Time Series Datasets Research Articles

Lidar cloud and aerosol layer detection method based on point cloud filtering

A point cloud filtering method is presented for atmospheric layer detection from lidar data. The method involves rising edge event recognition based on a wavelet transform function. Density-based clustering was then utilized to separate the real boundary from the original noisy point clouds based on continuous distribution characteristics of cloud and aerosol layer. Tests were carried out to verify the performance of our algorithm with synthetic lidar signals with noise. The layer base detection error within ± 5 bins was achieved for signals with SNRs higher than 3. Even for SNRs higher than 1, high consistency was still observed between retrieved results with our method and a visual analysis. These results indicate that our algorithm is suitable for unsupervised detection with large time-series datasets, such as CALIOP.

A variational approach to quantum gated recurrent units

Quantum Recurrent Neural Networks are receiving an increased attention thanks to their enhanced generalization capabilities in time series analysis. However, their performances were bottlenecked by long training times and unscalable architectures. In this paper, we propose a novel Quantum Recurrent Neural Network model based on Quantum Gated Recurrent Units. It uses a learnable Variational Quantum Layer to process temporal data, interspersed with two classical layers to properly match the dimensionality of the input and output vectors. Such an architecture has fewer quantum parameters than existing Quantum Long Short-Term Memory models. Both the quantum networks were evaluated on periodic and real-world time series datasets, together with the classical counterparts. The quantum models exhibited superior performances compared to the classical ones in all the test cases. The Quantum Gated Recurrent Units outperformed the Quantum Long Short-Term Memory network despite having a simpler internal configuration. Moreover, the Quantum Gated Recurrent Units network demonstrated to be about 25% faster during the training and inference procedure over the Quantum Long Short-Term Memory. This improvement in speed comes with one less quantum circuit to be executed, suggesting that our model may offer a more efficient alternative for implementing Quantum Recurrent Neural Networks on both simulated and real quantum hardware.

Short-term load forecasting of mixed industrial users in distribution networks using DARTS-LSTM method

Abstract Load forecasting has perennially stood as a fundamental pillar within the power industry, with precision in prognostication constituting a paramount necessity for critical functions including grid management, optimization of energy storage, and response to demand fluctuations. In the present study, we introduce a comprehensive global modeling prediction framework aimed at forecasting large-scale power demand within time series data, alongside the development of a novel short-term power load forecasting model. By integrating all time series data pertaining to the same regression task and employing a unified univariate prediction function across the entire time series dataset, we efficaciously forecast large-scale heterogeneous loads aggregated at various levels within distribution networks. The proposed framework leverages the cutting-edge DARTS (Differentiable Architecture Search) deep learning framework for implementation. Through comprehensive experimentation on real-world data, our model framework demonstrates superior performance in terms of overall prediction accuracy, scalability, and robustness in handling missing and non-stationary time series data.

Neural Hierarchical Interpolation for Standardized Precipitation Index Forecasting

In the context of climate change, studying changes in rainfall patterns is a crucial area of research, remarkably so in arid and semi-arid regions due to the susceptibility of human activities to extreme events such as droughts. Employing predictive models to calculate drought indices can be a useful method for the effective characterization of drought conditions. This study applies two type of machine learning methods—long short-term memory (LSTM) and Neural Hierarchical Interpolation for Time Series Forecasting (N-HiTS)—to develop and deploy artificial neural network models with the aim of predicting the regional standardized precipitation index (SPI) in four regions of Zacatecas, Mexico. The predictor variables were a set of climatological time series data spanning from 1964 to 2020. The results suggest that the N-HiTS model outperforms the LSTM model in the prediction and forecasting of SPI time series for all regions in terms of performance metrics: the Mean Squared Error, Mean Absolute Error, Coefficient of Determination and ξ correlation coefficient range from 0.0455 to 0.5472, from 0.1696 to 0.6661, from 0.9162 to 0.9684 and from 0.9222 to 0.9368, respectively, for the regions under study. Consequently, the outcomes revealed the successful performance of the N-HiTS models in accurately predicting the SPI across the four examined regions.

