Search

Class noise is a common issue that affects the performance of classification techniques on real-world data sets. Class noise appears when a class variable in data sets has incorrect class labels. In the case of noisy data, the robustness of classification techniques against noise could be more important than the performance results on noise-free data sets. The decision tree method is one of the most popular techniques for classification tasks. The C4.5, CART, and random forest (RF) algorithms are considered to be three of the most used algorithms in decision trees. The aim of this paper is to reach conclusions on which decision tree algorithm is better to use for building decision trees in terms of its performance and robustness against class noise. In order to achieve this aim, we study and compare the performance of the models when applied to class variables with noise. The results obtained indicate that the RF algorithm is more robust to data sets with noisy class variable than other algorithms. [ABSTRACT FROM AUTHOR]

With the systematization, informatization, and intelligence of the management and operation of communication enterprises, the amount of audit content and business data is increasing. The traditional audit work is mainly offline audit, the task is heavy and temporary, there are few auditors, and there is a lack of unified data mart and analysis model support. In order to solve many problems in traditional auditing, this paper realizes the interface of data dictionary attributes through effective domain management of multisystem data, so as to achieve the purpose of system support and effective management to improve data quality. Furthermore, the platform is extensible, which can gradually expand the data coverage and support more management fields and business systems. A method of obtaining continuous attribute clustering center points by K-means algorithm and using triangular fuzzy numbers to fuzzify process continuous data is proposed. At the same time, a visual fuzzy decision-making system based on Fuzzy ID3 and Min-Ambiguity algorithm was designed. Adding the C4.5 and CART algorithms implemented in the Weka open source data mining software, through experimental analysis, the differences in the classification accuracy and the number of rules generated by the four decision tree algorithms are compared. The experiment found that the Fuzzy ID3 algorithm has a high accuracy rate on each dataset and the number of rules is large. The number of rules generated by the CART algorithm is the least because of the model characteristics of its binary tree and the characteristics of using the Gini index as the criterion for the selection of splitting attributes. Comparing Fuzzy ID3 and Min-Ambiguity fuzzy decision tree algorithms, it is found that the overall performance of the former is better than that of the latter, and the effect of authenticity on the two algorithms is analyzed experimentally. This paper constructs the internal audit flowchart in the big data environment and studies the optimization of the internal audit process in the big data environment. We evaluate traditional internal audit processes using process optimization theory. The internal audit process in this paper refers to the internal audit process under the risk-oriented model that is widely used at this stage. The optimization research on the internal audit process in this paper is mainly aimed at the internal audit process of general enterprises with a certain level of informatization, at a large scale, and certain foundation and requirements for the analysis and use of data. The optimization of internal audit process is feasible in research and practical application and has universal applicability and reference value. [ABSTRACT FROM AUTHOR]

With the research of machine learning technology and big data intelligent processing technology in engineering application becoming more and more mature, people gradually combine machine learning technology and big data intelligent processing technology. Aiming at the problem of innovative employment in colleges and universities, this paper proposes a dynamic decision tree algorithm based on these two technologies and constructs a dynamic model of graduates' behavior. Through the analysis of dynamic decision tree algorithm, a big data analysis system is formed. Finally, simulation experiments verify whether the dynamic model can correctly reflect the behavior of college graduates. The results show that the big data integration system based on big data and dynamic decision tree algorithm has high adaptability. Incremental adaptive optimization of the traditional decision tree model can significantly improve the prediction effect and prediction time of dynamic data and provide theoretical support for the industrialization and social significance of big data technology. The dynamic decision tree algorithm of college employment proposed in this paper has good predictability and provides a certain theoretical reference for college graduates' entrepreneurship. [ABSTRACT FROM AUTHOR]

Financial crises can have a negative impact on business operations, and in serious cases, they directly affect the survival and growth of a company. Therefore, the study of financial early warning based on financial indicators is particularly important. However, there are still some shortcomings in the current research on financial early warning, for example, it still evaluates the scoring method or only uses a single model to participate in the construction of financial early warning algorithm. In view of the above problems, this study will mainly use the random forest method combined with the decision tree algorithm to study the financial early warning problem of listed companies in China. Firstly, this paper uses the literature review method to analyse the relevant literature and generate the financial indicator system for this study. Subsequently, by collecting the financial data of A-share listed companies in China from 2013 to 2018 as the research object, the importance ranking of financial indicators was generated by using random forest modelling after data preprocessing. On this basis, CART decision tree modelling was applied to generate financial indicator early warning determination rules and analyse them. The results of the study show the importance ranking of financial indicators and the six financial warning rules based on the CART decision tree. Through this research, it is expected to achieve the objective of providing early warning for the risk of financial crisis and to provide constructive financial warning solutions for relevant stakeholders. [ABSTRACT FROM AUTHOR]

