Time Warping Distance Research Articles

Combining Numerical Simulation and Deep Learning for Landslide Displacement Prediction: An Attempt to Expand the Deep Learning Dataset

Effective landslide hazard prevention requires accurate landslide prediction models, and the data-driven approaches based on deep learning models are gradually becoming a hot research topic. When training deep learning models, it is always preferable to have a large dataset, while most available landslide monitoring data are limited. For data missing or data sparseness problems, conventional interpolation methods based on mathematical knowledge lack mechanism interpretability. This paper proposes that numerical simulations can be used to expand the deep learning dataset we need. Taking the Jiuxianping landslide in the Three Gorges Reservoir Area (TGRA) as the geological background, a finite element numerical model was established, and the landslide displacement time series data were solved considering the boundary conditions of reservoir water level change and precipitation. Next, based on three metrics: Euclidean distance, cosine similarity, and dynamic time warping (DTW) distance, the time series similarity between the displacement data obtained from simulation and data obtained from actual monitoring were verified. Finally, the combined deep learning model was built to predict the displacement of the Jiuxianping landslide. The model was trained on both the simulated and monitored datasets and tested by the last 12 monitored data points. Prediction results with the testing set showed that the models trained using the expanded training set from numerical simulations exhibited lower prediction errors, and the errors had a more concentrated distribution. The results suggest that this landslide displacement prediction method combining numerical simulation and deep learning can solve the problem of inadequate datasets due to low monitoring frequency, as well as provide an interpretation of the physical mechanism for data vacancy filling.

A Faster Reduction of the Dynamic Time Warping Distance to the Longest Increasing Subsequence Length

The similarity between a pair of time series, i.e., sequences of indexed values in time order, is often estimated by the dynamic time warping (DTW) distance, instead of any in the well-studied family of measures including the longest common subsequence (LCS) length and the edit distance. Although it may seem as if the DTW and the LCS(-like) measures are essentially different, we reveal that the DTW distance can be represented by the longest increasing subsequence (LIS) length of a sequence of integers, which is the LCS length between the integer sequence and itself sorted. For a given pair of time series of length n such that the dissimilarity between any elements is an integer between zero and c, we propose an integer sequence that represents any substring-substring DTW distance as its band-substring LIS length. The length of the produced integer sequence is O(c n^2), which can be translated to O(n^2) for constant dissimilarity functions. To demonstrate that techniques developed under the LCS(-like) measures are directly applicable to analysis of time series via our reduction of DTW to LIS, we present time-efficient algorithms for DTW-related problems utilizing the semi-local sequence comparison technique developed for LCS-related problems.

Malicious data injection attacks risk mitigation strategy of cyber–physical power system based on hybrid measurements attack detection and risk propagation

Malicious data injection attacks (MDIAs) can tamper with the measurements of the supervisory control and data acquisition (SCADA) system and wide area measurement system (WAMS), which seriously affect the state estimation of cyber–physical power system (CPPS). The existing researches mostly focus on SCADA system, ignoring the vulnerability of phasor measurement unit (PMU) in WAMS, and not considering that MDIA can propagate through the communication network of CPPS as a kind of network attack, which seriously affects the safe operation of CPPS. Aiming at these problem, a risk mitigation strategy based on SCADA/WAMS hybrid measurement networks is proposed. Firstly, a malicious data detection mechanism based on dynamic time warping (DTW) distance is proposed, which includes two parts: state predictor and malicious data identifier. Then, based on the attack detection results, a risk mitigation model is proposed which minimizes the risk level of the system by disabling the damaged meters caused by the propagation of malicious data, while maintaining the observability of CPPS. Finally, the cooperation mechanism of MDIA detection and multi-scenario risk mitigation strategy for hybrid measurement networks is proposed. Case study based on IEEE 14-bus system and IEEE 57-bus system verifies the effectiveness of the risk mitigation strategy proposed in this article.

