Fastest-growing source prediction of US electricity production based on a novel hybrid model using wavelet transform

People communicate their views, arguments and emotions about their everyday life on social media (SM) platforms (e.g. Twitter and Facebook). Twitter stands as an international micro-blogging servic...

A gradient boosted decision tree-based sentiment classification of twitter data

To understand daily events accurately, adaptive pose estimation (APE) systems require a robust context-aware model and optimal feature selection methods. In this paper, we propose a novel gait event detection (GED) system that consists of saliency silhouette detection, a robust body parts model and a 2D stick-model followed by a hierarchical optimization algorithm. Furthermore, the most prominent context-aware features such as energy, 0–180° intensity and distinct moveable features are proposed by focusing on invariant and localized characteristics of human postures in different event classes. Finally, we apply Grey Wolf optimization and a genetic algorithm to discriminate complex postures and to provide appropriate labels to each event. In order to evaluate the performance of proposed GED, two public benchmark datasets, UCF101 and YouTube, are examined via the n-fold cross validation method. For the two benchmark datasets, our proposed method detects the human body key points with 82.4% and 83.2% accuracy respectively. Also, it extracts the context-aware features and finally recognizes the gait events with 82.6% and 85.0% accuracy, respectively. Compared with other well-known statistical and state-of-the-art methods, our proposed method outperforms other similarly tasked methods in terms of posture detection and recognition accuracy.

Adaptive Pose Estimation for Gait Event Detection Using Context-Aware Model and Hierarchical Optimization

Routing and hotspot are considered among the key issues in Wireless Sensor Networks (WSNs). Several routing protocols have been proposed; yet they suffer from the lack of fault tolerance, the uneven load balancing and the local optimal solution issues. To fix these issues, we propose in this paper an algorithm named Energy-efficient Cluster-based Routing Protocol using Unequal Clustering and improved Ant Colony Optimization (ACO) techniques (ECRP-UCA). ECRP-UCA divides the network into unequal clusters based on residual energy, distance to the sink, number of neighbor nodes, and a new parameter named number of backward relay nodes in previous round to properly balance the load among Cluster Heads (CHs). We propose also a batch-based clustering method that allows the network to function several rounds without requiring control overhead for its configuration. Additionally, we devise an improved ACO based routing technique for efficient and reliable inter-cluster routing from CHs to the sink. In this algorithm, the heuristic function is formulated considering the energy of next hop sensor node, distance from the current sensor node to the next sensor node and the latter to the destination, and the new parameter Number of Preferred Probable Relay Nodes (NPPRN). By using NPPRN, the ants have the probability to follow other paths, which improves the algorithm’s ability to obtain the optimal global solution and protect the algorithm from converging quickly into the optimal local solution. Moreover, the ant searching direction is improved. The proposed routing protocol is intensively experimented and compared with recent and relevant existing protocols. The simulation results show that the proposed ECRP-UCA outperforms these protocols in terms of various interesting metrics.

An energy-efficient cluster-based routing protocol using unequal clustering and improved ACO techniques for WSNs

Learning features from covariance matrix of gabor wavelet for face recognition under adverse conditions

In this paper, we propose a quadtree based approach to capture the spatial information of medical images for explaining nonlinear SVM prediction. In medical image classification, interpretability becomes important to understand why the adopted model works. Explaining an SVM prediction is difficult due to implicit mapping done in kernel classification is uninformative about the position of data points in the feature space and the nature of the separating hyperplane in the original space. The proposed method finds ROIs which contain the discriminative regions behind the prediction. Localization of the discriminative region in small boxes can help in interpreting the prediction by SVM. Quadtree decomposition is applied recursively before applying SVMs on sub images and model identified ROIs are highlighted. Pictorial results of experiments on various medical image datasets prove the effectiveness of this approach. We validate the correctness of our method by applying occlusion methods.

/pdf/interpreting-svm-for-medical-images-using-quadtree-3yjc66esvg.pdf

Interpreting SVM for medical images using Quadtree.

