Online Classification Research Articles

Sensor attack online classification for UAVs using machine learning

Unmanned Aerial Vehicle (UAV) sensors play a vital role in maintaining flight safety and stability. However, the increasing frequency and complexity of sensor attacks have emerged as a critical threat to UAV systems. The current lack of robust multi-classification methods for detecting sensor attacks limits the effectiveness and completeness of existing defense strategies. This research addresses these challenges by leveraging machine learning (ML) techniques to classify various sensor attacks using heterogeneous sensor data and control parameters, thereby enhancing UAV system security. In this study, we design and implement multiple sensor attack scenarios targeting gyroscopes, accelerometers, barometers, and GPS. Comprehensive datasets are collected during UAV flight, integrating diverse sensor readings, flight states, and control parameters. By analyzing the characteristics of sensor attacks and their impact on position estimation and attitude control, we identify and extract key features. To optimize the classification model, we employ feature importance analysis, correlation analysis, and ablation experiments, significantly reducing data dimensionality and enhancing model training efficiency. The experimental results demonstrate the proposed ML-based multi-classification model’s superior performance, achieving a detection rate of 89.38%, significantly outperforming traditional single-attack detection methods in terms of generalization capability. Our approach efficiently handles complex multi-sensor attack scenarios. Moreover, deploying the optimized model on UAV firmware enables real-time monitoring and classification, achieving an online detection rate of 74% with a response time of approximately 0.495ms per detection. The model’s lightweight design, requiring only 48KB of storage, makes it ideal for resource-constrained UAV environments. These contributions highlight the potential of our approach to enhance real-time anomaly detection and improve UAV system resilience against diverse sensor attacks.

Improved motor imagery training for subject's self-modulation in EEG-based brain-computer interface.

For the electroencephalogram- (EEG-) based motor imagery (MI) brain-computer interface (BCI) system, more attention has been paid to the advanced machine learning algorithms rather than the effective MI training protocols over past two decades. However, it is crucial to assist the subjects in modulating their active brains to fulfill the endogenous MI tasks during the calibration process, which will facilitate signal processing using various machine learning algorithms. Therefore, we propose a trial-feedback paradigm to improve MI training and introduce a non-feedback paradigm for comparison. Each paradigm corresponds to one session. Two paradigms are applied to the calibration runs of corresponding sessions. And their effectiveness is verified in the subsequent testing runs of respective sessions. Different from the non-feedback paradigm, the trial-feedback paradigm presents a topographic map and its qualitative evaluation in real time after each MI training trial, so the subjects can timely realize whether the current trial successfully induces the event-related desynchronization/event-related synchronization (ERD/ERS) phenomenon, and then they can adjust their brain rhythm in the next MI trial. Moreover, after each calibration run of the trial-feedback session, a feature distribution is visualized and quantified to show the subjects' abilities to distinguish different MI tasks and promote their self-modulation in the next calibration run. Additionally, if the subjects feel distracted during the training processes of the non-feedback and trial-feedback sessions, they can execute the blinking movement which will be captured by the electrooculogram (EOG) signals, and the corresponding MI training trial will be abandoned. Ten healthy participants sequentially performed the non-feedback and trial-feedback sessions on the different days. The experiment results showed that the trial-feedback session had better spatial filter visualization, more beneficiaries, higher average off-line and on-line classification accuracies than the non-feedback session, suggesting the trial-feedback paradigm's usefulness in subject's self-modulation and good ability to perform MI tasks.

Heartbeat-related spectral perturbation of electroencephalogram reflects dynamic interoceptive attention states in the trial-by-trial classification analysis

Attending to heartbeats for interoceptive awareness initiates distinct electrophysiological responses synchronized with the R-peaks of an electrocardiogram (ECG), such as the heartbeat-evoked potential (HEP). Beyond HEP, this study proposes heartbeat-related spectral perturbation (HRSP), a time–frequency map of the R-peak locked electroencephalogram (EEG), and explores its characteristics in identifying interoceptive attention states using a classification approach. HRSPs of EEG brain components specified by independent component analysis (ICA) were used for the offline and online classification of interoceptive states. A convolutional neural network (CNN) designed specifically for HRSP was applied to publicly available data from a binary-state experiment (attending to self-heartbeats and white noise) and data from our four-state classification experiment (attending to self-heartbeats, white noise, time passage, and toe) with diverse input feature conditions of HRSP. From the dynamic state perspective, we evaluated the primary frequency bands of HRSP and the minimal number of averaging epochs required to reflect changing interoceptive attention states without compromising accuracy. We also assessed the utility of group ICA and models for classifying HRSP in new participants. The CNN for trial-by-trial HRSP with actual R-peaks demonstrated significantly higher classification accuracy than HRSP with sham, i.e., randomly positioned, R-peaks. Gradient-weighted class activation mapping highlighted the prominent role of theta and alpha bands between 200–600 ms post-R-peak—features absent in classifications using sham HRSPs. Online classification benefits from employing a group ICA and classification model, ensuring reliable accuracy without individual EEG precollection. These results suggest HRSP’s potential to reflect interoceptive attention states, proposing transformative implications for clinical applications.

