Search results

The purpose of this paper is to present a preliminary solution to address the problem of estimating human pose and trajectory by an aerial robot with a monocular camera in near real time.

The distinguishing feature of the solution is a dynamic classifier selection architecture. Each video frame is corrected for perspective using projective transformation. Then, a silhouette is extracted as a Histogram of Oriented Gradients (HOG). The HOG is then classified using a dynamic classifier. A class is defined as a pose-viewpoint pair, and a total of 64 classes are defined to represent a forward walking and turning gait sequence. The dynamic classifier consists of a Support Vector Machine (SVM) classifier C64 that recognizes all 64 classes, and 64 SVM classifiers that recognize four classes each – these four classes are chosen based on the temporal relationship between them, dictated by the gait sequence.

The solution provides three main advantages: first, classification is efficient due to dynamic selection (4-class vs 64-class classification). Second, classification errors are confined to neighbors of the true viewpoints. This means a wrongly estimated viewpoint is at most an adjacent viewpoint of the true viewpoint, enabling fast recovery from incorrect estimations. Third, the robust temporal relationship between poses is used to resolve the left-right ambiguities of human silhouettes.

Experiments conducted on both fronto-parallel videos and aerial videos confirm that the solution can achieve accurate pose and trajectory estimation for these different kinds of videos. For example, the “walking on an 8-shaped path” data set (1,652 frames) can achieve the following estimation accuracies: 85 percent for viewpoints and 98.14 percent for poses.

Ensemble models that combine multiple base classifiers have been widely used to improve prediction performance in credit risk evaluation. However, an arbitrary selection of base classifiers is problematic. The purpose of this paper is to develop a framework for selecting base classifiers to improve the overall classification performance of an ensemble model.

In this study, selecting base classifiers is treated as a feature selection problem, where the output from a base classifier can be considered a feature. The proposed correlation-based classifier selection using the maximum information coefficient (MIC-CCS), a correlation-based classifier selection under the maximum information coefficient method, selects the features (classifiers) using nonlinear optimization programming, which seeks to optimize the relationship between the accuracy and diversity of base classifiers, based on MIC.

The empirical results show that ensemble models perform better than stand-alone ones, whereas the ensemble model based on MIC-CCS outperforms the ensemble models with unselected base classifiers and other ensemble models based on traditional forward and backward selection methods. Additionally, the classification performance of the ensemble model in which correlation is measured with MIC is better than that measured with the Pearson correlation coefficient.

The study provides an alternate solution to effectively select base classifiers that are significantly different, so that they can provide complementary information and, as these selected classifiers have good predictive capabilities, the classification performance of the ensemble model is improved.

This paper introduces MIC to the correlation-based selection process to better capture nonlinear and nonfunctional relationships in a complex credit data structure and construct a novel nonlinear programming model for base classifiers selection that has not been used in other studies.

Ensemble methods have been widely used in the field of pattern recognition due to the difficulty of finding a single classifier that performs well on a wide variety of problems. Despite the effectiveness of these techniques, studies have shown that ensemble methods generate a large number of hypotheses and that contain redundant classifiers in most cases. Several works proposed in the state of the art attempt to reduce all hypotheses without affecting performance.

In this work, the authors are proposing a pruning method that takes into consideration the correlation between classifiers/classes and each classifier with the rest of the set. The authors have used the random forest algorithm as trees-based ensemble classifiers and the pruning was made by a technique inspired by the CFS (correlation feature selection) algorithm.

The proposed method CES (correlation-based Ensemble Selection) was evaluated on ten datasets from the UCI machine learning repository, and the performances were compared to six ensemble pruning techniques. The results showed that our proposed pruning method selects a small ensemble in a smaller amount of time while improving classification rates compared to the state-of-the-art methods.

CES is a new ordering-based method that uses the CFS algorithm. CES selects, in a short time, a small sub-ensemble that outperforms results obtained from the whole forest and the other state-of-the-art techniques used in this study.

Influenced by the constantly changing manufacturing environment, no single dispatching rule (SDR) can consistently obtain better scheduling results than other rules for the dynamic job-shop scheduling problem (DJSP). Although the dynamic SDR selection classifier (DSSC) mined by traditional data-mining-based scheduling method has shown some improvement in comparison to an SDR, the enhancement is not significant since the rule selected by DSSC is still an SDR.

This paper presents a novel data-mining-based scheduling method for the DJSP with machine failure aiming at minimizing the makespan. Firstly, a scheduling priority relation model (SPRM) is constructed to determine the appropriate priority relation between two operations based on the production system state and the difference between their priority values calculated using multiple SDRs. Subsequently, a training sample acquisition mechanism based on the optimal scheduling schemes is proposed to acquire training samples for the SPRM. Furthermore, feature selection and machine learning are conducted using the genetic algorithm and extreme learning machine to mine the SPRM.

