A neural network model forecasting for prediction of daily maximum ozone concentration in an industrialized urban area

Modeling the air pollution process using a novel multi-site and multi-scale method with adaptive utilization of spatio-temporal information

This study focuses on modeling air quality with an adaptive utilization of spatio-temporal information from multiple air quality monitoring stations under a multi-scale framework. To this end, it is necessary to consider different strategies to combine different methods to decompose the given series and to fuse multi-site information. Based on a systematic comparative analysis, we propose a novel multi-scale and multi-site modeling method named the multivariate empirical mode decomposition and spatial cosine-attention-based long short-term memory (MEMD-SCA-LSTM). The MEMD-SCA-LSTM first employs MEMD to decompose the multi-site air quality series into the scale-aligned components and then models the components at different scales. The high-frequency components are modeled by a novel SCA-LSTM, which employs LSTM with residual blocks to extract the temporal information and utilizes an attention mechanism based on the cosine similarity to adaptively extract interactions among different sites. Other relatively regular components are modeled by the LSTM. Empirical study on PM2.5 in Hong Kong has demonstrated the effectiveness of fusing multi-site information using the spatial attention (SA) mechanism under the multi-scale framework with MEMD. The proposed MEMD-SCA-LSTM can improve the one-day ahead modeling performance with the mean absolute error and the root mean square error reduced over 10%, compared to the baseline modeling methods. For the two-day and three-day ahead performance, the MEMD-SCA-LSTM is still the best one. Furthermore, by visualizing the attention weights, we illustrate that our proposed SCA-LSTM can overcome some limitations of the traditional attention mechanisms and that the attention weights exhibit more informative patterns which could be used to analysis the transport of air pollutant between sites. The proposed modeling method is a general method, which is feasible and applicable to other pollutants in other cities or regions.

The application of machine learning to air pollution research: A bibliometric analysis

Machine learning (ML) is an advanced computer algorithm that simulates the human learning process to solve problems. With an explosion of monitoring data and the increasing demand for fast and accurate prediction, ML models have been rapidly developed and applied in air pollution research. In order to explore the status of ML applications in air pollution research, a bibliometric analysis was made based on 2962 articles published from 1990 to 2021. The number of publications increased sharply after 2017, comprising approximately 75% of the total. Institutions in China and United States contributed half of all publications with most research being conducted by individual groups rather than global collaborations. Cluster analysis revealed four main research topics for the application of ML: chemical characterization of pollutants, short-term forecasting, detection improvement and optimizing emission control. The rapid development of ML algorithms has increased the capability to explore the chemical characteristics of multiple pollutants, analyze chemical reactions and their driving factors, and simulate scenarios. Combined with multi-field data, ML models are a powerful tool for analyzing atmospheric chemical processes and evaluating the management of air quality and deserve greater attention in future.

Prediction of ground-level ozone by SOM-NARX hybrid neural network based on the correlation of predictors

Current approaches to ozone prediction using hybrid neural networks are numerous but not perfect. Decomposition algorithms ignore the correlation between predictors and ozone, and feature extraction methods rarely select appropriate predictors in terms of correlation, especially for VOCs. Therefore, this study proposes a hybrid neural network model SOM-NARX based on the correlation of predictors. The model is based on MIC to filter predictors, using SOM to make predictors as feature sequences and using NARX networks to make predictions. Data from the JCDZURI site were used for training, testing, and validation. The results show that the correlation of the predictors, classification numbers of SOM, neuron numbers, and delay steps can affect prediction accuracy. Model comparison shows that the SOM-NARX model has 13.82, 10.60, 6.58% and 12.05, 9.44, 68.14% RMSE, MAE, and MAEP in winter and summer, which is smaller than CNN-LSTM, CNN-BiLSTM, CNN-GRU, SOM-LSTM, SOM-BiLSTM, and SOM-GRU.

A DEA-ANN-based analytical framework to assess and predict the efficiency of Canadian universities in a service supply chain context

Purpose

This research is about embedding service-based supply chain management (SCM) concepts in the education sector. Due to Canada's competitive education sector, the authors focus on Canadian universities.

Design/methodology/approach

The authors develop a framework for evaluating and forecasting university performance using data envelopment analysis (DEA) and artificial neural networks (ANNs) to assist education policymakers. The application of the proposed framework is illustrated based on information from 16 Canadian universities and by investigating their teaching and research performance.

