Multi-Graph Heterogeneous Interaction Fusion for Social Recommendation

With the rapid development of online social recommendation system, substantial methods have been proposed. Unlike traditional recommendation system, social recommendation performs by integrating social relationship features, where there are two major challenges, i.e., early summarization and data sparsity. Thus far, they have not been solved effectively. In this article, we propose a novel social recommendation approach, namely Multi-Graph Heterogeneous Interaction Fusion (MG-HIF), to solve these two problems. Our basic idea is to fuse heterogeneous interaction features from multi-graphs, i.e., user–item bipartite graph and social relation network, to improve the vertex representation learning. A meta-path cross-fusion model is proposed to fuse multi-hop heterogeneous interaction features via discrete cross-correlations. Based on that, a social relation GAN is developed to explore latent friendships of each user. We further fuse representations from two graphs by a novel multi-graph information fusion strategy with attention mechanism. To the best of our knowledge, this is the first work to combine meta-path with social relation representation. To evaluate the performance of MG-HIF, we compare MG-HIF with seven states of the art over four benchmark datasets. The experimental results show that MG-HIF achieves better performance.

Backward Trajectory and Multifractal Analysis of Air Pollution in Zhengzhou Region of China

With the continuous promotion of industrialization and urbanization, China's environmental pollution is becoming increasingly serious, which has caused considerable damage to the natural balance. Air pollution seriously harms people's physical and mental health, the ecological environment, and the social sustainable development of society. In this study, the backward trajectory model and multifractal methods were adopted to analyze air pollution in Zhengzhou. The backward trajectory analysis showed that most clusters of air pollution were from southern Hebei, eastern Shandong, and mid-western Henan, which were then transported to Zhengzhou. For the PSCF and CWT analyses, we selected four representative cities to explore how close the air pollution of Zhengzhou is to other areas on the basis of air polluted concentration. The results of several multifractal methods indicated that multifractality existed in the AQI time series of Zhengzhou and cross-correlations between Zhengzhou and each of the four cities. The widths of multifractal spectra showed that the air pollution in Zhengzhou was closest to that in Jinan, followed by Shijiazhuang, Zibo, and Luoyang. The CDFA analysis showed that carbon monoxide (CO), nitrogen dioxide (NO2), and inhalable particulate matter (PM10) had important influences on air pollution in Zhengzhou. These findings offer a useful reference for air pollution sources and their potential contributions in Zhengzhou, which can support policy makers in environmental governance and in achieving sustainable urban development.

Spectral density of individual trajectories of an active Brownian particle

We study analytically the single-trajectory spectral density (STSD) of an active Brownian motion as exhibited, for example, by the dynamics of a chemically-active Janus colloid. We evaluate the standardly-defined spectral density, {\it i.e.} the STSD averaged over a statistical ensemble of trajectories in the limit of an infinitely long observation time $T$, and also go beyond the standard analysis by considering the coefficient of variation $\gamma$ of the distribution of the STSD. Moreover, we analyse the finite-$T$ behaviour of the STSD and $\gamma$, determine the cross-correlations between spatial components of the STSD, and address the effects of translational diffusion on the functional forms of spectral densities. The exact expressions that we obtain unveil many distinctive features of active Brownian motion compared to its passive counterpart, which allow to distinguish between these two classes based solely on the spectral content of individual trajectories.

Cryptocurrency Market Consolidation in 2020–2021

Time series of price returns for 80 of the most liquid cryptocurrencies listed on Binance are investigated for the presence of detrended cross-correlations. A spectral analysis of the detrended correlation matrix and a topological analysis of the minimal spanning trees calculated based on this matrix are applied for different positions of a moving window. The cryptocurrencies become more strongly cross-correlated among themselves than they used to be before. The average cross-correlations increase with time on a specific time scale in a way that resembles the Epps effect amplification when going from past to present. The minimal spanning trees also change their topology and, for the short time scales, they become more centralized with increasing maximum node degrees, while for the long time scales they become more distributed, but also more correlated at the same time. Apart from the inter-market dependencies, the detrended cross-correlations between the cryptocurrency market and some traditional markets, like the stock markets, commodity markets, and Forex, are also analyzed. The cryptocurrency market shows higher levels of cross-correlations with the other markets during the same turbulent periods, in which it is strongly cross-correlated itself.

