“…For graph structure, each node is allowed to have multiple parents simultaneously. [21][22][23] For the tree structure, each node only has one parent node, [24][25][26] TA B L E 1 Summary of notation…”
Section: Hierarchical Tree Structurementioning
confidence: 99%
“…In hierarchical classification tasks, 19 hierarchical structures can be split into direct acyclic graph structure and tree structure, 20 respectively. For graph structure, each node is allowed to have multiple parents simultaneously 21‐23 . For the tree structure, each node only has one parent node, 24‐26 and in this article, we consider the tree structure between categories.…”
Hierarchical classification learning is a hot research topic in machine learning and data mining domains, and many feature selection algorithms with category hierarchy have been proposed. However, existing algorithms assume that the feature space of data is completely obtained in advance, and ignore its uncertainty and dynamicity. To address these problems, we propose an online streaming feature selection framework with a hierarchical structure to solve the above two problems simultaneously. First, we apply the hierarchical relationship between nodes in a hierarchical structure to the Relief algorithm, so that it can be used to compute the weights of dynamic features. Second, we dynamically select important features for each internal node via comparing the magnitude of the weights of features on these nodes with their parent and sibling nodes. In addition, we perform redundancy analysis of features by calculating the covariance between features to obtain a superior online feature subset for each internal node. Finally, the proposed algorithm is compared with six online streaming feature selection methods on six hierarchical data sets, and experimental results shows that the classification performance of the proposed algorithm is effective.
“…For graph structure, each node is allowed to have multiple parents simultaneously. [21][22][23] For the tree structure, each node only has one parent node, [24][25][26] TA B L E 1 Summary of notation…”
Section: Hierarchical Tree Structurementioning
confidence: 99%
“…In hierarchical classification tasks, 19 hierarchical structures can be split into direct acyclic graph structure and tree structure, 20 respectively. For graph structure, each node is allowed to have multiple parents simultaneously 21‐23 . For the tree structure, each node only has one parent node, 24‐26 and in this article, we consider the tree structure between categories.…”
Hierarchical classification learning is a hot research topic in machine learning and data mining domains, and many feature selection algorithms with category hierarchy have been proposed. However, existing algorithms assume that the feature space of data is completely obtained in advance, and ignore its uncertainty and dynamicity. To address these problems, we propose an online streaming feature selection framework with a hierarchical structure to solve the above two problems simultaneously. First, we apply the hierarchical relationship between nodes in a hierarchical structure to the Relief algorithm, so that it can be used to compute the weights of dynamic features. Second, we dynamically select important features for each internal node via comparing the magnitude of the weights of features on these nodes with their parent and sibling nodes. In addition, we perform redundancy analysis of features by calculating the covariance between features to obtain a superior online feature subset for each internal node. Finally, the proposed algorithm is compared with six online streaming feature selection methods on six hierarchical data sets, and experimental results shows that the classification performance of the proposed algorithm is effective.
“…The logistic regression method can balance the accuracy and interpretability of the model and is widely used in the diagnosis and prediction of diseases such as Alzheimer's disease 27 and cancer. 28 Frequent data mining methodologies have been used to modeling complex clinical temporal processes such as onset of adverse events following immunizations, 29 drug-drug interactions, 30 treatment of acute coronary syndrome, 31 patient management of diabetes mellitus 32,33 and other chronic diseases. 34 Finally, Sheetrit 35 have developed temporal probabilistic profiles (TPF) for sepsis onset prediction that model frequently repeating temporal patterns of multivariate ICU time-series.…”
Large collections of electronic clinical data today provide us with a vast source of information on medical practice. However, the utilization of those data for exploratory analysis to support clinical decisions is still limited. It is particularly challenging for extracting useful disease progression patterns from such data because it is longitudinal, incomplete, irregular, and heterogeneous of the patient conditions. In this article, we propose an integrated clinical event prediction model medical concept integrated residual short-long temporal convolutional networks (SL-TCN) to address these challenges. Compared to existing models, our model has three-fold advantages: (1) it learns a compact set of medical concepts as the bridge between the hidden progression process and the observed medical evidence. ( 2) it learns a continuous-time progression model from discrete-time observations with nonequal intervals and high-dimensional features. (3) We fuse the temporal convolutional network, the long short-term memory network, and the residual connector so as to capture the local and global dependency of the sequence and make the clinical event predictions more robust. Through extensive experiments on the MIMIC III dataset, we demonstrate that our SL-TCN achieves higher precision in clinical event prediction and derives some interesting clinical insights.
“…With the rapid development of intelligent algorithms, big data, the internet of things and so forth, [8][9][10][11][12][13][14] intelligent control methods for manipulators have emerged, which to a certain extent improve the adaptive capacity of traditional control methods and the stability of the manipulator during movement. However, as ordinary intelligent algorithms have the disadvantage of having a single control means, they are still not adaptable to changing demands and environments.…”
A manipulator is a complex electromechanical system that is nonlinear, strongly coupled, and uncertain. Achieving its precise and high-quality trajectory control is difficult.Sliding mode control (SMC) is one of the common control methods for manipulators.However, discontinuities in SMC can cause jitter and vibration in the manipulator system, leading to a reduction in the performance of the control system. For the self-adaptive capability jitter vibration problem of SMC, the Dobot magician manipulator is treated as the research object in this article. The dynamics equations of the manipulator are established by Lagrange method, and a simplified model of the manipulator dynamics is constructed. The method of self-adaptive sliding mode control is proposed. Self-adaptive parameters are added to the SMC to achieve self-adaptive adjustment of the SMC parameters. In the MATLAB/Simulink simulation environment analysis show that the self-adaptive SMC method has better self-tuning ability and trajectory tracking ability than the traditional SMC, and weakens the jitter phenomenon existing in the traditional SMC.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.