MARC: a robust method for multiple-aspect trajectory classification via space,time, and semantic embeddings

ABSTRACT

The increasing popularity of Location-Based Social Networks (LBSNs) and the semantic enrichment of mobility data in several contexts in the last years has led to the generation of large volumes of trajectory data. In contrast to GPS-based trajectories, LBSN and context-aware trajectories are more complex data, having several semantic textual dimensions besides space and time, which may reveal interesting mobility patterns. For instance, people may visit different places or perform different activities depending on the weather conditions. These new semantically rich data, known as multiple-aspect trajectories, pose new challenges in trajectory classification, which is the problem that we address in this paper. Existing methods for trajectory classification cannot deal with the complexity of heterogeneous data dimensions or the sequential aspect that characterizes movement. In this paper we propose MARC, an approach based on attribute embedding and Recurrent Neural Networks (RNNs) for classifying multiple-aspect trajectories, that tackles all trajectory properties: space, time, semantics, and sequence. We highlight that MARC exhibits good performance especially when trajectories are described by several textual/categorical attributes. Experiments performed over four publicly available datasets considering the Trajectory-User Linking (TUL) problem show that MARC outperformed all competitors, with respect to accuracy, precision, recall, and F1-score.     

Correction to: Spatial distribution and estimation of rainfall trends and erosivity in the Epitácio Pessoa reservoir catchment,Paraíba,Brazil

Natural Hazards - The article “Spatial distribution and estimation of rainfall trends and erosivity in the Epitácio Pessoa reservoir catchment, Paraíba, Brazil”, written by da...     

Urban forest loss using a GIS-based approach and instruments for integrated urban planning:A case study of Jo(a)o Pessoa,Brazil

Urban forests play an important role in the thermal comfort and overall life of local populations in large-and medium-sized cities.This study analyzes urban forest loss and maps land use and land cover (LULC) changes between 1991 and 2018 by evaluating the use of urban planning instruments for the mitigation of urban forest loss in Jo～o Pessoa,Brazil.For this purpose,satellite-derived LULC images from 1991,2006,2010 and 2018 and data on urban forest loss areas obtained using the Google Earth Engine were used.In addi-tion,this paper also discusses the instruments used for integrated urban planning,which are(a) the legal sector,responsibility and nature;(b) the urban expansion process;and (c) the elements of urban infrastructure.The results show a clear shift in land use in the study area.The major changes in LULC classes occurred in urban areas and herbaceous vegetation,while the greatest loss was in arboreal/shrub vegetation.Thus,an increase in the pressure to occupy zones intended for environmental preservation could be estimated.Our results showed similar accuracies with other studies and more spatial details.The characteristics of the patterns,traces,and hotspots of urban expansion and forest cover loss were explored.We highlighted the potential use of this proposed framework to be applied and validated in other parts of the world to help better understand and quantify various aspects of ur-ban-related problems such as urban forest loss mapping using instruments for integrated urban planning and low-cost approaches.     

Studies of Plasma Dynamics in a Small Plasma Focus Operating in Hydrogen-Argon Mixtures

We report on experimental observations in PFP-I, a small 3.8 kJ plasma focus, which is operated in Hydrogen-Argon mixtures to investigate the effect of parameter modifications on the overall performance of the device. An extensive array of diagnostics is been used, which includes voltage and current probes in the external circuit, a novel small magnetic probes array located along the cathode rods, filtered PIN diodes located side and end on, and filtered multi-pinhole and slit-wire X-ray camera. Extended operating range from below 0.2 Torr upwards has been achieved. Hot spot formation has been investigated as a function of H₂-Ar mixing ratio. Hot spot sizes around 150 m in the soft X-ray region, have been inferred from the slit-wire measurements. Pin-hole time integrated X-ray pictures and time resolved PIN diode measurements have been used to determine characteristic hot-spot temperatures in the 350 to 450 eV range.     

Impacts of climate change on building energy demands in the intra-Americas region

It was recently reported a regional warming in the intra-Americas region where sea surface temperature exhibited increases exceeding 0.15 °C/decade and an accelerated air temperature rise that could impact building energy demands per capita (EDC). Reanalysis data is used herein to quantify the impacts of these warming trends on EDC. Results of the analysis depict a Southern Greater Antilles and inland South America with a positive annual EDC rate of 1–5 kWh per year. The Intergovernmental Panel on Climate Change (IPCC) Representative Concentration Pathways (RCP) 2.6 and 4.5 scenarios were selected to analyze energy demand changes in the twenty-first century. A multi-model ensemble forecasts an EDC increase of 9.6 and 23 kWh/month in the RCP2.6 and RCP4.5 at the end of the twenty-first century, which may increase average building cooling loads in the region by 7.57 GW (RCP2.6) and 8.15 GW (RCP4.5), respectively. Furthermore, 4 of 9 (RCP2.6) and 7 of 9 (RCP4.5) of the major countries in this region have EDCs ranging between 1887 and 2252 kWh/year at the end of this century. Therefore, increased energy production and improved energy infrastructure will be required to maintain ideal indoor building conditions at the end of the twenty-first century in these tropical coastal regions as consequence of a warmer climate.     

