Understanding Dilated Mathematical Relationship between Image Features and the Convolutional Neural Network’s Learnt Parameters

Deep learning, in general, was built on input data transformation and presentation, model training with parameter tuning, and recognition of new observations using the trained model. However, this came with a high computation cost due to the extensive input database and the length of time required in training. Despite the model learning its parameters from the transformed input data, no direct research has been conducted to investigate the mathematical relationship between the transformed information (i.e., features, excitation) and the model’s learnt parameters (i.e., weights). This research aims to explore a mathematical relationship between the input excitations and the weights of a trained convolutional neural network. The objective is to investigate three aspects of this assumed feature-weight relationship: (1) the mathematical relationship between the training input images’ features and the model’s learnt parameters, (2) the mathematical relationship between the images’ features of a separate test dataset and a trained model’s learnt parameters, and (3) the mathematical relationship between the difference of training and testing images’ features and the model’s learnt parameters with a separate test dataset. The paper empirically demonstrated the existence of this mathematical relationship between the test image features and the model’s learnt weights by the ANOVA analysis.

A systems genomics approach uncovers molecular associates of RSV severity

Respiratory syncytial virus (RSV) infection results in millions of hospitalizations and thousands of deaths each year. Variations in the adaptive and innate immune response appear to be associated with RSV severity. To investigate the host response to RSV infection in infants, we performed a systems-level study of RSV pathophysiology, incorporating high-throughput measurements of the peripheral innate and adaptive immune systems and the airway epithelium and microbiota. We implemented a novel multi-omic data integration method based on multilayered principal component analysis, penalized regression, and feature weight back-propagation, which enabled us to identify cellular pathways associated with RSV severity. In both airway and immune cells, we found an association between RSV severity and activation of pathways controlling Th17 and acute phase response signaling, as well as inhibition of B cell receptor signaling. Dysregulation of both the humoral and mucosal response to RSV may play a critical role in determining illness severity.

Multifeature Metric Learning Based on Enhanced Equidistance Embedding for Electroencephalogram Recognition of Epilepsy

Mobile edge computing (MEC) has the ability of pattern recognition and intelligent processing of real-time data. Electroencephalogram (EEG) is a very important tool in the study of epilepsy. It provides rich information that can not be provided by other physiological methods. In the automatic classification of EEG signals by intelligent algorithms, feature extraction and the establishment of classifiers are both very important steps. Different feature extraction methods, such as time domain, frequency domain, and nonlinear dynamic feature methods, contain independent and diverse specific information. Using multiple forms of features at the same time can improve the accuracy of epilepsy recognition. In this paper, we apply metric learning to epileptic EEG signal recognition. Inspired by the equidistance constrained metric learning algorithm, we propose multifeature metric learning based on enhanced equidistance embedding (MMLE3) for EEG recognition of epilepsy. The MMLE3 algorithm makes use of various forms of EEG features, and the feature weights are adaptively weighted. It is a big advantage that the feature weight vector can be adjusted adaptively, without manual adjustment. The MMLE3 algorithm maximizes the distance between the samples constrained by the cannot-link, and the samples of different classes are transformed into equidistant; meanwhile, MMLE3 minimizes the distance between the data constrained by the must-link, and the samples of the same class are compressed to one point. Under the premise that the various feature classification tasks are consistent, MMLE3 can fully extract the associated and complementary information hidden between the features. The experimental results on the CHB-MIT dataset verify that the MMLE3 algorithm has good generalization performance.

A novel weighted fuzzy c-means based on feature weight learning

In feature weighted fuzzy c-means algorithms, there exist two challenges when the feature weighting techniques are used to improve their performances. On one hand, if the values of feature weights are learnt in advance, and then fixed in the process of clustering, the learnt weights might be lack of flexibility and might not fully reflect their relevance. On the other hand, if the feature weights are adaptively adjusted during the clustering process, the algorithms maybe suffer from bad initialization and lead to incorrect feature weight assignment, thus the performance of the algorithms may degrade the in some conditions. In order to ease these problems, a novel weighted fuzzy c-means based on feature weight learning (FWL-FWCM) is proposed. It is a hybrid of fuzzy weighted c-means (FWCM) algorithm with Improved FWCM (IFWCM) algorithm. FWL-FWCM algorithm first learns feature weights as priori knowledge from the data in advance by minimizing the feature evaluation function using the gradient descent technique, then iteratively optimizes the clustering objective function which integrates the within weighted cluster dispersion with a term of the discrepancy between the weights and the priori knowledge. Experiments conducted on an artificial dataset and real datasets demonstrate the proposed approach outperforms the-state-of-the-art feature weight clustering methods. The convergence property of FWL-FWCM is also presented.

