Enhanced evaluation of a Lagrangian-particle air pollution model based on a Šaleška region field data set

AbstractThe predictive performance of a machine learning model highly depends on the corresponding hyper-parameter setting. Hence, hyper-parameter tuning is often indispensable. Normally such tuning requires the dedicated machine learning model to be trained and evaluated on centralized data to obtain a performance estimate. However, in a distributed machine learning scenario, it is not always possible to collect all the data from all nodes due to privacy concerns or storage limitations. Moreover, if data has to be transferred through low bandwidth connections it reduces the time available for tuning. Model-Based Optimization (MBO) is one state-of-the-art method for tuning hyper-parameters but the application on distributed machine learning models or federated learning lacks research. This work proposes a framework $$\textit{MODES}$$ MODES that allows to deploy MBO on resource-constrained distributed embedded systems. Each node trains an individual model based on its local data. The goal is to optimize the combined prediction accuracy. The presented framework offers two optimization modes: (1) $$\textit{MODES}$$ MODES -B considers the whole ensemble as a single black box and optimizes the hyper-parameters of each individual model jointly, and (2) $$\textit{MODES}$$ MODES -I considers all models as clones of the same black box which allows it to efficiently parallelize the optimization in a distributed setting. We evaluate $$\textit{MODES}$$ MODES by conducting experiments on the optimization for the hyper-parameters of a random forest and a multi-layer perceptron. The experimental results demonstrate that, with an improvement in terms of mean accuracy ($$\textit{MODES}$$ MODES -B), run-time efficiency ($$\textit{MODES}$$ MODES -I), and statistical stability for both modes, $$\textit{MODES}$$ MODES outperforms the baseline, i.e., carry out tuning with MBO on each node individually with its local sub-data set.

Download Full-text

Monte Carlo methods for sensitivity studies of large-scale air pollution model

10.1063/5.0034848 ◽

2020 ◽

Author(s):

Tz. Ostromsky ◽

V. Todorov ◽

I. Dimov

Keyword(s):

Air Pollution ◽

Monte Carlo ◽

Monte Carlo Methods ◽

Large Scale ◽

Sensitivity Studies ◽

Air Pollution Model

Download Full-text

Adaptive sampling of potential-field data: A direct approach to compressive inversion

Geophysics ◽

10.1190/geo2013-0087.1 ◽

2014 ◽

Vol 79 (1) ◽

pp. IM1-IM9 ◽

Cited By ~ 20

Author(s):

Nathan Leon Foks ◽

Richard Krahenbuhl ◽

Yaoguo Li

Keyword(s):

Potential Field ◽

Adaptive Sampling ◽

Field Data ◽

Computational Cost ◽

Large Field ◽

Magnetic Data ◽

Data Sampling ◽

Data Types ◽

Data Set ◽

Potential Field Data

Compressive inversion uses computational algorithms that decrease the time and storage needs of a traditional inverse problem. Most compression approaches focus on the model domain, and very few, other than traditional downsampling focus on the data domain for potential-field applications. To further the compression in the data domain, a direct and practical approach to the adaptive downsampling of potential-field data for large inversion problems has been developed. The approach is formulated to significantly reduce the quantity of data in relatively smooth or quiet regions of the data set, while preserving the signal anomalies that contain the relevant target information. Two major benefits arise from this form of compressive inversion. First, because the approach compresses the problem in the data domain, it can be applied immediately without the addition of, or modification to, existing inversion software. Second, as most industry software use some form of model or sensitivity compression, the addition of this adaptive data sampling creates a complete compressive inversion methodology whereby the reduction of computational cost is achieved simultaneously in the model and data domains. We applied the method to a synthetic magnetic data set and two large field magnetic data sets; however, the method is also applicable to other data types. Our results showed that the relevant model information is maintained after inversion despite using 1%–5% of the data.

Download Full-text

Research of Personalized Recommendation Technology Based on Knowledge Graphs

Applied Sciences ◽

10.3390/app11157104 ◽

2021 ◽

Vol 11 (15) ◽

pp. 7104

Author(s):

Xu Yang ◽

Ziyi Huan ◽

Yisong Zhai ◽

Ting Lin

Keyword(s):

