Time Series Data Mining Research Articles

A novel two-enhancive aspect module in convolutional neural networks for multivariate time series classification

Multivariate Time Series Classification (MTSC) presents a significant challenge in time series data mining. While many methods have been proposed, Convolutional Neural Networks (CNNs) still face limitations in effectively capturing multivariate dependencies. In this paper, we introduce the Two-Enhancive Aspect Module (TEAM), a novel attention mechanism that integrates both Temporal Attention (TEAM_T) and Channel Attention (TEAM_C) to enhance feature extraction in CNNs. In TEAM_T, the calculation of attention weights considers both the sample’s content and its relative positional placement, forming a pseudo-Gaussian distribution that better reflects the relative importance of time samples. In TEAM_C, the Discrete Cosine Transform (DCT) is used for time-frequency transformation, building channel attention in the frequency domain, it better captures the interdependencies between channels. Evaluated on the UEA benchmark with 15 MTSC datasets, TEAM consistently outperforms 14 baseline methods and 5 alternative attention mechanisms, significantly boosting classification performance and surpassing the current SOTA in mean accuracy.

A shared multi-scale lightweight convolution generative network for few-shot multivariate time series forecasting

Time series forecasting is an important time series data mining technique. Among them, multivariate time series (MTS) forecasting has received extensive attention in many fields. However, many existing MTS forecasting models usually rely on a large amount of labeled data for model training, and data collection and labeling are difficult in real systems. The insufficient amount of data makes it difficult for the model to fully learn the intrinsic patterns and features of the data, which not only increases the prediction error, but also makes it hard to obtain satisfactory prediction results. To address this challenge, we propose a shared multi-scale lightweight convolution generative (SMLCG) network for few-shot multivariate time series forecasting by using samples generation strategy. The overall goal is to design a shared multi-scale feature generation prediction framework that generates data highly similar to the original sample and enriches the training sample to improve prediction accuracy. Specifically, the MTS is divided into different scales, and the multi-scale feature fusion module is utilized to capture and fuse the MTS information in different spatial dimensions to eliminate the heterogeneity among the data. Then, the key information in the multi-scale features is captured by a lightweight convolution generative network, and the feature weights are dynamically assigned to explore the change information. In addition, a spatio-temporal memory module is designed based on the parameter sharing strategy to capture the spatio-temporal dynamic relationship of sequences by learning the common knowledge in multi-scale features, thus improving the robustness and generalization ability. Through comprehensive experiments on four publicly available datasets and comparisons with other reported models, it is demonstrated that the SMLCG model can efficiently generate approximate samples in the few-shot case and provide excellent prediction results. The architecture of SMLCG serves as a valuable reference for practical solutions to address the few-shot problem in multivariate time series.

Accelerating time series similarity search under Move-Split-Merge distance via dissimilarity space embedding

The ubiquity sensory devices across scientific and industrial settings have made time-series data mining an important research topic in the IoT domain. Among all time-series mining techniques, the similarity search is considered the most essential and noteworthy because it is frequently employed as a fundamental subroutine for other procedures. The Dynamic Time Warping (DTW) was once considered the best similarity measure in time-series data mining community; however, two recent studies have demonstrated that the Move-Split-Merge (MSM) performs better than DTW in both supervised and unsupervised settings. Computing MSM between two time-series incurs costly computation because MSM has quadratic time complexity. Regrettably, there is little research on accelerating time-series similarity search under MSM. This paper provides an efficient method named DSEMSM for speeding up this procedure. DSEMSM utilizes dissimilarity space embedding to transform time-series into low-dimensional vectors. The filter-and-verification framework is subsequently employed to reduce the large number of actual MSM computations and to identify the final search results from the candidate set. The furthest prototype selection strategy and lower bound based technique are applied in DSEMSM to improve its retrieval accuracy and efficiency, respectively. We validate the effectiveness of the proposed method on a large number of datasets. The experimental results indicate that, in comparison to its relevant competitors, DSEMSM achieves a more stable speedup rate and higher retrieval accuracy.