Clustering based on dynamic time warping to extract typical daily patterns from long-term operation data of a ground source heat pump system

While clustering has been commonly used to profile the building electricity consumption data, its application to HVAC system data is relatively less. Based on the operation data of a residential ground source heat pump (GSHP) system, a pre-processing procedure was set up, including acquisition, cleaning, missing-data fill-in, and standardization. Then hierarchical clustering was used, based on the dynamic time warping (DTW) distance calculation method. A new index, the sum of squares of errors based on DTW, was used to determine the best cluster number. The patterns were extracted based on clustering results.The heat exchange on the user-side during one cooling season was processed as a case study. We obtained 5 valid clusters and extracted patterns from each. For the studied case, the result based on the DTW method yields a better homogeneity compared to the Euclidean method. The method is then applied to a five-year operation data, where 9 and 6 patterns were obtained for the cooling and heating seasons, respectively. They mainly differed in shape, and the cooling patterns fluctuate more. Overall, clustering and pattern extraction can reduce the data dimension, and provide a data-driven view of how the system supplies the cooling and heating.

Weight Loss Trajectories in Healthy Weight Coaching: Cohort Study.

BackgroundAs global obesity prevalence continues to increase, there is a need for accessible and affordable weight management interventions, such as web-based programs.ObjectiveThis paper aims to assess the outcomes of healthy weight coaching (HWC), a web-based obesity management program integrated into standard Finnish clinical care.MethodsHWC is an ongoing, structured digital 12-month program based on acceptance and commitment therapy. It includes weekly training sessions focused on lifestyle, general health, and psychological factors. Participants received remote one-on-one support from a personal coach. In this real-life, single-arm, prospective cohort study, we examined the total weight loss, weight loss profiles, and variables associated with weight loss success and program retention in 1189 adults (963 women) with a BMI >25 kg/m² among participants of the program between October 2016 and March 2019. Absolute (kg) and relative (%) weight loss from the baseline were the primary outcomes. We also examined the weight loss profiles, clustered based on the dynamic time-warping distance, and the possible variables associated with greater weight loss success and program retention. We compared different groups using the Mann-Whitney test or Kruskal-Wallis test for continuous variables and the chi-squared test for categorical variables. We analyzed changes in medication using the McNemar test.ResultsAmong those having reached the 12-month time point (n=173), the mean weight loss was 4.6% (SE 0.5%), with 43% (n=75) achieving clinically relevant weight loss (≥5%). Baseline BMI ≥40 kg/m² was associated with a greater weight loss than a lower BMI (mean 6.6%, SE 0.9%, vs mean 3.2%, SE 0.6%; P=.02). In addition, more frequent weight reporting was associated with greater weight loss. No significant differences in weight loss were observed according to sex, age, baseline disease, or medication use. The total dropout rate was 29.1%. Dropouts were slightly younger than continuers (47.2, SE 0.6 years vs 49.2, SE 0.4 years; P=.01) and reported their weight less frequently (3.0, SE 0.1 entries per month vs 3.3, SE 0.1 entries per month; P<.001).ConclusionsA comprehensive web-based program such as HWC is a potential addition to the repertoire of obesity management in a clinical setting. Heavier patients lost more weight, but weight loss success was otherwise independent of baseline characteristics.

Does LSTM outperform 4DDTW-KNN in lane change identification based on eye gaze data?

