Location Prediction Research Articles

DeepmRNALoc: A Novel Predictor of Eukaryotic mRNA Subcellular Localization Based on Deep Learning.

The subcellular localization of messenger RNA (mRNA) precisely controls where protein products are synthesized and where they function. However, obtaining an mRNA's subcellular localization through wet-lab experiments is time-consuming and expensive, and many existing mRNA subcellular localization prediction algorithms need to be improved. In this study, a deep neural network-based eukaryotic mRNA subcellular location prediction method, DeepmRNALoc, was proposed, utilizing a two-stage feature extraction strategy that featured bimodal information splitting and fusing for the first stage and a VGGNet-like CNN module for the second stage. The five-fold cross-validation accuracies of DeepmRNALoc in the cytoplasm, endoplasmic reticulum, extracellular region, mitochondria, and nucleus were 0.895, 0.594, 0.308, 0.944, and 0.865, respectively, demonstrating that it outperforms existing models and techniques.

SanMove: next location recommendation via self-attention network

PurposeThis paper aims to address the following issues: (1) most existing methods are based on recurrent network, which is time-consuming to train long sequences due to not allowing for full parallelism; (2) personalized preference generally are not considered reasonably; (3) existing methods rarely systematically studied how to efficiently utilize various auxiliary information (e.g. user ID and time stamp) in trajectory data and the spatiotemporal relations among nonconsecutive locations.Design/methodology/approachThe authors propose a novel self-attention network–based model named SanMove to predict the next location via capturing the long- and short-term mobility patterns of users. Specifically, SanMove uses a self-attention module to capture each user's long-term preference, which can represent her personalized location preference. Meanwhile, the authors use a spatial-temporal guided noninvasive self-attention (STNOVA) module to exploit auxiliary information in the trajectory data to learn the user's short-term preference.FindingsThe authors evaluate SanMove on two real-world datasets. The experimental results demonstrate that SanMove is not only faster than the state-of-the-art recurrent neural network (RNN) based predict model but also outperforms the baselines for next location prediction.Originality/valueThe authors propose a self-attention-based sequential model named SanMove to predict the user's trajectory, which comprised long-term and short-term preference learning modules. SanMove allows full parallel processing of trajectories to improve processing efficiency. They propose an STNOVA module to capture the sequential transitions of current trajectories. Moreover, the self-attention module is used to process historical trajectory sequences in order to capture the personalized location preference of each user. The authors conduct extensive experiments on two check-in datasets. The experimental results demonstrate that the model has a fast training speed and excellent performance compared with the existing RNN-based methods for next location prediction.

Prediction of apoptosis protein subcellular location based on amphiphilic pseudo amino acid composition.

Introduction: Apoptosis proteins play an important role in the process of cell apoptosis, which makes the rate of cell proliferation and death reach a relative balance. The function of apoptosis protein is closely related to its subcellular location, it is of great significance to study the subcellular locations of apoptosis proteins. Many efforts in bioinformatics research have been aimed at predicting their subcellular location. However, the subcellular localization of apoptotic proteins needs to be carefully studied. Methods: In this paper, based on amphiphilic pseudo amino acid composition and support vector machine algorithm, a new method was proposed for the prediction of apoptosis proteins\x{2019} subcellular location. Results and Discussion: The method achieved good performance on three data sets. The Jackknife test accuracy of the three data sets reached 90.5%, 93.9% and 84.0%, respectively. Compared with previous methods, the prediction accuracies of APACC_SVM were improved.

Predicting Location of Tweets Using Machine Learning Approaches

Twitter, one of the most popular microblogging platforms, has tens of millions of active users worldwide, generating hundreds of millions of posts every day. Twitter posts, referred to as “tweets”, the short and the noisy text, bring many challenges with them, such as in the case of some emergency or disaster. Predicting the location of these tweets is important for social, security, human rights, and business reasons and has raised noteworthy consideration lately. However, most Twitter users disable the geo-tagging feature, and their home locations are neither standardized nor accurate. In this study, we applied four machine learning techniques named Logistic Regression, Random Forest, Multinomial Naïve Bayes, and Support Vector Machine with and without the utilization of the geo-distance matrix for location prediction of a tweet using its textual content. Our extensive experiments on our vast collection of Arabic tweets From Saudi Arabia with different feature sets yielded promising results with 67% accuracy.