Introducing Mplots: scaling time series recurrence plots to massive datasets

Time series similarity matrices (informally, recurrence plots or dot-plots), are useful tools for time series data mining. They can be used to guide data exploration, and various useful features can be derived from them and then fed into downstream analytics. However, time series similarity matrices suffer from very poor scalability, taxing both time and memory requirements. In this work, we introduce novel ideas that allow us to scale the largest time series similarity matrices that can be examined by several orders of magnitude. The first idea is a novel algorithm to compute the matrices in a way that removes dependency on the subsequence length. This algorithm is so fast that it allows us to now address datasets where the memory limitations begin to dominate. Our second novel contribution is a multiscale algorithm that computes an approximation of the matrix appropriate for the limitations of the user’s memory/screen-resolution, then performs a local, just-in-time recomputation of any region that the user wishes to zoom-in on. Given that this largely removes time and space barriers, human visual attention then becomes the bottleneck. We further introduce algorithms that search massive matrices with quadrillions of cells and then prioritize regions for later examination by either humans or algorithms. We will demonstrate the utility of our ideas for data exploration, segmentation, and classification in domains as diverse as astronomy, bioinformatics, entomology, and wildlife monitoring.

Application of big data technology in emergency management platform informatization construction

AbstractThe emergence of emergencies has impacted China’s politics, economy, and other aspects to varying degrees, seriously threatening social stability, economic development, and people’s happiness. The emergency management platform plays an important role in the handling of emergencies. The establishment of the emergency management platform can help leaders make correct decisions and reduce the impact of emergencies on the safety of people’s lives and property. However, some problems are gradually exposed in the process of establishing the emergency management platform. How to make it play its role better has become an important goal of platform informatization. With the advent of the big data era, big data technology has become a new way of information construction of emergency management platforms. This article mainly discussed the information construction of the emergency management platform. This article proposed to use big data technology to carry out the information construction of the emergency management platform and use the time series data mining algorithm based on the Artificial Neural Network to achieve the prediction function of the platform. The experimental results in this article showed that the prediction rate was 98.0% when the emergencies of the input platform were 50 and 99.0% when the emergencies of the input platform were 300. It can be seen that the emergency management platform designed in this article has a high ability to predict emergencies, so that emergency management measures can be taken in advance to reduce the occurrence of emergencies.

Analyzing variation of water inflow to inland lakes under climate change: Integrating deep learning and time series data mining

The alarming depletion of global inland lakes in recent decades makes it essential to predict water inflow from rivers to lakes (WIRL) trend and unveil the dominant influencing driver, particularly in the context of climate change. The raw time series data contains multiple components (i.e., long-term trend, seasonal periodicity, and random noise), which makes it challenging for traditional machine/deep learning techniques to effectively capture long-term trend information. In this study, a novel FactorConvSTLnet (FCS) method is developed through integrating STL decomposition, convolutional neural networks (CNN), and factorial analysis into a general framework. FCS is more robust in long-term WIRL trend prediction through separating trend information as a modeling predictor, as well as unveiling predominant drivers. FCS is applied to typical inland lakes (the Aral Sea and the Lake Balkhash) in Central Asia, and results indicate that FCS (Nash-Sutcliffe efficiency = 0.88, root mean squared error = 67m³/s, mean relative error = 10%) outperforms the traditional CNN. Some main findings are: (i) during 1960–1990, reservoir water storage (WSR) was the dominant driver for the two lakes, respectively contributing to 71% and 49%; during 1991–2014 and 2015–2099, evaporation (EVAP) would be the dominant driver, with the contribution of 30% and 47%; (ii) climate change would shift the dominant driver from human activities to natural factors, where EVAP and surface snow amount (SNW) have an increasing influence on WIRL; (iii) compared to SSP1-2.6, the SNW contribution would decrease by 26% under SSP5-8.5, while the EVAP contribution would increase by 9%. The findings reveal the main drivers of shrinkage of the inland lakes and provide the scientific basis for promoting regional ecological sustainability.

