Alignment Of Time Series Research Articles

Analysis of the development of organic crop production in Latin America

The paper is devoted to the study of the process of development of organic production as a global trend in agriculture. The objects of the research are the countries of Latin America, as territories that are promising in terms of large-scale and rapidly developing organic agriculture for export and domestic consumption — Uruguay, Paraguay, Brazil, Argentina, Chile, Mexico, Dominican Republic. The countries have large land funds and favorable agroclimatic conditions in most of their territories. In the work, based on the application of methods of time series alignment and the graphical method, trends in demand for the purchase and subsequent use of pesticides, the prevalence of malnutrition in Latin America and in the world, as well as prospects for the development of organic crop production by such an indicatosr as the total area of organic lands and number of producers are identified. The simple linear regression model constructed by the authors made it possible to establish the relationship between the volumes of greenhouse gas emissions and the average annual temperature, which made it possible to establish the need to reduce the use of synthetic fertilizers and pesticides to achieve sustainable development goals. Calculation of the linear and quadratic coefficients of structural shifts, as well as the index of differences, made it possible to identify structural changes in priority groups of crops for growing organic crops.

Analysis of the prospects for the development of organic pig farming in Russia

The paper is devoted to the study of factors that form the conditions for the development of organic pig production. The relevance of this topic is due to the growing interest of consumers both domestically and abroad in organic products as the basis for a healthy lifestyle and the resource potential available in Russia, which is sufficient to form a sector of producers — competitive participants in the world market. Based on the use of statistical methods for time series alignment, the authors have identified positive trends in traditional pig production, meat consumption and outpacing growth rates of demand for pig products. It has been determined that stable growth of real incomes of the population in all constituent entities of the Russian Federation can become the basis for the growth of effective demand for organic pig products. The paper summarizes the requirements for the organization of organic production in pig farming, taking into account internal regulatory documentation and foreign experience. Previous studies conducted by the authors showed that the existing soil and climatic conditions in the regions of the Russian Federation make it possible to produce organic feed for this industry, and the expansion of production is one of the target indicators of the Strategy for the Development of Organic Agriculture in the Russian Federation until 2030. The authors identified a number of risks in the work that must be taken into account when creating projects for the development of organic pig farming at the micro and macro levels.

Population of Russia: demographic situation, problems and prospects

Analysis of the size, structure and age-sex composition of the population of Russia is the most important task of modern statistics, providing government agencies with the necessary information for developing programs and plans for the development of key sectors of the economy. In times of economic downturns and recessions, it is important for authorities to have reliable and complete information about the state of the population, since it is the population size and its structure that are the basis of human resources for the development of sectors of the economy, including agriculture, which ensures food security and preservation of the population, decides strategic development tasks for the country as a whole. The study is aimed at analyzing the state and identifying patterns of changes in the Russian population, population distribution, identifying trends in the dynamics of the working-age and elderly population, and analyzing the gender and age pyramid of Russia for 2023. To conduct a comprehensive study, the following methods were used: calculation of demographic coefficients, alignment of time series. The results of the study made it possible to propose directions for improving measures within the framework of the existing demographic policy of the state to ensure uniform distribution of the population throughout the country, reduce the tax burden on individuals, and increase labor productivity.

The EDIBLES Survey

Context. There have been many attempts to identify families of diffuse interstellar bands (DIBs) with perfectly correlating band strengths. Although major efforts have been made to classify broadly based DIB families and important insights have been gained, no family has been identified with sufficient accuracy or statistical significance to prove that a series of selected DIBs originates from the same carrier. This can be attributed in part to the exclusive use of equivalent widths to establish DIB families. Aims. In a change of strategy, we search for DIBs that are highly correlated in both band strength and profile shape. This approach increases the chance of correlating DIBs being members of one family and originating from the same carrier molecule. We also search for correlations between DIB profile families and atomic interstellar lines, with the goal of further chemically constraining possible DIB carriers. Methods. We adapted the well-known method of time-series alignment to perform a spectral alignment; that is, DIB alignment. In a second step, we analysed the alignment results using a clustering analysis. This method required a statistically significant data set of DIB sight lines. The ESO Diffuse Interstellar Bands Large Exploration Survey (EDIBLES) data were perfectly suited for this application. Results. We report eight DIB families with correlating strengths and profiles, as well as four previously unreported DIBs in the visual range, found using DIB alignment. All profile family members show Pearson correlation coefficients in band strength higher than 0.9. In particular, we report the 6614–6521 Å DIB pair, in which both DIBs show the same triple-peak substructure and an unprecedented band strength Pearson correlation coefficient of 0.9935. The presented approach opens up new perspectives that can guide the laboratory search for DIB carriers.