Currently, Customers are struggling to retain their business in today's competitive markets. Thus, the issue of customer churn becomes a significant challenge for the industries. In order to achieve this, it is vital to have an efficient churn prediction system. In this paper, we discuss methods for reducing features using PCA, Autoencoders, LDA, T-SNE, and Xgboost. In this paper, a model for predicting light GBM churn is proposed. The model consists of five steps. The first step is to preprocess the data so that missing and corrupt values can be handled and the data can be scaled. Secondly, implementing a comprehensive feature reduction system based on popular algorithms reduces the features and selects the most suitable one. In the third step, light GBM's hyperparameter is tuned using Bayesian hyperparameter optimization and genetic optimization algorithms. Lastly, interpreting the model and evaluating the impact of the features on model outputs by using the SHAP method, and finally ranking the churners by customer lifetime value. Aside from evaluating and choosing the best feature reduction methods, the proposed method is also evaluated using four famous datasets. It outperforms other ensemble and ML algorithms like AdaBoost, SVM, and decision tree on over seven evaluation metrics: accuracy, area under the curve (AUC), Kappa, Mathews correlation coefficient (MCC), Brier score, F1 score, and EMPC. In light of the evaluation metrics, our model shows a significant improvement in handling imbalanced datasets in churn prediction. Finally, in this paper, interpretability and how the features affect the model's output are presented by the SHAP method. Then CLV ranking is suggested for better decision-making. [ABSTRACT FROM AUTHOR]

Topping is an important part in cotton field management, the spraying time has a great impact on cotton quality. In agricultural production, the strategy of timing the cotton topping mainly relies on manual inspections and experience, which is lack of efficiency and science. To solve the problem, this paper uses a drone equipped with a multispectral camera to collect the multispectral information of the cotton canopy of 12 days which includes before and after the topping operation in Shihezi. At the same time, the information of cotton plant height, the number of fruiting branches, and flower buds are collected. Compare multiple band combinations and vegetation index; the combined data of 550 + 730 + 790 nm band is selected as the model input. AdaBoost + decision tree method is proposed as a fitting model, the fitting results show that the coefficient of determination (R2) between multispectrum and cotton plant height is 0.96, and the average prediction error (RMSEP) is 0.40 cm, the coefficient of determination (R2) between multispectrum information and the fruiting branches is 0.97, the prediction mean error (RMSEP) is 0.54, and the correlation determination (R2) with the flower buds is 0.84, and the prediction mean error is (RMSEP) 0.49. The output data of the fitting model is used as the input of the topping time discriminant model, and the discriminant model can obtain an accuracy of 94.03%. The method in this paper can effectively monitor the growth status of cotton in the topping time and provide a technical path to scientifically determine the cotton topping time. [ABSTRACT FROM AUTHOR]

Both inter- and intraorganizational networks draw the attention of researchers and practitioners from various disciplines who view them as the fabric of the socioeconomic world. The network perspective is believed to successfully model most of the socioeconomic phenomena, which, in combination with the prospects of continuously advancing tools for automated data mining and machine learning, gives a tempting promise to effectively forecast socioeconomic events occurring in our societies and businesses. Despite their significance, the topic of event forecasting in the context of organizational networks appears unexplored. Therefore, the objective of this study was (1) to fill the theoretical gap by proposing a mathematical model for organizational network event forecasting, rooted in the social science to remain consistent with the theory, and (2) to experimentally evaluate how the model performs on real data and validate if the results support its use in practical applications. An implementation of the proposed model, based on a decision tree classifier, achieved a prediction accuracy of 87% on a longitudinal data sample and thus demonstrated the practical usability of the model. [ABSTRACT FROM AUTHOR]