DeepQC: A deep learning system for automatic quality control of in-situ soil moisture sensor time series data

Amidst changing climate, real-time soil moisture monitoring is vital for the development of in-season decision support tools to help farmers manage weather-related risks in agriculture. Precision Sustainable Agriculture (PSA) recently established a real-time soil moisture monitoring network across the central, Midwest, and eastern U.S., but continuous field-scale sensor observations often come with data gaps and anomalies. To maintain the data quality and continuity needed for developing decision tools, a quality control system is necessary.The International Soil Moisture Network (ISMN) introduced the Flagit module for anomaly detection in soil moisture time series observations. However, under certain conditions, Flagit's threshold and spectral based quality control approaches may underperform in identifying anomalies. Recently, deep learning methods have been successfully applied to detect time series anomalies in time series data in various disciplines. However, their use in agriculture for anomaly detection in time series datasets has not been yet investigated. This study focuses on developing a Bi-directional Long Short-Term Memory (LSTM) model, referred to as DeepQC, to identify anomalies in soil moisture time series data. Manual flagged PSA observations were used for training, validation, and testing the model, following an 80:10:10 split. The study then compared the DeepQC and Flagit-based estimates to assess their relative performance.Flagit correctly flagged 95.8 % of the correct observations and 50.3 % of the anomaly observations, indicating its limitations in identifying anomalies, particularly at sites consists of more than 30 % anomalies. On the other hand, the DeepQC correctly flagged 89.8 % of the correct observations and 99.5 % of the anomalies, with overall accuracy of 95.6 %, in significantly less time, demonstrating its superiority over Flagit approach. Importantly, the performance of the DeepQC remained consistent regardless of the number of anomalies in site observations. Given the promising results obtained with the DeepQC, future studies will focus on implementing and finetuning this model on national and global soil moisture networks.

Local structure and effective dimensionality of time series data sets

The goal of this paper is to develop novel tools for understanding the local structure of systems of functions, e.g. time-series data points. The proposed tools include a total correlation function, the Cohen class of the data set, the data operator and the average lack of concentration. The Cohen class of the data operator gives a time-frequency representation of the data set. Furthermore, we show that the von Neumann entropy of the data operator captures local features of the data set and that it is related to the notion of effective dimensionality. The accumulated Cohen class of the data operator gives us a low-dimensional representation of the data set and we quantify this in terms of the average lack of concentration and the von Neumann entropy of the data operator and an improvement of the Berezin-Lieb inequality using the projection functional of the data augmentation operator. The framework for our approach is provided by quantum harmonic analysis.

Research on time-series financial data prediction and analysis based on deep recurrent neural network

Time series data is widely available in a variety of industries. By forecasting time series, decision-makers can better grasp future trends and make more effective decisions. Financial time series data exhibit non-stationarity and high volatility. High-frequency fluctuations in financial products such as exchange rates, bonds and equities may reflect external shocks and risks in global financial markets, which are potentially dangerous and may threaten national economic security or even trigger financial crises. For financial time series data, a deep recurrent neural network first progressively processes each data point in the time series through its recurrent unit. Each recurring unit can adjust its own weights to better predict or analyze future values. Over time, these recurrent units continuously update their internal state, resulting in a comprehensive understanding of the characteristics of the entire data sequence. In addition, we add a gating mechanism to further improve the network's ability to control the flow of information, so that the model is more effective when retaining long-term dependencies, so as to improve the accuracy of prediction and the stability of the model. Experimental results show that our recurrent neural network model shows higher prediction accuracy and stability than other baseline models on financial time series datasets.

Analyzing and explaining privacy risks on time series data: ongoing work and challenges

Currently, privacy risks assessment is mainly performed as audits conducted by data privacy analysts. In the TAILOR project, we promote a more systematic and automatic approach based on interpretable metrics and formal methods to evaluate privacy risks and to control the tension between data privacy and utility. In this paper, we focus on privacy risks raised by publishing time series datasets, and we survey the methods developed in TAILOR to analyze and quantify privacy risks depending on different publisher and attacker models.

Temporal Trends and Future Projections: A Deep Dive into India’s Buffalo Milk Production Through Time Series Modelling

Background: Buffalo milk production in India plays a significant role in the global dairy market, with a rich history deeply intertwined with the country’s economy and culture. Over six decades, the dynamics of buffalo farming have been pivotal in shaping India’s dairy landscape. Methods: This paper delves into the subject by analysing a comprehensive time series dataset spanning six decades. The focus lies on understanding the economic and cultural significance of buffalo farming, particularly in relation to milk production. Four forecasting models-ARIMA, SES, Seasonal Naive and ETS-are employed to discern temporal patterns in buffalo milk production. Result: The study reveals that the ARIMA and ETS models outperform SES and Seasonal Naive models in capturing and elucidating data behaviour. Their superior performance underscores their efficacy in predicting buffalo milk production trends accurately. These findings offer valuable insights for policymakers and stakeholders aiming to optimize buffalo milk production and foster long-term growth in India’s dairy sector.

