More Like this

1. Learning Incident Prediction Models Over Large Geographical Areas for Emergency Response Systems

   Vazirizade, Sayyed; Mukhopadhyay, Ayan; Pettet, Geoffrey; El Said, Said; Baroud, Hiba; Dubey, Abhishek (January 2021, IEEE Conference on Smart Computing. SmartComp 2021.)

   null (Ed.)

   Principled decision making in emergency response management necessitates the use of statistical models that predict the spatial-temporal likelihood of incident occurrence. These statistical models are then used for proactive stationing which allocates first responders across the spatial area in order to reduce overall response time. Traditional methods that simply aggregate past incidents over space and time fail to make useful short-term predictions when the spatial region is large and focused on fine-grained spatial entities like interstate highway networks. This is partially due to the sparsity of incidents with respect to the area in consideration. Further, accidents are affected by several covariates, and collecting, cleaning, and managing multiple streams of data from various sources is challenging for large spatial areas. In this paper, we highlight how this problem is being solved for the state of Tennessee, a state in the USA with a total area of over 100,000 sq. km. Our pipeline, based on a combination of synthetic resampling, non-spatial clustering, and learning from data can efficiently forecast the spatial and temporal dynamics of accident occurrence, even under sparse conditions. In the paper, we describe our pipeline that uses data related to roadway geometry, weather, historical accidents, and real-time traffic congestion to aid accident forecasting. To understand how our forecasting model can affect allocation and dispatch, we improve upon a classical resource allocation approach. Experimental results show that our approach can significantly reduce response times in the field in comparison with current approaches followed by first responders. 

   more » « less
2. Multivariate Spatial Feature Selection in fMRI

   https://doi.org/10.1093/scan/nsab010  

   Jolly, E; Chang, L J (January 2021, Social Cognitive and Affective Neuroscience)

   null (Ed.)

   Abstract Multivariate neuroimaging analyses constitute a powerful class of techniques to identify psychological representations. However, not all psychological processes are represented at the same spatial scale throughout the brain. This heterogeneity is apparent when comparing hierarchically organized local representations of perceptual processes to flexible transmodal representations of more abstract cognitive processes such as social and affective operations. An open question is how the spatial scale of analytic approaches interacts with the spatial scale of the representations under investigation. In this article, we describe how multivariate analyses can be viewed as existing on a spatial spectrum, anchored by searchlights used to identify locally distributed patterns of information on one end, whole brain approach used to identify diffuse neural representations at the other, and region-based approaches in between. We describe how these distinctions are an important and often overlooked analytic consideration and provide heuristics to compare these different techniques to choose based on the analyst’s inferential goals. 

   more » « less
3. Time‐varying feature selection for longitudinal analysis

   https://doi.org/10.1002/sim.8412  

   Xue, Lan; Shu, Xinxin; Shi, Peibei; Wu, Colin_O; Qu, Annie (November 2019, Statistics in Medicine)

   We propose time‐varying coefficient model selection and estimation based on the spline approach, which is capable of capturing time‐dependent covariate effects. The new penalty function utilizes local‐region information for varying‐coefficient estimation, in contrast to the traditional model selection approach focusing on the entire region. The proposed method is extremely useful when the signals associated with relevant predictors are time‐dependent, and detecting relevant covariate effects in the local region is more scientifically relevant than those of the entire region. Our simulation studies indicate that the proposed model selection incorporating local features outperforms the global feature model selection approaches. The proposed method is also illustrated through a longitudinal growth and health study from National Heart, Lung, and Blood Institute. 

   more » « less
4. Semantic embedding for regions of interest

   https://doi.org/10.1007/s00778-020-00647-0  

   Paul, Debjyoti; Li, Feifei; Phillips, Jeff M. (May 2021, The VLDB Journal)

   null (Ed.)

   Abstract The available spatial data are rapidly growing and also diversifying. One may obtain in large quantities information such as annotated point/place of interest (POIs), check-in comments on those POIs, geo-tagged microblog comments, and demarked regions of interest (ROI). All sources interplay with each other, and together build a more complete picture of the spatial and social dynamics at play in a region. However, building a single fused representation of these data entries has been mainly rudimentary, such as allowing spatial joins. In this paper, we extend the concept of semantic embedding for POIs (points of interests) and devise the first semantic embedding of ROIs, and in particular ones that captures both its spatial and its semantic components. To accomplish this, we develop a multipart network model capturing the relationships between the diverse components, and through random-walk-based approaches, use this to embed the ROIs. We demonstrate the effectiveness of this embedding at simultaneously capturing both the spatial and semantic relationships between ROIs through extensive experiments. Applications like popularity region prediction demonstrate the benefit of using ROI embedding as features in comparison with baselines. 

   more » « less
5. A Physics-Informed, Machine Learning Emulator of a 2D Surface Water Model: What Temporal Networks and Simulation-Based Inference Can Help Us Learn about Hydrologic Processes

   https://doi.org/10.3390/w13243633  

   Maxwell, Reed M.; Condon, Laura E.; Melchior, Peter (December 2021, Water)

   While machine learning approaches are rapidly being applied to hydrologic problems, physics-informed approaches are still relatively rare. Many successful deep-learning applications have focused on point estimates of streamflow trained on stream gauge observations over time. While these approaches show promise for some applications, there is a need for distributed approaches that can produce accurate two-dimensional results of model states, such as ponded water depth. Here, we demonstrate a 2D emulator of the Tilted V catchment benchmark problem with solutions provided by the integrated hydrology model ParFlow. This emulator model can use 2D Convolution Neural Network (CNN), 3D CNN, and U-Net machine learning architectures and produces time-dependent spatial maps of ponded water depth from which hydrographs and other hydrologic quantities of interest may be derived. A comparison of different deep learning architectures and hyperparameters is presented with particular focus on approaches such as 3D CNN (that have a time-dependent learning component) and 2D CNN and U-Net approaches (that use only the current model state to predict the next state in time). In addition to testing model performance, we also use a simplified simulation based inference approach to evaluate the ability to calibrate the emulator to randomly selected simulations and the match between ML calibrated input parameters and underlying physics-based simulation. 

   more » « less