Prediction of the probabilistic water leak location in underground pipelines using Monte Carlo simulation

Abstract The water loss rates in Brazil are still very concerning and cause huge financial, social, and environmental damages. Although great advances have been made in acoustic leak detection methods, some of them are intrusive or at least demand convenient access points to the pipe's surface. Furthermore, some methods are expensive and require a highly experienced and qualified operator. Thus, the goal of this paper is to establish a benchmark using a simplistic analytical model for the ground surface analysis for leaks location, simplifying the search process carried out in water networks, and requiring minimal system interference. For leak location prediction, the proposed method considered two seismic wave paths – the compressional body waves and the Rayleigh superficial waves. The traveltime curves of the compressional waves are calculated as a function of the surface wave's experimental information. Monte Carlo Simulations were carried out to test each model's precision considering uncertainties related to the input parameters of the problem. At last, through a parametric study, the two seismic events were combined to obtain better accuracy in predicting the leakage epicenter location. In the results, the maximum absolute error found was 15 cm, and the minimum improvement obtained with the parametric study was 27.4%.

Impact Ionization Coefficient Prediction of a Lateral Power Device Using Deep Neural Network

Nowadays, the impact ionization coefficient in the avalanche breakdown theory is obtained using 1-D PN junctions or SBDs, and is considered to be a constant determined by the material itself only. In this paper, the impact ionization coefficient of silicon in a 2D lateral power device is found to be no longer a constant, but instead a function of the 2D coupling effects. The impact ionization coefficient of silicon that considers the 2D depletion effects in real-world devices is proposed and extracted in this paper. The extracted impact ionization coefficient indicates that the conventional empirical impact ionization in the Fulop equation is not suitable for the analysis of 2D lateral power devices. The veracity of the proposed impact ionization coefficient is validated by the simulations obtained from TCAD tools. Considering the complexity of direct modeling, a new prediction method using deep neural networks is proposed. The prediction method demonstrates 97.67% accuracy for breakdown location prediction and less than 6% average error for the impact ionization coefficient prediction compared with the TCAD simulation.

Preconditioning of Summer Melt Ponds From Winter Sea Ice Surface Temperature

AbstractComparing helicopter‐borne surface temperature maps in winter and optical orthomosaics in summer from the year‐long Multidisciplinary drifting Observatory for the Study of Arctic Climate expedition, we find a strong geometric correlation between warm anomalies in winter and melt pond location the following summer. Warm anomalies are associated with thinner snow and ice, that is, surface depression and refrozen leads, that allow for water accumulation during melt. Warm surface temperature anomalies in January were 0.3–2.5 K warmer on sea ice that later formed melt ponds. A one‐dimensional steady‐state thermodynamic model shows that the observed surface temperature differences are in line with the observed ice thickness and snow depth. We demonstrate the potential of seasonal prediction of summer melt pond location and coverage from winter surface temperature observations. A threshold‐based classification achieves a correct classification for 41% of the melt ponds.

Areas of the Terai Arc landscape in Nepal at risk of forest fire identified by fuzzy analytic hierarchy process

Forest fire frequency has increased in South Asia in recent decades, with a growing impact on forest ecosystems. Nepal's Terai Arc Landscape (TAL) is one of the most ecologically important landscapes in Asia, hosting a great diversity of endangered flora and fauna. To better predict the threat of fire to forest ecosystems in the TAL, we identified fire-prone areas using fuzzy AHP methods. We produced a fire risk map by applying the weighted linear combination method using topographic (slope, aspect, and elevation), climatic (temperature, precipitation, and wind speed), biophysical (normalised difference vegetation index and landcover classes), and anthropogenic variables (distance to road and proximity to settlements). We then validated the map with records of past fires by applying a confusion matrix. The accuracy of our technique of fire location prediction was above 95%, with a kappa coefficient of 0.93 and AUC value of 0.83. Locations of medium to very high forest fire risk were found in around 51% of the study area, usually in the vicinity of areas affected by anthropogenic factors. This landscape-level study shows the potential of multicriteria prediction models to inform the preparation of national and regional forest fire management strategies and plans.

Driver Intent Prediction and Collision Avoidance With Barrier Functions

For autonomous vehicles and driver assist systems, path planning and collision avoidance algorithms benefit from accurate predictions of future location of other vehicles and intent of their drivers. In the literature, the algorithms that provide driver intent belong to two categories: those that use physics based models with some type of filtering, and machine learning based approaches. In this paper we employ barrier functions (BF) to decide driver intent. BFs are typically used to prove safety by establishing forward invariance of an admissible set. Here, we decide if the “target” vehicle is violating one or more possibly fictitious (i.e., non-physical) barrier constraints determined based on the context provided by the road geometry. The algorithm has a very small computational footprint and better false positive and negative rates than some of the alternatives. The predicted intent is then used by a control barrier function (CBF) based collision avoidance system to prevent unnecessary interventions, for either an autonomous or human-driven vehicle.