Lane change identification based on long short-term memory (LSTM) neural networks has received increasing attention. The most commonly known disadvantage of the approach is the “black box” nature. This study, to improve the interpretability, proposed a novel method for identifying lane changes. The philosophy was to measure the similarity of the glance behavior during the maneuver. The glance behavior in the same maneuver is more similar, while that in different maneuvers is different. First, a driving simulator was used to collect driving behavior data. The eye gaze time series were captured by the eye tracker. The median of the eye gaze sequence of 3 s before the initial moment of lane change was obtained. With the median as the center, the driver's visual field plane was divided into several grids with the side length of 600 pixels. A sliding space-time cuboid algorithm was proposed to extract the scanning path. Second, four-dimensional dynamic time warping (4DDTW) distance was used to measure the similarity of the scanning paths. Third, the K nearest neighbor (KNN) algorithm was used to classify the driving maneuver into the lane-keeping (LK), the right lane change (RLC), and the left lane change (LLC) based on the 4DDTW distance of the scan paths. LSTM was also used to classify the driving maneuver into LK, RLC, and LLC. The performances of 4DDTW-KNN and LSTM were compared. The accuracies of 4DDTW-KNN and LSTM were 86.50% and 86.33%, respectively. LSTM is not better than 4DDTW-KNN in lane change identification based on eye gaze data with a comprehensive consideration of the time efficiency, the accuracy and the interpretability.

Building-Level Urban Functional Area Identification Based on Multi-Attribute Aggregated Data from Cell Phones—A Method Combining Multidimensional Time Series with a SOM Neural Network

Portraying functional urban areas provides useful insights for understanding complex urban systems and formulating rational urban plans. Mobile phone user trajectory data are often used to infer the individual activity patterns of people and for functional area identification, but they are difficult to obtain because of personal privacy issues and have the drawback of a sparse spatial and temporal distribution. Deep learning models have been widely utilized in functional area recognition but are limited by the difficulty of acquiring training samples with large data volumes. This paper aims to achieve a fast and automatic identification of large-scale urban functional areas without prior knowledge. This paper uses Nanjing city as a test area, and a self-organizing map (SOM) neural network model based on an improved dynamic time warping (Ndim-DTW) distance is used to automatically identify the function of each building using mobile phone aggregated data containing work and residence attributes. The results show that the recognition accuracy reaches 88.7%, which is 12.4% higher than that of the K-medoids method based on the DTW distance using a single attribute and 7.8% higher than that of the K-medoids method based on the Ndim-DTW distance with multiple attributes, confirming the effectiveness of the multi-attribute mobile phone aggregated data and the SOM model based on the Ndim-DTW distance. Furthermore, at the traffic analysis zone (TAZ) level, this paper detects that Nanjing has seven functional area hotspots with a high degree of mixing. The results can provide a data basis for urban studies on, for example, the urban spatial structure, the separation of occupations and residences, and environmental suitability evaluation.

Faulty Line Selection Method Based on Comprehensive Dynamic Time Warping Distance in a Flexible Grounding System

A flexible grounding system is a system in which the neutral point of the power supply is grounded via the arc suppression coil in parallel with a low-resistance resistor. When operating normally or a temporary ground fault occurs, the arc suppression coil is used for grounding, whereas the small resistance is switched on when a permanent ground fault occurs. At present, the problem of low protection sensitivity when a high-resistance ground fault occurs in a flexible grounding system has not been solved yet. According to the characteristics of low waveform similarity between the faulty line and the non-faulty line when a single-phase grounding fault occurred, a new faulty line selection method based on a combination of Dynamic Time Warping (DTW) distance and the transient projection method is proposed in this paper. Firstly, the fault transient signal is extracted by a digital filter as a basis for faulty line selection. Secondly, the transient zero-sequence current of each line is projected onto the busbar transient zero-sequence voltage, and the projected DTW distance of each line is calculated. Finally, according to the calculation formula of waveform comprehensive similarity coefficient, the Comprehensive DTW (CDTW) distance is obtained, and the top three CDTW distance values are selected to determine the faulty line. If the maximum value is greater than the sum of the other two CDTW distance values, the line corresponding to the maximum value is judged as the faulty line; otherwise, it is judged as a busbar fault. The simulation results based on MATLAB/Simulink and field data test show that the method can accurately determine the faulty line under diverse fault conditions.