Google Trends as a predictive tool for COVID-19 vaccinations in Italy: a retrospective infodemiological analysis.

Background: Google Trends is an infoveillance tool widely used by the scientific community to investigate different user behaviors related to COVID-19. However, several limitations regarding its adoption are reported in the literature. Objective: This brief paper aims to provide an effective and efficient approach to investigating vaccine adherence against COVID-19 via Google Trends. Methods: Through the cross-correlational analysis of well-targeted hypotheses, we investigate the predictive capacity of web searches related to COVID-19 towards vaccinations in Italy from November 2020 to November 2021. The keyword "vaccine reservation" (VRQ) was chosen as it reflects a real intention of being vaccinated (V). Furthermore, the impact of the second-largest Italian national newspaper on vaccines-related web searches was investigated to evaluate the role of the mass media as a confounding factor. Results: Simple and generic keywords are more likely to identify the actual web interest in COVID-19 vaccines than specific and elaborated keywords. Cross-correlations between VRQ and V were very strong and significant (min r^2 = .460, P<.001, lag = 0 weeks; max r^2 = .903, P < .001, lag = 6 weeks). Cross-correlations between VRQ and news about COVID-19 vaccines have been markedly lower and characterized by greater lags (min r^2 = .190, P=.001, lag = 0 weeks; max r^2 = .493, P < .001, lag = -10 weeks). No correlation between news and vaccinations was sought since the lag would have been too high. Conclusions: This research provides strong evidence in favor of using Google Trends as a surveillance and prediction tool for vaccine adherence against COVID-19 in Italy. These findings prove that the search for suitable keywords is a fundamental step to reduce confounding factors. Additionally, targeting hypotheses helps diminish the likelihood of spurious correlations. It is recommended that Google Trends be leveraged as a complementary infoveillance tool by government agencies to monitor and predict vaccine adherence in this and future crises by following the methods proposed in this manuscript.

Google Trends as a Predictive Tool for COVID-19 Vaccinations in Italy: a Retrospective Infodemiological Analysis (Preprint)

BACKGROUND Google Trends is an infoveillance tool widely used by the scientific community to investigate different user behaviors related to COVID-19. However, several limitations regarding its adoption are reported in the literature. OBJECTIVE This brief paper aims to provide an effective and efficient approach to investigating vaccine adherence against COVID-19 via Google Trends. METHODS Through the cross-correlational analysis of well-targeted hypotheses, we investigate the predictive capacity of web searches related to COVID-19 towards vaccinations in Italy from November 2020 to November 2021. The keyword "vaccine reservation" (VRQ) was chosen as it reflects a real intention of being vaccinated (V). Furthermore, the impact of the second most read Italian newspaper on vaccines-related web searches was investigated to evaluate the role of the mass media as a confounding factor. RESULTS Simple and generic keywords are more likely to identify the actual web interest in COVID-19 vaccines than specific and elaborated keywords. Cross-correlations between VRQ and V were very strong and significant (min r² = .460, P<.001, lag = 0 weeks; max r² = .903, P < .001, lag = 6 weeks). Cross-correlations between VRQ and news about COVID-19 vaccines have been markedly lower and characterized by greater lags (min r² = .190, P=.001, lag = 0 weeks; max r² = .493, P < .001, lag = -10 weeks). No correlation between news and vaccinations was sought since the lag would have been too high. CONCLUSIONS This research provides strong evidence in favor of using Google Trends as a surveillance and prediction tool for vaccine adherence against COVID-19 in Italy. These findings prove that the search for suitable keywords is a fundamental step to reduce confounding factors. Additionally, targeting hypotheses helps diminish the likelihood of spurious correlations. It is recommended that Google Trends be leveraged as a complementary infoveillance tool by government agencies to monitor and predict vaccine adherence in this and future crises by following the methods proposed in this manuscript.