Communication network security situation analysis based on time series data mining technology

AbstractCommunication network security situation analysis prevents unauthorized users from accessing and stealing sensitive information. Network security analysis aims to monitor, discover, and eradicate security flaws by carefully examining the network’s architecture, data, and traffic to ensure safety. In time series data mining analysis by cyber terrorism, specialists must pay attention to cyber security, which involves identifying the elements contributing to long-term trends or systemic patterns via pattern-matching algorithms and other types of inferential processing on large datasets. The challenging characteristics of communication network security situation analysis are data loss, security breaches, hacking, and viruses. Hence, in this research, attention mechanism-based convolutional neural network-enabled practical byzantine fault tolerant (AMBCNN-PBFT) has been designed to improve communication network security situation analysis in time series data mining. AMBCNN-PBFT helps to increase communication network security usage and support the expansion during the evaluation system by optimizing the time series data mining. AMBCNN-PBFT effectively predicts the rise in the communication network, associated with faster times series benefits data mining approach. The study concludes that the AMBCNN-PBFT efficiently indicates and validates the communication network security in time series data mining during the evaluation system. The experimental analysis of AMBCNN-PBFT outperforms the data mining time series in terms of accuracy, efficiency, performance, and prediction.

Application of Recurrence Plot Analysis to Examine Dynamics of Biological Molecules on the Example of Aggregation of Seed Mucilage Components.

The goal of the research is to describe the aggregation process inside the mucilage produced by plant seeds using molecular dynamics (MD) combined with time series algorithmic analysis based on the recurrence plots. The studied biological molecules model is seed mucilage composed of three main polysaccharides, i.e. pectins, hemicellulose, and cellulose. The modeling of biological molecules is based on the assumption that a classical-quantum passage underlies the aggregation process in the mucilage, resulting from non-covalent interactions, as they affect the macroscopic properties of the system. The applied recurrence plot approach is an important tool for time series analysis and data mining dedicated to analyzing time series data originating from complex, chaotic systems. In the current research, we demonstrated that advanced algorithmic analysis of seed mucilage data can reveal some features of the dynamics of the system, namely temperature-dependent regions with different dynamics of increments of a number of hydrogen bonds and regions of stable oscillation of increments of a number of hydrophobic-polar interactions. Henceforth, we pave the path for automatic data-mining methods for the analysis of biological molecules with the intermediate step of the application of recurrence plot analysis, as the generalization of recurrence plot applications to other (biological molecules) datasets is straightforward.

Time series clustering with random convolutional kernels

Time series data, spanning applications ranging from climatology to finance to healthcare, presents significant challenges in data mining due to its size and complexity. One open issue lies in time series clustering, which is crucial for processing large volumes of unlabeled time series data and unlocking valuable insights. Traditional and modern analysis methods, however, often struggle with these complexities. To address these limitations, we introduce R-Clustering, a novel method that utilizes convolutional architectures with randomly selected parameters. Through extensive evaluations, R-Clustering demonstrates superior performance over existing methods in terms of clustering accuracy, computational efficiency and scalability. Empirical results obtained using the UCR archive demonstrate the effectiveness of our approach across diverse time series datasets. The findings highlight the significance of R-Clustering in various domains and applications, contributing to the advancement of time series data mining.