Anomaly Detection Method of Aircraft System using Multivariate Time Series Clustering and Classification Techniques

The paper presents an anomaly detection method that identifies and explains anomalies in an aircraft system based on explainable multivariate time series clustering techniques. The method considers the cyclicity of each variable within the flight phases and selects those behind the anomalies. It combines the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) and a modified Dynamic Time Warping (DTW) distance algorithms to detect abnormal behavioral profiles within the collected flight phases without any prior knowledge on the system's behavior. The proposed method explains those abnormal profiles compared to normal profiles using a new importance score. Profiles are detected using the Time Series Forest (TSF) and the silhouette criterion. The method is trained and tested using a sample from the Bombardier's Aircraft Health Monitoring System. It distinguishes the normal and abnormal behaviors by achieving a clustering silhouette score of 0.95 and detects unknown profiles with a precision of 89%.

Time-Series Clustering Benchmark on Regional Economic Indicator

This paper presents a benchmark study on time-series clustering using regional economic data from the World Bank Open Data (WBOD) repository. It serves as a paradigm for future researchers. This study compares the effectiveness of twenty different techniques for time series grouping. They combine three clustering algorithms (partitional, hierarchical, and fuzzy), two centroids (K-means and K-medoids), and four distance measurements (distance between two points on a graph) (Dynamic time warping, Euclidean, shape-based distance, and global triagonal kernel alignment). The internal clustering validation index will be used to compare the performance of various techniques. Additionally, statistical tests are run on the performance of the pair of approaches to establish whether they can be compared. Across all clustering algorithms evaluated, it was discovered that utilizing K-means as centroids outperformed using K-medoids. When it comes to distance measurements, all clustering algorithms perform optimally, but the Triagonal Global Alignment Kernel is the best of these (except for the fuzzy C-means). Another conclusion reached in this study is that no solution utilizing Dynamic Time Warping and Euclidean distance measures can be compared to another (insignificant Wilcoxon test result). Simultaneously, Shape-Based Distance consistently beats all other approaches to clustering in terms of consistency.

Large-Area Land-Cover Changes Monitoring With Time-Series Remote Sensing Images Using Transferable Deep Models

Dense time-series remote sensing images have transformed the traditional bitemporal land-cover change detection to continuous monitoring. Previous work mostly employs linear fitting, prediction, or decomposition methods, and the detection accuracy is not high. The latest progress of deep learning (DL) shows its advantages in time-series change monitoring. However, DL models are computationally expensive and require lots of labeled samples, resulting in often employed prediction-threshold-based unsupervised change detection method. However, the determination of a reasonable threshold has always been a big problem. Therefore, we proposed the similarity-measurement-based deep transfer learning for time-series adaptive change detection (SDTL-TSACD) model. First, a standard dynamic time warping (SDTW) distance was proposed and used to cluster large-scale time series into multiple subcategories with high time-series similarity. Second, a time convolutional network (TCN) was used for nonlinear time-series fitting and prediction, and an early stop strategy was used to prevent overfitting. Then, the trained TCN model would be transferred and performed pixel-by-pixel time-series prediction within the same category, and the SDTW was also used to evaluate the prediction accuracy. Finally, the Otsu adaptive threshold was used to detect change points, and the spatial neighbor relationship was used to eliminate the pseudo-change points. Change detection results using 132 benchmark datasets showed that the SDTL-TSACD performed well in both accuracy and efficiency. In addition, the MOD13Q1-EVI images from 2001 to 2020 were used to study the land-cover change of the Loess Plateau, and the SDTL-TSACD also showed a good ability to solve practical problems.