The framework of outcomes-based education(OBE) has become a central issue for global university education, which is benefited to drive the education development by a series of assessments for historical teaching data, especially student course score, and employment information. The issue of how to timely update the talent training plans for computer major in a university has received considerable critical attention. It is becoming extremely difficult to ignore the requirement of fast shortened update cycle in IT area. One of the main obstacles is that the teaching inertia and the fixed awareness of a major training plan may delay the feedback of teaching problems. There is still a contradiction between the plan rationality and the real-time needs of contemporary IT enterprises. Hence, this paper puts forward a novel data-based framework to evaluate the relevance between the major courses, employment rate, and enterprise needs through the decision tree expression, thus providing reliable data support for systematic curriculum reform. On top of that, A recommendation algorithm is proposed to automatically generate the computer course group that satisfies the staff requirements of IT enterprises. Finally, teaching and employment data of Xihua University in China is applied as an example to undertake course optimization and recommendation. The consequences have an obvious positive effect on student employment and enterprise feedback. [ABSTRACT FROM AUTHOR]

This study aims to survey the bibliometric properties of distributed decisions, decentralized decisions, and fuzzy articles published between 1995 and 2023 in the Web of Science (WoS) database. During the analysis process, the keywords "distributed decision, decentralized decision, and fuzzy" were scanned in all languages, both in the titles and the content of all publication types. As a result of the analysis, 79 articles in all fields comprised the dataset. The most used keywords in the articles were related to the distributed decision, decentralized decision, and fuzzy logic, and the most frequently cited publications were examined using the social network analysis method, which uses VOSviewer (version 1.6.19) to visualize the relationships. The study's goal on "active researchers, active journals, journal metrics, title document type, active countries, and active institutions" was to look at the words most frequently used in articles on distributed, decentralized, and fuzzy logic. The social network analysis represented the relationships between these keywords and the most frequently cited publications. The findings demonstrated a significant correlation between using these keywords in academic literature and their contribution to this field's research. These results can assist researchers in finding potential partners and keeping up with current research trends. Overall, this study offers important new perspectives on the state of research on fuzzy logic, distributed decision making, and decentralized decision making. [ABSTRACT FROM AUTHOR]

In view of the current problems in the visual aggregation of the financial service industry, the random forest graph model is applied to the spatial recognition of the financial service industry's scientific calculation visual aggregation in this paper. Based on the detailed analysis of the characteristics of the financial service industry in the past, combined with the characteristics of finance, this method is used to construct a random forest graph model, which mainly optimizes parameters from different aspects such as model structure, data characteristics, and dynamic changes of the model to obtain optimal parameter values of the random forest graph model. Finally, through the analysis of the experimental results, it can be seen that, according to the spatial state of the financial service industry, the method proposed in this paper can be used for visual aggregation analysis. This method can effectively improve the timeliness of the scientific calculation of spatial recognition in the financial service industry. [ABSTRACT FROM AUTHOR]

The purpose of this interdisciplinary study is to investigate the applications of modern computational technologies in the field of credit investigation and discuss the related legal governance measurements in order to unblock the bottlenecks in the future development of credit investigation market. By analyzing the computational technologies and algorithms most commonly used in credit data collection and data storage, data transmission and data access, data analysis and processing, data calculation, result output and effect evaluation, this paper summarizes and proposes a unified general process of modern credit investigation, pointing out that in this general process, low data quality, privacy violation, algorithmic bias are the main challenges in the big data era, and countermeasures like data quality control, privacy protection, and algorithm governance need to be to be taken seriously into account in order to further explore the great potential of the credit investigation under the legal framework. [ABSTRACT FROM AUTHOR]

With the advancement of information technology, the informatization construction of internal control has evolved into crucial fundamental support for the construction of internal control in colleges and universities. The informatization construction of internal control is a systematic project of which the selection of software development mode is the key link. In this paper, a detailed analysis is carried out on the equilibrium strategy of the two parties in the game between the informatization construction department of internal control in colleges and universities as well as software development companies by establishing a game model and a software maintenance risk management model for both sides, in order to probe into the selection of the software development mode in colleges and universities. The model test and case study prove that the fully commissioned development mode is the optimal mode for the informatization construction of internal control in colleges and universities, and relevant suggestions are made. This research has provided sufficient theoretical support for the informatization construction of internal control in colleges and universities. [ABSTRACT FROM AUTHOR]