On cavitation in the radial flow of a thin lubricating film between two overlying disks

This study is focused on the characterization and modeling of aviation fuel cavitation physics in radial flow in a thin layer between two disks—a geometry highly relevant to aviation fuel pump systems. This involves a lower circular disk with a centrally located fuel injection port and a matching disk resting on top of the lower disk. In the described experiments, we have quantified and compared cavitating disk behavior for distilled water and JP-5 fuel at various inlet supply pressures via high-speed imaging and radial pressure measurements. High-speed imaging data were used to quantify the radial collapse location of cavitation voids. An enhanced gradient shadowgraphy method was employed to obtain the spatial–temporal evolution of propagating bubbly shock waves. This technique revealed unsteady shock waves propagating in a spiral motion in JP-5 fuel, while a standing bubbly shock was observed in water. In our modeling efforts, the Rayleigh–Plesset equation was adapted to a spatial form in order to predict the radial location of cavitation bubble collapse. Further work incorporated the spatial Rayleigh–Plesset equation into the barotropic model that has been used previously for cavitating nozzle flows and generalized it to radial flow geometry in order to reproduce radial pressure profiles obtained from the experiment. The model predictions of the radial location of bubble collapse and the radial pressure profiles are shown to be in excellent agreement with the experiments. This approach will be valuable for predicting aviation fuel cavitation in complex fuel system geometries.

A Federated Learning Framework for Fingerprinting-Based Indoor Localization in Multibuilding and Multifloor Environments

The participatory nature of federated learning (FL) makes it attractive for fingerprinting-based indoor localization in multibuilding and multifloor environments. A group of sensing clients can collaboratively leverage their private, local fingerprint data to help their edge server update a location prediction model. However, it is challenging to jointly handle the two involved issues, i.e., building-floor classification (BFC) and latitude–longitude regression (LLR), in a wide 3-D space through enabling FL on decentralized yet heterogeneous data and over an imperfect wireless network. In this article, we confront these challenges and propose an FL framework, FedLoc3D, for both BFC and LLR. Specifically, the former issue is addressed by an FedDSC-BFC approach, which generates a multilabel classification model based on a convolutional neural network with depthwise separable convolutions. The latter issue is addressed by an FedADA-LLR approach, which develops a multitarget regression model based on a deep neural network with autoencoder and data augmentation. Extensive experiments on a real-world data set of WiFi fingerprints are carried out, and our approaches with enhanced capabilities of feature extraction, generalization, and convergence are validated to improve both localization accuracy and learning efficiency under data heterogeneity and network instability.

Fusing high-resolution multispectral image with trajectory for user next travel location prediction

Fusing high-resolution multispectral image with trajectory for user next travel location prediction

Validation of a multiple marker test for early pregnancy outcome prediction.

To validate the use of a multiple biomarker test panel for predicting first trimester pregnancy outcome in a multi-center cohort. A case-control study of women presenting with pain and bleeding in early pregnancy at 5-10 weeks gestational age was performed at three academic centers. Sera from women with ectopic pregnancy (EP), viable intrauterine pregnancy (IUP), and miscarriage (SAB) were analyzed via immunoassay for Activin A (AA), Progesterone (P4), A Disintegrin And Metalloprotease-12 (ADAM12), pregnancy-associated plasma protein A (PAPP-A), glycodelin (Glyc), and human chorionic gonadotropin (hCG). Biomarkers were assessed for reproducibility using medians, ranges, standard deviations, and area under receiver-operating characteristic curve (AUC) and accuracy in early pregnancy outcome classification compared to a previous derivation population. In 192 pregnancies, the biomarkers demonstrated good reproducibility with similar medians, ranges, and AUCs when compared to the derivation population except glycodelin. Pregnancy location was conclusively classified in 53% (n = 94) of the whole study sample with 78% accuracy. Pregnancy viability was conclusively classified in 58% (n = 112) of the new sample with 89% accuracy. Results were similar with subsequent model revisions where glycodelin was excluded and in the subgroups of subjects with a hCG below 2000 mIU/mL and a gestational age less than 6 weeks. The use of a panel of biomarkers to maximize test accuracy of a prediction of pregnancy location and prediction of pregnancy viability was reproducible and validated in an external population from which it was derived, but clinical utility is limited based on the test characteristics obtained.