Empirical Evaluation of the Time Series Forecasting Method by Combining ARIMA with RBFNN under the Additive Model

Time series data is a series of values observed through repeated measurements at different times. Time series data is a type of data present in almost all different fields of life. Time series prediction is an significant problem in time series data mining. Accurate forecasting is crucial to support decision making in many areas of life. Therefore, improving the precision of time series predicting is a interesting mission for experts in this field. Many models for predicting time series have been proposed from traditional time series models as Auto Regressive Integrated Moving Average (ARIMA) model to artificial neural network (ANN) models. ARIMA is a linear model therefore it can only take the linear characteristics in time series. In contrast, Radial Basis Function Neural Network (RBFNN) is a non-linear model therefore it can not predict effectively seasonal or trend changes in time series. To combine the strengths of these two models, in this study, we experimentally evaluate the hybrid method between ARIMA and RBFNN on real time series data from different fields. Experimental results demonstrate that the combined method outperforms each model used individually in terms of accuracy.

Matrix Profile data mining for BGP anomaly detection

The Border Gateway Protocol (BGP), acting as the communication protocol that binds the Internet, remains vulnerable despite Internet security advancements. This is not surprising, as the Internet was not designed to be resilient to cyber-attacks, therefore the detection of anomalous activity was not of prime importance to the Internet creators. Detection of BGP anomalies can potentially provide network operators with an early warning system to focus on protecting networks, systems, and infrastructure from significant impact, improve security posture and resilience, while ultimately contributing to a secure global Internet environment. In this paper, we present a novel technique for the detection of BGP anomalies in different events. This research uses publicly available datasets of BGP messages collected from the repositories, Route Views and Réseaux IP Européens (RIPE). Our contribution is the application of a time series data mining approach, Matrix Profile (MP), to detect BGP anomalies in all categories of BGP events. Advantages of the MP detection technique compared to extant approaches include that it is domain agnostic, is assumption-free, requires few parameters, does not require training data, and is scalable and storage efficient. The single hyper-parameter analyzed in MP shows it is robust to change. Our results indicate the MP detection scheme is competitive against existing detection schemes. A novel BGP anomaly detection scheme is also proposed for further research and validation.

Analysis and Design of Digital Transformation Factors in the Teaching Process of the "Wise Traditional Music Classroom"

This study aims to explore the digital technology-enabled traditional music classroom education and teaching ecological design based on cultural heritage protection and inheritance, in order to improve the efficiency of cultural heritage classroom teaching and digital reform. We chose the excellent traditional Chinese music culture "Poyang Fishing Song" as a case study for analysis and demonstration. Mathematical modelling methods, time series analysis, data mining and optimisation models were used to quantify the efficiency of cultural heritage transmission and the impact of digital reform. Meanwhile, the application of blockchain technology in music education, especially in the field of copyright management, is explored to support the digital transformation of the education ecosystem. By digging deeper and revealing the potentials and limitations of digital transformation, it provides tools for educational institutions, policy makers and researchers to learn from in order to better address new challenges and seize new opportunities.

Research on the evaluation of equipment condition in assembly line workshop based on time series analysis

Abstract Intelligent manufacturing requires the full use of modern information technology in traditional production scenarios, and assembly line workshops also need intelligent industrial upgrading in production management. To this end, this paper proposes a time series-based evaluation study of the state of equipment in an assembly line workshop. Through the theoretical analysis of the characteristics of assembly line workshop equipment state data, time series data mining technology is used to collect research and analysis data, which is then pre-processed. The differential autoregressive average model is used to construct the assembly line workshop equipment state evaluation model, and the simulation analysis of the assembly line workshop equipment state is carried out by combining the corresponding parameters with the research data. The data shows that the MAPE value of the ARIMA model for equipment failure is −0.0706%, which is more significant compared to the prediction effect of the traditional CNN model. In addition, the error range of the ARIMA model prediction of equipment state parameters of the assembly line workshop is 0.01~0.02. This study can accurately assess the equipment state of the assembly line workshop and lay the foundation for its intelligent transformation.