Impact of data model and point density on aboveground forest biomass estimation from airborne LiDAR

Background

Accurate estimation of aboveground forest biomass (AGB) and its dynamics is of paramount importance in understanding the role of forest in the carbon cycle and the effective implementation of climate change mitigation policies. LiDAR is currently the most accurate technology for AGB estimation. LiDAR metrics can be derived from the 3D point cloud (echo-based) or from the canopy height model (CHM). Different sensors and survey configurations can affect the metrics derived from the LiDAR data. We evaluate the ability of the metrics derived from the echo-based and CHM data models to estimate AGB in three different biomes, as well as the impact of point density on the metrics derived from them.

Results

Our results show that differences among metrics derived at different point densities were significantly different from zero, with a larger impact on CHM-based than echo-based metrics, particularly when the point density was reduced to 1 point m^?2. Both data models-echo-based and CHM-performed similarly well in estimating AGB at the three study sites. For the temperate forest in the Sierra Nevada Mountains, California, USA, R² ranged from 0.79 to 0.8 and RMSE (relRMSE) from 69.69 (35.59%) to 70.71 (36.12%) Mg ha^?1 for the echo-based model and from 0.76 to 0.78 and 73.84 (37.72%) to 128.20 (65.49%) Mg ha^?1 for the CHM-based model. For the moist tropical forest on Barro Colorado Island, Panama, the models gave R² ranging between 0.70 and 0.71 and RMSE between 30.08 (12.36%) and 30.32 (12.46) Mg ha^?1 [between 0.69–0.70 and 30.42 (12.50%) and 61.30 (25.19%) Mg ha^?1] for the echo-based [CHM-based] models. Finally, for the Atlantic forest in the Sierra do Mar, Brazil, R² was between 0.58–0.69 and RMSE between 37.73 (8.67%) and 39.77 (9.14%) Mg ha^?1 for the echo-based model, whereas for the CHM R² was between 0.37–0.45 and RMSE between 45.43 (10.44%) and 67.23 (15.45%) Mg ha^?1.

Conclusions

Metrics derived from the CHM show a higher dependence on point density than metrics derived from the echo-based data model. Despite the median of the differences between metrics derived at different point densities differing significantly from zero, the mean change was close to zero and smaller than the standard deviation except for very low point densities (1 point m^?2). The application of calibrated models to estimate AGB on metrics derived from thinned datasets resulted in less than 5% error when metrics were derived from the echo-based model. For CHM-based metrics, the same level of error was obtained for point densities higher than 5 points m^?2. The fact that reducing point density does not introduce significant errors in AGB estimates is important for biomass monitoring and for an effective implementation of climate change mitigation policies such as REDD + due to its implications for the costs of data acquisition. Both data models showed similar capability to estimate AGB when point density was greater than or equal to 5 point m^?2.

     

Cenozoic tectonics of Amazon Mouth Basin

 The Cenozoic shelf margin of the Amazon Mouth Basin is characterized by a thick prograding prism of siliciclastic sediments. This prism, composed mainly of Upper Miocene and younger sediments, overlies a Lower Tertiary carbonate shelf. Two tectonic–sedimentary models for the area were developed with the aid of new deep-reflection seismic data. Gravitational tectonics dominate the regional geological framework. Tensional stresses are created near the shelf margin, and compressional features dominate at the base of the slope. The morphology of this compressional zone is closely associated with the St. Paul Fracture Zone and the boundary between continental and oceanic crusts. Received: 20 August 1996 / Revision received: 11 June 1998     

Flash flood in the mountainous region of Rio de Janeiro state (Brazil) in 2011: part I—calibration watershed through hydrological SMAP model

Natural Hazards - Flooding and water logging possess severe hazards to human population in many parts of the world. Mumbai, the study area, is one of the cities in India and has been frequently...     

Geological Facies Recovery Based on Weighted $$\ell _1$$ ℓ 1 -Regularization

Mathematical Geosciences - A weighted compressed sensing (WCS) algorithm is proposed for the problem of channelized facies reconstruction from pixel-based measurements. This strategy integrates...     

Application of open tools and datasets to probabilistic modeling of road traffic disruptions due to earthquake damage

The disruption of a transportation network can have a high social and economic impact on the welfare of a society, as it can significantly affect the daily routines of a community. Although many studies have focused on the estimation of physical risk in the components that compose these networks, only a limited number have analyzed their interconnections and impact in the traffic flow. The present study analyzes how earthquake damage can disrupt the road network in an urban environment, and how this will influence the ability of the population to travel. Traffic due to daily commutes is modeled for different layouts of the network, corresponding to possible disruptions caused by earthquake damage. The duration and length of each trip were calculated both for the undamaged network conditions and for the disrupted network. The increase in the median duration and length of each trip allows estimating the economic loss for each event due to drivers' delay. By combining the probability of a specific road being blocked with its number of users, the average number of affected vehicles was estimated, and the most critical segments identified. The methodology was applied to a case study concerning the road network of the area around the Italian city of Messina in Sicily. The results were calculated for both a repetition of the well-known historical event of 1908 and a set of simulated earthquakes consistent with the national probabilistic seismic hazard model of Italy.