Warp: Use of Phase of Fourier Descriptors and Time Warping Distance for Accurate Retrieval of Shapes

Warp: Use of Phase of Fourier Descriptors and Time Warping Distance for Accurate Retrieval of Shapes

Grasshopper optimization algorithm based improved variational mode decomposition technique for muscle artifact removal in ECG using dynamic time warping

The removal of unwanted noise from electrocardiogram (ECG) recordings is a difficult procedure in biomedical signal analysis. It alters the signal and affects the accurate interpretation of the signal. A major distortion produced while recording is due to muscle artifact (MA) or electromyographic (EMG) noise. To achieve a high-quality ECG recording by eliminating this MA noise, our research proposes a novel filtering technique using the grasshopper optimization algorithm (GOA) based variational mode decomposition (VMD) method with the dynamic time warping (DTW) distance concept. GOA is utilized to identify the best optimal parameters for the VMD process, and then the DTW technique is applied to discover the relevant modes that include MA noise. To successfully remove noise, these modes are denoised using the discrete wavelet transform (DWT) method. ECG signal analysis is performed on a real-time MIT-BIH arrhythmia database and is compared with the performance of recent existing techniques using signal characteristics such as signal-to-noise ratio (SNR), mean square error (MSE), correlation coefficient (CC), and so on. As an outcome of these computations, our proposed technique outperforms those existing methods when it comes to denoising MA noise.

Quantitative analysis of fundamental frequency in Spanish (L2) and Brazilian Portuguese (L1)

This paper analyzes the intonation of Spanish and Brazilian Portuguese (BP) produced by monolingual speakers of both languages and bilingual BP speakers that lived in Spain on average for 6 years. Bilinguals produced data in both Spanish L2 (BL2) and BP L1 (BL1). Speech materials are sentences in different modalities (declaratives, yes-no and wh-questions) and reading styles (isolated sentences and storytelling). Fundamental frequency (f0) contours were analyzed to assess learning in Spanish L2 and language attrition in the L1 production of bilinguals. Variability in the f0 contours of the four language conditions was gauged by means of three indices (peak rate, peak range and global standard deviation). Dynamic time warping (DTW) distances between pairs of f0 contours were also measured as a way to measure within- and between-language differences in intonation patterns. The main results are: 1) BL2 and BL1 contours are significantly more variable than the monolingual ones both quantitatively and qualitatively; 2) BL2 contours partially converge towards the patterns of Spanish monolinguals, showing that there is learning; 3) there is evidence for language attrition in the form of transfer of Spanish patterns to BP contours produced by bilinguals; 4) Learning and attrition levels are different depending on sentence modality, such that learning is greater in modalities that differ less between BP and Spanish and attrition is greater in modalities that differ the most.

Research on Semantic Similarity of Short Text Based on Bert and Time Warping Distance

The research on semantic similarity of short text plays an important role in machine translation, emotion analysis, information retrieval and other AI business applications. However, according to existing short text similarity research, the characteristics of ambiguous vocabularies are difficult to be effectively analyzed, the solution of the problem caused by words order needs to be further optimized as well. This paper proposes a short text semantic similarity calculation method that combines BERT and time warping distance algorithm, in order to solve the problem of vocabulary ambiguity. The model first uses the pre trained Bert model to extract the semantic features of the short text from the whole level, and obtains a 768 dimensional short text feature vector. Then, it transforms the extracted feature vector into a point sequence in space, uses the CTW algorithm to calculate the time warping distance between the curves connected by the point sequence, and finally uses the weight function designed by the analysis, according to the smaller the time warpage distance is, the higher the degree of small similarity is, to calculate the similarity between short texts. The experimental results show that this model can mine the feature information of ambiguous words, and calculate the similarity of short texts with lexical ambiguity effectively. Compared with other models, it can distinguish the semantic features of ambiguous words more accurately.

Identifying App-Based Meditation Habits and the Associated Mental Health Benefits: Longitudinal Observational Study.