Using artificial intelligence to analyze and classify music emotion

With the rapid development of music digitization and online streaming services, automatic analysis and classification of music content has become an urgent need. This research focuses on music sentiment analysis, which is the identification and classification of emotions expressed by music through algorithms. The study defines and classifies possible emotions in music. Then, advanced artificial intelligence techniques, including traditional machine learning and deep learning methods, were employed to perform sentiment analysis on music fragments. In the process of creating and validating the model, the combination of convolutional neural network and long term memory network shows excellent performance in various performance indicators. However, for some complex or culturally ambiguous music fragments, the model may also suffer from misclassification problems. This provides the direction for further optimization of future research aimed at achieving more accurate music emotion analysis.

Electrotactile BCI for Top-Down Somatosensory Training: Clinical Feasibility Trial of Online BCI Control in Subacute Stroke Patients.

This study investigates the feasibility of a novel brain-computer interface (BCI) device designed for sensory training following stroke. The BCI system administers electrotactile stimuli to the user's forearm, mirroring classical sensory training interventions. Concurrently, selective attention tasks are employed to modulate electrophysiological brain responses (somatosensory event-related potentials-sERPs), reflecting cortical excitability in related sensorimotor areas. The BCI identifies attention-induced changes in the brain's reactions to stimulation in an online manner. The study protocol assesses the feasibility of online binary classification of selective attention focus in ten subacute stroke patients. Each experimental session includes a BCI training phase for data collection and classifier training, followed by a BCI test phase to evaluate online classification of selective tactile attention based on sERP. During online classification tests, patients complete 20 repetitions of selective attention tasks with feedback on attention focus recognition. Using a single electroencephalographic channel, attention classification accuracy ranges from 70% to 100% across all patients. The significance of this novel BCI paradigm lies in its ability to quantitatively measure selective tactile attention resources throughout the therapy session, introducing a top-down approach to classical sensory training interventions based on repeated neuromuscular electrical stimulation.

Tracking full posterior in online Bayesian classification learning: a particle filter approach

The rapid growth of data volume and velocity is challenging traditional methods of classification, making it impossible to store so much data in memory. Developing online classification methods is becoming increasingly important. Online classification approaches have been subject to several assumptions in recent years, such as an independent and consistent distribution of initial values. In this paper, we present a general online Bayesian classification algorithm that can be adapted to handle streaming data based on the Monte Carlo fusion method. Rather than storing the entire raw data, we only update the posterior with historical data and the current batch of data. In addition, the proposed method has the capability of being applied to a wide range of situations with a high degree of accuracy, even in the case of data imbalances and discrepancies between sub-posterior distributions. The proposed algorithm was validated and evaluated through a comprehensive simulation study.

Online network traffic classification based on external attention and convolution by IP packet header

Network traffic classification is an important part of network monitoring and network management. Three traditional methods for network traffic classification are flow-based, session-based, and packet-based, while flow-based and session-based methods cannot meet the real-time requirements and existing packet-based methods will violate user’s privacy. To solve the above problems, we propose a network traffic classification method only by the IP packet header, which satisfies the requirements of both the user’s privacy protection and online classification performances. Through statistical analyses, we find that IP packet header information is effective on the network traffic classification tasks and this conclusion is also demonstrated by experiments. Furthermore, we propose a novel external attention and convolution mixed (ECM) model for online network traffic classification. This model adopts both low-computational complexity external attention and convolution to respectively extract the byte-level and packet-level characteristics for traffic classification. Therefore, it can achieve high classification accuracy and low time consumption. The experiments show that ECM can reach over 96% classification accuracy on four datasets and the classification time is 0.36 ms per packet which can meet the real-time requirements. The code is available at https://github.com/CNZZQ1030/ECM-for-Network-Traffic-Classification.

