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.     

Multidimensional Similarity Measuring for Semantic Trajectories

Most existing approaches aiming at measuring trajectory similarity are focused on two‐dimensional sequences of points, called raw trajectories. However, recent proposals have used background geographic information and social media data to enrich these trajectories with a semantic dimension, giving rise to the concept of semantic trajectories. Only a few works have proposed similarity measures for semantic trajectories or multidimensional sequences, having limitations such as predefined weight of the dimensions, sensitivity to noise, tolerance for gaps with different sizes, and the prevalence of the worst dimension similarity. In this article we propose MSM, a novel similarity measure for multidimensional sequences that overcomes the aforementioned limitations by considering and weighting the similarity in all dimensions. MSM is evaluated through an extensive experimental study that, based on a seed trajectory, creates sets of semantic trajectories with controlled transformations to introduce different kinds and levels of dissimilarity. For each set, we compute the similarity between the seed and the transformed trajectories, using different measures. The results showed that MSM was more robust and efficient than related approaches in the domain of semantic trajectories.     

Imposex levels and concentrations of organotin compounds (TBT and its metabolites) in Nassarius nitidus from the Lagoon of Venice

Specimens of Nassarius nitidus were collected in seven stations of the Venice Lagoon to assess the levels of tributyltin (TBT) and its metabolites monobutyltin and dibutyltin in the tissues and monitor their effect on organisms, in particular the phenomenon of imposex (superimposition of male sexual characteristics on females). The following values of population indices were found: vas deferens sequence: 1.2+/-0.7-4.0+/-0.5; relative penis length: 6-47%. The least impacted station was situated in the northern part of the Lagoon, where females without imposex were found and Butyltin (BuTs) concentrations in the organisms (average sum of BuTs=43+/-14 ngSng(-1)w.) were significantly lower than in the other stations (range of average sum of BuTs: 101+/-22-217+/-27 ngSng(-1)d.w.). Population indices were found to be related to the TBT content in the tissues. In particular VDSI had a significant logarithmic correlation: r=0.95, n=8, p<0.05.     

Observations of Comet 9P/Tempel 1 around the Deep Impact event by the OSIRIS cameras onboard Rosetta

The OSIRIS cameras on the Rosetta spacecraft observed Comet 9P/Tempel 1 from 5 days before to 10 days after it was hit by the Deep Impact projectile. The Narrow Angle Camera (NAC) monitored the cometary dust in 5 different filters. The Wide Angle Camera (WAC) observed through filters sensitive to emissions from OH, CN, Na, and OI together with the associated continuum. Before and after the impact the comet showed regular variations in intensity. The period of the brightness changes is consistent with the rotation period of Tempel 1. The overall brightness of Tempel 1 decreased by about 10% during the OSIRIS observations. The analysis of the impact ejecta shows that no new permanent coma structures were created by the impact. Most of the material moved with . Much of it left the comet in the form of icy grains which sublimated and fragmented within the first hour after the impact. The light curve of the comet after the impact and the amount of material leaving the comet ( of water ice and a presumably larger amount of dust) suggest that the impact ejecta were quickly accelerated by collisions with gas molecules. Therefore, the motion of the bulk of the ejecta cannot be described by ballistic trajectories, and the validity of determinations of the density and tensile strength of the nucleus of Tempel 1 with models using ballistic ejection of particles is uncertain.     

Temperature,fluid content and rheology of localized ductile shear zones in subsolidus cooling plutons