Efficient calibration of cost-efficient particulate matter sensors using machine learning and time-series alignment

Atmospheric particulate matter (PM) poses a significant threat to human health, infiltrating the lungs and brain and leading to severe issues such as heart and lung diseases, cancer, and premature death. The main sources of PM pollution are vehicular and industrial emissions, construction and agricultural activities, and natural phenomena such as wildfires. Research underscores the absence of a safe threshold for particulate exposure, highlighting the crucial need for monitoring PM levels to develop and implement effective risk mitigation measures. Notwithstanding, accurate measurement of PM concentration relies on expensive and cumbersome equipment. Despite the rising popularity of low-cost alternatives, their reliability remains questionable, given their sensitivity to environmental conditions, inherent instability, and manufacturing imperfections. This article proposes a novel approach to efficient correction of low-cost PM sensors. The primary calibration model is a feedforward artificial neural network (ANN), which directly renders predicted output of the corrected sensor based on environmental variables such as temperature, humidity, and atmospheric pressure. The ANN hyper-parameters are identified by aligning time series of prior reference and low-cost sensor readings, which enables the network to learn typical temporal changes of the sensor outcome as a function of the aforementioned parameters as well as operational relationships between the sensor and the reference device. The architecture of the ANN is optimized in terms of the number of neurons in each layer to enhance its generalization capability. Our methodology has been demonstrated using a custom-designed portable monitoring platform and reference data acquired from public stations in Gdansk. The results are indicative of excellent calibration reliability. The achieved correlation coefficients w.r.t. the reference readings are 0.86, 0.88, and 0.72 for PM1, PM10, and PM2.5, respectively, whereas RMSE values are only 3.0, 3.9, and 5.4 µg/m³.

A pre-whitening with block-bootstrap cross-correlation procedure for temporal alignment of data sampled by eddy covariance systems

The eddy covariance (EC) method is a standard micrometeorological technique for monitoring the exchange rate of the main greenhouse gases across the interface between the atmosphere and ecosystems. One of the first EC data processing steps is the temporal alignment of the raw, high frequency measurements collected by the sonic anemometer and gas analyser. While different methods have been proposed and are currently applied, the application of the EC method to trace gases measurements highlighted the difficulty of a correct time lag detection when the fluxes are small in magnitude. Failure to correctly synchronise the time series entails a systematic error on covariance estimates and can introduce large uncertainties and biases in the calculated fluxes. This work aims at overcoming these issues by introducing a new time lag detection procedure based on the assessment of the cross-correlation function (CCF) between variables subject to (i) a pre-whitening based on autoregressive filters and (ii) a resampling technique based on block-bootstrapping. Combining pre-whitening and block-bootstrapping facilitates the assessment of the CCF, enhancing the accuracy of time lag detection between variables with correlation of low order of magnitude (i.e. lower than -1\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$-1$$\\end{document}) and allowing for a proper estimate of the associated uncertainty. We expect the proposed procedure to significantly improve the temporal alignment of the EC time-series measured by two physically separate sensors, and to be particularly beneficial in centralised data processing pipelines of research infrastructures (e.g. the Integrated Carbon Observation System, ICOS-RI) where the use of robust and fully data-driven methods, like the one we propose, constitutes an essential prerequisite.

Research on Dual GRU model driving by time series alignment for multi-step prediction of ash content of dense medium clean coal

ABSTRACT Dense medium cyclone separation is a widely used coal processing method. However, duetween the ash content measurement and the other sensors. However, existing studies either do not consider the impact of time lag or only conduct single-step ash prediction, making it difficult to consider the systematic changes and uncertainties of multiple time steps in the future entirely. To solve these problems, first, this study uses time lag correlation to quantify and eliminate the lag effect between each variable and the ash content. Secondly, this study designs a Dual GRU model combining time series and regression prediction. Finally, this approach is adapted to some of the time series forecasting models proposed in recent years to compare with the model proposed in this study. Experiments conducted on production data from a coal preparation plant showed that input and prediction lengths were determined to be 7 based on the time lag correlation. Then, the mean absolute error of the 7-step ash prediction was 0.1495. The MAE value was reduced by 4.66% compared to the model without time series alignment. Compared to the model without aligned process variables, the MAE value was reduced by 6.77%. Compared to Informer, the best performer among the Transformer-based models, Dual GRU accuracy is improved by about 7.94%. Time series alignment and Dual GRU provide technical support for realizing density control based on ash content multi-step prediction.

