ABSTRACT
With the onset of COVID-19, OTT platforms have become popular. With this added popularity, many production companies tend to release their content on platforms like Netflix, Amazon Prime, Disney + Hotstar, etc. Through this research work, we tend to check the impact of different classical factors like genre, age certification, time of release, the platform of release, etc. as well as various social factors like the sentiment of the audience around the trailer, songs, and success of the previous season in predicting the success of the pre-release season of an English web series by creating our dataset. This will enhance the business strategies that production houses can use to improve their profits. We have trained different classification models like Decision Tree, Support Vector Machine, Multinomial Naive Bayes, and hyper tuned the parameters of Random Forest and K-Nearest Neighbours. We have also created a Multi-Layer Perceptron model and an ensemble classifier and trained them on our dataset. The best accuracy of 76.66% was achieved by the Hard Voting type ensemble classifier. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
ABSTRACT
With the onset of COVID-19, OTT platforms have become popular. With this added popularity, many production companies tend to release their content on platforms like Netflix, Amazon Prime, Disney + Hotstar, etc. Through this research work, we tend to check the impact of different classical factors like genre, age certification, time of release, the platform of release, etc. as well as various social factors like the sentiment of the audience around the trailer, songs, and success of the previous season in predicting the success of the pre-release season of an English web series by creating our dataset. This will enhance the business strategies that production houses can use to improve their profits. We have trained different classification models like Decision Tree, Support Vector Machine, Multinomial Naive Bayes, and hyper tuned the parameters of Random Forest and K-Nearest Neighbours. We have also created a Multi-Layer Perceptron model and an ensemble classifier and trained them on our dataset. The best accuracy of 76.66% was achieved by the Hard Voting type ensemble classifier. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
ABSTRACT
This paper examines the effect of Covid-19 on currency exchange rate behaviour by taking a sample of 37 countries over a period from 4th January 2020 to 30th April 2021. Three variables, such as daily confirmed cases, daily deaths, and the world pandemic uncertainty index (WPUI), are taken as the measure of Covid-19. By applying fixed-effect regression, the study documents that the exchange rate behaves positively to the Covid-19 outbreak, particularly to daily confirmed cases and daily deaths, which implies that the value of other currencies against the US dollar has been depreciated. However, the impact of WPUI is insignificant. On studying the time-varying impact of the pandemic, the study reveals that the Covid-19 has an asymmetric impact on exchange rate over different time frames. Further, it is observed that though daily confirmed cases and daily deaths show a uniform effect, WPUI puts an asymmetric effect on the exchange rate owing to the nature of economies.
ABSTRACT
As we head toward more than 34 million cases of Coronavirus disease of 2019 and a million deaths worldwide, there is an urgent need to control the spread and recover the patients as soon as possible. Technology has resolved many complex problems of human beings. Machine learning and deep learning have been intensely used in areas like medical image diagnosis, drug discovery, and manufacturing. It has achieved success in complex problems like the detection of cancerous tumors, skin cancer, and diabetic retinopathy. In this study, considering the power and ability of deep learning, we will study the possibilities and role of deep learning in detecting COVID-19. Several deep learning models can be used to monitor, detect, and predict the spread of the virus. The convolutional neural network has great potential to detect COVID-19 as it is currently used to detect pneumonia using chest X-ray images. This knowledge is used to build a model to detect the presence of COVID-19 which gives a high accuracy with minimum error percentage. The model is highly capable of detecting patients with COVID-19. Deep learning can be implemented widely to tackle this pandemic. © 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
ABSTRACT
To evaluate the present situation concerning the epidemic trend associated with the COVID-19 in Indian demography, the dynamics of the case rise has been analyzed from the perspective of the different index. The index for the analysis has been chosen in terms of the Case Recovery Rate (CRR), CASE Fatality Rate (CFR), as well as Mortality rate (MR). The study includes the rise of the case related to the pandemic in the different demographic regions of India as well as deep analysis and calculation of the indexes considered for the study. The analysis of the rising cases has also been investigated to relax the imposed rule so that economy of the country will not get affected adversely. Several preventive and control initiative has been taken by the central and state government in collaboration. The result of this paper can be taken as an input to decide further policy in the fight against the COVID-19.