In this paper, we address the lack of interpretability of Support Vector Machine (SVM) via rules based on support vectors. The lack of intuitive explanation of the rules in the domain of medical whole slide image classification by determining a reduced subset of Scale-Invariant Feature Transform (SIFT) features and linear dimensionality reduction. This reduced subset of SIFT features that participate in decision making provide an intuitive interpretation of the decision rules via a visual presentation of the features in a slide image. In our work, we have taken whole-slide images of sentinel lymph node to predict the presence of cancer using SVM and developed an equivalent Fuzzy model for better interpretability. The SVM model consisted of 50 support vectors which achieved an accuracy of 94.59%. A two IF-THEN rules equivalent Fuzzy Rule Based Model (FRBS) was created using all 50 support vectors which classified each WSI in either {−1, 1} class for non-cancerous and cancerous respectively. FRBS achieved 91.89% accuracy which shows that it is able to represent SVM working accurately in a human interpretable way.

A compact fuzzy rule interpretation of SVM classifier for medical whole slide images

Machine learning algorithms are inherently not interpretable, and this poses a problem in risk-averse applications of machine learning. Mammographic images are widely used tool to predict breast cancer. Various machine learning algorithms like SVM, RBFNN are used to detect the mass in the mammographic images and classify for cancer, but the classification by SVMs are not intuitive. Our aim is to counter this problem by employing a novel method of using multiple SVMs to elucidate the area affected by cancer. We also color-code the patches for further clarification.

An Interpretable SVM Based Model for Cancer Prediction in Mammograms

In manifold learning, the intrinsic geometry of the manifold is explored and preserved by identifying the optimal local neighborhood around each observation. It is well known that when a Riemannian manifold is unfolded correctly, the observations lying spatially near to the manifold, should remain near on the lower dimension as well. Due to the nonlinear properties of manifold around each observation, finding such optimal neighborhood on the manifold is a challenge. Thus, a sub-optimal neighborhood may lead to erroneous representation and incorrect inferences. In this paper, we propose a rotation-based affinity metric for accurate graph Laplacian approximation. It exploits the property of aligned tangent spaces of observations in an optimal neighborhood to approximate correct affinity between them. Extensive experiments on both synthetic and real world datasets have been performed. It is observed that proposed method outperforms existing nonlinear dimensionality reduction techniques in low-dimensional representation for synthetic datasets. The results on real world datasets like COVID-19 prove that our approach increases the accuracy of classification by enhancing Laplacian regularization.

/pdf/a-rotation-based-regularization-method-for-semi-supervised-1fa9z9z3rr.pdf

A rotation based regularization method for semi-supervised learning.

In this paper, an Ensemble based Graph Convolutional Network (EGCN) is proposed for relational datasets. EGCN encodes the pairwise similarity between data points for semi supervised learning and feature representation. The GCN model with Laplacian filter is not able to completely explore the local geometry of the data distribution as the null space of Laplacian remains constant along the manifold. This leads to poor extrapolating ability of the GCN model. In order to improve this GCN model, Hessian filter is introduced. The Hessian filter is more efficient and has richer null space. It can encode the relationship among the data points better by fusing structural properties with original features. In our EGCN, two parallel GCNs are executed, one with Laplacian filter and other with Hessian filter. In EGCN model Laplacian, filter accumulate the information from a data point to all its neighbors and Hessian filter gathers the information between adjacent data points that lie in the neighborhood of a data point. Extensive experiment conducted on various relational datasets indicates that EGCN out performs both Laplacian and Hessian filter based GCN models by a significant margin on various relational datasets.

Prashant Shukla

Papers

Interpreting SVM for medical images using Quadtree.

A compact fuzzy rule interpretation of SVM classifier for medical whole slide images

An Interpretable SVM Based Model for Cancer Prediction in Mammograms

A rotation based regularization method for semi-supervised learning.

Ensemble Based Graph Convolutional Network for Semi Supervised Learning