Kind Of Classification Algorithm Research Articles

Ankle foot motion recognition based on wireless wearable sEMG and acceleration sensors for smart AFO

Ankle joint is one of the important anatomical structures of the human body, smart ankle-foot orthosis(AFO) can assist human walking and improve the ankle motion for patients. This study focused on ankle foot movements recognition based on data fusion via sEMG and acceleration sensors. A wireless signal acquisition system (WAS) was designed specifically, forming a platform to demonstrate and record individual sEMG and acceleration data simultaneously. In the experimental tests, three channel sEMG signals from Tibialis Anterior (TA), Gastrocnemius (GM) and Soleus (SO), as well as three-axis acceleration data of the ankle joints, were collected when subjects performed four kinds of typical motions including dorsiflexion, plantar flexion, eversion and inversion. A total of 21,600 frames of sEMG /acceleration action data were constructed and then different kinds of classification algorithms were studied to classify the motions by the principal component analysis (PCA) based data fusion signal features. Results showed that the classification accuracy of bi-directional long short-term memory (BiLSTM) algorithm performed the best compared with traditional networks such as support vector machine(SVM), artificial neural network (ANN) and reached 99.8 %. These results demonstrated the potential application for accurate ankle foot intent identification by sEMG and acceleration sensors, which provided the basis for further implementation of subsequent smart AFO manipulation.

Evaluation of Sampling-Based Ensembles of Classifiers on Imbalanced Data for Software Defect Prediction Problems

Defect prediction in software projects plays a crucial role to reduce quality-based risk and increase the capability of detecting faulty program modules. Hence, classification approaches to anticipate software defect proneness based on static code characteristics have become a hot topic with a great deal of attention in recent years. While several novel studies show that the use of a single classifier causes the performance bottleneck, ensembles of classifiers might effectively enhance classification performance compared to a single classifier. However, the class imbalance property of software defect data severely hinders the classification efficiency of ensemble learning. To cope with this problem, resampling methods are usually combined into ensemble models. This paper empirically assesses the importance of sampling with regard to ensembles of various classifiers on imbalanced data in software defect prediction problems. Extensive experiments with the combination of seven different kinds of classification algorithms, three sampling methods, and two balanced data learning schemata were conducted over ten datasets. Empirical results indicated the positive effects of combining sampling techniques and the ensemble learning model on the performance of defect prediction regarding datasets with imbalanced class distributions.

Leveraging Contextual Sentences for Text Classification by Using a Neural Attention Model.

We explored several approaches to incorporate context information in the deep learning framework for text classification, including designing different attention mechanisms based on different neural network and extracting some additional features from text by traditional methods as the part of representation. We propose two kinds of classification algorithms: one is based on convolutional neural network fusing context information and the other is based on bidirectional long and short time memory network. We integrate the context information into the final feature representation by designing attention structures at sentence level and word level, which increases the diversity of feature information. Our experimental results on two datasets validate the advantages of the two models in terms of time efficiency and accuracy compared to the different models with fundamental AM architectures.

Learning from Multi-annotator Data

In the field of sentiment analysis and emotion detection in social media, or other tasks such as text classification involving supervised learning, researchers rely more heavily on large and accurate labelled training datasets. However, obtaining large-scale labelled datasets is time-consuming and high-quality labelled datasets are expensive and scarce. To deal with these problems, online crowdsourcing systems provide us an efficient way to accelerate the process of collecting training data via distributing the enormous tasks to various annotators to help create large amounts of labelled data at an affordable cost. Nowadays, these crowdsourcing platforms are heavily needed in dealing with social media text, since the social network platforms (e.g., Twitter) generate huge amounts of data in textual form everyday. However, people from different social and knowledge backgrounds have different views on various texts, which may lead to noisy labels. The existing noisy label aggregation/refinement algorithms mostly focus on aggregating labels from noisy annotations, which would not guarantee their effectiveness on the subsequent classification/ranking tasks. In this article, we propose a noise-aware classification framework that integrates the steps of noisy label aggregation and classification. The aggregated noisy crowd labels are fed into a classifier for training, while the predicted labels are employed as feedback for adjusting the parameters at the label aggregating stage. The classification framework is suitable for directly running on crowdsourcing datasets and applies to various kinds of classification algorithms. The feedback strategy makes it possible for us to find optimal parameters instead of using known data for parameter selection. Simulation experiments demonstrate that our method provide significant label aggregation performance for both binary and multiple classification tasks under various noisy environments. Experimenting on real-world data validates the feasibility of our framework in real noise data and helps us verify the reasonableness of the simulated experiment settings.

RESEARCH ON REMOTE SENSING IMAGE CLASSIFICATION BASED ON FEATURE LEVEL FUSION

Abstract. Remote sensing image classification, as an important direction of remote sensing image processing and application, has been widely studied. However, in the process of existing classification algorithms, there still exists the phenomenon of misclassification and missing points, which leads to the final classification accuracy is not high. In this paper, we selected Sentinel-1A and Landsat8 OLI images as data sources, and propose a classification method based on feature level fusion. Compare three kind of feature level fusion algorithms (i.e., Gram-Schmidt spectral sharpening, Principal Component Analysis transform and Brovey transform), and then select the best fused image for the classification experimental. In the classification process, we choose four kinds of image classification algorithms (i.e. Minimum distance, Mahalanobis distance, Support Vector Machine and ISODATA) to do contrast experiment. We use overall classification precision and Kappa coefficient as the classification accuracy evaluation criteria, and the four classification results of fused image are analysed. The experimental results show that the fusion effect of Gram-Schmidt spectral sharpening is better than other methods. In four kinds of classification algorithms, the fused image has the best applicability to Support Vector Machine classification, the overall classification precision is 94.01 % and the Kappa coefficients is 0.91. The fused image with Sentinel-1A and Landsat8 OLI is not only have more spatial information and spectral texture characteristics, but also enhances the distinguishing features of the images. The proposed method is beneficial to improve the accuracy and stability of remote sensing image classification.

Optimal multivariate method for Raman spectroscopy based diagnosis of nasopharyngeal carcinoma

In this paper, we evaluated four kinds of classification algorithms on Raman spectra for nasopharyngeal carcinoma (NPC) diagnosis: Bayesian classification (BC), Linear discriminate analysis (LDA), Mahalanobis distance after the principal component analysis (PCA); as well the Genetic algorithm-LDA. A total of 225 Raman spectra were acquired from 120 tissue sites of 63 patients, in which 56 Raman spectra were from normal tissue, whereas 171 Raman spectra were from cancer nasopharyngeal tissue. The averaged Raman spectrum of NPC could be distinguished from that of the control group by the above multivariate analysis. Discrimination analysis of PCA-BC revealed that the highest sensitivity, specificity and overall accuracy of cancer diagnosis were 98% (1/56), 99% (1/171), and 99%, respectively. The results showed that Raman spectroscopy in combination with Bayesian classification had high enough sensitivity and specificity to accurately detect and diagnose NPC.

A Kind of Classification Algorithms of Data Mining and Quantitative Analysis

C4.5, Bayesian network and Sequential Minimal Optimization (SMO) are three typical classification algorithms in data mining. Using cross-validation method with 10 folds get analysis and calculation results of the experiments for the three classification algorithms in the same training set and test set. The main metrics include accuracy, precision, speed, robustness, scalability and comprehensibility, we use margin curve show these. It provides a theoretical and experimental basis for users to select a proper classification algorithm with different training sets in quality and amount.