Findings

The major findings are (1) applying the service SCM concept to develop a performance evaluation and prediction framework, (2) demonstrating the application of DEA-ANN for computing and predicting the efficiency of service SCM in Canadian universities, and (3) generating insights to enable universities to improve their research and teaching performances considering critical inputs and outputs.

Research limitations/implications

This paper presents a new framework for universities' performance assessment and performance prediction. DEA and ANN are integrated to aid decision-makers in evaluating the performances of universities.

Practical implications

The findings suggest that higher education policymakers should monitor attrition rates at graduate and undergraduate levels and provide financial support to facilitate research and concentrate on Ph.D. programs. Additionally, the sensitivity analysis indicates that selecting inputs and outputs is critical in determining university rankings.

Originality/value

This research proposes a new integrated DEA and ANN framework to assess and forecast future teaching and research efficiencies applying the service supply chain concept. The findings offer policymakers insights such as paying close attention to the attrition rates of undergraduate and postgraduate programs. In addition, prioritizing internal research support and concentrating on Ph.D. programs is recommended.

Feature importance: Opening a soil-transmitted helminth machine learning model via SHAP

In the field of landscape epidemiology, the contribution of machine learning (ML) to modeling of epidemiological risk scenarios presents itself as a good alternative. This study aims to break with the ”black box” paradigm that underlies the application of automatic learning techniques by using SHAP to determine the contribution of each variable in ML models applied to geospatial health, using the prevalence of hookworms, intestinal parasites, in Ethiopia, where they are widely distributed; the country bears the third-highest burden of hookworm in Sub-Saharan Africa. XGBoost software was used, a very popular ML model, to fit and analyze the data. The Python SHAP library was used to understand the importance in the trained model, of the variables for predictions. The description of the contribution of these variables on a particular prediction was obtained, using different types of plot methods. The results show that the ML models are superior to the classical statistical models; not only demonstrating similar results but also explaining, by using the SHAP package, the influence and interactions between the variables in the generated models. This analysis provides information to help understand the epidemiological problem presented and provides a tool for similar studies.

Temporal variation analysis, impact of COVID-19 on air pollutant concentrations, and forecasting of air pollutants over the cities of Bangalore and Delhi in India

Indian cities are highly vulnerable to atmospheric pollution in recent years, due to exponential growth in urbanisation and industrialisation, and the increased pollution has been made to focus on the temporal variation analysis and forecasting of air pollutants over major Indian cities like Delhi and Bangalore. PM_2.5 concentrations are nearly 60.5% less than the annual average value during monsoon season while 76.3% more during the winter months. Ozone concentrations increase during the summer months (~ 46.3% more than the annual average) in Delhi, whereas in Bangalore, ozone concentrations are more (~ 75% more than the annual average) during the winter months. Variations of carbon monoxide and nitrogen oxides are significantly less comparatively. COVID-19 lockdown has a substantial positive impact on air pollution. Air pollutant concentrations are reduced during phase I and phase II of the lockdown. Pollutants, especially NOx and PM_2.5 concentrations, are drastically reduced compared to the previous years. NOx concentrations are reduced by ~ 20% in Bangalore, whereas ~ 50% in Delhi. PM_2.5 concentrations are reduced by ~ 41% in Delhi and ~ 55% in Bangalore. Forecasting of pollutants will be helpful in providing the valuable information for the optimal air pollution control strategies. It has been observed that linear model gives better results compared to ARIMA and Exponential Smoothening models. By forecasting, the concentration of NO₂ is 115.288 µg/m³, the ozone is 30.636 µg/m³, SO₂ is 11.798 µg/m³, and CO is 2.758 mg/m³ over Delhi in 2021. All the pollutants during forecasting showed a rising trend except sulphur dioxide.

An intelligent clustering method for devising the geochemical fingerprint of underground aquifers

Geochemical fingerprinting is a rapidly expanding discipline in the earth and environmental sciences, anchored in the recognition that geological processes leave behind physical, chemical and sometimes also isotopic patterns in the samples. Furthermore, the geochemical fingerprinting of natural cycles (water, carbon, soil and biota fingerprinting) are influenced by the anthropogenic impact and by the climate change. So, their monitoring is a tool of resilience and adaptation. In recent years, computational statistics and artificial intelligence methods have started to be used to help the process of geochemical fingerprinting. In this paper we consider data from 57 wells located in the province of Ferrara (Italy), all belonging to the same geological group and separated into 4 different aquifers. The aquifer from which each well extracts its water is known only in 18 of the 57 cases, while in other 39 cases it can be only hypothesized based on geological considerations. We devise a novel technique for geochemical fingerprinting of groundwater by means of which we are able to identify the exact aquifer from which a sample is extracted with a sufficiently high accuracy. Then, we experimentally prove that out method is sensibly more accurate than typical statistical approaches, such as principal component analysis, for this particular problem.