Phase Shift Between Age-Specific COVID-19 Incidence Curves Points to a Potential Epidemic Driver Function of Kids and Juveniles in Germany

Mutual phase shifts between three German COVID-19 incidence curves corresponding to the age classes of children, juveniles and adults, respectively, are calculated by means of delay-cross-correlations. At the country level, a phase shift of -5 weeks during the first half of the epidemic between the incidence curves corresponding to the juvenile age class and the curve corresponding to the adult class is observed. The children's incidence curve is shifted by -3 weeks with respect to the adults' curve. On the regional level of the 411 German districts (Landkreise) the distributions of observed time lags are inclined towards negative values. Regarding the incidence time series of the juvenile sub-population, 20% of the German districts exhibit negative phase shifts and only 3% show positive shifts versus the incidence curves of the adult sub-population. Similarly for the children with 6% positive shifts. Thus, children's and juveniles' epidemic activity is ahead of the adults' activity. The correlation coefficients of shifted curves are large (> 0.9 for juveniles versus adults on the country level) which indicates that aside from the phase shift the sub-populations follow a similar epidemic dynamics. Negative phase shifts of the children's incidence curves during the first and second epidemic waves are predictors for high incidences during the current fourth wave with respect to the corresponding districts.

Determining seismic shear-velocity from ambient noise sources at regional and local scales

<p>We present three projects that use different bandwidths of the ambient noise spectrum to solve geophysical problems. Specifically, we use signals within the noise field to determine surface and shear wave velocities, image the shallow and deep crust, and monitor time-dependent deformation resulting from geothermal fluid injection and extraction. Harrat Al-Madinah, a Cenozoic bimodal alkaline volcanic field in west-central Saudi Arabia, is imaged using shear-velocities obtained from natural ambient seismic noise. To our knowledge, this project is the first analysis of Saudi Arabia structure using ambient noise methods. Surface wave arrivals are extracted from a year's worth of station-pair cross-correlations, which are approximations of the empirical Green's function of the interstation path. We determine group and phase velocity surface wave dispersion maps with a 0.1 decimal degree resolution and resolve a zone of slow surface wave velocity south-east of the city of Medina, which is spatially correlated with the most recent historical eruption (the 1256 CE Medina eruption). Dispersion curves are calculated at each grid-point of the surface-wave velocity maps and inverted to obtain measurements of shear-velocity with depth. The 1D velocity models are then used to produce average shear-velocity models for the volcanic field. A shear-velocity increase ranging from 0.5 to 1.0 km/s, suggesting a layer interface, is detected at approximately 20 km depth and compared to P-wave measurement from a previous refraction study. We compute cross-section profiles by interpolating the inversions into a pseudo-3D model and resolve a zone of slow shear-velocity below the 1256 CE eruption location. These areas are also spatially correlated with low values of Bouguer gravity. We hypothesize that the low shear-velocity and gravity measurements are caused by fluids and fractures created from prior volcanic eruptions. We use the coda of cross-correlations extracted from ambient noise to determine shear-velocity changes at Rotokawa and Ngatamariki, two electricity producing geothermal fields located in the North Island of New Zealand. Stacks of cross correlations between stations prior to the onset of production are compared to cross correlations of moving stacks in time periods of well stimulation and the onset of electricity production using the Moving Window Cross Spectral technique. An increase between 0.05% to 0.1% of shear-velocity is detected at Rotokawa coinciding with an increase of injection. The shear-velocity subsequently decreases by approximately 0.1% when the rate of production surpasses the rate of injection. A similar amplitude shear-velocity increase is detected at Ngatamariki during the beginning of injection. After the initial increase, the shear-velocity at Ngatamariki fluctuates in response to differences in injection and production rates. A straight-ray pseudo-tomography analysis is conducted at the geothermal fields, which reveals that localized positive velocity changes are co-located with injection wells. Lastly, we use ambient noise and active sources at the Ngatamariki geothermal field to determine the structure of the top 200 meters using the Refraction Microtremor technique. We deployed a linear 72-channel array of vertical geophones with ten meter spacing at two locations of the geothermal field and determine average 1D and 2D shear-velocity profiles. We were able to image depths between 57 to 93 meters for 2D profiles and up to 165 meters for 1D profiles. A shear-velocity anomaly was detected across one of the lines that coincided with the inferred location of a fault determined from nearby well logs. This suggests that the method can be used to cheaply and quickly constrain near-surface geology at geothermal fields, where ambient noise is abundant and typical reflection and refraction surveys require large inputs of energy and are hindered by attenuation and scattering in near-surface layers.</p>