BackgroundBehavioral habits are often initiated by contextual cues that occur at approximately the same time each day; so, it may be possible to identify a reflexive habit based on the temporal similarity of repeated daily behavior. Mobile health tools provide the detailed, longitudinal data necessary for constructing such an indicator of reflexive habits, which can improve our understanding of habit formation and help design more effective mobile health interventions for promoting healthier habits.ObjectiveThis study aims to use behavioral data from a commercial mindfulness meditation mobile phone app to construct an indicator of reflexive meditation habits based on temporal similarity and estimate the association between temporal similarity and meditation app users’ perceived health benefits.MethodsApp-use data from June 2019 to June 2020 were analyzed for 2771 paying subscribers of a meditation mobile phone app, of whom 86.06% (2359/2771) were female, 72.61% (2012/2771) were college educated, 86.29% (2391/2771) were White, and 60.71% (1664/2771) were employed full-time. Participants volunteered to complete a survey assessing their perceived changes in physical and mental health from using the app. Receiver operating characteristic curve analysis was used to evaluate the ability of the temporal similarity measure to predict future behavior, and variable importance statistics from random forest models were used to corroborate these findings. Logistic regression was used to estimate the association between temporal similarity and self-reported physical and mental health benefits.ResultsThe temporal similarity of users’ daily app use before completing the survey, as measured by the dynamic time warping (DTW) distance between app use on consecutive days, significantly predicted app use at 28 days and at 6 months after the survey, even after controlling for users’ demographic and socioeconomic characteristics, total app sessions, duration of app use, and number of days with any app use. In addition, the temporal similarity measure significantly increased in the area under the receiver operating characteristic curve (AUC) for models predicting any future app use in 28 days (AUC=0.868 with DTW and 0.850 without DTW; P<.001) and for models predicting any app use in 6 months (AUC=0.821 with DTW and 0.802 without DTW; P<.001). Finally, a 1% increase in the temporal similarity of users’ daily meditation practice with the app over 6 weeks before the survey was associated with increased odds of reporting mental health improvements, with an odds ratio of 2.94 (95% CI 1.832-6.369).ConclusionsThe temporal similarity of the meditation app use was a significant predictor of future behavior, which suggests that this measure can identify reflexive meditation habits. In addition, temporal similarity was associated with greater perceived mental health benefits, which demonstrates that additional mental health benefits may be derived from forming reflexive meditation habits.

A novel automatic phenology learning (APL) method of training sample selection using multiple datasets for time-series land cover mapping

The long record of Landsat imagery, which is the cornerstone of Earth observation, provides an opportunity to monitor land use and land cover (LULC) change and understand the interactions between the climate and earth system through time. A few change detection algorithms such as Continuous Change Detection and Classification (CCDC) have been developed to utilize all available Landsat images for change detection and characterization at local or global scales. However, the reliable, rapid, and reproducible collection of training samples have become a challenge for time series land cover classification at a large scale. To meet the challenge, we proposed an automatic phenology learning (APL) method with the assumption that the temporal profiles of samples within the same land cover type are the same or similar at a local scale to generate evenly distributed training samples automatically. We designed the method to build land cover patterns for each category based on consensus samples derived from multiple existing scientific datasets including LANDFIRE's (LF) Existing Vegetation Type (EVT), USGS National Land Cover Database (NLCD), National Agricultural Statistics Service (NASS) Cropland Data Layer (CDL), and National Wetlands Inventory (NWI). Then we calculated the Time-Weighted Dynamic Time Warping (twDTW) distance between any undefined samples and land cover patterns in the same geographical region as prior knowledge. Finally, we selected the optimal land cover category for each undefined sample from the land cover products based on the designed criteria iteratively using the twDTW distance as an indicator. The method was applied in the footprint of 10 selected Landsat Analysis Ready Data (ARD) tiles in the eastern and western conterminous United States (CONUS) to produce annual land cover maps from 1985 to 2017. The accuracy assessment and visual comparison revealed that the APL method can generate reliable training samples without any manual interpretation, producing better land cover results especially for the grass/shrub and wetland land cover classes. Applying the APL method, the overall accuracy of the annual land cover maps was improved by 2% over the accuracy of Land Change Monitoring, Assessment, and Projection (LCMAP) Collection 1.0 Science Products in the research regions. Our results also indicate that the APL method provides an approach for best use of different land cover products and meets the requirement of intensive sampling for training data collection.