Temperature Prediction Using the Missing Data Refinement Model Based on a Long Short-Term Memory Neural Network

In this paper, we propose a new temperature prediction model based on deep learning by using real observed weather data. To this end, a huge amount of model training data is needed, but these data should not be defective. However, there is a limitation in collecting weather data since it is not possible to measure data that have been missed. Thus, the collected data are apt to be incomplete, with random or extended gaps. Therefore, the proposed temperature prediction model is used to refine missing data in order to restore missed weather data. In addition, since temperature is seasonal, the proposed model utilizes a long short-term memory (LSTM) neural network, which is a kind of recurrent neural network known to be suitable for time-series data modeling. Furthermore, different configurations of LSTMs are investigated so that the proposed LSTM-based model can reflect the time-series traits of the temperature data. In particular, when a part of the data is detected as missing, it is restored by using the proposed model’s refinement function. After all the missing data are refined, the LSTM-based model is retrained using the refined data. Finally, the proposed LSTM-based temperature prediction model can predict the temperature through three time steps: 6, 12, and 24 h. Furthermore, the model is extended to predict 7 and 14 day future temperatures. The performance of the proposed model is measured by its root-mean-squared error (RMSE) and compared with the RMSEs of a feedforward deep neural network, a conventional LSTM neural network without any refinement function, and a mathematical model currently used by the meteorological office in Korea. Consequently, it is shown that the proposed LSTM-based model employing LSTM-refinement achieves the lowest RMSEs for 6, 12, and 24 h temperature prediction as well as for 7 and 14 day temperature prediction, compared to other DNN-based and LSTM-based models with either no refinement or linear interpolation. Moreover, the prediction accuracy of the proposed model is higher than that of the Unified Model (UM) Local Data Assimilation and Prediction System (LDAPS) for 24 h temperature predictions.

Applying machine learning to forecast daily Ambrosia pollen using environmental and NEXRAD parameters

Approximately 50 million Americans have allergic diseases. Airborne plant pollen is a significant trigger for several of these allergic diseases. Ambrosia (ragweed) is known for its abundant production of pollen and its potent allergic effect in North America. Hence, estimating and predicting the daily atmospheric concentration of pollen (ragweed pollen in particular) is useful for both people with allergies and for the health professionals who care for them. In this study, we show that a suite of variables including meteorological and land surface parameters, as well as next-generation radar (NEXRAD) measurements together with machine learning can be used to estimate successfully the daily pollen concentration. The supervised machine learning approaches we used included random forests, neural networks, and support vector machines. The performance of the training is independently validated using 10% of the data partitioned using the holdout cross-validation method from the original dataset. The random forests (R= 0.61, R²= 0.37), support vector machines (R= 0.51, R²= 0.26), and neural networks (R= 0.46, R²= 0.21) effectively predicted the daily Ambrosia pollen, where the correlation coefficient (R) and R-squared (R²) values are given in brackets. Three independent approaches-the random forests, correlation coefficients, and interaction information-were employed to rank the relative importance of the available predictors.

Forecasting of Surface Ozone Concentration by Using Artificial Neural Networks in Rural and Urban Areas in Central Poland

This paper presents the development of artificial neural network models for the prediction of the daily maximum hourly mean of surface ozone concentration for the next day at rural and urban locations in central Poland. The models were generated with six input variables: forecasted basic meteorological parameters and the maximum O3 concentration recorded on the previous day and number of the month. The training data set covered the period from April 2015 to September 2015. An analogous data set of input variables, for the 2014 year, not used during the process of training the networks, was used as test data to examine the quality of these models. From the results of simulations for the year 2014, the average relative error values were equal to 15.3% and 15.7% for Belsk and Warsaw stations, respectively. The mean error (ME) value indicates a tendency to overestimate the predicted values by 4.8 µg/m3 for Belsk station and to underestimate the predicted values by 0.9 µg/m3 for Warsaw station. The analysis of days when the relative error value was >50% revealed that all predictions with extremely high relative error value were associated with relatively low daily maximum surface ozone concentration values that occurred suddenly due to a sharp drop in day-to-day ozone concentration values.