Neural Network ◽

Hot Spot ◽

Experimental Results ◽

Graph Representation ◽

Personalized Recommendation ◽

Knowledge Graph ◽

Learning Technology ◽

Data Set ◽

Model Based ◽

Knowledge Graphs

Nowadays, personalized recommendation based on knowledge graphs has become a hot spot for researchers due to its good recommendation effect. In this paper, we researched personalized recommendation based on knowledge graphs. First of all, we study the knowledge graphs’ construction method and complete the construction of the movie knowledge graphs. Furthermore, we use Neo4j graph database to store the movie data and vividly display it. Then, the classical translation model TransE algorithm in knowledge graph representation learning technology is studied in this paper, and we improved the algorithm through a cross-training method by using the information of the neighboring feature structures of the entities in the knowledge graph. Furthermore, the negative sampling process of TransE algorithm is improved. The experimental results show that the improved TransE model can more accurately vectorize entities and relations. Finally, this paper constructs a recommendation model by combining knowledge graphs with ranking learning and neural network. We propose the Bayesian personalized recommendation model based on knowledge graphs (KG-BPR) and the neural network recommendation model based on knowledge graphs(KG-NN). The semantic information of entities and relations in knowledge graphs is embedded into vector space by using improved TransE method, and we compare the results. The item entity vectors containing external knowledge information are integrated into the BPR model and neural network, respectively, which make up for the lack of knowledge information of the item itself. Finally, the experimental analysis is carried out on MovieLens-1M data set. The experimental results show that the two recommendation models proposed in this paper can effectively improve the accuracy, recall, F1 value and MAP value of recommendation.

Download Full-text

Improvement of METANET Model Based on Critical Density Modeling for Traffic Control

Canadian Journal of Civil Engineering ◽

10.1139/cjce-2020-0735 ◽

2021 ◽

Author(s):

Lina Fu ◽

Jie Fang ◽

Yunjie Lyu ◽

Huahui Xie

Keyword(s):

Traffic Safety ◽

Traffic Control ◽

Traffic Congestion ◽

Field Data ◽

Critical Density ◽

Predictive Performance ◽

Density Model ◽

Model Based ◽

Freeway Control ◽

Density Modeling

Freeway control has been increasingly used as an innovative approach to ease traffic congestion, improve traffic safety and reduce exhaust emissions. As an important predictive model involved in freeway control, the predictive performance of METANET greatly influences the effect of freeway control. This paper focuses on modifying the METANET model by modeling the critical density. Firstly, the critical density model is deduced based on the catastrophe theory. Then, the perturbation wave and traveling wave that are obtained using the macro and micro data, respectively, have been developed to modify the above proposed critical density model. Finally, the numerical simulation is established to evaluate the effectiveness of the modified METANET model based on the field data from the realistic motorway network. The results show that overall, the predicted data from the modified METANET model are closer to the field data than those obtained from the original model.

Download Full-text

Building training data set for deep learning-based P- and S-wave separation: Field data case

10.1190/segam2021-3583348.1 ◽

2021 ◽

Author(s):

Yanwen Wei ◽

Yunyue Elita Li ◽

Haohuan Fu

Keyword(s):

Deep Learning ◽

Field Data ◽

Training Data ◽

S Wave ◽

Data Set ◽

Wave Separation ◽

Separation Field

Download Full-text

A Differentiated Spatial Assessment of Urban Ecosystem Services Based on Land Use Data in Halle, Germany

Land ◽

10.3390/land7030101 ◽

2018 ◽

Vol 7 (3) ◽

pp. 101 ◽

Cited By ~ 7

Author(s):

Janis Arnold ◽

Janina Kleemann ◽

Christine Fürst

Keyword(s):

Land Use ◽

Air Pollution ◽

Ecosystem Services ◽

Urban Areas ◽

Urban Ecosystem ◽

Data Set ◽

Urban Ecosystem Services ◽

Climate Regulation ◽

Land Use Types ◽

Cooling Effects

Urban ecosystem services (ES) contribute to the compensation of negative effects caused by cities by means of, for example, reducing air pollution and providing cooling effects during the summer time. In this study, an approach is described that combines the regional biotope and land use data set, hemeroby and the accessibility of open space in order to assess the provision of urban ES. Hemeroby expresses the degree of naturalness of land use types and, therefore, provides a differentiated assessment of urban ES. Assessment of the local capacity to provide urban ES was conducted with a spatially explicit modeling approach in the city of Halle (Saale) in Germany. The following urban ES were assessed: (a) global climate regulation, (b) local climate regulation, (c) air pollution control, (d) water cycle regulation, (e) food production, (f) nature experience and (g) leisure activities. We identified areas with high and low capacity of ES in the urban context. For instance, the central parts of Halle had very low or no capacity to provide ES due to highly compact building styles and soil sealing. In contrast, peri-urban areas had particularly high capacities. The potential provision of regulating services was spatially limited due to the location of land use types that provide these services.

Download Full-text