Determining seismic shear-velocity from ambient noise sources at regional and local scales

<p>We present three projects that use different bandwidths of the ambient noise spectrum to solve geophysical problems. Specifically, we use signals within the noise field to determine surface and shear wave velocities, image the shallow and deep crust, and monitor time-dependent deformation resulting from geothermal fluid injection and extraction. Harrat Al-Madinah, a Cenozoic bimodal alkaline volcanic field in west-central Saudi Arabia, is imaged using shear-velocities obtained from natural ambient seismic noise. To our knowledge, this project is the first analysis of Saudi Arabia structure using ambient noise methods. Surface wave arrivals are extracted from a year's worth of station-pair cross-correlations, which are approximations of the empirical Green's function of the interstation path. We determine group and phase velocity surface wave dispersion maps with a 0.1 decimal degree resolution and resolve a zone of slow surface wave velocity south-east of the city of Medina, which is spatially correlated with the most recent historical eruption (the 1256 CE Medina eruption). Dispersion curves are calculated at each grid-point of the surface-wave velocity maps and inverted to obtain measurements of shear-velocity with depth. The 1D velocity models are then used to produce average shear-velocity models for the volcanic field. A shear-velocity increase ranging from 0.5 to 1.0 km/s, suggesting a layer interface, is detected at approximately 20 km depth and compared to P-wave measurement from a previous refraction study. We compute cross-section profiles by interpolating the inversions into a pseudo-3D model and resolve a zone of slow shear-velocity below the 1256 CE eruption location. These areas are also spatially correlated with low values of Bouguer gravity. We hypothesize that the low shear-velocity and gravity measurements are caused by fluids and fractures created from prior volcanic eruptions. We use the coda of cross-correlations extracted from ambient noise to determine shear-velocity changes at Rotokawa and Ngatamariki, two electricity producing geothermal fields located in the North Island of New Zealand. Stacks of cross correlations between stations prior to the onset of production are compared to cross correlations of moving stacks in time periods of well stimulation and the onset of electricity production using the Moving Window Cross Spectral technique. An increase between 0.05% to 0.1% of shear-velocity is detected at Rotokawa coinciding with an increase of injection. The shear-velocity subsequently decreases by approximately 0.1% when the rate of production surpasses the rate of injection. A similar amplitude shear-velocity increase is detected at Ngatamariki during the beginning of injection. After the initial increase, the shear-velocity at Ngatamariki fluctuates in response to differences in injection and production rates. A straight-ray pseudo-tomography analysis is conducted at the geothermal fields, which reveals that localized positive velocity changes are co-located with injection wells. Lastly, we use ambient noise and active sources at the Ngatamariki geothermal field to determine the structure of the top 200 meters using the Refraction Microtremor technique. We deployed a linear 72-channel array of vertical geophones with ten meter spacing at two locations of the geothermal field and determine average 1D and 2D shear-velocity profiles. We were able to image depths between 57 to 93 meters for 2D profiles and up to 165 meters for 1D profiles. A shear-velocity anomaly was detected across one of the lines that coincided with the inferred location of a fault determined from nearby well logs. This suggests that the method can be used to cheaply and quickly constrain near-surface geology at geothermal fields, where ambient noise is abundant and typical reflection and refraction surveys require large inputs of energy and are hindered by attenuation and scattering in near-surface layers.</p>