Simple Extensible Deep Learning Model for Automatic Arabic Diacritization

Automatic diacritization is an Arabic natural language processing topic based on the sequence labeling task where the labels are the diacritics and the letters are the sequence elements. A letter can have from zero up to two diacritics. The dataset used was a subset of the preprocessed version of the Tashkeela corpus. We developed a deep learning model composed of a stack of four bidirectional long short-term memory hidden layers of the same size and an output layer at every level. The levels correspond to the groups that we classified the diacritics into (short vowels, double case-endings, Shadda, and Sukoon). Before training, the data were divided into input vectors containing letter indexes and outputs vectors containing the indexes of diacritics regarding their groups. Both input and output vectors are concatenated, then a sliding window operation with overlapping is performed to generate continuous and fixed-size data. Such data is used for both training and evaluation. Finally, we realize some tests using the standard metrics with all of their variations and compare our results with two recent state-of-the-art works. Our model achieved 3% diacritization error rate and 8.99% word error rate when including all letters. We have also generated the confusion matrix to show the performances per output and analyzed the mismatches of the first 500 lines to classify the model errors according to their linguistic nature.

Human Weight Compensation With a Backdrivable Upper-Limb Exoskeleton: Identification and Control

Active exoskeletons are promising devices for improving rehabilitation procedures in patients and preventing musculoskeletal disorders in workers. In particular, exoskeletons implementing human limb’s weight support are interesting to restore some mobility in patients with muscle weakness and help in occupational load carrying tasks. The present study aims at improving weight support of the upper limb by providing a weight model considering joint misalignments and a control law including feedforward terms learned from a prior population-based analysis. Three experiments, for design and validation purposes, are conducted on a total of 65 participants who performed posture maintenance and elbow flexion/extension movements. The introduction of joint misalignments in the weight support model significantly reduced the model errors, in terms of weight estimation, and enhanced the estimation reliability. The introduced control architecture reduced model tracking errors regardless of the condition. Weight support significantly decreased the activity of antigravity muscles, as expected, but increased the activity of elbow extensors because gravity is usually exploited by humans to accelerate a limb downwards. These findings suggest that an adaptive weight support controller could be envisioned to further minimize human effort in certain applications.

On the Use of 210Pb-based Records of Sedimentation Rates and Activity Concentrations for Tracking Past Environmental Changes

Lead-210 from natural atmospheric fallout is widely used in multidisciplinary studies to date recent sediments. Some of the 210Pb-based dating models can produce historical records of sediment accumulation rates (SAR) and initial activity concentrations ( ). The former have been profusely used to track past changes in the sedimentary conditions. Both physical magnitudes are differently affected by model errors (those arising for the partial or null accomplishment of some model assumptions). This work is aimed at assessing the effects on SAR and of model errors in the CRS, CS, PLUM and TERESA dating models, due to random variability in 210Pb fluxes, which is a usual sedimentary condition. Synthetic cores are used as virtual laboratories for this goal. Independently of the model choice, SARs are largely affected by model errors, resulting in some large and spurious deviations from the true values. This questions their general use for tracking past environmental changes. are less sensitive to model errors and their trends of change with time may reflect real changes in sedimentary conditions, as it is shown with some real cores from varved sediments.

Advancing stream classification and hydrologic modeling of ungaged basins for environmental flow management in coastal southern California

Environmental streamflow management can improve the ecological health of streams by returning modified flows to more natural conditions. The Ecological Limits of Hydrologic Alteration (ELOHA) framework for developing regional environmental flow criteria has been implemented to reverse hydromodification across the heterogenous region of coastal southern California (So. CA) by focusing on two elements of the flow regime: streamflow permanence and flashiness. Within ELOHA, classification groups streams by hydrologic and geomorphic similarity to stratify flow-ecology relationships. Analogous grouping techniques are used by hydrologic modelers to facilitate streamflow prediction in ungaged basins (PUB) through regionalization. Most watersheds, including those needed for stream classification and environmental flow development, are ungaged. Furthermore, So. CA is a highly heterogeneous region spanning a gradient of urbanization, which presents a challenge for regionalizing ungaged basins. In this study, we develop a novel classification technique for PUB modeling that uses an inductive approach to group regional streams by modeled hydrologic similarity followed by deductively determining class membership with hydrologic model errors and watershed metrics. As a new type of classification, this “Hydrologic Model-based Classification” (HMC) prioritizes modeling accuracy, which in turn provides a means to improve model predictions in ungaged basins, while complementing traditional classifications and improving environmental flow management. HMC is developed by calibrating a regional catalog of process-based rainfall-runoff models, quantifying the hydrologic reciprocity of calibrated parameters that would be unknown in ungaged basins, and grouping sites according to hydrologic and physical similarity. HMC was applied to 25 USGS streamflow gages in the south coast region of California and was compared to other hybrid PUB approaches combining inductive and deductive classification. Using an Average Cluster Error metric, results show HMC provided the most hydrologically similar groups according to calibrated parameter reciprocity. Hydrologic Model-based Classification is relatively complex and time-consuming to implement, but it shows potential for advancing ungaged basin management. This study demonstrates the benefits of thorough stream classification using multiple approaches, and suggests that Hydrologic Model-based Classification has advantages for PUB and building the hydrologic foundation for environmental flow management.