ECCO Grant Untangling the gut phagosome dynamics of the relapse/remission cycle in ulcerative colitis

Abstract Background and Aims The course of ulcerative colitis (UC) is usually characterized by intertwining periods of remission and relapse. Most microbiome UC studies focus on the gut bacteriome. However, recent studies on animal models suggest that phage-mediated bacterial cell lysis may play a role in sustaining gut inflammation, proving the phageome (collection of bacteriophages) should not be ignored in this disease. Therefore, this project aims to: 1) characterize the gut phageome of patients with UC who suffer from frequent relapses, 2) compare temporal dynamics of gut the phageome between patients and healthy controls, 3) To model interactions between bacteriome, mycobiome, phageome on one hand, and lifestyle factors (diet, stress levels, medication) on the other 4) To infer host status (relapse versus remission of the disease) based on phageome composition, and using interactions inferred in Objective 3. Methods The cohort comprises 24 patients with UC (14 of whom suffered from relapse) and 10 healthy controls, with stool samples collected monthly for a year. Phageome profiling will be achieved via shotgun sequencing of phage genomes from the same stool samples. This dataset will be interrogated together with bacterial (16S rRNA) and fungal (ITS) amplicon data, which have just been generated. Available metadata include disease symptoms, diet, stress levels, work productivity, medication, and faecal calprotectin. Dynamic time-wrapping algorithms will be used to align time series of microbial members with similar temporal dynamics. Dynamic Bayesian Network modelling will infer interactions between microbiome members and predict future relapse. Anticipated Impact The phageome is the often neglected (“dark matter”) part of the microbiome, but has defining interactions with bacteria. We will expand knowledge on inter-species dynamics during relapse and remission cycles of UC. This will potentially lead to new microbiome-related relapse predictors, which may contribute to new prognostic tools in UC, and potentially inform therapeutic avenues.

ESDTW: Extrema-based shape dynamic time warping

As a nonlinear distance measurement, Dynamic time warping (DTW) is widely used to solve the alignment problem of time series. However, due to local distortion of signal, noise interference, and high computational cost, DTW and its variants present some limitations in terms of classification accuracy or speed. In this paper, we develop the extrema-based shape dynamic time warping (ESDTW) for the purpose of fast and accurate alignment of time series. The proposed ESDTW introduces local extrema to represent the original time series, and further uses DTW to align the descriptors of extrema shape. By considering the location information of local extrema, the proposed measure avoids unreliable time series alignment caused by local maxima and minima separately in LEDTW. The experiments on some artificially simulated pairs of aligned sequences demonstrate that ESDTW can obtain more accurate warping paths than comparison methods. Moreover, combining with the nearest neighbor classifier, ESDTW achieves higher classification accuracy than other distance measures on the 84 UCR time series datasets. We also confirm significant computational efficiency gains on long time series.

Somtimes: self organizing maps for time series clustering and its application to serious illness conversations

There is demand for scalable algorithms capable of clustering and analyzing large time series data. The Kohonen self-organizing map (SOM) is an unsupervised artificial neural network for clustering, visualizing, and reducing the dimensionality of complex data. Like all clustering methods, it requires a measure of similarity between input data (in this work time series). Dynamic time warping (DTW) is one such measure, and a top performer that accommodates distortions when aligning time series. Despite its popularity in clustering, DTW is limited in practice because the runtime complexity is quadratic with the length of the time series. To address this, we present a new a self-organizing map for clustering TIME Series, called SOMTimeS, which uses DTW as the distance measure. The method has similar accuracy compared with other DTW-based clustering algorithms, yet scales better and runs faster. The computational performance stems from the pruning of unnecessary DTW computations during the SOM’s training phase. For comparison, we implement a similar pruning strategy for K-means, and call the latter K-TimeS. SOMTimeS and K-TimeS pruned 43% and 50% of the total DTW computations, respectively. Pruning effectiveness, accuracy, execution time and scalability are evaluated using 112 benchmark time series datasets from the UC Riverside classification archive, and show that for similar accuracy, a 1.8×\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ imes$$\\end{document} speed-up on average for SOMTimeS and K-TimeS, respectively with that rates vary between 1×\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ imes$$\\end{document} and 18×\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ imes$$\\end{document} depending on the dataset. We also apply SOMTimeS to a healthcare study of patient-clinician serious illness conversations to demonstrate the algorithm’s utility with complex, temporally sequenced natural language.