Temperature and fluid content are critical parameters that control rock rheology and strain localization in the continental crust. Here, we determine by thermodynamic modelling the of localized ductile shearing during cooling of three different granitoid plutons: the Rieserferner and the Adamello plutons in the Italian Alps, and the Lake Edison pluton in the Sierra Nevada—USA. Shear zones exploited precursor joints, associated veins and alteration zones. and P−T phase diagram sections were computed with Perple_X in the system MnO−Na₂O−CaO −K₂O−FeO−MgO−Al₂O₃−SiO₂−H₂O−Fe₂O₃. The phase diagram sections show that the nucleation of the brittle precursors (joints, veins) occurred at T» 450°C at fluid-saturated conditions. Localized ductile shearing likely occurred at temperature ranging between 420 and 460°C evolving from initially fluid-saturated to fluid-undersaturated conditions in a closed system. In this temperature range, granitoid rocks are potentially subject to a series of retrograde metamorphic reactions replacing the load-bearing feldspars with weaker phyllosilicates. Metamorphic reactions occurred in spatial association with the precursory structures, leading to localized shearing. Decreasing temperature and fluid-undersaturated conditions likely hampered progressive strain accommodation in shear zones by slowing down metamorphic reactions, thermally activated dislocation creep processes, fluid-mediated deformation mechanisms and weakening mechanisms. Polyphase granitoid ultramylonite and mylonitic quartz veins have been affected differently by the fluid-undersaturated conditions of the system, as consequence of different dominant deformation mechanisms and syn-kinematic paragenesis during localized shearing. Localized ductile shearing in cooling plutons effectively occurs in a limited temperature range (420–460°C) in which the strain accommodation capacity of the shear zone is controlled by the negative feedback between the cooling rate, the kinetics of metamorphic reactions and deformation mechanisms, and the consumption of the limited amount of available fluids.     

Genesis of a mottled horizon by Fe-depletion within a laterite cover in the Amazon BasinGenèse d'un horizon tacheté par déferruginisation dans une couverture à latérite du Bassin amazonien

A mottled horizon in a laterite cover (without any duricrust) was studied by microscopy and quantitative chemical microanalysis. Apart from the voids, light red spots consisting of Fe-rich particles (≈2 μm) are set in clayey plasma. Dark red spots consisted of concentrations of Fe-rich particles. These patterns are inherited. On the border of structural or biological voids, where Fe-depletion features are systematic, gray or yellow spots result from dissolution of the Fe-rich particles and impregnation of the plasma by iron, respectively. The present Fe-depletion is the dominant process that explains the mottled differentiation and the absence of lateritic duricrust. To cite this article: V. Rosolen et al., C. R. Geoscience 334 (2002) 187–195.     

Imposex levels and concentrations of organotin compounds (TBT and its metabolites) in Nassarius nitidus from the Lagoon of Venice

Specimens of Nassarius nitidus were collected in seven stations of the Venice Lagoon to assess the levels of tributyltin (TBT) and its metabolites monobutyltin and dibutyltin in the tissues and monitor their effect on organisms, in particular the phenomenon of imposex (superimposition of male sexual characteristics on females). The following values of population indices were found: vas deferens sequence: 1.2 ± 0.7–4.0 ± 0.5; relative penis length: 6–47%. The least impacted station was situated in the northern part of the Lagoon, where females without imposex were found and Butyltin (BuTs) concentrations in the organisms (average sum of BuTs = 43 ± 14 ng Sn g⁻¹ w.) were significantly lower than in the other stations (range of average sum of BuTs: 101 ± 22–217 ± 27 ng Sn g⁻¹ d.w.). Population indices were found to be related to the TBT content in the tissues. In particular VDSI had a significant logarithmic correlation: r = 0.95, n = 8, p < 0.05.     

Formation and early history of Lakes Pepin and St. Croix of the upper Mississippi River