Results from numerical experiments demonstrate that the SPRM, mined by the proposed method, not only achieves better scheduling results in most manufacturing environments but also maintains a higher level of stability in diverse manufacturing environments than an SDR and the DSSC.

This paper constructs a SPRM and mines it based on data mining technologies to obtain better results than an SDR and the DSSC in various manufacturing environments.

Feature space heterogeneity exists widely in various application fields of classification techniques, such as customs inspection decision, credit scoring and medical diagnosis. This paper aims to study the relationship between feature space heterogeneity and classification performance.

A measurement is first developed for measuring and identifying any significant heterogeneity that exists in the feature space of a data set. The main idea of this measurement is derived from a meta-analysis. For the data set with significant feature space heterogeneity, a classification algorithm based on factor analysis and clustering is proposed to learn the data patterns, which, in turn, are used for data classification.

The proposed approach has two main advantages over the previous methods. The first advantage lies in feature transform using orthogonal factor analysis, which results in new features without redundancy and irrelevance. The second advantage rests on samples partitioning to capture the feature space heterogeneity reflected by differences of factor scores. The validity and effectiveness of the proposed approach is verified on a number of benchmarking data sets.

Measurement should be used to guide the heterogeneity elimination process, which is an interesting topic in future research. In addition, to develop a classification algorithm that enables scalable and incremental learning for large data sets with significant feature space heterogeneity is also an important issue.

Measuring and eliminating the feature space heterogeneity possibly existing in the data are important for accurate classification. This study provides a systematical approach to feature space heterogeneity measurement and elimination for better classification performance, which is favorable for applications of classification techniques in real-word problems.

A measurement based on meta-analysis for measuring and identifying any significant feature space heterogeneity in a classification problem is developed, and an ensemble classification framework is proposed to deal with the feature space heterogeneity and improve the classification accuracy.

The aim of continuous learning is to obtain and fine-tune information gradually without removing the already existing information. Many conventional approaches in streaming data classification assume that all arrived new data is completely labeled. To regularize Neural Networks (NNs) by merging side information like user-provided labels or pair-wise constraints, incremental semi-supervised learning models need to be introduced. However, they are hard to implement, specifically in non-stationary environments because of the efficiency and sensitivity of such algorithms to parameters. The periodic update and maintenance of the decision method is the significant challenge in incremental algorithms whenever the new data arrives.

Hence, this paper plans to develop the meta-learning model for handling continuous or streaming data. Initially, the data pertain to continuous behavior is gathered from diverse benchmark source. Further, the classification of the data is performed by the Recurrent Neural Network (RNN), in which testing weight is adjusted or optimized by the new meta-heuristic algorithm. Here, the weight is updated for reducing the error difference between the target and the measured data when new data is given for testing. The optimized weight updated testing is performed by evaluating the concept-drift and classification accuracy. The new continuous learning by RNN is accomplished by the improved Opposition-based Novel Updating Spotted Hyena Optimization (ONU-SHO). Finally, the experiments with different datasets show that the proposed learning is improved over the conventional models.

From the analysis, the accuracy of the ONU-SHO based RNN (ONU-SHO-RNN) was 10.1% advanced than Decision Tree (DT), 7.6% advanced than Naive Bayes (NB), 7.4% advanced than k-nearest neighbors (KNN), 2.5% advanced than Support Vector Machine (SVM) 9.3% advanced than NN, and 10.6% advanced than RNN. Hence, it is confirmed that the ONU-SHO algorithm is performing well for acquiring the best data stream classification.

This paper introduces a novel meta-learning model using Opposition-based Novel Updating Spotted Hyena Optimization (ONU-SHO)-based Recurrent Neural Network (RNN) for handling continuous or streaming data. This is the first work utilizes a novel meta-learning model using Opposition-based Novel Updating Spotted Hyena Optimization (ONU-SHO)-based Recurrent Neural Network (RNN) for handling continuous or streaming data.

Currently, ubiquitous smartphones embedded with various sensors provide a convenient way to collect raw sequence data. These data bridges the gap between human activity and multiple sensors. Human activity recognition has been widely used in quite a lot of aspects in our daily life, such as medical security, personal safety, living assistance and so on.

To provide an overview, the authors survey and summarize some important technologies and involved key issues of human activity recognition, including activity categorization, feature engineering as well as typical algorithms presented in recent years. In this paper, the authors first introduce the character of embedded sensors and dsiscuss their features, as well as survey some data labeling strategies to get ground truth label. Then, following the process of human activity recognition, the authors discuss the methods and techniques of raw data preprocessing and feature extraction, and summarize some popular algorithms used in model training and activity recognizing. Third, they introduce some interesting application scenarios of human activity recognition and provide some available data sets as ground truth data to validate proposed algorithms.