This study focuses on the volatility prediction and option volatility investment. By investigating the traditional Volatility Prediction Model and machine learning algorithms, this study tries to merge these two aspects together. This work setup a bridge of previous financial studies and machine learning studies by proposing an algorithm integrating neural network and three traditional volatility models, called "Quantile based neural network and model integration combination algorithm." The algorithm effectively lowers the volatility prediction error (measured by root of mean square error, shorted for RMSE: 0.319724) and beat the Wavenet (RMSE: 0.44) which is the benchmark and surpasses integrated model (RMSE: 0.348346) in test set. In terms of option investment strategy, this paper constructs a CSI 300 index option portfolio which hedges the underlying asset price risk and exposes the volatility risk. Then propose the "Option strategy of volatility prediction with dynamic thresholds." With the new algorithm above, the strategy improves the return-risk ratio in test set (measured by Sharpe ratio: 1.99–2.07). [ABSTRACT FROM AUTHOR]

In this paper, a stochastic traffic assignment model for networks is proposed for the study of discrete dynamic Bayesian algorithms. In this paper, we study a feasible method and theoretical system for implementing traffic engineering in networks based on Bayesian algorithm theory. We study the implementation of traffic assignment engineering in conjunction with the network stochastic model: first, we study the Bayesian algorithm theoretical model of control layer stripping in the network based on the discrete dynamic Bayesian algorithm theory and analyze the resource-sharing mechanism in different queuing rules; second, we study the extraction and evaluation theory of traffic assignment for the global view obtained by the control layer of the network and establish the Bayesian algorithm analysis model based on the traffic assignment; subsequently, the routing of bandwidth guarantee and delay guarantee in the network is studied based on Bayesian algorithm model and Bayesian algorithm network random traffic allocation theory. In this paper, a Bayesian algorithm estimation model based on Bayesian algorithm theory is constructed based on network random observed traffic assignment as input data. The model assumes that the roadway traffic distribution follows the network random principle, and based on this assumption, the likelihood function of the roadway online traffic under the network random condition is derived; the prior distribution of the roadway traffic is derived based on the maximum entropy principle; the posterior distribution of the roadway traffic is solved by combining the likelihood function and the prior distribution. The corresponding algorithm is designed for the model with roadway traffic as input, and the reliability of the algorithm is verified in the arithmetic example. [ABSTRACT FROM AUTHOR]

Word sense disambiguation (WSD) is an important research topic in natural language processing, which is widely applied to text classification, machine translation, and information retrieval. In order to improve disambiguation accuracy, this paper proposes a WSD method based on the graph convolutional network (GCN). Word, part of speech, and semantic category are extracted from contexts of the ambiguous word as discriminative features. Discriminative features and sentence containing the ambiguous word are used as nodes to construct the WSD graph. Word2Vec tool, Doc2Vec tool, pointwise mutual information (PMI), and TF-IDF are applied to compute embeddings of nodes and edge weights. GCN is used to fuse features of a node and its neighbors, and the softmax function is applied to determine the semantic category of the ambiguous word. Training corpus of SemEval-2007: Task #5 is adopted to optimize the proposed WSD classifier. Test corpus of SemEval-2007: Task #5 is used to test the performance of WSD classifier. Experimental results show that average accuracy of the proposed method is improved. [ABSTRACT FROM AUTHOR]

Environmental quality assessment is an important way to promote the improvement of urban environmental quality. Environmental performance is usually used to evaluate the improvement of environmental quality, and residents' satisfaction with environmental quality is also an important method to evaluate environmental improvement. At present, in many cities in China, the results of the two evaluation methods vary greatly. Residents' environmental satisfaction is not high in some cities with good environmental performance; however, in cities with poor environmental performance, residents' environmental satisfaction is higher. Here, based on the environmental subjective assessment of more than 4,600 independent samples from 56 cities in 2014, this paper constructed an index between subjective and objective scores for each sample and its city, separating the total samples into two groups. In order to analyze the differences between groups, firstly, the important factors driving the differences were extracted by random forest. Secondly, the key individual characteristics were identified by the model based on conditional inference tree. Finally, the regional heterogeneity was analyzed by nonmetric multidimensional scaling. The results show that population density is the main factor that affects the difference between subjective and objective evaluations. Furthermore, in those cities with low population density, investment increasing in transportation infrastructure helps to improve urban air quality, which can bring about more perceptual environmental optimization to people. As individuals, education is the key factor for residents when it comes to environment evaluation, but it is not a simple linear relationship. In terms of regional heterogeneity, the consistency of important factors among regions is not obvious, and the situation that "neighboring" cities share the same factors is not significant. [ABSTRACT FROM AUTHOR]

