep learning models of LSTM, GRU and Bi-LSTM”, 2020, deep learning models of LSTM, GRU and Bi-LSTM”, 2020,
Chaos, Solitons &Fractals [7] Anwar Jarndal, Saddam Chaos, Solitons & Fractals [7] Anwar Jarndal, Saddam
Husain, Omar Zaatar, Talal Al Gumaei, Amar Hamadeh, Husain, Omar Zaatar, Talal Al Gumaei, Amar Hamadeh,
“GPR and ANN based Prediction Models for COVID-19 “GPR and ANN based Prediction Models for COVID-19
Death Cases”, 2020, International Conference on Death Cases”, 2020, International Conference on
Communications, Computing, Cybersecurity, and Communications, Computing, Cybersecurity, and
Informatics(CCCI) [8] Pratima Kumari, Durga Toshniwal, Informatics (CCCI) [8] Pratima Kumari, Durga Toshniwal,
“Real-time estimation of COVID-19 cases using machine “Real-time estimation of COVID-19 cases using machine
learning and mathematical models - The case of India”, learning and mathematical models - The case of India”,
2020, IEEE 15th International Conference on Industrial 2020, IEEE 15th International Conference on Industrial
and Information Systems(ICIIS) [9] Yasin Khan, Pritam and Information Systems (ICIIS) [9] Yasin Khan, Pritam
Khan, Sudhir Kumar, Jawar Singh, Rajesh M. Hegde, Khan, Sudhir Kumar, Jawar Singh, Rajesh M. Hegde,
“Detection and Spread Prediction of COVID-19 from “Detection and Spread Prediction of COVID-19 from
Chest X-ray Images using Convolutional Neural Network- Chest X-ray Images using Convolutional Neural Network-
Gaussian Mixture Model”, 2020, IEEE 17th India Council Gaussian Mixture Model”, 2020, IEEE 17th India Council
International Conference (INDICON) [10] Choujun Zhan, International Conference (INDICON) 19 [10] Choujun
Yufan Zheng, Haijun Zhang, Quansi Wen, “Random- Zhan, Yufan Zheng, Haijun Zhang, Quansi Wen, “Random-
Forest-Bagging Broad Learning System with Applications Forest-Bagging Broad Learning System with Applications
for COVID-19 Pandemic”, 2021, IEEE Internet of Things for COVID-19 Pandemic”, 2021, IEEE Internet of Things
Journal [11] Yifan Yang, Wenwu Yu, Duxin Chen, Journal [11] Yifan Yang, Wenwu Yu, Duxin Chen,
“Prediction of COVID-19 spread via LSTM and the “Prediction of COVID-19 spread via LSTM and the
deterministic SEIR model”, 2020, 39th Chinese Control deterministic SEIR model”, 2020, 39th Chinese Control
Conference(CCC) [12] Leonardo Sestrem de Oliveira, Conference (CCC) [12] Leonardo Sestrem de Oliveira,
Sarah Beatriz Gruetzmacher, João Paulo Teixeira, “COVID- Sarah Beatriz Gruetzmacher, João Paulo Teixeira, “COVID-
19 Time Series Prediction”, 2021, Procedia 19 Time Series Prediction”, 2021, Procedia
ComputerScience [13] Durga Prasad Kavadi, Rizwan Patan, Computer Durga Prasad Kavadi, Rizwan Patan,
Amir H. Gandomi, “Partial derivative Nonlinear Global Amir H. Gandomi, “Partial derivative Nonlinear Global
Pandemic Machine Learning prediction of COVID 19”, Pandemic Machine Learning prediction of COVID 19”,
2020, Chaos, Solitons & Fractals [14] Ankan Ghosh 2020, Chaos, Solitons & Fractals [14] Ankan Ghosh
Dastider, Farhan Sadik, Shaikh Anowarul Fattah, “An Dastider, Farhan Sadik, Shaikh Anowarul Fattah, “An
integrated autoencoder-based hybrid CNN-LSTM model integrated autoencoder-based hybrid CNN-LSTM model
for COVID-19 severity prediction from lung ultrasound”, for COVID-19 severity prediction from lung ultrasound”,
13/15
2021, Computers in Biology andMedicine [15] Abhishek 2021, Computers in Biology and Abhishek Dixit,
Dixit, Ashish Mani, Rohit Bansal, “CoV2-Detect-Net: Ashish Mani, Rohit Bansal, “CoV2-Detect-Net: Design of
Design of COVID-19 COVID-19
Information Sciences [16] Elena Casiraghi, Dario Information Sciences [16] Elena Casiraghi, Dario
Malchiodi, Gabriella Trucco, Marco Frasca, Luca Malchiodi, Gabriella Trucco, Marco Frasca, Luca
Cappelletti, Tommaso Fontana,“ Cappelletti, Tommaso Fontana,
Explainable Machine Learning for Early Assessment of Explainable Machine Learning for Early Assessment of
COVID-19 Risk Prediction in Emergency Departments”, COVID-19 Risk Prediction in Emergency Departments”,
2020, IEEEAccess [17] Anwar Jarndal, Saddam Husain, 2020, IEEE Anwar Jarndal, Saddam Husain, Omar
Omar Zaatar, Talal Al Gumaei, Amar Hamadeh, “GPR and Zaatar, Talal Al Gumaei, Amar Hamadeh, “GPR and ANN
ANN based Prediction Models for COVID-19 Death based Prediction Models for COVID-19 Death Cases”,
Cases”, 2020, International Conference on 2020, International Conference on Communications,
Communications, Computing, Cybersecurity, and Computing, Cybersecurity, and Informatics (CCCI) [18]
Informatics(CCCI) [18] Huan Zhao, Yichi Li, Shenglan Chu, Huan Zhao, Yichi Li, Shenglan Chu, Suling Zhao, Chenglin
Suling Zhao, Chenglin Liu, “A COVID-19 Prediction Liu, “A COVID-19 Prediction Optimization Algorithm
Optimization Algorithm Based on Real-time Neural Based on Real-time Neural Network Training—Taking Italy
Network Training—Taking Italy as an Example”, 2021, IEEE as an Example”, 2021, IEEE Asia-Pacific Conference on
Asia-Pacific Conference on Image Processing, Electronics Image Processing, Electronics and Computers (IPEC) 20
and Computers(IPEC) [19] Nanning Zheng, Shaoyi Du, [19] Nanning Zheng, Shaoyi Du, Jianji Wang, He Zhang,
Jianji Wang, He Zhang, Wenting Cui, Zijian Kang, Tao Wenting Cui, Zijian Kang, Tao Yang, Bin Lou, Yuting Chi,
Yang, Bin Lou, Yuting Chi, Hong Long, Mei Ma, Qi Yuan, Hong Long, Mei Ma, Qi Yuan, Shupei Zhang, Dong Zhang,
Shupei Zhang, Dong Zhang,
14/15
IEEE Transactions on Cybernetics [20] Safa Bahri, IEEE Transactions on Cybernetics [20] Safa Bahri,
MoetezKdayem, NesrineZoghlami, “Deep Learning for MoetezKdayem, NesrineZoghlami, “Deep Learning for
COVID-19 prediction”, 2020 4th International Conference COVID-19 prediction”, 2020 4th International Conference
on Advanced Systems and Emergent on Advanced Systems and Emergent Technologies
Technologies(IC_ASET) [21] Teddy Mantoro, (IC_ASET) [21] Teddy Mantoro, RahmadyaTriasHandayanto,
RahmadyaTriasHandayanto, Media Anugerah Ayu, Jelita Media Anugerah Ayu, Jelita Asian, “Prediction of COVID-
Asian, “Prediction of COVID-19 Spreading Using Support 19 Spreading Using Support Vector Regression and
Vector Regression and Susceptible Infectious Recovered Susceptible Infectious Recovered Model”, 2020, 6th
Model”, 2020, 6th International Conference on International Conference on Computing Engineering and
Computing Engineering and Design(ICCED) [22] B. Design (ICCED) [22] B. Prabha, Sandeep Kaur,
Prabha, Sandeep Kaur, HarikumarPallathadka, “Intelligent HarikumarPallathadka, “Intelligent predictions of Covid
predictions of Covid
Proceedings [23] Wencheng Zou, “The COVID-19 Proceedings [23] Wencheng Zou, “The COVID-19
Pandemic Prediction in the US Based on Machine Pandemic Prediction in the US Based on Machine
Learning”, 2020, International Conference on Public Learning”, 2020, International Conference on Public
Health and Data Science(ICPHDS) Health and Data Science (ICPHDS) [24]
15/15