Modeling Dengue vector population using remotely sensed data and machine learning

Mosquitoes are vectors of many human diseases. In particular, Aedes ægypti (Linnaeus) is the main vector for Chikungunya, Dengue, and Zika viruses in Latin America and it represents a global threat. Public health policies that aim at combating this vector require dependable and timely information, which is usually expensive to obtain with field campaigns. For this reason, several efforts have been done to use remote sensing due to its reduced cost. The present work includes the temporal modeling of the oviposition activity (measured weekly on 50 ovitraps in a north Argentinean city) of Aedes ægypti (Linnaeus), based on time series of data extracted from operational earth observation satellite images. We use are NDVI, NDWI, LST night, LST day and TRMM-GPM rain from 2012 to 2016 as predictive variables. In contrast to previous works which use linear models, we employ Machine Learning techniques using completely accessible open source toolkits. These models have the advantages of being non-parametric and capable of describing nonlinear relationships between variables. Specifically, in addition to two linear approaches, we assess a support vector machine, an artificial neural networks, a K-nearest neighbors and a decision tree regressor. Considerations are made on parameter tuning and the validation and training approach. The results are compared to linear models used in previous works with similar data sets for generating temporal predictive models. These new tools perform better than linear approaches, in particular nearest neighbor regression (KNNR) performs the best. These results provide better alternatives to be implemented operatively on the Argentine geospatial risk system that is running since 2012.

Air Pollution Forecasting Model Based on Chance Theory and Intelligent Techniques

A novel approach of weighted support vector regression (WSVR) technique with applied chance theory was proposed to build a robust forecasting model, called the chance weighted support vector regression (chWSVR) model. In order to forecast the particulate matter air pollutant of diameter less than 10 micrometers (PM10) one hour advance in the Greater Cairo Metropolitan Area (GCMA) in Egypt. The chance theory has advanced concepts pertinent to treat cases where both randomness and fuzziness play simultaneous roles at one time. The basic idea based on the proposed chWSVR model is assigning the chance weight value of the target variable, based on the chance theory, to its corresponding dataset point to become minimized in the objective function making that point more significant during the training process. Measuring data were collected and reprocessed from four monitoring stations located in GCMA and relative to the springs during the period from 2007 to 2010. The results of such model compared to similar ones built by other machine learning techniques, Random Forest and Bootstrap aggregating techniques. In all stations, comparing such models revealed that the proposed chWSVR model findings were promising in the forecasting of PM10 hourly concentration.

Daily air quality index forecasting with hybrid models: A case in China

Air quality is closely related to quality of life. Air pollution forecasting plays a vital role in air pollution warnings and controlling. However, it is difficult to attain accurate forecasts for air pollution indexes because the original data are non-stationary and chaotic. The existing forecasting methods, such as multiple linear models, autoregressive integrated moving average (ARIMA) and support vector regression (SVR), cannot fully capture the information from series of pollution indexes. Therefore, new effective techniques need to be proposed to forecast air pollution indexes. The main purpose of this research is to develop effective forecasting models for regional air quality indexes (AQI) to address the problems above and enhance forecasting accuracy. Therefore, two hybrid models (EMD-SVR-Hybrid and EMD-IMFs-Hybrid) are proposed to forecast AQI data. The main steps of the EMD-SVR-Hybrid model are as follows: the data preprocessing technique EMD (empirical mode decomposition) is utilized to sift the original AQI data to obtain one group of smoother IMFs (intrinsic mode functions) and a noise series, where the IMFs contain the important information (level, fluctuations and others) from the original AQI series. LS-SVR is applied to forecast the sum of the IMFs, and then, S-ARIMA (seasonal ARIMA) is employed to forecast the residual sequence of LS-SVR. In addition, EMD-IMFs-Hybrid first separately forecasts the IMFs via statistical models and sums the forecasting results of the IMFs as EMD-IMFs. Then, S-ARIMA is employed to forecast the residuals of EMD-IMFs. To certify the proposed hybrid model, AQI data from June 2014 to August 2015 collected from Xingtai in China are utilized as a test case to investigate the empirical research. In terms of some of the forecasting assessment measures, the AQI forecasting results of Xingtai show that the two proposed hybrid models are superior to ARIMA, SVR, GRNN, EMD-GRNN, Wavelet-GRNN and Wavelet-SVR. Therefore, the proposed hybrid models can be used as effective and simple tools for air pollution forecasting and warning as well as for management.

