Inferring trip purposes and uncovering travel patterns from taxi trajectory data

Global positioning system-enabled vehicles provide an efficient way to obtain large quantities of movement data for individuals. However, the raw data usually lack activity information, which is highly valuable for a range of applications and services. This study provides a novel and practical framework for inferring the trip purposes of taxi passengers such that the semantics of taxi trajectory data can be enriched. The probability of points of interest to be visited is modeled by Bayes’ rules, which take both spatial and temporal constraints into consideration. Combining this approach with Monte Carlo simulations, we conduct a study on Shanghai taxi trajectory data. Our results closely approximate the residents’ travel survey data in Shanghai. Furthermore, we reveal the spatiotemporal characteristics of nine daily activity types based on inference results, including their temporal regularities, spatial dynamics, and distributions of trip lengths and directions. In the era of big data, we encounter the dilemma of “trajectory data rich but activity information poor” when investigating human movements from various data sources. This study presents a promising step toward mining abundant activity information from individuals’ trajectories.     

Modelling the influence of environmental and weather factors on the density of the invasive polychaete Boccardia proboscidea

The inter‐tidal zone around sewage discharges in a Southwest Atlantic shore (Mar del Plata, Argentina) is currently colonized by extensive inter‐tidal reefs of the invasive spionid Boccardia proboscidea. Understanding the links between both human and natural disturbances and the massive development of non‐indigenous species will help prevent marine bioinvasions, which are already favoured by global oceanic trade. We present herein predictive models for variations in the density of B. proboscidea around sewage discharges of Mar del Plata, using environmental (pH, turbidity, temperature, salinity and total organic matter content), weather (wind direction and storm records), spatial (sites) and temporal (season and year) variables. Density variations were modelled by generalized linear models, and model averaging (multimodel inference) was used to obtain predicted values. The highest predicted values of B. proboscidea density occurred at sites to the south of the sewage effluent in spring. These sites are more affected by urban effluent discharges and they showed increased B. proboscidea density when the north wind was predominant. In addition, B. proboscidea density values were higher in sites with 20–22 °C (seawater temperature), high total organic matter content in sediments and low salinity. The averaged model was only a good ‘predictive model’ for sites to the north of the outfall, but was useful as an ‘explanatory model’ in all sites. Such predictions may help to back up conservation and management policies and decisions.     

圆弧破坏边坡反演设计的ANN方法与ANFIS方法的比较研究

圆弧破坏边坡反演设计方法,是根据边坡的岩土力学参数及边坡高度,在确保安全系数满足要求的前提下,对圆弧破坏边坡的边坡角进行设计的一种新方法。建立圆弧破坏边坡反演设计方法的关键,就是要对大量稳定圆弧破坏边坡实例加以集成,以建立从{ , c, , H, F}到 的映射。文中收集了大量稳定圆弧破坏边坡实例,采用ANN和ANFIS对其进行了集成,分别建立了圆弧破坏边坡反演设计的ANN方法和ANFIS方法,进而又采用圆弧破坏边坡实例对两者的反演结果进行了比较。结果表明,ANFIS更适合于用来建立圆弧破坏边坡反演设计方法。     

Delineation of hydrologically similar units in a watershed based on fuzzy classification of soil hydraulic properties

Evaluation of flow and transport processes in a watershed‐scale requires that the watershed be divided into homogenous spatial units referred to as hydrologically similar units (HSUs). Although a few discretization schemes are already in use, a universally acceptable method of obtaining HSUs is yet to emerge. In this study, we developed a fuzzy inference system (FIS) to classify the saturated hydraulic conductivity (K_s) and two water‐retention parameters α and n into fuzzy logic‐based soil hydrologic classes (FSHCs). Analysis of these classes showed that soil properties within an FSHC have less variability and those between two FSHCs have large variability. This result suggested that soils belonging to a specific FSHC may be more similar than those across different FSHCs and may be grouped together to represent an HSU. Soils within a specific hydrologic class were aggregated to delineate HSUs within the watershed. For the Dengei Pahad micro‐watershed (DPW), this approach showed five distinct regions representing a discretized zone having similar soil hydraulic properties. Application of this approach on a larger international database of soil hydraulic properties revealed that the developed hydrologic classes are quite comparable across different databases. The delineated HSUs based on these FSHCs were also better than the soil series map of the watershed in maintaining the soil heterogeneity of the watershed. Moreover, this new discretization scheme using the SWAT modelling environment showed better performance than the soil series‐based discretization approach. Copyright © 2010 John Wiley & Sons, Ltd.     