Interpretable predictive modeling of non-stationary long time series

Highly accurate and interpretable forecasting of non-stationary long-time series is a challenge. It is often difficult for existing models to capture the trends of non-stationary time series and achieve interpretable forecasts. To this end, by integrating causal learning and high-order fuzzy cognitive graphs (HFCM), this paper proposes the Causal-HFCM framework. Firstly, the Peter-Clark (PC) algorithm is used to obtain the causal relationships of data, so that the relationships between data can be interpretable and causal learning breaks the limitation that the fuzzy cognitive graph only considers the correlation of data. Secondly, to capture the non-stationary property of time series and improve the prediction ability of fuzzy cognitive maps, a method combining the learning weight matrix algorithm of Huber regression with high-order fuzzy cognitive maps is proposed. Finally, the effectiveness of the proposed method is verified by conducting experiments with three different non-stationary time series datasets of traffic, air, and bearings.

New two‐sample test utilizing interpoint distance discrepancy

In this paper, we propose a novel two‐sample test for multivariate sample space. The test statistic calculates the mean of absolute difference of average interpoint distance. We utilize a permutation procedure to establish the critical value for the test. Through comprehensive simulation studies, we compare the performance of our proposed test with that of the K‐nearest neighbour test and the energy test. The results demonstrate that our proposed test exhibits advantages over the other two tests, particularly in high‐dimensional sample spaces. This superiority is further validated by its application to UCR time series datasets.

TGGS network: A multi-task learning network for gradient-guided knowledge sharing

Multi-task learning (MTL) has been widely used in various fields, such as time series data prediction and image classification. Most existing deep MTL methods achieve joint learning by sharing common knowledge with task relevance. However, they rely solely on self-exploration to mine common knowledge across the tasks, which is unguided and purposeless. Unlike these methods, in this paper, we propose a novel Task Gradient Guided Sharing (TGGS) MTL network for purposefully guiding the model to share knowledge. Specifically, we define a confidence matrix for each task for selecting effective knowledge from the common knowledge to share. To realize an effective guidance, we align the confidence matrix with a class gradient mapping (CGM) computed based on the task-specific gradients. Consequently, our method effectively employs task-specific gradients to guide the model in sharing useful knowledge. More importantly, the confidence matrix explains the complex correlations among tasks in MTL. As our method exclusively focuses on guided knowledge sharing, the TGGS network can seamlessly integrate with diverse MTL sharing strategies. We conduct extensive experiments on time series datasets and image datasets. The experimental results demonstrate our method significantly improves the performance of these MTL strategies.

POCKET: Pruning random convolution kernels for time series classification from a feature selection perspective

In recent years, two competitive time series classification models, namely, ROCKET and MINIROCKET, have garnered considerable attention due to their low training cost and high accuracy. However, they rely on a large number of random 1-D convolutional kernels to comprehensively capture features, which is incompatible with resource-constrained devices. Despite the development of heuristic algorithms designed to recognize and prune redundant kernels, the inherent time-consuming nature of evolutionary algorithms hinders efficient evaluation. To efficiently prune models, this paper eliminates feature groups contributing minimally to the classifier, thereby discarding the associated random kernels without direct evaluation. To this end, we incorporate both group-level (l2,1-norm) and element-level (l2-norm) regularizations to the classifier, formulating the pruning challenge as a group elastic net classification problem. An ADMM-based algorithm is initially introduced to solve the problem, but it is computationally intensive. Building on the ADMM-based algorithm, we then propose our core algorithm, POCKET, which significantly speeds up the process by dividing the task into two sequential stages. In Stage 1, POCKET utilizes dynamically varying penalties to efficiently achieve group sparsity within the classifier, removing features associated with zero weights and their corresponding kernels. In Stage 2, the remaining kernels and features are used to refit a l2-regularized classifier for enhanced performance. Experimental results on diverse time series datasets show that POCKET prunes up to 60% of kernels without a significant reduction in accuracy and performs 11× faster than its counterparts. Our code is publicly available at https://github.com/ShaowuChen/POCKET.