Variation of surface ozone in Campo Grande, Brazil: meteorological effect analysis and prediction

The effect of meteorological variables on surface ozone (O3) concentrations was analysed based on temporal variation of linear correlation and artificial neural network (ANN) models defined by genetic algorithms (GAs). ANN models were also used to predict the daily average concentration of this air pollutant in Campo Grande, Brazil. Three methodologies were applied using GAs, two of them considering threshold models. In these models, the variables selected to define different regimes were daily average O3 concentration, relative humidity and solar radiation. The threshold model that considers two O3 regimes was the one that correctly describes the effect of important meteorological variables in O3 behaviour, presenting also a good predictive performance. Solar radiation, relative humidity and rainfall were considered significant for both O3 regimes; however, wind speed (dispersion effect) was only significant for high concentrations. According to this model, high O3 concentrations corresponded to high solar radiation, low relative humidity and wind speed. This model showed to be a powerful tool to interpret the O3 behaviour, being useful to define policy strategies for human health protection regarding air pollution.

Evaluation of the temporal scaling variability in forecasting ground-level ozone concentrations obtained from multiple linear regressions

Ozone is a highly unpredictable pollutant which severely affects living conditions in urban and surrounding areas in the Mediterranean basin. This secondary pollutant periodically reaches extremely high concentrations, damaging human health. Multiple linear regression has been widely used in previous works due to the fact that it is a simple and versatile method for forecasting ozone concentrations. However, these models usually prove their validity using fulfillment of statistical constraints, ignoring other intrinsic characteristics existing in the time series, such as the temporal scaling behavior and the data distribution over different time scales. In previous works, it has been demonstrated that observed ozone time series are of a multifractal nature, meaning that the data distribution can be described by using the multifractal spectrum. This work focuses on the capacity of a forecasting model to reproduce the scaling features existing in an observed time series when several chemical and meteorological explanatory variables are introduced following the stepwise procedure. A comparison between the observed spectrum and the simulated ones for each step is used to check which explanatory variables better reproduce the multifractal nature in real ozone time series. It has been confirmed that a model with few explanatory variables allows reproducing the multifractal nature in the simulated time series with an acceptable accuracy without compromising the values of the coefficient of determination and root-mean-squared error, which were used as performance indicators.

Predicting submicron air pollution indicators: a machine learning approach

The regulation of air pollutant levels is rapidly becoming one of the most important tasks for the governments of developing countries, especially China. Submicron particles, such as ultrafine particles (UFP, aerodynamic diameter ≤ 100 nm) and particulate matter ≤ 1.0 micrometers (PM1.0), are an unregulated emerging health threat to humans, but the relationships between the concentration of these particles and meteorological and traffic factors are poorly understood. To shed some light on these connections, we employed a range of machine learning techniques to predict UFP and PM1.0 levels based on a dataset consisting of observations of weather and traffic variables recorded at a busy roadside in Hangzhou, China. Based upon the thorough examination of over twenty five classifiers used for this task, we find that it is possible to predict PM1.0 and UFP levels reasonably accurately and that tree-based classification models (Alternating Decision Tree and Random Forests) perform the best for both these particles. In addition, weather variables show a stronger relationship with PM1.0 and UFP levels, and thus cannot be ignored for predicting submicron particle levels. Overall, this study has demonstrated the potential application value of systematically collecting and analysing datasets using machine learning techniques for the prediction of submicron sized ambient air pollutants.

Optimization of artificial neural network models through genetic algorithms for surface ozone concentration forecasting

INTRODUCTION

This study proposes three methodologies to define artificial neural network models through genetic algorithms (GAs) to predict the next-day hourly average surface ozone (O(3)) concentrations. GAs were applied to define the activation function in hidden layer and the number of hidden neurons.

METHODS

Two of the methodologies define threshold models, which assume that the behaviour of the dependent variable (O(3) concentrations) changes when it enters in a different regime (two and four regimes were considered in this study). The change from one regime to another depends on a specific value (threshold value) of an explanatory variable (threshold variable), which is also defined by GAs. The predictor variables were the hourly average concentrations of carbon monoxide (CO), nitrogen oxide, nitrogen dioxide (NO(2)), and O(3) (recorded in the previous day at an urban site with traffic influence) and also meteorological data (hourly averages of temperature, solar radiation, relative humidity and wind speed). The study was performed for the period from May to August 2004.