Trends in the development of dairy farming and its state financial support in Ukraine

The importance of dairy farming for food security and the formation of added value in agriculture necessitates a rational combination of market self-regulation mechanisms and instruments of state intervention in economic processes in the sector. The purpose of this study was to assess the development of dairy farming in Ukraine and identify ways to improve its state support. The following methods of scientific cognition were employed in the study: analysis and synthesis, induction and deduction, abstract-logical, historical, comparative analysis, monographic, computational-constructive, graphical, tabular, statistical analysis, generalisation, causal and retrospective analysis, analytical alignment of time series, economic modelling, and forecasting. Based on the results of using these methods, the study summarised the current trends in the development of dairy farming in Ukraine and its state financial support using tax and budgetary regulation instruments. The study covered the specifics of the existing structure of dairy farming in Ukraine, which, unlike the countries of the European Union, is represented by three categories of economic entities. It was found that since the early 2010s, Ukraine has been experiencing processes of concentration in the field of dairy farming. The study found a direct correlation between the size of the number of cows kept and the profitability of dairy farming. It was determined that enterprises with up to 100 cows gradually reduce the volume of activities in the dairy farming sector due to its unprofitability. The study assessed changes in the structure of dairy farms in European countries from the mid-1990s to the early 2020s and found that they are also characterised by trends in the concentration of dairy farming. The efficiency of the state financial support for the development of dairy cattle breeding in Ukraine through the instruments of tax and budgetary regulation in the pre-war period and in the martial law regime was assessed. The paper described the advantages and disadvantages of the applied instruments of budgetary support for dairy farming, substantiated proposals for the use of instruments of tax and budgetary stimulation of dairy farming development in the period of post-war economic recovery of Ukraine, considering the existing trends in the development of dairy farming in Ukraine. The practical significance of this study lies in the possibility of using the obtained results in the drafting of policy documents on the post-war development of the agro-industrial complex of Ukraine in general and specifically its dairy subcomplex

Dynamic time warping to quantify age distortion in firn cores impacted by melt processes

Abstract As warming intensifies across the Greenland ice sheet, an increasing number of shallow coring and radar studies are targeting the melt-impacted firn column to investigate meltwater processes. Highly inhomogeneous infiltration and refreezing, however, redistributes mass, distorting age–depth relationships and confounding comparisons between different cores. Here, we utilize a dynamic time warping algorithm for time series alignment to investigate and quantify the heterogeneous impact of melt processes on nine closely spaced (within 50 m) firn core-density profiles. The 10 m cores were collected relatively high in Greenland's percolation zone, where melt alteration is minimal compared to lower elevation. Our analysis demonstrates the effectiveness of dynamic time warping as a tool for assessing heterogeneity between ice core records. We find that the optimal alignment of density profiles in the nine cores requires vertical stretching and compression of individual profiles, ranging from, on average, <1 to ~16% of the core lengths. We identified four depth zones of mass redistribution that appear to coincide with observed ice layers. Further, ~75% of density measurements from each core do not align with an age model-derived density profile that assumes no mass redistribution of meltwater, indicating the pervasive impact of melt processes.

Cross-Chamber Data Transferability Evaluation for Fault Detection and Classification in Semiconductor Manufacturing

Unit-to-unit variation among the production chambers is a long-lasting challenge for Fault Detection and Classification (FDC) development in the semiconductor industry. Currently, various methods are applied for knowledge transfer among chambers and generalized FDC model development. However, the existing methods cannot give a quantitative or qualitative measure for cross-chamber data transferability evaluation. This research proposes a novel methodology for data transferability evaluation and important sensor screening, which can serve as a data quality evaluation tool for any FDC model. In this research, firstly, Time Series Alignment Kernel (TSAK) is incorporated into Multidomain Discriminant Analysis (MDA) algorithm to achieve sensor-based domain generalization. Then, domain-invariant features are directly extracted for sensor visualization. After that, Fisher’s criterion ratios of the labeled good wafer samples and defective ones are computed based on the domain-invariant features of each sensor to quantitatively estimate how easy it is to transfer knowledge of each sensor among chambers, i.e., data transferability evaluation. Lastly, the proposed method develops a Recursive Feature Elimination (RFE)-based sensor selection algorithm to qualitatively analyze the importance of each sensor channel and identify the critical sensor subset. In this study, validation of the proposed method is based on two open-source datasets from real production lines.