Forest Fire Prediction Based on Time Series Networks and Remote Sensing Images

Protecting forest resources and preventing forest fires are vital for social development and public well-being. However, current research studies on forest fire warning systems often focus on extensive geographic areas like states, counties, and provinces. This approach lacks the precision and detail needed for predicting fires in smaller regions. To address this gap, we propose a Transformer-based time series forecasting model aimed at improving the accuracy of forest fire predictions in smaller areas. Our study focuses on Quanzhou County, Guilin City, Guangxi Province, China. We utilized time series data from 2021 to 2022, along with remote sensing images and ArcGIS technology, to identify various factors influencing forest fires in this region. We established a time series dataset containing twelve influencing factors, each labeled with forest fire occurrences. By integrating these data with the Transformer model, we generated forest fire danger level prediction maps for Quanzhou County. Our model’s performance is compared with other deep learning methods using metrics such as RMSE, and the results reveal that the proposed Transformer model achieves higher accuracy (ACC = 0.903, MAPE = 0.259, MAE = 0.053, RMSE = 0.389). This study demonstrates that the Transformer model effectively takes advantage of spatial background information and the periodicity of forest fire factors, significantly enhancing predictive accuracy.

Enhancing bridge damage assessment: Adaptive cell and deep learning approaches in time-series analysis

In recent years, the application of Deep Learning (DL) for damage detection in Structural Health Monitoring (SHM) using time-series data has garnered significant attention from the global scientific community. By incorporating numerous filters and units in their architecture, these models facilitate feature extraction making the processing of large time-series datasets in SHM more effective. However, optimizing of the number of filters and units is a challenge, often relying on scientists' experience or previous models without a systematic method for measurement or automatic selection. Therefore, this study proposes a novel framework to enhance the accuracy and efficiency of SHM for damage detection in bridge structures by using metaheuristic algorithms, Electric Eel Foraging Optimization (EEFO) algorithm, to optimize hyperparameters of DL model. This DL model combines the advantages of each traditional model. Specifically, 1D Convolutional Neural Network (1DCNN) is employed for features extraction, Gated Recurrent Units (GRU) for identifying long-term dependencies, and Residual Networks (ResNet) for avoiding vanishing gradient problem which often happens in DL model during training, referred to as 1DCNN-GRU-ResNet. The time-series dataset generated from Cua Rao bridge is used to demonstrate the effectiveness of the proposed method. 1DCNN-GRU-ResNet after optimization (1DCNN-GRU-ResNet-opt) achieves 91.6 %, significantly surpassing traditional 1DCNN (82.5 %), GRU (79.9 %), 1DCNN-GRU (85.7 %), and 1DCNN-GRU-ResNet (89.3 %) in the test set. The proposed 1DCNN-GRU-ResNet-opt approach demonstrates considerable potential in practical applications for SHM, offering high accuracy and efficiency.

The privacy-explainability trade-off: unraveling the impacts of differential privacy and federated learning on attribution methods.

Since the advent of deep learning (DL), the field has witnessed a continuous stream of innovations. However, the translation of these advancements into practical applications has not kept pace, particularly in safety-critical domains where artificial intelligence (AI) must meet stringent regulatory and ethical standards. This is underscored by the ongoing research in eXplainable AI (XAI) and privacy-preserving machine learning (PPML), which seek to address some limitations associated with these opaque and data-intensive models. Despite brisk research activity in both fields, little attention has been paid to their interaction. This work is the first to thoroughly investigate the effects of privacy-preserving techniques on explanations generated by common XAI methods for DL models. A detailed experimental analysis is conducted to quantify the impact of private training on the explanations provided by DL models, applied to six image datasets and five time series datasets across various domains. The analysis comprises three privacy techniques, nine XAI methods, and seven model architectures. The findings suggest non-negligible changes in explanations through the implementation of privacy measures. Apart from reporting individual effects of PPML on XAI, the paper gives clear recommendations for the choice of techniques in real applications. By unveiling the interdependencies of these pivotal technologies, this research marks an initial step toward resolving the challenges that hinder the deployment of AI in safety-critical settings.