Exploring target-related information with reliable global pixel relationships for robust RGB-T tracking

RGB-T Siamese trackers have drawn continuous interest in recent years due to their proper trade-off between accuracy and speed. However, they are sensitive to the background distractors in some challenging cases, thereby inducing unreliable response positions. To overcome such drawbacks, we advance a new RGB-T Siamese tracker, named SiamTIH, which will advance the RGB-T Siamese trackers’ discriminability against distractors by exploiting target-related information and reliable global pixel relationships within multi-modal data. Specifically, we propose a target-related feature enhancement module (TFE) to highlight such areas in the detection branch that are similar to the templates and suppress those background distractor regions that are significantly different from the templates but are greatly informative. Then, we propose an intra- and mutual-modal attention based multi-modal feature fusion module (IMA-MF) to capture the reliable global pixel relationships within multi-modal data. Especially, the intra-modal attention is used to capture the global pixel relationships within each single modality data, and the mutual-modal attention is utilized to enhance the feature representation of the current modality by overall pixel relationships as well as modality-specific relationships. Finally, we propose a hard-focused online classifier (HFOC) that combines an offline classifier and an online classifier to further improve the robustness of our tracker. Besides, the proposed framework is further extended to a Transformer based tracker to verify its generality. Extensive experiments on three RGB-T benchmarks demonstrate that our new RGB-T tracker outperforms the existing ones and maintains real-time performance, exceeding on average 30 frames per second (FPS). The code will be available at https://github.com/Tianlu-Zhang/SiamTIH.

A robust event-driven approach to always-on object recognition

We propose a neuromimetic architecture capable of always-on pattern recognition, i.e. at any time during processing. To achieve this, we have extended an existing event-based algorithm (Lagorce et al., 2017), which introduced novel spatio-temporal features as a Hierarchy Of Time-Surfaces (HOTS). Built from asynchronous events captured by a neuromorphic camera, these time surfaces allow to encode the local dynamics of a visual scene and to create an efficient event-based pattern recognition architecture. Inspired by neuroscience, we have extended this method to improve its performance. First, we add a homeostatic gain control on the activity of neurons to improve the learning of spatio-temporal patterns (Grimaldi et al., 2021). We also provide a new mathematical formalism that allows an analogy to be drawn between the HOTS algorithm and Spiking Neural Networks (SNN). Following this analogy, we transform the offline pattern categorization method into an online and event-driven layer. This classifier uses the spiking output of the network to define new time surfaces and we then perform the online classification with a neuromimetic implementation of a multinomial logistic regression. These improvements not only consistently increase the performance of the network, but also bring this event-driven pattern recognition algorithm fully online. The results have been validated on different datasets: Poker-DVS (Serrano-Gotarredona and Linares-Barranco, 2015), N-MNIST (Orchard, Jayawant et al., 2015) and DVS Gesture (Amir et al., 2017). This demonstrates the efficiency of this bio-realistic SNN for ultra-fast object recognition through an event-by-event categorization process.

Neuron Circuit Based on a Split-gate Transistor with Nonvolatile Memory for Homeostatic Functions of Biological Neurons.

To mimic the homeostatic functionality of biological neurons, a split-gate field-effect transistor (S-G FET) with a charge trap layer is proposed within a neuron circuit. By adjusting the number of charges trapped in the Si3N4 layer, the threshold voltage (Vth) of the S-G FET changes. To prevent degradation of the gate dielectric due to program/erase pulses, the gates for read operation and Vth control were separated through the fin structure. A circuit that modulates the width and amplitude of the pulse was constructed to generate a Program/Erase pulse for the S-G FET as the output pulse of the neuron circuit. By adjusting the Vth of the neuron circuit, the firing rate can be lowered by increasing the Vth of the neuron circuit with a high firing rate. To verify the performance of the neural network based on S-G FET, a simulation of online unsupervised learning and classification in a 2-layer SNN is performed. The results show that the recognition rate was improved by 8% by increasing the threshold of the neuron circuit fired.

RadarTCN: Lightweight Online Classification Network for Automotive Radar Targets Based on TCN.

Automotive radar is one of the key sensors for intelligent driving. Radar image sequences contain abundant spatial and temporal information, enabling target classification. For existing radar spatiotemporal classifiers, multi-view radar images are usually employed to enhance the information of the target and 3D convolution is employed for spatiotemporal feature extraction. These models consume significant hardware resources and are not applicable to real-time applications. In this paper, RadarTCN, a novel lightweight network, is proposed that achieves high-accuracy online target classification using single-view radar image sequences only. In RadarTCN, 2D convolution and 3D-TCN are employed to extract spatiotemporal features sequentially. To reduce data dimensionality and computational complexity, a multi-layer max pooling down-sampling method is designed in a 2D convolution module. Meanwhile, the 3D-TCN module is improved through residual pruning and causal convolution is introduced for leveraging the performance of online target classification. The experimental results demonstrate that RadarTCN can achieve high-precision online target recognition for both range-angle and range-Doppler map sequences. Compared to the reference models on the CARRADA dataset, RadarTCN exhibits better classification performance, with fewer parameters and lower computational complexity.

