COVID-19 cases prediction in multiple areas via shapelet learning

Wang, Zhijin; Cai, Bing

doi:10.1007/s10489-021-02391-6

Cited by 16 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The methods for obtaining shapelets can be divided into two categories: shapelet discovery and shapelet learning. The basic step of shapelet discovery is to consider all sequences from training time series data which are evaluated with a scoring function to estimate how well they predict given class labels [14,28]. The shapelet discovery is inefficient because time series usually have many candidate segments.…”

Section: The Prior Information and Shapelet Learningmentioning

confidence: 99%

“…And the extracted shapelet can well assist in capturing the important dynamic features of time series. In order to realize the effectiveness and prediction of the covid-19 case, a method on how to recognize and effectively use the shape information provided [14,15]. The key data points of the observed time series were described by the shapelets, then a regression task was built to predict the upcoming COVID-19 cases.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A combination prediction model based on ELM and prior information of shapelet for algal bloom

Liu

Zhang

Wang

et al. 2022

Preprint

View full text Add to dashboard Cite

Prediction is crucial to prevent the outbreaks of algal bloom. However, due to the time-varying and nonlinear characteristics of algal bloom, which brings challenges for accurate prediction. Aiming at solving the problem, we consider both prior information and data-driven, and propose an error compensation combination prediction model based on the manifold regularized extreme learning machine fused shapelet (ECM-MRELM-FS). First, the prior information is extracted as the upward shapelet set to discover and predict the value of algal bloom at the typical evolution mode. Meanwhile, a data-driven model of manifold regularized extreme learning machine (MRELM) is used to predict the value of algal bloom at other atypical evolution modes. This switching combination builds a prediction model by MRELM fused the upward shapelet, which can effectively utilize the advantages of prior information and data-driven. Then, an optimized T-S fuzzy inference systems as error compensation model (ECM) based on improved fuzzy c-means clustering is proposed to promote the accuracy of the prediction further. The experiments are performed by two real water quality datasets from USA and China. The final results demonstrate that the proposed combination prediction model can effectively predict algal bloom and provide a feasible and scientific early warning support for the water quality management in lake and reservoir.

show abstract

Section: The Prior Information and Shapelet Learningmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A combination prediction model based on ELM and prior information of shapelet for algal bloom

Liu

Zhang

Wang

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The outstanding performance of DL in industry and academia is widely recognized. The medical tasks are being automated using DL methods [20,35], including cancer detection, tumour classification, vessel segmentation, etc. The high accuracy is impressive; however, models often act as black-boxes with unexplainable complicated layers.…”

Section: Introductionmentioning

confidence: 99%

Mitigating adversarial evasion attacks by deep active learning for medical image classification

Ahmed

Lin

Srivastava

2022

Multimed Tools Appl

View full text Add to dashboard Cite

In the Internet of Medical Things (IoMT), collaboration among institutes can help complex medical and clinical analysis of disease. Deep neural networks (DNN) require training models on large, diverse patients to achieve expert clinician-level performance. Clinical studies do not contain diverse patient populations for analysis due to limited availability and scale. DNN models trained on limited datasets are thereby constraining their clinical performance upon deployment at a new hospital. Therefore, there is significant value in increasing the availability of diverse training data. This research proposes institutional data collaboration alongside an adversarial evasion method to keep the data secure. The model uses a federated learning approach to share model weights and gradients. The local model first studies the unlabeled samples classifying them as adversarial or normal. The method then uses a centroid-based clustering technique to cluster the sample images. After that, the model predicts the output of the selected images, and active learning methods are implemented to choose the sub-sample of the human annotation task. The expert within the domain takes the input and confidence score and validates the samples for the model’s training. The model re-trains on the new samples and sends the updated weights across the network for collaboration purposes. We use the InceptionV3 and VGG16 model under fabricated inputs for simulating Fast Gradient Signed Method (FGSM) attacks. The model was able to evade attacks and achieve a high accuracy rating of 95%.

show abstract