Multifocus object detector for vehicle tracking in smart cities using spatiotemporal attention map

In recent days, an unrivaled proliferation of autonomous driving becomes an active area of research in the field of smart city applications. Millions of autonomous cars emerge as ubiquitous in smart cities in the foreseeable future. Vehicle detection and tracking is one of the important cases in smart city applications. It clarifies that the efficient scene parsing methods will be a huge potential. There were several traditional methods employed for the detection of object for the purpose of vehicle detection and road segmentation. However, there were some drawbacks such as reduced rate of detection accuracy, improper road segmentations, and missing detections of occluded objects exist. To overcome the existing issues, we propose an effective multi focus object detector (MFOD) that retrieves spatiotemporal data and considers multiple cues of an object for detection and tracking. The proposed MFOD performs lane detection, vehicle detection, and tracking. To accomplish this task, modified Hough transform is proposed for lane detection and road segmentation. Residual attention unit (RAU) with shared convolutional neural network layers is deployed for vehicle detection and tracking. It makes use of motion model to update the network and creates visualization map to explore the visibility of occluded objects and spatial attention map for accurate prediction of object location. The proposed MFOD is also extended for scene parsing task. We performed extensive analysis of the proposed MFOD with standard evaluation metrics on two different datasets, such as KITTI for object detection and tracking task and cityscapes for scene parsing task. The experimental results demonstrate that the proposed MFOD achieves 3% to 4% improvement compared with the existing methods.

Damage evolution process of fiber-reinforced backfill based on acoustic emission three-dimensional localization

The spatial localization of the damage evolution caused by a fiber-reinforced backfill is critical for the prediction of the rupture location in a mining area filling. Acoustic emission (AE) three-dimensional localization tests under uniaxial compression were conducted on various fiber-reinforced backfills to study the damage evolution process. The general characteristics of energy count and the law of evolution for the spatial localization damage of fiber-reinforced backfills were derived by analyzing the time sequence evolution process of energy count and the spatial distribution of the AE events. The results show that the time-series evolution process of the backfill energy count can be approximately divided into the three periods: rising, quiet, and active. The damage mode of the fiber-reinforced backfill in uniaxial compression is determined as ductile damage, and the penetration mode is determined as shear + tension penetration. The evolution of the spatial distribution of the AE events of a backfill can not only reflect the spatial location of damage during the entire loading process but also determine the degree of damage at a certain location according to the density of spatial points. The results of this study can provide a theoretical basis for the prediction of rupture localization for mine filling.

Climatological predictions of the auroral zone locations driven by moderate and severe space weather events

Auroral zones are regions where, in an average sense, aurorae due to solar activity are most likely spotted. Their shape and, similarly, the geographical locations most vulnerable to extreme space weather events (which we term ‘danger zones’) are modulated by Earth’s time-dependent internal magnetic field whose structure changes on yearly to decadal timescales. Strategies for mitigating ground-based space weather impacts over the next few decades can benefit from accurate forecasts of this evolution. Existing auroral zone forecasts use simplified assumptions of geomagnetic field variations. By harnessing the capability of modern geomagnetic field forecasts based on the dynamics of Earth’s core we estimate the evolution of the auroral zones and of the danger zones over the next 50 years. Our results predict that space-weather related risk will not change significantly in Europe, Australia and New Zealand. Mid-to-high latitude cities such as Edinburgh, Copenhagen and Dunedin will remain in high-risk regions. However, northward change of the auroral and danger zones over North America will likely cause urban centres such as Edmonton and Labrador City to be exposed by 2070 to the potential impact of severe solar activity.

VMLH: Efficient Video Moment Location via Hashing

Video-moment location by query is a hot topic in video understanding. However, most of the existing methods ignore the importance of location efficiency in practical application scenarios; video and query sentences have to be fed into the network at the same time during the retrieval, which leads to low efficiency. To address this issue, in this study, we propose an efficient video moment location via hashing (VMLH). In the proposed method, query sentences and video clips are, respectively, converted into hash codes and hash code sets, in which the semantic similarity between query sentences and video clips is preserved. The location prediction network is designed to predict the corresponding timestamp according to the similarity among hash codes, and the videos do not need to be fed into the network during the process of retrieval and location. Furthermore, different from the existing methods, which require complex interactions and fusion between video and query sentences, the proposed VMLH method only needs a simple XOR operation among codes to locate the video moment with high efficiency. This paper lays the foundation for fast video clip positioning and makes it possible to apply large-scale video clip positioning in practice. The experimental results on two public datasets demonstrate the effectiveness of the method.