Exposing and explaining fake news on-the-fly

Social media platforms enable the rapid dissemination and consumption of information. However, users instantly consume such content regardless of the reliability of the shared data. Consequently, the latter crowdsourcing model is exposed to manipulation. This work contributes with an explainable and online classification method to recognize fake news in real-time. The proposed method combines both unsupervised and supervised Machine Learning approaches with online created lexica. The profiling is built using creator-, content- and context-based features using Natural Language Processing techniques. The explainable classification mechanism displays in a dashboard the features selected for classification and the prediction confidence. The performance of the proposed solution has been validated with real data sets from Twitter and the results attain 80% accuracy and macro F-measure. This proposal is the first to jointly provide data stream processing, profiling, classification and explainability. Ultimately, the proposed early detection, isolation and explanation of fake news contribute to increase the quality and trustworthiness of social media contents.

Non-invasive real-time diagnosis of PMSM faults implemented in motor control software for mission critical applications

The machine’s health should be continuously monitored, or a test should be applied on a regular basis to predict failure before fatal damage is incurred. Unexpected failures, particularly in mission-critical applications, can cause irreversible damage to the system, or even human life. This paper introduces a novel non-intrusive, real-time, online Condition Monitoring (CM), and Fault Diagnosis (FD) system for Permanent Magnet Synchronous Machines (PMSMs). Only the motor drive’s built-in sensors, such as current and position sensors, are used to detect three types of faults: inter-turn short circuit, partial demagnetization, and static eccentricity. It encompasses the implementation of algorithms within a motor drive system and the creation of failure mode models. The proposed solution adopts a hardware-free approach, utilizing current/voltage signature analysis for cost-effectiveness. It requires a small memory and short execution time, allowing it to be implemented on a simple motor controller with limited memory and calculation power. The drive system is intended for mission critical applications, therefore, computation load, code size, memory allocation, run-time optimization, etc. are the key focuses for real-time operation. It offers immediate insights into motor’s health without interrupting the drive operation. Additionally, it ensures rapid processing with modest computational requirements, making it adaptable for implementation on any PMSM controller. The non-intrusive nature of this diagnostic approach has the potential to enhance safety in systems reliant on PMSM drives. The proposed method has a high detection accuracy of 98%, is computationally efficient and can detect and classify the fault accurately. Simulation and experimental results demonstrate the efficiency of the FD algorithm for online identification and classification of machine faults. Theoretical hypotheses are proven based on experimental data.

Robust Drug Use Detection on X: Ensemble Method with a Transformer Approach

There is a growing trend for groups associated with drug use to exploit social media platforms to propagate content that poses a risk to the population, especially those susceptible to drug use and addiction. Detecting drug-related social media content has become important for governments, technology companies, and those responsible for enforcing laws against proscribed drugs. Their efforts have led to the development of various techniques for identifying and efficiently removing drug-related content, as well as for blocking network access for those who create it. This study introduces a manually annotated Twitter dataset consisting of 112,057 tweets from 2008 to 2022, compiled for use in detecting associations connected with drug use. Working in groups, expert annotators classified tweets as either related or unrelated to drug use. The dataset was subjected to exploratory data analysis to identify its defining features. Several classification algorithms, including support vector machines, XGBoost, random forest, Naive Bayes, LSTM, and BERT, were used in experiments with this dataset. Among the baseline models, BERT with textual features achieved the highest F1-score, at 0.9044. However, this performance was surpassed when the BERT base model and its textual features were concatenated with a deep neural network model, incorporating numerical and categorical features in the ensemble method, achieving an F1-score of 0.9112. The Twitter dataset used in this study was made publicly available to promote further research and enhance the accuracy of the online classification of English-language drug-related content.

SA-O2DCA: Seasonal Adapted Online Outlier Detection and Classification Approach for WSN

SA-O2DCA: Seasonal Adapted Online Outlier Detection and Classification Approach for WSN

Online semi-supervised active learning ensemble classification for evolving imbalanced data streams

Concept drift is a core challenge in classification tasks of data streams. Although many drift adaptation methods have been presented, most of them assume that labels of all data are available, which is impractical in many real-world applications. Additionally, the absence of label makes the imbalance ratio of an imbalanced data stream difficultly being obtained in time, providing the inaccurate guidance for resampling and causing poor generalization. To tackle the joint challenges, an online semi-supervised active learning method is proposed to classifier imbalanced data streams with concept drift. A newly-arrived data is first added to the sliding window, and then assigned a pseudo label in terms of its nearest cluster. Meanwhile, semi-supervised clustering algorithm offers its predicted label. Based on the above two predictive labels, cluster-based query strategy provides the criteria for the evaluation and selection of representative instances. More especially, the uncertainty and importance of instances are defined to synthetically evaluate its representativeness. After obtaining true labels of typical ones, ensemble classifier is updated by all instances in current sliding window. Experimental results on 13 synthetic and real data streams indicate that the proposed method outperforms six comparative methods on both G-mean and Recall under various labeling budgets.