RESULTS AND DISCUSSION

Several models were achieved and only the best model of each methodology was analysed. In threshold models, the variables selected by GAs to define the O(3) regimes were temperature, CO and NO(2) concentrations, due to their importance in O(3) chemistry in an urban atmosphere.

CONCLUSION

In the prediction of O(3) concentrations, the threshold model that considers two regimes was the one that fitted the data most efficiently.

ANN application for prediction of atmospheric nitrogen deposition to aquatic ecosystems

The occurrences of increased atmospheric nitrogen deposition (ADN) in Southeast Asia during smoke haze episodes have undesired consequences on receiving aquatic ecosystems. A successful prediction of episodic ADN will allow a quantitative understanding of its possible impacts. In this study, an artificial neural network (ANN) model is used to estimate atmospheric deposition of total nitrogen (TN) and organic nitrogen (ON) concentrations to coastal aquatic ecosystems. The selected model input variables were nitrogen species from atmospheric deposition, Total Suspended Particulates, Pollutant Standards Index and meteorological parameters. ANN models predictions were also compared with multiple linear regression model having the same inputs and output. ANN model performance was found relatively more accurate in its predictions and adequate even for high-concentration events with acceptable minimum error. The developed ANN model can be used as a forecasting tool to complement the current TN and ON analysis within the atmospheric deposition-monitoring program in the region.

Prediction of daily ground-level ozone concentration maxima over New Delhi

The pollution levels in New Delhi from industrial, residential, and transportation sources are continuously growing. As one of the major pollutants, ground-level ozone is responsible for various adverse effects on both humans and foliage. The present study aims to predict daily ground-level ozone concentration maxima over a site situated in New Delhi through neural networks (NN) and multiple-regression (MR) analysis. Although these methodologies are case and site specific, they are being developed and used widely. Therefore, to test these methodologies for New Delhi where no such study is available for ground-level ozone, six models have been developed based on NNs and MR using the same input data set. The changes in the performance capability of the two methods are sensitive to the selection of input parameters. The results are encouraging, and remarkable improvements in the performance of the models have been observed.

Prediction of daily maximum ground ozone concentration using support vector machine

The accurate predictions of ground ozone concentrations are required for proper management, control, and making public warning strategies. Due to the difficulties in handling phenomenological models that are based on complex chemical reactions of ozone production, neural network models gained popularity in the last decade. These models also have some limitations due to problems of overfitting, local minima, and tuning of network parameters. In this study, the predictions of daily maximum ozone concentrations are attempted using support vector machines (SVMs). The comparison between the accuracy of SVM and neural network predictions is performed to evaluate their performance. For this, the daily maximum ozone concentration data observed during 2002-2004 at a site in Delhi is utilized. The models are developed using the available meteorological parameters. The results indicated the promising performance of SVM over neural networks in predicting daily maximum ozone concentrations.

Improving the accuracy of low level quantum chemical calculation for absorption energies: the genetic algorithm and neural network approach

The combination of genetic algorithm and back-propagation neural network correction approaches (GABP) has successfully improved the calculation accuracy of absorption energies. In this paper, the absorption energies of 160 organic molecules are corrected to test this method. Firstly, the GABP1 is introduced to determine the quantitative relationship between the experimental results and calculations obtained by using quantum chemical methods. After GABP1 correction, the root-mean-square (RMS) deviations of the calculated absorption energies reduce from 0.32, 0.95 and 0.46 eV to 0.14, 0.19 and 0.18 eV for B3LYP/6-31G(d), B3LYP/STO-3G and ZINDO methods, respectively. The corrected results of B3LYP/6-31G(d)-GABP1 are in good agreement with experimental results. Then, the GABP2 is introduced to determine the quantitative relationship between the results of B3LYP/6-31G(d)-GABP1 method and calculations of the low accuracy methods (B3LYP/STO-3G and ZINDO). After GABP2 correction, the RMS deviations of the calculated absorption energies reduce to 0.20 and 0.19 eV for B3LYP/STO-3G and ZINDO methods, respectively. The results show that the RMS deviations after GABP1 and GABP2 correction are similar for B3LYP/STO-3G and ZINDO methods. Thus, the B3LYP/6-31G(d)-GABP1 is a better method to predict absorption energies and can be used as the approximation of experimental results where the experimental results are unknown or uncertain by experimental method. This method may be used for predicting absorption energies of larger organic molecules that are unavailable by experimental methods and by high-accuracy theoretical methods with larger basis sets. The performance of this method was demonstrated by application to the absorption energy of the aldehyde carbazole precursor.