Weakening of the Atlantic Niño variability under global warming

The Atlantic Niño is one of the most important tropical patterns of interannual climate variability, with major regional and global impacts. How global warming will influence the Atlantic Niño has been hardly explored, because of large climate model errors. We show for the first time that the state-of-the-art climate models robustly predict that equatorial Atlantic Niño variability will weaken in response to global warming. This is primarily because subsurface and surface temperature variations decouple as the upper equatorial Atlantic Ocean warms. The weakening is predicted by most (>80%) models following the highest emission scenarios in the Coupled Model Intercomparison Project Phases 5 and 6 considered here. These indicate a reduction in variability by the end of the century of 12-17%, and as much as 25% when accounting for model errors. Weaker Atlantic Niño variability will have major consequences for global climate and the skill of seasonal predictions.

Inversion method for measuring multi-hole probe surface pressure distribution of supersonic compressible atmospheric flow field

The conventional method of measuring a multi-hole probe is based on Bernoulli’s equation and suffers from certain model errors. A computational fluid dynamics (CFD)-based method was used in this study to reduce the theoretical error and establish a parametric model of the surface pressure of a spherical multi-hole pressure probe for measuring compressible flow fields at supersonic velocities. A flow field inversion method based on the parametric model is proposed herein. Numerical simulations were conducted to validate the proposed method. The experiment results show that in the compressible atmospheric flow field within Mach 1.2–1.7, the measurement errors of the inversion method were 1.3% and 2.35% for velocity and angle, respectively, thus verifying the feasibility of the method. Thus, a new method of measuring multi-hole pressure probe atmospheric flow fields was demonstrated.

Dynamic Thresholds for a Reliable Diagnosis of Switched Systems

Safety, availability and reliability are the main concern of many industries. Thus, fault detection and isolation of industrial machines, which are in most cases switched systems, is a primary task in many companies. The presented paper proposes a new diagnostic approach for switched systems using two powerful tools: bond graph and observer. A diagnostic layer detects model errors using bond graph, and a smart algorithm identifies and locates faults using observer. Although observers serve as fault detectors, they also have their own errors caused by convergence delay of calculations; even in the case of no sensor defect, the residue does not converge to zero. In this paper, we propose a new method to solve this problem by integrating dynamic thresholds in the detection procedure, which helped to avoid false alarms and ensure a highly reliable diagnosis.

A Comparison of National Water Model Retrospective Analysis Snow Outputs at SNOTEL Sites Across the Western U.S.

This study compares the U.S. National Water Model (NWM) reanalysis snow outputs to observed snow water equivalent (SWE) and snow-covered area fraction (SCAF) at SNOTEL sites across the Western U.S. SWE was obtained from SNOTEL sites, while SCAF was obtained from MODIS observations at a nominal 500 m grid scale. Retrospective NWM results were at a 1000 m grid scale. We compared results for SNOTEL sites to gridded NWM and MODIS outputs for the grid cells encompassing each SNOTEL site. Differences between modeled and observed SWE were attributed to both model errors, as well as errors in inputs, notably precipitation and temperature. The NWM generally under-predicted SWE, partly due to precipitation input differences. There was also a slight general bias for model input temperature to be cooler than observed, counter to the direction expected to lead to under-modeling of SWE. There was also under-modeling of SWE for a subset of sites where precipitation inputs were good. Furthermore, the NWM generally tends to melt snow early. There was considerable variability between modeled and observed SCAF as well as the binary comparison of snow cover presence that hampered useful interpretation of SCAF comparisons. This is in part due to the shortcomings associated with both model SCAF parameterization and MODIS observations, particularly in vegetated regions. However, when SCAF was aggregated across all sites and years, modeled SCAF tended to be more than observed using MODIS. These differences are regional with generally better SWE and SCAF results in the Central Basin and Range and differences tending to become larger the further away regions are from this region. These findings identify areas where predictions from the NWM involving snow may be better or worse, and suggest opportunities for research directed towards model improvements.

The Performance of Three Exponential Decay Models in Estimating Tropical Cyclone Intensity Change After Landfall Over China

In this study, the performance of three exponential decay models in estimating intensity change of tropical cyclones (TCs) after landfall over China is evaluated based on the best-track TC data during 1980–2018. Results indicate that the three models evaluated can reproduce the weakening trend of TCs after landfall, but two of them (M1 and M2) tend to overestimate TC intensity and one (M3) tends to overestimate TC intensity in the first 12 h and underestimate TC intensity afterwards. M2 has the best performance with the smallest errors among the three models within 24 h after landfall. M3 has better performance than M1 in the first 20 h after landfall, but its errors increase largely afterwards. M1 and M2 show systematic positive biases in the southeastern China likely due to the fact that they have not explicitly included any topographic effect. M3 has better performance in the southeastern China, where it was originally attempted, but shows negative biases in the eastern China. The relative contributions of different factors, including landfall intensity, translational speed, 850-hPa moist static energy, and topography, to model errors are examined based on classification analyses. Results indicate that the landfall intensity contributes about 18%, translational speed, moist static energy and topography contribute equally about 15% to the model errors. It is strongly suggested that the TC characteristics and the time-dependent decay constant determined by environmental conditions, topography and land cover properties, should be considered in a good exponential decay model of TC weakening after landfall.

Parametric Predictions for Pure Electric Vehicles

Demand for pure electric vehicles has been found to be increasing over the years. This has necessitated the development of a model that would serve as a predicting machine for manufacturing different types of pure electric vehicles. Direct Artificial Neural Network approach was used for predictions of nine different parameters commonly found in pure electric cars. Predictions were found to be of high degree of accuracy while using unit and overall model errors as the basis of performance measurement. The mean absolute error, mean square error and root mean square error of the model were 0.109, 0.218 and 0.467, respectively, when the combined electric charge consumption was used for modeling. For the model formation, using the same variable, the losses for the training and testing were 3.9132 × 10−6 and 9.698 × 10−7, respectively. The model was also evaluated using redefined datasets. The developed model can be used by manufacturers and engineers to simulate future designs when certain parameters are given.