This paper combines discrete wavelet transform (DWT), autoregressive moving average (ARMA), and XGBoost algorithm to propose a weighted hybrid algorithm named DWTs-ARMA-XGBoost (DAX) on long-time temperature series analysis. Firstly, this paper chooses the temperature data of February 1 to 20 from 1967 to 2016 of northern mountainous area in North China as the observed data. Then, we use 10 different discrete wavelet functions to decompose and reconstruct the observed data. Next, we build ARMA models on all the reconstructed data. In the end, we regard the calculations of 10 DWT-ARMA (DA) algorithms and the observed data as the labels and target of the XGBoost algorithm, respectively. Through the data training and testing of the XGBoost algorithm, the optimal weights and the corresponding output of the hybrid DAX model can be calculated. Root mean squared error (RMSE) was followed as the criteria for judging the precision. This paper compared DAX with an equal-weighted average (EWA) algorithm and 10 DA algorithms. The result shows that the RMSE of the two hybrid algorithms is much lower than that of the DA algorithms. Moreover, the bigger decrease in RMSE of the DAX model than the EWA model represents that the proposed DAX model has significant superiority in combining models which proves that DAX has significant improvement in prediction as well. [ABSTRACT FROM AUTHOR]

This paper aims to explore the factors influencing passengers' satisfaction toward bus transit and develop the improvement strategy for the small-medium city. Data including individual attributes, travel activity attributes, and perceived service attributes were collected in the city of Weinan, China. The ordered logit (OL) model and ordered Probit (OP) model were employed to explore the significant factors associated with passengers' satisfaction toward bus transit. The odds ratio (OR) was applied to quantitatively measure the effects of the significant factors. Improving strategies of bus transit service were proposed based on the model results using the three-factor theory. Results show that the OL model outperforms the OP model. The age, daily average waiting time, perceived waiting time, transferring convenience, the attitude of the driver, intelligent travel information service, hygienic environment inside the bus, ticket price, bus route setting, and bus stop setting significantly affect the passengers' satisfaction. Among them, the ticket price, perceived waiting time, bus stop setting, intelligent travel information service, transferring convenience, and bus route setting were identified as exciting factors. It is recommended that optimization of bus route and bus stop setting, building bus dedicated lanes, optimizing dynamic charging system, and providing intelligent travel information service could be effective strategies to improve passengers' satisfaction toward bus transit in Chinese small-medium cities. [ABSTRACT FROM AUTHOR]

By combining the similarity matching (SM) method with the utilities additives discriminates (UTADIS) method, we propose a hybrid SM-UTADIS approach to detect falsified financial statements (FFS) of listed companies. To evaluate the performance of this hybrid approach, we conduct experiments using the annual financial ratios of listed traditional Chinese medicine (TCM) companies in China. There are three stages in the detection procedure. First, we use the cosine similarity matching method to select matched companies for each considered company, derive the deviation data of each considered company as a sample dataset to capture the intrinsic law of the financial data, and further divide these into training and testing datasets for the next two stages. Second, we put the training dataset into the UTADIS to train the SM-UTADIS model. Finally, we use the trained SM-UTADIS model to classify the testing dataset and evaluate the performance of the proposed method. Furthermore, we use other approaches, such as single UTADIS and logistic and SM-logistic regression models, to detect FFS. By comparing these results to those of the hybrid SM-UTADIS approach, we find that the proposed hybrid approach greatly improves the accuracy of FFS detection. [ABSTRACT FROM AUTHOR]

In the collaborative filtering (CF) recommendation applications, the sparsity of user rating data, the effectiveness of cold start, the strategy of item information neglection, and user profiles construction are critical to both the efficiency and effectiveness of the recommendation algorithm. In order to solve the above problems, a personalized recommendation approach combining semisupervised support vector machine and active learning (AL) is proposed in this paper, which combines the benefits of both TSVM (Transductive Support Vector Machine) and AL. Firstly, a "maximum-minimum segmentation" of version space-based AL strategy is developed to choose the most informative unlabeled samples for human annotation; it aims to choose the least data which is enough to train a high-quality model. And then, an AL-based semisupervised TSVM algorithm is proposed to make full use of the distribution characteristics of unlabeled samples by adding a manifold regularization into objective function, which is helpful to make the proposed algorithm to overcome the traditional drawbacks of TSVM. Furthermore, during the procedure of recommendation model construction, not only user behavior information and item information, but also demographic information is utilized. Due to the benefits of the above design, the quality of unlabeled sample annotation can be improved; meanwhile, both the data sparsity and cold start problems are alleviated. Finally, the effectiveness of the proposed algorithm is verified based on UCI datasets, and then it is applied to personalized recommendation. The experimental results show the superiority of the proposed method in both effectiveness and efficiency. [ABSTRACT FROM AUTHOR]