CASTLE: A CONTEXT-AWARE SPATIAL-TEMPORAL LOCATION EMBEDDING PRE-TRAINING MODEL FOR NEXT LOCATION PREDICTION

Abstract. Next location prediction is helpful for service recommendation, public safety, intelligent transportation, and other location-based applications. Existing location prediction methods usually use sparse check-in trajectories and require massive historical data to capture complex spatial-temporal correlations. High spatial-temporal resolution trajectories have rich information. However, obtaining personal trajectories with long time series and high spatio-temporal resolution usually proves challenging. Herein, this paper proposes a two-stage Context-Aware Spatial-Temporal Location Embedding (CASTLE) model, a multi-modal pre-training model for sequence-to- sequence prediction tasks. The method is built in two steps. First, large-scale location datasets, which are sparse but easier to be acquired (i.e., check-in and anomalous navigation data), are used for pre-training location embedding to capture the multi-functional properties under different contexts. After that, the learned contextual embedding is used for downstream location prediction in small-scale but higher spatio-temporal resolution trajectory datasets. Specifically, the CASTLE model combines Bidirectional and Auto-Regressive Transformers to generate contextual embedding vectors rather than a fixed vector for each location. Furthermore, we introduce a location and time-aware encoder to reflect the spatial distances between locations and visit times. Experiments are conducted on two real trajectory datasets. The results show that the CASTLE model can pre-train beneficial location embedding and outperforms the model without pre-training by 4.6–7.1%. The proposed method is expected to improve the next location prediction accuracy without massive historical data, which will greatly drive the use of trajectory data.

Wheel impact test by deep learning: prediction of location and magnitude of maximum stress

For ensuring vehicle safety, the impact performance of wheels during wheel development must be ensured through a wheel impact test. However, manufacturing and testing a real wheel requires a significant time and money because developing an optimal wheel design requires numerous iterative processes to modify the wheel design and verify the safety performance. Accordingly, wheel impact tests have been replaced by computer simulations such as finite element analysis (FEA); however, it still incurs high computational costs for modeling and analysis, and requires FEA experts. In this study, we present an aluminum road wheel impact performance prediction model based on deep learning that replaces computationally expensive and time-consuming 3D FEA. For this purpose, 2D disk-view wheel image data, 3D wheel voxel data, and barrier mass values used for the wheel impact test were utilized as the inputs to predict the magnitude of the maximum von Mises stress, corresponding location, and the stress distribution of the 2D disk-view. The input data were first compressed into a latent space with a 3D convolutional variational autoencoder (cVAE) and 2D convolutional autoencoder (cAE). Subsequently, the fully connected layers were used to predict the impact performance, and a decoder was used to predict the stress distribution heatmap of the 2D disk-view. The proposed model can replace the impact test in the early wheel-development stage by predicting the impact performance in real-time and can be used without domain knowledge. The time required for the wheel development process can be reduced using this mechanism.

APPLYING 2D NUMERICAL MODEL TO CALCULATE THE SCOUR AT THE CUA DAI BRIDGE, QUANG NAM PROVINCE

Abstract: When the Cua Dai bridge is built across the Thu Bon River, it causes a narrowing of the flow leading to scour, which seriously threatens the pier and endangers the structure. In order to limit the influence of scour the bridge, it is necessary to ensure that the scour under the bridge is within the allowable range. At the same time, it is necessary to take measures to limit and overcome the causes of scour. Therefore, the study and prediction of locations that are likely to cause scour are of great significance in choosing effective and versatile prevention solutions for works on this river section. Most of the scour formulas for bridge use the 1D average velocity to calculate; The calculated results are much different from the real distribution of flow velocity in the river. Therefore, the results of scour calculation often have big deviations from reality, especially when the river cross-section has no flats or curved rivers... The article focuses on researching scour at the bridge pier based on the two-dimensional horizontal velocity in front of the pier by using RIVER 2D model, built from the mathematical model of two horizontal flows and solved by the finite element method. The results of the paper show that the scour depth of pier by the one-dimensional (1D) average velocity gives much larger than when applying the two-dimensional horizontal velocity to calculate the scour.