Study of Lake Pepin and Lake St. Croix began more than a century ago, but new information has permitted a closer look at the geologic history of these two riverine lakes located on the upper Mississippi River system. Drainages from large proglacial lakes Agassiz and Duluth at the end of the last glaciation helped shape the current valleys. As high-discharge outlet waters receded, tributary streams deposited fans of sediment in the incised river valleys. These tributary fans dammed the main river, forming riverine lakes. Lake Pepin was previously thought to be a single long continuous lake, extending for 80 km from its dam at the Chippewa River fan all the way up to St. Paul, with an arm extending up the St. Croix valley. Recent borings taken at bridge and dam locations show more than a single section of lake sediments, indicating a more complex history. The Minnesota and Mississippi Rivers did not always follow their current paths. Valleys cut into bedrock but now buried by glacial sediment indicate former river courses, with the most recent of these from the last interglacial period marked at the surface by chains of lakes. The morphology of the Mississippi valley bottom, and thus the morphology of Lake Pepin as it filled the valley, is reflect in part by the existence of these old valleys but also by the presence of glacial outwash terraces and the alluvial fans of tributary streams. A sediment core taken in Lake Pepin near Lake City had a piece of wood in gravels just below lake sediments that dated to 10.3 ka cal. BP, indicating that the lake formed as the Chippewa River fan grew shortly after the floodwaters of Lakes Agassiz and Duluth receded. Data from new borings indicate small lakes were dammed behind several tributary fans in the Mississippi River valley between the modern Lake Pepin and St. Paul. One tributary lake, here called Early Lake Vermillion, may have hydraulically dammed the St. Croix River, creating an incipient Lake St. Croix. The tributary fans from the Vermillion River, the Cannon River, and the Chippewa River all served to segment the main river valley into a series of riverine lakes. Later the growth of the Chippewa fan surpassed that of the Vermillion and Cannon fans to create a single large lake, here called late Lake Pepin, which extended upstream to St. Paul. Sediment cores taken from Lake Pepin did not have significant organic matter to develop a chronology from radiocarbon dating. Rather, magnetic features were matched with those from a Lake St. Croix core, which did have a known radiocarbon chronology. The Pepin delta migration rate was then estimated by projecting the elevations of the top of the buried lake sediments to the dated Lake Pepin core, using an estimated slope of 10 cm/km, the current slope of Lake Pepin sediment surface. By these approximations, the Lake Pepin delta prograded past Hastings 6.0 ka cal BP and Red Wing 1.4 ka cal BP. This is one of eight papers dedicated to the “Recent Environmental History of the Upper Mississippi River” published in this special issue of the Journal of Paleolimnology. D. R. Engstrom served as guest editor of the special issue.     

Climate-related variations in crustal trace elements in Dome C (East Antarctica) ice during the past 672 kyr

Cr, Fe, Rb, Ba and U were determined by inductively coupled plasma sector field mass spectrometry (ICP-SFMS) in various sections of the 3,270 m deep ice core recently drilled at Dome C on the high East Antarctic plateau as part of the EPICA program. The sections were dated from 263 kyr bp (depth of 2,368 m) to 672 kyr bp (depth of 3,062 m). When combined with the data previously obtained by Gabrielli and co-workers for the upper 2,193 m of the core, it gives a detailed record for these elements during a 672-kyr period from the Holocene back to Marine Isotopic Stage (MIS) 16.2. Concentrations and fallout fluxes of all elements are found to be highly variable with low values during the successive interglacial periods and much higher values during the coldest periods of the last eight climatic cycles. Crustal enrichment factors indicates that rock and soil dust is the dominant source for Fe, Rb, Ba and U whatever the period and for Cr during the glacial maxima. The relationship between Cr, Fe, Rb, Ba and U concentrations and the deuterium content of the ice appears to be similar before and after the Mid-Brunhes Event (MBE, around 430 kyr bp). Mean concentration values observed during the successive interglacials from the Holocene to MIS 15.5 appear to vary from one interglacial to another at least for part of the elements. Concentrations observed during the successive glacial maxima suggest a decreasing trend from the most recent glacial maxima (MIS 2.2 and 4.2) to the oldest glacial maxima such as MIS 14.2, 14.4 and 16.2, which could be linked with changes in the size distribution of dust particles transported from mid-latitude areas to the East Antarctic ice cap.     

Numerical study of partially drained penetration and pore pressure dissipation in piezocone test

The piezocone penetration test (CPTU) is commonly used as a fast and economical tool to identify soil profile and to estimate relevant material properties in soils ranging from fine to coarse-grained. Moreover, in the case of fine-grained soils (clays and silts), the consolidation coefficient and the permeability can be estimated through the dissipation test. Undrained conditions are commonly assumed for the interpretation of CPTU in fine-grained soils, but in soils such as silts, penetration may occur in partially drained conditions. This aspect is often neglected in data interpretation thus leading to an inaccurate estimate of soil properties. This paper investigates numerically the effect of partial drainage during penetration on the measured tip resistance and the subsequent pore pressure dissipation response contributing to a more accurate interpretation of field data. A realistic simulation of the cone penetration is achieved with the two-phase Material Point Method, modelling the soil response with the modified Cam-Clay model. The approach takes into account large soil deformations induced by the advancing cone, soil–water, and soil–structure interactions, as well as nonlinear soil behavior.