On the use of matrix profiles and optimal transport theory for multivariate time series anomaly detection within structural health monitoring

In order for a practical application of structural health monitoring to be considered successful, not only is the detection of anomalies important but so is the tracking of various other behaviors, which may be induced by changing operational conditions or environmental variations. Rather than being problems of anomaly detection, these are known as problems of changing contexts or changing semantics. In this paper, we propose a novel, and efficient method which is able to perform both tasks: anomaly detection, and semantic analysis, simultaneously, and which is also able to scale to a multi-sensor network array. The proposed methodology works to combine a recent development from the field of time series data mining known as matrix profiling, as well as ideas from a popular area of mathematics known as optimal transport theory. In particular, matrix profiles will be used to generate curves of semantic segmentation, over multivariate time series data. From here, it will be demonstrated that by finding the unbalanced optimal transport barycenter one can successfully pool together information from a sensor array network in such a way that the presence of false positive transition points is reduced significantly when compared to the baseline mean. Moreover, this will be demonstrated to occur in a hierarchical fashion by employing a variational mode decomposition across each sensor in the network array. Overall, the proposed methodology: (i) Can work in the small data context without requiring a large amount of data for model training and validation, (ii) Applies readily to unsupervised and semi-supervised learning settings, (iii) Can extend to the online learning setting, (iv) Is adaptable and robust for sensor array network problems, (v) Does not require specifying any threshold, and (vi) Is highly interpretable. Finally, the proposed methodology will be validated on three different levels of investigation: (i) A numerical dataset of an aerospace model based on a two-dimensional plate model, (ii) An experimental dataset of a building from Los Alamos National Laboratories, and finally, (iii) A field investigation on a cable-stayed bridge located in Sydney, Australia.

Taobao transaction data mining based on time series evaluation under the background of big data

With the emergence of e-commerce, more and more people conduct transactions through the Internet, thus resulting in a large number of transaction data. Data mining is to decompose a large amount of data according to data rules, and analyze network transaction data, so as to provide necessary digital links for companies to analyze the market and develop business. Although time series data mining is smaller than other types of data mining, it is also an important issue of data mining. In the real world, the correlation between data and time is very common. The study of time series model plays a very important role in data mining. Due to different purposes, Taobao data analysis is also different. In addition to statistics, at present, the in-depth research and analysis of Taobao data are relatively insufficient, and the analysis of Taobao transaction data based on time series is rare. In order to improve the accuracy of Taobao transaction data mining and better formulate Taobao marketing strategy, this paper used time series data mining technology to mine Taobao transaction data. This paper first introduced the role of Taobao transaction data mining, and then described the calculation method of time series data mining, including the re-description of time series and the similarity measurement of time series. Finally, through a series of processes such as data collection, data processing and data feature extraction, the data mining model for Taobao transaction was established, and two data prediction evaluation indicators, namely prediction accuracy and entropy, were proposed. The experimental part verified the effect of Taobao transaction data mining. The experimental results showed that the data mining model moment had good data prediction accuracy and entropy. The average data prediction accuracy was 94.26%, and the data mining ability was strong.

빅데이터 활용한 시계열 데이터 텍스트 마이닝 및 시각화

빅데이터 활용한 시계열 데이터 텍스트 마이닝 및 시각화

An autocorrelation incremental fuzzy clustering framework based on dynamic conditional scoring model