On maximum likelihood estimation of parameters in incorrectly specified models of covariance for spatial data

The analysis of, and from, models of spatial data usually proceeds under the assumption, often implicit, that the correct model has been specified. However, any model identification procedures based on sample data are subject to error, and consequences of such errors then permeate subsequent analysis. Thus, an attempt to quantify some of these consequences is of interest. A standard framework for analysis is extended here, by introduction of information theory, to permit the study of effects of model misspecification on maximum likelihood estimators of parameters of model covariance. Asymptotically valid theoretical results are presented, and the relevance of these results to samples of finite sizes met in practice is assessed in a series of simulation experiments. The effect of model misspecification, and use of estimators of parameters of misspecified covariance models, on the practical problem of prediction at a previously unsampled location is considered briefly, and further areas for possible investigation are outlined.     

基于Sugeno型模糊推理的震中烈度预测模型

以Sugeno型模糊推理系统为基础,建立了以震级和震源深度为输入,以震中烈度为输出的震中烈度预测模型,并以四川地区震例数据为例,对模型构建的关键环节进行了详细的说明.通过与参考文献中的拟合模型进行预测数据对比分析,可得本文的推理预测模型精度更高,误差更小,且有更强的扩展性.     

Pumping-rate control for contaminated groundwater with VOCs by using fuzzy inference model

In this paper, a new method is proposed for the remediation of contaminated groundwater. The method is based on fuzzy inference and risk evaluation. The effectiveness of the presented method is assessed in numerical simulations. The main results of the study are as follows: 1) A pump-and-treat control algorithm was proposed by using the fuzzy inference and the human health risk calculation model, 2) It was found that applying the proposed algorithm was likely to reduce the pumped quantity, and 3) It was found that the proposed model can be used to notify residents of the human health risk; the model is based on the ASTM RBCA model for residents. A new fuzzy control system for contaminated groundwater can be used as a useful model for characterising the effects of contaminants on human health and providing helpful information on the human risk assessment of the contaminated groundwater site.     

A 24‐h Forecast of Oxidant Concentration in Tokyo Using Neural Network and Fuzzy Learning Approach

In this study, several types of adaptive network‐based fuzzy inference system (ANFIS) with different membership functions (MFs) and artificial neural network (ANN) were employed to predict hourly photochemical oxidants that were oxidizing substances such as ozone and peroxiacetyl nitrate produced by photochemical reactions. The results indicated that ANFIS statistically outperforms ANN in terms of hourly oxidant prediction. The minimum mean absolute percentage errors (MAPEs) of 4.99% could be achieved using ANFIS with bell shaped MFs. The maximum correlation coefficient, the minimum mean square errors, and the minimum root mean square errors were 0.99, 0.15, and 0.39, respectively. ANFIS's architecture consists of both ANN and fuzzy logic including linguistic expression of MFs and if‐then rules, so it can overcome the limitations of traditional neural network and increase the prediction performance.     

Modelling of infiltration using artificial intelligence techniques in semi-arid Iran

ABSTRACT

Infiltration plays a fundamental role in streamflow, groundwater recharge, subsurface flow, and surface and subsurface water quality and quantity. In this study, adaptive neuro-fuzzy inference system (ANFIS), support vector machine (SVM) and random forest (RF) models were used to determine cumulative infiltration and infiltration rate in arid areas in Iran. The input data were sand, clay, silt, density of soil and soil moisture, while the output data were cumulative infiltration and infiltration rate, the latter measured using a double-ring infiltrometer at 16 locations. The results show that SVM with radial basis kernel function better estimated cumulative infiltration (RMSE = 0.2791 cm) compared to the other models. Also, SVM with M4 radial basis kernel function better estimated the infiltration rate (RMSE = 0.0633 cm/h) than the ANFIS and RF models. Thus, SVM was found to be the most suitable model for modelling infiltration in the study area.