A Predictive Model of Dimensional Deviation Based on Regeneration PSO-SVR with Cutting Feature Weight in Milling

Support vector regression (SVR) optimized by particle swarm optimization (PSO) has low predictive accuracy and premature convergence in milling. To solve this problem, A PSO-SVR model combined with the cutting feature weight was proposed in this paper. Firstly, basing on the SVR, the feature weight was integrated with the kernel function, and added the premature judging to the PSO to improve the global searching ability. Secondly, the mathematical model composed of the cutting force, temperature and cutting vibration was built based on the datasets obtained by experiment. The covariance was calculated to get the characteristic weights of process parameters, which promoted the incremental data in turn. Finally, the predictive model of the dimensional deviation was established based on the promoted PSO-SVR and the result was compared with the general PSO-SVR. The accuracy of the predictive model reached 97.5%. And compared with the predictive model of the general PSO-SVR without feature weighting, the dimensional deviation predictive accuracy and generalization ability of the regeneration PSO-SVR predictive model with feature weighting was improved by 37.75% and 24.5%.

Sentiment Analysis of Healthcare Reviews Using Context-Based Feature Weight Embedding Technique

Healthcare reviews play a major role in providing feedback to consumers as well as medical care information to users. Historically, the sentiment analysis of clinical documents will help patients in analyzing the medicines and identifying the relevant medicines. Existing methods of word embeddings use only the context of words; hence, they ignore the sentiment of texts. Medical review analysis is important due to several reasons. Patients will know the results of using medicines since such information is not easily obtained from any other source. Historical results of predictive analysis say that among people aged 55-80, the death rate from 2005 to 2015 in the US was at the top for the deadliest disease, which increased exponentially. Traditional machine learning techniques use a lexical approach for feature extraction. In this paper, baseline algorithms are checked with the proposed work of the recurrent network, and results show that the method outperforms baseline methods by a significant improvement in terms of precision, recall, f-score, and accuracy.

A Multiattention-Based Supervised Feature Selection Method for Multivariate Time Series

Feature selection is a known technique to preprocess the data before performing any data mining task. In multivariate time series (MTS) prediction, feature selection needs to find both the most related variables and their corresponding delays. Both aspects, to a certain extent, represent essential characteristics of system dynamics. However, the variable and delay selection for MTS is a challenging task when the system is nonlinear and noisy. In this paper, a multiattention-based supervised feature selection method is proposed. It translates the feature weight generation problem into a bidirectional attention generation problem with two parallel placed attention modules. The input 2D data are sliced into 1D data from two orthogonal directions, and each attention module generates attention weights from their respective dimensions. To facilitate the feature selection from the global perspective, we proposed a global weight generation method that calculates a dot product operation on the weight values of the two dimensions. To avoid the disturbance of attention weights due to noise and duplicated features, the final feature weight matrix is calculated based on the statistics of the entire training set. Experimental results show that this proposed method achieves the best performance on compared synthesized, small, medium, and practical industrial datasets, compared to several state-of-the-art baseline feature selection methods.

Multi-label feature selection based on logistic regression and manifold learning

Like traditional single label learning, multi-label learning is also faced with the problem of dimensional disaster.Feature selection is an effective technique for dimensionality reduction and learning efficiency improvement of high-dimensional data. In this paper, Logistic regression, manifold learning and sparse regularization were combined to construct a joint framework for multi-label feature selection (LMFS). Firstly, the sparsity of the eigenweight matrix is constrained by the $L_{2,1}$-norm. Secondly, the feature manifold and label manifold can constrain the feature weight matrix to make it fit the data information and label information better. An iterative updating algorithm is designed and the convergence of the algorithm is proved.Finally, the LMFS algorithm is compared with DRMFS, SCLS and other algorithms on eight classical multi-label data sets. The experimental results show the effectiveness of LMFS algorithm.