This paper focuses on the real-time dynamic clustering analysis of power load data based on the dynamic conditional score (DCS) model, and an autocorrelation increment fuzzy C-means clustering algorithm based on the DCS model is proposed. (1) The paper addresses the problem that current power load clustering methods, when performing time series data mining, tend to focus on capturing the mean structure while ignoring the variance characteristics of the data, making it difficult to effectively capture the structural information of time series data. The DCS model is used as the statistical model basis for clustering analysis, and the time series is clustered based on the estimated conditional moment characteristics of the model, dynamically capturing data features such as the mean, variance, and sequence correlation of time series data, effectively improving the clustering performance. (2) This paper also addresses the issue that current power load clustering methods tend to focus on static datasets of user power loads and cannot effectively handle the data stream clustering problem with time series characteristics in practical applications. The DCS model parameter dataset and the autocorrelation increment fuzzy clustering algorithm are used to conduct a dynamic data flow analysis of user electricity behaviour evolution and pattern continuous updating research for power loads. The clustering results are dynamically updated based on the user's power load data stream using the proposed algorithm, achieving research on a universal clustering model and secure and efficient algorithms in a big data environment. (3) The paper verifies the clustering performance of the proposed method using power load time series data provided by a Chinese power supply company as a case dataset. The clustering evaluation index shows that the proposed algorithm has high clustering accuracy and good clustering performance. Additionally, different power supply recommendations are proposed for different customer electricity types in the obtained clustering results to provide more personalized power services.

Time Series Data Mining: A Unifying View

Time series data are ubiquitous; large volumes of such data are routinely created in scientific, industrial, entertainment, medical and biological domains. Examples include ECG data, gait analysis, stock market quotes, machine health telemetry, search engine throughput volumes etc. VLDB has traditionally been home to much of the community's best research on time series, with three to eight papers on time series appearing in the conference each year. What do we want to do with such time series? Everything! Classification, clustering, joins, anomaly detection, motif discovery, similarity search, visualization, summarization, compression, segmentation, rule discovery etc. Rather than a deep dive in just one of these subtopics, in this tutorial I will show a surprisingly small set of high-level representations, definitions, distance measures and primitives can be combined to solve the first 90 to 99.9% of the problems listed above. The tutorial will be illustrated with numerous real-world examples created just for this tutorial, including examples from robotics, wearables, medical telemetry, astronomy, and (especially) animal behavior. Moreover, all sample datasets and code snippets will be released so that the tutorial attendees (and later, readers) can first reproduce the results demonstrated, before attempting similar analysis on their data.

Prediction and analysis of time series data based on granular computing.

The advent of the Big Data era and the rapid development of the Internet of Things have led to a dramatic increase in the amount of data from various time series. How to classify, correlation rule mining and prediction of these large-sample time series data has a crucial role. However, due to the characteristics of high dimensionality, large data volume and transmission lag of sensor data, large sample time series data are affected by multiple factors and have complex characteristics such as multi-scale, non-linearity and burstiness. Traditional time series prediction methods are no longer applicable to the study of large sample time series data. Granular computing has unique advantages in dealing with continuous and complex data, and can compensate for the limitations of traditional support vector machines in dealing with large sample data. Therefore, this paper proposes to combine granular computing theory with support vector machines to achieve large-sample time series data prediction. Firstly, the definition of time series is analyzed, and the basic principles of traditional time series forecasting methods and granular computing are investigated. Secondly, in terms of predicting the trend of data changes, it is proposed to apply the fuzzy granulation algorithm to first convert the sample data into coarser granules. Then, it is combined with a support vector machine to predict the range of change of continuous time series data over a period of time. The results of the simulation experiments show that the proposed model is able to make accurate predictions of the range of data changes in future time periods. Compared with other prediction models, the proposed model reduces the complexity of the samples and improves the prediction accuracy.

Multivariate Time Series Clustering based on Graph Convolutional Network

Multivariable time series (MTS) clustering is an important topic in time series data mining. The major challenge of MTS clustering is to capture the temporal correlations and the dependencies between multiple variables. In this paper, we propose a novel MTS clustering approach based on graph convolutional network (GCN), which is a powerful feature extractor for graph structure data. We regard each variable in MTS as a node in the graph and construct edges through the correlation between variables. Furthermore, GCN and deep learning back-ropagation technology are used to continuously learn the relationship between multiple variables. Combining the learned variables with the characteristics of the time dimensions, the comprehensive features can be fused to form effective representation for MTS clustering task. We carry out extensive experimental analysis on four open time series data sets and six benchmark algorithms, which shows the superiority of the proposed method.