Decision support methods for the detection of adverse events in post-marketing data

Spontaneous reporting is a crucial component of post-marketing drug safety surveillance despite its significant limitations. The size and complexity of some spontaneous reporting system databases represent a challenge for drug safety professionals who traditionally have relied heavily on the scientific and clinical acumen of the prepared mind. Computer algorithms that calculate statistical measures of reporting frequency for huge numbers of drug-event combinations are increasingly used to support pharamcovigilance analysts screening large spontaneous reporting system databases. After an overview of pharmacovigilance and spontaneous reporting systems, we discuss the theory and application of contemporary computer algorithms in regular use, those under development, and the practical considerations involved in the implementation of computer algorithms within a comprehensive and holistic drug safety signal detection program.

A comparative study on various statistical techniques predicting ozone concentrations: implications to environmental management

The objective of the present work is to compare various techniques for modeling the dependence of the tropospheric ozone concentrations on several meteorological and pollutant parameters. The study focuses on two different sites in the metropolitan area of Athens, Greece; one in the city centre and another one in the suburbs. It is found that although simple Linear Regression Analysis fails to construct accurate equations due to the existence of multicollinearity among the independent variables, still various combinations of a Multivariate Method (PCA) and Stepwise Regression Analysis manage to produce equations free of the multicollinearity issue. The derived formulas are validated and prove to have R(2) values in the order of 0.8 approximately. However, the equations are found to be unsuccessful in case of severe episodes. For this reason, a new procedure is followed for estimating the ozone values in case of episodes exclusively. The new R(2) value is estimated to be 0.9, approximately.

Interval estimation of urban ozone level and selection of influential factors by employing automatic relevance determination model

In this work, we focus on simulating the ground-level ozone (O3) time series and its daily maximum concentration in Hong Kong urban air by employing the multilayer perceptron (MLP) model combined with the automatic relevance determination (ARD) method (for simplicity, we name it as MLP-ARD model). Two air quality monitoring sites in Hong Kong, i.e., Tsuen Wan and Tung Chung, are selected for the numerical experiments. The MLP-ARD model based on Bayesian evidence framework can provide reliable interval estimation of real observation as well as offering efficient strategy to avoid over-fitting. The performance comparisons between MLP-ARD model and traditional artificial neural network (ANN) model based on maximum likelihood indicate that MLP-ARD model is more powerful to capture the wild fluctuation of O3 level especially during O3 episodes than the traditional model. Furthermore, it can assess and rank the input variables for the prediction according to their relative importance to the output variable, i.e., the daily maximum O3 concentration in this study. The preliminary experimental results indicate that nitric oxide (NO) and solar radiation are the most important input variables for O3 prediction at both selected sites. In addition, the previous daily maximum O3 level is also important for Tung Chung site. In this regard, MLP-ARD model is a feasible tool to interpret the real physical and chemical process of urban O3 variation.

Potential assessment of the "support vector machine" method in forecasting ambient air pollutant trends

Monitoring and forecasting of air quality parameters are popular and important topics of atmospheric and environmental research today due to the health impact caused by exposing to air pollutants existing in urban air. The accurate models for air pollutant prediction are needed because such models would allow forecasting and diagnosing potential compliance or non-compliance in both short- and long-term aspects. Artificial neural networks (ANN) are regarded as reliable and cost-effective method to achieve such tasks and have produced some promising results to date. Although ANN has addressed more attentions to environmental researchers, its inherent drawbacks, e.g., local minima, over-fitting training, poor generalization performance, determination of the appropriate network architecture, etc., impede the practical application of ANN. Support vector machine (SVM), a novel type of learning machine based on statistical learning theory, can be used for regression and time series prediction and have been reported to perform well by some promising results. The work presented in this paper aims to examine the feasibility of applying SVM to predict air pollutant levels in advancing time series based on the monitored air pollutant database in Hong Kong downtown area. At the same time, the functional characteristics of SVM are investigated in the study. The experimental comparisons between the SVM model and the classical radial basis function (RBF) network demonstrate that the SVM is superior to the conventional RBF network in predicting air quality parameters with different time series and of better generalization performance than the RBF model.