Phenology Alignment-Based PolSAR Crop Classification Considering Polarimetric Statistical and Time-Varying Curve Characteristics

The uncertainty of crop phenological cycle is an important issue in crop classification with time series PolSAR data. The time series alignment algorithm represented by dynamic time warping (DTW) can supply a potential solution, which realigns curves based on shape matching, dealing with the distortion of feature curves caused by uncertain crop phenological development. However, previous studies mainly focused on shape characteristics of time-varying feature curves, which is hard to comprehensively evaluate the similarity degree of crop phenological cycles. Furthermore, it ignored the differences in scattering signal and polarimetric statistical distribution of crops, which limited the accuracy of crop classification. In this letter, a novel crop classification method based on phenology alignment is proposed. Firstly, the dual-branch time series alignment method is proposed, including the time-weighted dynamic time warping (TWDTW) alignment and the Wishart distance-based TWDTW (WD-TWDTW) alignment, which combines the feature curve characteristics and the polarimetric statistical information to correctly describe the similarity degree of phenological cycles. Secondly, a multi-similarity measure (including shape similarity, feature similarity and polarimetric similarity) is defined to improve capacity of crop discrimination. The multi-similarity measure can describe the differences of crop types from three aspects, including crop growth trend, growth status, and statistical distribution. The proposed method is evaluated with time series full-polarization Radarsat-2 data in Flevoland area. The results show that our method is superior to traditional method with single TWDTW alignment and shape similarity, and the corresponding overall accuracy is improved by 6%.

Learning an autoencoder to compress EEG signals via a neural network based approximation of DTW

In this paper, we propose a method to compress electroencephalography (EEG) data using a differentiable neural network trained with the dynamic time warping (DTW) algorithm. DTW is a popular method for aligning time series data, but it is computationally expensive and not differentiable, which makes it difficult to optimize using gradient-based methods. To circumvent this difficulty, we use a neural network based approximation of DTW as the reconstruction loss to learn a convolutional auto-encoder to compress EEG signals. We show that using a custom DTW as reconstruction loss for the autoencoder improves the sleep staging score of the reconstructed data, suggesting that it allows to better preserve important features of EEG data. We also show that these findings can be extended to multiple datasets and compression ratios.

Spatial-temporal alignment of time series with different sampling rates based on cellular multi-objective whale optimization

Aligning time series of different sampling rates is an important but challenging task. Current commonly used dynamic time warping methods usually suffer from pathological temporal singularity problem. In order to overcome this, we transform the alignment task to a spatial-temporal multi-objective optimization (MOO) problem. Existing MOO algorithms are always inefficient in finding Pareto optimal alignment solutions due to their insufficiency in maintaining convergence and diversity among the obtained Pareto solutions. In light of this, we propose a novel and efficient MOO algorithm Cell-MOWOA which integrates Cellular automata with the rising Whale Optimization Algorithm to find Pareto optimal alignment solutions. Innovative multi-variate non-linear cell state evolutionary rules are designed within Pareto solution external archive to improve the convergence and diversity of the Pareto solutions, and novel whale population updating mechanism is designed to accelerate the convergence to the Pareto front. Besides, new integer whale updating mechanism is presented to transform real-number whale solutions to integer whale solutions. Experimental results on 85 gold-standard UCR time series datasets showed that Cell-MOWOA outperformed six state-of-the-art baselines by 24.53% in average in increasing alignment accuracy and 42.66% in average in reducing singularity. Besides, it achieved outstanding runtime efficiency, especially on long time series datasets.

Temporal Characterization and Filtering of Sensor Data to Support Anomaly Detection

We present an approach for characterizing complex temporal behavior in the sensor measurements of a system in order to support detection of anomalies in that system. We first characterize typical behavior by extending a hidden Markov model-based approach to time series alignment. We then use a trace of that learned behavior to develop a particle filter that enables efficient estimation of the filtering distribution on the state space. This produces filtered residuals that can then be used in an anomaly detection framework. Our motivating example is the daily behavior of a building’s heating, ventilation, and air conditioning (HVAC) system, using sensor measurements that arrive every minute and induce a state space with 15,120 states. We provide an end-to-end demonstration of our approach showing improved performance of anomaly detection after application of alignment and filtering compared to the unaligned data. The proposed model is implemented as a computationally efficient R package alignts (align time series) built with R and Fortran 95 with OpenMP support.