Efficient semi-supervised clustering with pairwise constraint propagation for multivariate time series

Semi-supervised clustering is an effective method, which improves the clustering performance based on pairwise constraints. However, state-of-the-art methods suffer from two issues: 1) due to the high dimensionality and multiple variables of multivariate time series (MTS) data, the competitive similarity approach DTW is time consuming on large-scale MTS data. 2) Traditional semi-supervised clustering methods could not make full use of pairwise constraints information, which affects clustering performance. To deal with these issues, we propose an efficient semi-supervised clustering with pairwise constraint propagation for MTS data. First, two approximate distance measure methods are designed based on dynamic time warping (DTW) from the perspectives of boundary and dimension reduction, which greatly improve the efficiency of clustering without sacrificing its accuracy. Then, a graph-based clustering method with pairwise constraints propagation (GCPCP) is advanced on multivariate time series data. In GCPCP, the pairwise constraint propagation matrix and the affinity matrix are jointly optimized to exploit the dependence between them, which finally improves the clustering performance. Experimental results on 12 multivariate time series datasets show the effectiveness and efficiency of our proposed method.

TFFS: A trainable federal fusion strategy for multistep time series forecasting

This paper presents a trainable federal fusion strategy (TFFS) for multistep time series forecasting (MSTSF) that unifies the current strategies. First, we designed a feature fusion embedding scheme (FFES) that integrates sharing features and all future steps in a trainable mode to overcome the drawback of cumulative errors. Second, multitask learning was designed to train each step in a parallel mode, enabling each step to share a common feature with the others while keeping each step’s personality. Finally, a multiple-input multiple-output scheme (MIMOS) was developed to forecast future step values in semi-series mode, which increases the flexibility. Owing to this design, TFFS provides the following advantages for MSTSF: 1) TFFS is free of cumulative errors in future horizons during training. 2) TFFS provides accurate multistep forecasting results, requiring only one model training. 3) Sharing common features while retaining each step’s personality increases forecasting accuracy. We conducted many experiments on univariate and multivariate time-series datasets to validate the effectiveness of TFFS with seven common deep learning structures. The experimental results confirmed the state-of-the-art performance of the proposed method. Additionally, we extensively and systematically compared each strategy for MSTSF with deep learning.

Temporal dynamics and extreme events in solar, wind, and wave energy complementarity: Insights from the Polish Exclusive Economic Zone

The Polish Exclusive Economic Zone (EEZ) in the Baltic Sea is an area of increasing strategic importance for Poland's pursuit of renewable energy, especially offshore wind. This research investigates the complementarity among solar, wind, and wave energy resources within the Polish EEZ to examine these energy sources' temporal dynamics, correlations, and extremes. The primary data source corresponds to a 31-year hourly time series dataset from the ERA5 reanalysis, whose reliability was evaluated through performance metrics. The results from complementarity metrics indicate varying levels of association among the three variable renewable energy resources (VRES) in the EEZ, spanning from weak similarity to weak complementarity. The findings of this research indicate that blackouts are most probable at offshore locations during winter and autumn for renewable power systems integrating wind and solar energy, with over 70 % of occurrences within these seasons. The investigation of extreme events highlights critical elements when evaluating VRES and their complementarity. This understanding aids in effectively planning and managing renewable energy systems, ensuring resilience and reliability under challenging weather conditions. Furthermore, while the complementarity may be consistent throughout the entire Polish EEZ, the feasibility and cost of implementing hybrid power systems can significantly vary between locations.

Forecasting urban expansion of Vadodara City, Gujarat (India): An integrated modeling approach with spatial, economic and demographical parameters

Urbanization is the inevitable phenomenon associated with the development of the region. The cities and towns tend to expand their territory exerting pressure on the environment and resources. Often this growth turns into sprawl, owing to irrational planning policies, coarse data, and generalized approaches irrespective of the underlying characteristics of the region. Thus, the need for urban planners is to understand the impact of past policies and their future course. The spatial urban expansion models like Cellular automata (CA) prove to be effective to trace urban development along with predicting the future. It equips the inclusion of the multiple spatial parameters to understand its degree of influence on regional urbanization. The present study focuses on the prediction of urbanization for 2031 in the Vadodara urban development authority (VUDA) region in Gujarat (India) using the cellular automata model. The study includes multiple variables comprising the social, demography, and land price variation along with the physical and site-situation suitability. A total of 32 years of time series dataset for identification of the land use pattern was generated. The result shows the city of Vadodara predominantly service industries centric growth and having industrial development on the peri-urban stretch will push the city limit to expand and according to the calibrated CA model, around 25 sq. km of the area is going to be transformed into the built-up usage in the VUDA region by the year 2031. The expansion of the built-up cover will mostly transform agricultural land, scrubland, and open land usage. Thus, a balanced horizontal versus vertical growth of the city will be required to have sustainable city development.