The authors summarize their viewpoints on human activity recognition, discuss the main challenges and point out some potential research directions.

It is hoped that this work will serve as the steppingstone for those interested in advancing human activity recognition.

Feature selection is an important step for data pre-processing specially in the case of high dimensional data set. Performance of the data model is reduced if the model is trained with high dimensional data set, and it results in poor classification accuracy. Therefore, before training the model an important step to apply is the feature selection on the dataset to improve the performance and classification accuracy.

A novel optimization approach that hybridizes binary particle swarm optimization (BPSO) and differential evolution (DE) for fine tuning of SVM classifier is presented. The name of the implemented classifier is given as DEPSOSVM.

This approach is evaluated using 20 UCI benchmark text data classification data set. Further, the performance of the proposed technique is also evaluated on UCI benchmark image data set of cancer images. From the results, it can be observed that the proposed DEPSOSVM techniques have significant improvement in performance over other algorithms in the literature for feature selection. The proposed technique shows better classification accuracy as well.

The proposed approach is different from the previous work, as in all the previous work DE/(rand/1) mutation strategy is used whereas in this study DE/(rand/2) is used and the mutation strategy with BPSO is updated. Another difference is on the crossover approach in our case as we have used a novel approach of comparing best particle with sigmoid function. The core contribution of this paper is to hybridize DE with BPSO combined with SVM classifier (DEPSOSVM) to handle the feature selection problems.

During the production process of steel strip, some defects may appear on the surface, that is, traditional manual inspection could not meet the requirements of low-cost and high-efficiency production. The purpose of this paper is to propose a method of feature selection based on filter methods combined with hidden Bayesian classifier for improving the efficiency of defect recognition and reduce the complexity of calculation. The method can select the optimal hybrid model for realizing the accurate classification of steel strip surface defects.

A large image feature set was initially obtained based on the discrete wavelet transform feature extraction method. Three feature selection methods (including correlation-based feature selection, consistency subset evaluator [CSE] and information gain) were then used to optimize the feature space. Parameters for the feature selection methods were based on the classification accuracy results of hidden Naive Bayes (HNB) algorithm. The selected feature subset was then applied to the traditional NB classifier and leading extended NB classifiers.

The experimental results demonstrated that the HNB model combined with feature selection approaches has better classification performance than other models of defect recognition. Among the results of this study, the proposed hybrid model of CSE + HNB is the most robust and effective and of highest classification accuracy in identifying the optimal subset of the surface defect database.

The main contribution of this paper is the development of a hybrid model combining feature selection and multi-class classification algorithms for steel strip surface inspection. The proposed hybrid model is primarily robust and effective for steel strip surface inspection.

Text mining has been used for various knowledge discovery based applications, and thus, a lot of research has been contributed towards it. Latest trending research in the text mining is adopting the incremental learning data, as it is economical while dealing with large volume of information.

The primary intention of this research is to design and develop a technique for incremental text categorization using optimized Support Vector Neural Network (SVNN). The proposed technique involves four major steps, such as pre-processing, feature selection, classification and feature extraction. Initially, the data is pre-processed based on stop word removal and stemming. Then, the feature extraction is done by extracting semantic word-based features and Term Frequency and Inverse Document Frequency (TF-IDF). From the extracted features, the important features are selected using Bhattacharya distance measure and the features are subjected as the input to the proposed classifier. The proposed classifier performs incremental learning using SVNN, wherein the weights are bounded in a limit using rough set theory. Moreover, for the optimal selection of weights in SVNN, Moth Search (MS) algorithm is used. Thus, the proposed classifier, named Rough set MS-SVNN, performs the text categorization for the incremental data, given as the input.

For the experimentation, the 20 News group dataset, and the Reuters dataset are used. Simulation results indicate that the proposed Rough set based MS-SVNN has achieved 0.7743, 0.7774 and 0.7745 for the precision, recall and F-measure, respectively.

In this paper, an online incremental learner is developed for the text categorization. The text categorization is done by developing the Rough set MS-SVNN classifier, which classifies the incoming texts based on the boundary condition evaluated by the Rough set theory, and the optimal weights from the MS. The proposed online text categorization scheme has the basic steps, like pre-processing, feature extraction, feature selection and classification. The pre-processing is carried out to identify the unique words from the dataset, and the features like semantic word-based features and TF-IDF are obtained from the keyword set. Feature selection is done by setting a minimum Bhattacharya distance measure, and the selected features are provided to the proposed Rough set MS-SVNN for the classification.