Analytical time series alignment liquefaction number of corn starch mixture

The statistical description of the development of dynamic processes in time is carried out using time series. To eliminate random fluctuations and build an analytical function of the trend of the time series, an analytical alignment procedure is used. The choice of the type of the trend function is carried out by the method of finite differences, the calculation of the trend parameters ‒ by the method of least squares. The purpose of this work was the analytical alignment of the time series of the liquefaction number of the corn starch mixture obtained in the experiment on the PChP-99 device. The mechanism of starch gelatinization with a given liquefaction rate under such conditions requires additional theoretical and experimental study. Calculations have shown that the process of liquefaction of starch gel corresponds to the exponential trend equation: y = a • ebt, which is a particular case of an exponential trend. It has been experimentally established that an increase in the proportion of amylopectin starch in a corn starch mixture leads to an increase in the maximum viscosity of the resulting gel when the swelling water-starch suspension is heated. In the process of further gelatinization of corn starch mixture with an increase in the proportion of amylopectin starch, the strength of the gel decreases due to the preservation of the mobility of water molecules during the transition in the sol-gel system, which contributes to an increase in liquefaction number. Experimental data do not contain anomalous values; the error in approximating the regression equation for the trend of the time series is less than 5 %. The statistical significance of the coefficients of the linearized trend equation is proved in favor of the hypothesis of the existence of a time series. The obtained estimates of the regression equation make it possible to use it for predictive purposes, providing an accuracy of up to 95.42 % of the total variability of the liquefaction number in the absence of autocorrelation of first-order residues. Checking the normality of the distribution of the residual component according to the RS test showed the adequacy of the trend model, the hypothesis of the absence of heteroscedasticity according to the Spearman and Goldfeld-Quandt tests is accepted.

Impact of River-Reservoir Hybrid System on Zooplankton Community and River Connectivity

Anthropogenic connectivity regulation in rivers, such as via weirs and dams, affects the plankton community. We hypothesized that the longitudinal similarity of the zooplankton community in a river could change in a river–reservoir hybrid system (RRHS). The impact of weir construction on zooplankton communities in terms of species diversity, abundance, and community structure was examined biweekly at six sites on the Nakdong River for 14 years (before construction: 2002–2008; after construction: 2012–2018). We checked time-series alignment using a dynamic time-warping method between longitudinal survey sites. After RRHS, the zooplankton community showed an increasing number of species. However, RRHS decreased the longitudinal similarity in terms of number of zooplankton species and population density. Our results demonstrate the negative effect of lateral infrastructures on zooplankton populations due to river fragmentation and habitat alteration.

Non-Parallel Articulatory-to-Acoustic Conversion Using Multiview-Based Time Warping

In this paper, we propose a novel algorithm called multiview temporal alignment by dependence maximisation in the latent space (TRANSIENCE) for the alignment of time series consisting of sequences of feature vectors with different length and dimensionality of the feature vectors. The proposed algorithm, which is based on the theory of multiview learning, can be seen as an extension of the well-known dynamic time warping (DTW) algorithm but, as mentioned, it allows the sequences to have different dimensionalities. Our algorithm attempts to find an optimal temporal alignment between pairs of nonaligned sequences by first projecting their feature vectors into a common latent space where both views are maximally similar. To do this, powerful, nonlinear deep neural network (DNN) models are employed. Then, the resulting sequences of embedding vectors are aligned using DTW. Finally, the alignment paths obtained in the previous step are applied to the original sequences to align them. In the paper, we explore several variants of the algorithm that mainly differ in the way the DNNs are trained. We evaluated the proposed algorithm on a articulatory-to-acoustic (A2A) synthesis task involving the generation of audible speech from motion data captured from the lips and tongue of healthy speakers using a technique known as permanent magnet articulography (PMA). In this task, our algorithm is applied during the training stage to align pairs of nonaligned speech and PMA recordings that are later used to train DNNs able to synthesis speech from PMA data. Our results show the quality of speech generated in the nonaligned scenario is comparable to that obtained in the parallel scenario.