2D mapping by Kohonen networks of the air quality data from a large city

The 15-variable environmental data (7 concentrations: CO, SO2, O3, NOx, NO, NO2, particulate matter smaller than 10 micron (PM10), and 8 weather data: cloudiness, rainfall, insolation factor (Isfi), temperature, pressure at two locations, and wind intensity with direction) in a period of 45 days with 1-h intervals were extracted from a larger database of concentrations recorded in minute intervals for the same time period. The monitoring site was located in the City of Buenos Aires in a relatively heavy traffic crossroad of two avenues. The data required special pretreatment where the hourly content of rain, wind intensity, wind velocity, and cloudiness were concerned. The new variable named insolation factor (relative UV radiation) calculated on the basis of the general meteorological data, the geographic position of the monitoring site, cloudiness, date, and the time of the recording was composed. The relative intensity of UV radiation was modeled by a Gaussian function, multiplied by a cloudiness factor. Based on the 14-variable input and the 1-variable output (ozone) data, first, the clustering of all 980 data records was made. The top map clustering showing the ozone concentration was related to the maps of all 14 variables. The link between O3 clusters, NO2, and Isfi weight levels is shown and discussed. As a preliminary result of this study some of the most interesting correlations between the maps and remaining variables are given.

Potential assessment of a neural network model with PCA/RBF approach for forecasting pollutant trends in Mong Kok urban air, Hong Kong

The forecasting of air pollutant trends has received much attention in recent years. It is an important and popular topic in environmental science, as concerns have been raised about the health impacts caused by unacceptable ambient air pollutant levels. Of greatest concern are metropolitan cities like Hong Kong. In Hong Kong, respirable suspended particulates (RSP), nitrogen oxides (NOx), and nitrogen dioxide (NO2) are major air pollutants due to the dominant usage of diesel fuel by commercial vehicles and buses. Hence, the study of the influence and the trends relating to these pollutants is extremely significant to the public health and the image of the city. The use of neural network techniques to predict trends relating to air pollutants is regarded as a reliable and cost-effective method for the task of prediction. The works reported here involve developing an improved neural network model that combines both the principal component analysis technique and the radial basis function network and forecasts pollutant tendencies based on a recorded database. Compared with general neural network models, the proposed model features a more simple network architecture, a faster training speed, and a more satisfactory prediction performance. The improved model was evaluated with hourly time series of RSP, NOx and NO2 concentrations monitored at the Mong Kok Roadside Gaseous Monitory Station in Hong Kong during the year 2000 and proved to be effective. The model developed is a potential tool for forecasting air quality parameters and is superior to traditional neural network methods.

Comparative assessment of neural networks and regression models for forecasting summertime ozone in Athens

A comparison study has been performed with neural networks (NNs) and multiple linear regression models to forecast the next day's maximum hourly ozone concentration in the Athens basin at four representative monitoring stations that show very different behavior. All models use 11 predictors (eight meteorological and three persistence variables) and are developed and validated between April and October from 1992 to 1999. Performance results based on a wide set of forecast quality measures indicate that the NNs provide better estimates of ozone concentrations at the monitoring sites, whilst the more often used linear models are less efficient at accurately forecasting high ozone concentrations. The violation of the European information threshold of 180 microg/m(3) is successfully predicted by the NN in 72% of the cases on average. Results at all stations are consistent with similar ozone forecast studies using NNs in other European cities.

Prediction of maximum daily ozone level using combined neural network and statistical characteristics

Analysis and forecasting of air quality parameters are important topics of atmospheric and environmental research today due to the health impact caused by air pollution. As one of major pollutants, ozone, especially ground level ozone, is responsible for various adverse effects on both human being and foliage. Therefore, prediction of ambient ozone levels in certain environment, especially the ground ozone level in densely urban areas, is of great importance to urban air quality and city image. To date, though several ozone prediction models have been established, there is still a need for more accurate models to develop effective warning strategies. The development of such models is difficult because the meteorological variables and the photochemical reactions involved in ozone formation are very complex. The present work aims to develop an improved neural network model, which combines the adaptive radial basis function (ARBF) network with statistical characteristics of ozone in selected specific areas, and is used to predict the daily maximum ozone concentration level. The improved method is trained and testified by hourly time series data collected at three air pollutant-monitoring stations in Hong Kong during 1999 and 2000. The simulation results demonstrate the effectiveness and the reliability of the proposed method.