A time series clustering based approach for construction of real-world drive cycles

• A comprehensive, flexible approach and a generic framework for the construction of drive cycles. • A new definition of micro-trip to capture the stop-and-go driving behavior related emissions. • Time series shape based distances in the clustering process to retain real world transients. • The idea of a set of representative drive cycles for any region instead of one. Building representative real world drive cycles is an important component in the modelling of emissions, battery health of electric vehicles and autonomous vehicles. All these applications are sensitive to the transients and diversity present in real world driving patterns, which are not adequately captured by current approaches. To address this lacuna, we use clustering techniques involving time-series (shape) based distances on the raw data directly to obtain representative sets of real world drive cycles. We demonstrate the efficacy of our approach using experimental data from a fleet of eight motorcycles run across five locations in India. Dynamic Time Warping (DTW) distance based clustering gives optimal results. We give theoretical and experimental justification for our constructions. We believe that the constructed drive cycles using the proposed approach would help in assessing the impact of various policies aimed at building eco-friendly transportation systems.

Vegetable mapping using fuzzy classification of Dynamic Time Warping distances from time series of Sentinel-1A images

Vegetable production is important because of the food security, diet improvement and socio-economic value. Mapping the location and extent of vegetable fields is therefore important in agricultural policy, food security and farmer support. Dynamic Time Warping (DTW) is a common way to map crops from time series of satellite images. However, as all hard classifications, it does not show the spatial distribution of uncertainty in the classification. In fuzzy classification, where memberships to multiple classes are assigned to each pixel, differences in membership between the first and the runners-up class can be used to assess classification uncertainty at the pixel level. This research formulates a fuzzy classifier based upon Time-Weighted Dynamic Time Warping (TWDTW) distances to map vegetable types from time series of Sentinel-1A SAR images. For each pixel, the TWDTW distances to the classes was normalised by dividing them by the sum of all TWDTW distances to all the classes for that pixel. The normalized distances were then used to compute fuzzy memberships for each pixel to each class, using the Gaussian membership function. Based on these memberships, fuzzy measures such as Confusion Index (CI), Ambiguity Index (AI), fuzziness and fuzzy membership were calculated and different thresholds applied on each of the measures during subsequent defuzzification. The overall accuracy and kappa coefficient of the defuzzified output results were 0.86 and 0.83, respectively, which was an improvement with regard to the crisp Time-Weighted Dynamic Time Warping with SPRING strategy for subsequence searching (TWDTWS) algorithm with 0.73 and 0.68 for overall accuracy and kappa, respectively. This study concludes that this new approach improves classification accuracy in image classification by excluding pixels with high uncertainty, which is especially relevant when only a limited number of classes are sampled and mapped.

Fast data series indexing for in-memory data

Data series similarity search is a core operation for several data series analysis applications across many different domains. However, the state-of-the-art techniques fail to deliver the time performance required for interactive exploration, or analysis of large data series collections. In this work, we propose MESSI, the first data series index designed for in-memory operation on modern hardware. Our index takes advantage of the modern hardware parallelization opportunities (i.e., SIMD instructions, multi-socket and multi-core architectures), in order to accelerate both index construction and similarity search processing times. Moreover, it benefits from a careful design in the setup and coordination of the parallel workers and data structures, so that it maximizes its performance for in-memory operations. MESSI supports similarity search using both the Euclidean and dynamic time warping (DTW) distances. Our experiments with synthetic and real datasets demonstrate that overall MESSI is up to 4x faster at index construction and up to 11x faster at query answering than the state-of-the-art parallel approach. MESSI is the first to answer exact similarity search queries on 100GB datasets in \(\sim \)50 ms (30–75 ms across diverse datasets), which enables real-time, interactive data exploration on very large data series collections.