Health Analytics on COVID-19 Data with Few-Shot Learning

Leung, Carson K.; Fung, Daryl L. X.; Hoi, Calvin S. H.

doi:10.1007/978-3-030-86534-4_6

Cited by 10 publications

(2 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[12,43]. FSL scenarios frequently arise in domains where sample acquisition proves challenging, such as drug discovery [3], agriculture [44] and healthcare [26], thereby receiving increasing attention. A widely used framework for FSL is meta-learning, which follows the principle that "test and train conditions should match" [42] or "the process of improving a learning algorithm over multiple learning episodes" [17], aiming to learn transferable knowledge for inference.…”

Section: Introductionmentioning

confidence: 99%

TACDFSL: Task Adaptive Cross Domain Few-Shot Learning

2022

View full text Add to dashboard Cite

Cross Domain Few-Shot Learning (CDFSL) has attracted the attention of many scholars since it is closer to reality. The domain shift between the source domain and the target domain is a crucial problem for CDFSL. The essence of domain shift is the marginal distribution difference between two domains which is implicit and unknown. So the empirical marginal distribution measurement is proposed, that is, WDMDS (Wasserstein Distance for Measuring Domain Shift) and MMDMDS (Maximum Mean Discrepancy for Measuring Domain Shift). Besides this, pre-training a feature extractor and fine-tuning a classifier are used in order to have a good generalization in CDFSL. Since the feature obtained by the feature extractor is high-dimensional and left-biased, the adaptive feature distribution transformation is proposed, to make the feature distribution of each sample be approximately Gaussian distribution. This approximate symmetric distribution improves image classification accuracy by 3% on average. In addition, the applicability of different classifiers for CDFSL is investigated, and the classification model should be selected based on the empirical marginal distribution difference between the two domains. The Task Adaptive Cross Domain Few-Shot Learning (TACDFSL) is proposed based on the above ideas. TACDFSL improves image classification accuracy by 3–9%.

show abstract

Section: Introductionmentioning

confidence: 99%

TACDFSL: Task Adaptive Cross Domain Few-Shot Learning

2022

View full text Add to dashboard Cite

show abstract

“…Another data science solution for healthcare system is presented to perform analysis on COVID-19 data. The proposed system predicts and classifies the cases of COVID-19 using real-life COVID-19 data including routine blood test results from Brazilian COVID-19 patients [17].…”

Section: Related Workmentioning

confidence: 99%

Predictive analytics on COVID-19 data using Hive based on Hadoop cluster

Khaleel,

Kareem,

Mahdi

2023

IJEECS

View full text Add to dashboard Cite

COVID-19 pandemic has received a serious attention from academia, industry and governments to stop the huge number of deaths and economic disruptions around the world. Many techniques have been used to control the spread of the pandemic by understanding its characteristics and behavior. However, because of the large amounts and complex characteristics of COVID-19 data, the querying and analysis of such data using conventional tools have become a challenging task. As a result, powerful and distributed tools are highly required for querying and analyzing this data effectively. In this paper, distributed system using Hive based on Hadoop cluster is used to query and analyze COVID-19 data to obtain meaningful information. Hadoop is employed as a scalable and reliable framework to accommodate such large amounts of data. Hive is used as a data warehouse that run on Hadoop cluster to perform querying and predictive analytics on huge COVID-19 datasets. Several experiments are performed to evaluate the performance of proposed system. Experiments show that the proposed system outperforms relational database management system (RDBMS) in terms of query processing time. Experiments also show that the proposed system has a better efficiency in terms of data load, I/O operation, reading and writing data.

show abstract