Online Detection and Classification of Interturn and Groundwall Insulation Aging Based on Broadband Common-Mode Impedance Spectrum

Online condition monitoring for stator insulation is important for reliable operation of electrical machines. Recent studies show that the method based on common-mode (CM) impedance spectrum is effective in online monitoring of the overall insulation ageing condition. However, it is difficult to detect the inter-turn ageing condition since the groundwall insulation ageing has the similar impact on the monitoring results. In this paper, the relationship between the change in the common-mode impedance spectrum and the stator winding insulation ageing is studied based on the per-turn-unit broadband CM stator winding model. The dataset of impedance spectrum for different insulation ageing conditions is built using this model. Multiple points in the impedance spectrum are used to detect the change in the inter-turn and groundwall insulation parameters separately. The online experiment results prove the validity of the proposed method. The proposed method can detect the insulation ageing condition with the error about 6%. There is no other method which can quantify the ageing degree of inter-turn insulation in the presence of groundwall insulation ageing. Compared with the black-box data-driven methods, the proposed method is based on a model with clear physical meaning, which makes the results more reliable.

Identification and Verification of Error-Related Potentials Based on Cerebellar Targets.

The error-related potential (ErrP) is a weak explicit representation of the human brain for individual wrong behaviors. Previously, ErrP-related research usually focused on the design of automatic correction and the error correction mechanisms of high-risk pipeline-type judgment systems. Mounting evidence suggests that the cerebellum plays an important role in various cognitive processes. Thus, this study introduced cerebellar information to enhance the online classification effect of error-related potentials. We introduced cerebellar regional characteristics and improved discriminative canonical pattern matching (DCPM) in terms of data training and model building. In addition, this study focused on the application value and significance of cerebellar error-related potential characterization in the selection of excellent ErrP-BCI subjects (brain-computer interface). Here, we studied a specific ErrP, the so-called feedback ErrP. Thirty participants participated in this study. The comparative experiments showed that the improved DCPM classification algorithm proposed in this paper improved the balance accuracy by approximately 5-10% compared with the original algorithm. In addition, a correlation analysis was conducted between the error-related potential indicators of each brain region and the classification effect of feedback ErrP-BCI data, and the Fisher coefficient of the cerebellar region was determined as the quantitative screening index of the subjects. The screened subjects were superior to other subjects in the performance of the classification algorithm, and the performance of the classification algorithm was improved by up to 10%.

Research on multi-role classification task of online mall based on heterogeneous graph neural network

With the rapid development of e-commerce, online shopping malls have become an indispensable part of daily life. In order to better meet the needs of consumers, marketplace platforms need to accurately identify and categorize different user personas to provide personalized services and recommendations. In traditional role classification methods, basic information and behavioral data of users are typically used for classification. However, this approach often ignores the complex relationships between users and multiple heterogeneous data such as goods, reviews, social networks, and more. Therefore, we propose a new approach based on heterogeneous graphs to model different types of data in the form of graphs to better capture the connections between users and various elements in the marketplace. In this study, graph embedding technology is used to map nodes in heterogeneous graphs into low-dimensional vector spaces to capture similarities and relationships between nodes. Then, using the vector representation of these nodes, we can apply algorithms such as attention mechanisms for multi-role classification. Specifically, we use algorithms such as support vector machines to train classification models and use heterogeneous graph attention mechanisms to obtain the final feature representation of nodes. Experimental results show that our method shows significant advantages in multi-role classification tasks. Finally, the results of this study are discussed and summarized. We found that the classification model based on heterogeneous graph can effectively classify multiple roles in the online mall to provide personalized services and recommendations for the mall. At the same time, we also find that the construction of heterogeneous maps and the choice of graph embedding technology have important impacts on the classification results, which need further research and optimization. Therefore, multi-role task classification of online shopping malls based on heterogeneous graph neural networks is of great significance for improving the user experience and recommendation effect of online shopping malls, and also provides new ideas and methods for research in related fields.

StreamSoNGv2: Online Classification of Data Streams Using Growing Neural Gas

StreamSoNGv2: Online Classification of Data Streams Using Growing Neural Gas