Final Plag Report

Uploaded by

Copyright:

Available Formats

Final Plag Report

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Final Plag Report

Uploaded by

Copyright:

Available Formats

Document Information

Analyzed document 201_Deepti.docx (D132129220)

Submitted by Anil Gupta

Submitter email [email protected]

Analysis address [email protected]

Sources included in the report

Deepti Malhotra1, Gurinder Kaur Sodhi2

Corresponding Author EmailID: [email protected]

Keywords: covid19, prediction techniques, artificial intelligence, machine learning

100% MATCHING BLOCK 1/14 https://shodhgangotri.inflibnet.ac.in/jspui/bi ...

100% MATCHING BLOCK 14/14 https://shodhgangotri.inflibnet.ac.in/jspui/bi ...

97% MATCHING BLOCK 2/14 https://shodhgangotri.inflibnet.ac.in/jspui/bi ...

55% MATCHING BLOCK 4/14 https://shodhgangotri.inflibnet.ac.in/jspui/bi ...

89% MATCHING BLOCK 5/14 https://shodhgangotri.inflibnet.ac.in/jspui/bi ...

Logistic Growth Model Long short term memory

98% MATCHING BLOCK 6/14 https://shodhgangotri.inflibnet.ac.in/jspui/bi ...

98% MATCHING BLOCK 7/14 https://shodhgangotri.inflibnet.ac.in/jspui/bi ...

98% MATCHING BLOCK 9/14 https://shodhgangotri.inflibnet.ac.in/jspui/bi ...

98% MATCHING BLOCK 11/14 https://shodhgangotri.inflibnet.ac.in/jspui/bi ...

100% MATCHING BLOCK 12/14 https://shodhgangotri.inflibnet.ac.in/jspui/bi ...

100% MATCHING BLOCK 13/14 https://shodhgangotri.inflibnet.ac.in/jspui/bi ...

1/14 SUBMITTED TEXT 12 WORDS 100% MATCHING TEXT 12 WORDS

14/14 SUBMITTED TEXT 84 WORDS 100% MATCHING TEXT 84 WORDS

2/14 SUBMITTED TEXT 80 WORDS 97% MATCHING TEXT 80 WORDS

4/14 SUBMITTED TEXT 82 WORDS 55% MATCHING TEXT 82 WORDS

5/14 SUBMITTED TEXT 12 WORDS 89% MATCHING TEXT 12 WORDS

10/14 SUBMITTED TEXT 15 WORDS 100% MATCHING TEXT 15 WORDS

13/14 SUBMITTED TEXT 24 WORDS 100% MATCHING TEXT 24 WORDS

You might also like