Deep convolutional neural networks have been successfully applied to face detection recently. Despite making remarkable progress, most of the existing detection methods only localize each face using a bounding box, which cannot segment each face from the background image simultaneously. To overcome this drawback, we present a face detection and segmentation method based on improved Mask R-CNN, named G-Mask, which incorporates face detection and segmentation into one framework aiming to obtain more fine-grained information of face. Specifically, in this proposed method, ResNet-101 is utilized to extract features, RPN is used to generate RoIs, and RoIAlign faithfully preserves the exact spatial locations to generate binary mask through Fully Convolution Network (FCN). Furthermore, Generalized Intersection over Union (GIoU) is used as the bounding box loss function to improve the detection accuracy. Compared with Faster R-CNN, Mask R-CNN, and Multitask Cascade CNN, the proposed G-Mask method has achieved promising results on FDDB, AFW, and WIDER FACE benchmarks. [ABSTRACT FROM AUTHOR]

This paper presents a new hybrid discriminant analysis method, and this method combines the ideas of linearity and nonlinearity to establish a two-layer discriminant model. The first layer is a linear discriminant model, which is mainly used to determine the distinguishable samples and subsample; the second layer is a nonlinear discriminant model, which is used to determine the subsample type. Numerical experiments on real data sets show that this method performs well compared to other classification algorithms, and its stability is better than the common discriminant models. [ABSTRACT FROM AUTHOR]

Agent-based modelling has been proved to be extremely useful for learning about real world societies through the analysis of simulations. Recent agent-based models usually contain a large number of parameters that capture the interactions among microheterogeneous subjects and the multistructure of the complex system. However, this can result in the "curse of dimensionality" phenomenon and decrease the robustness of the model's output. Hence, it is still a great challenge to efficiently calibrate agent-based models to actual data. In this paper, we present a surrogate analysis method for calibration by combining supervised machine-learning and intelligent iterative sampling. Without any prior assumptions regarding the distribution of the parameter space, the proposed method can learn a surrogate model as the approximation of the original system with a relatively small number of training points, which will serve the needs of further sensitivity analysis and parameter calibration research. We take the heterogeneous asset pricing model as an example to evaluate the model's performance using actual Chinese stock market data. The results demonstrate the good capabilities of the surrogate model at modelling the observed reality, as well as the remarkable reduction of the computational time for validating the agent-based model. [ABSTRACT FROM AUTHOR]

The existence and persistence of financial statement fraud (FSF) are detrimental to the financial health of global capital markets. A number of detective and predictive methods have been used to prevent, detect, and correct FSF, but their practicability has always been a big challenge for researchers and auditors, as they do not address real-world problems. In this paper, both supervised and unsupervised approaches are employed for analysing the financial data obtained from China's stock market in detecting FSF. The variables used in this paper are 18 financial datasets, representing a fraud triangle. Additionally, this study examined the properties of five widely used supervised approaches, namely, multi-layer feed forward neural network (MFFNN), probabilistic neural network (PNN), support vector machine (SVM), multinomial log-linear model (MLM), and discriminant analysis (DA), applied in different real-life situations. The empirical results show that MFFNN yields the best classification results in detection of fraudulent data presented in financial statement. The outcomes of this study can be applied to different types of financial statement datasets, as they present a practical way for constructing predictive models using a combination of supervised and unsupervised approaches. [ABSTRACT FROM AUTHOR]

In order to reduce the air pollution impacts by aircraft operations around airports, a fast and accurate prediction of air quality related to aircraft operations is an essential prerequisite. This article proposes a new framework with a combination of the standard assessment procedure and machine learning methods for fast and accurate prediction of air quality in airports. Instead of taking some specific pollutant as concerned metric, we introduce the air quality index (AQI) for the first time to evaluate the air quality in airports. Then, following the standard assessment procedure proposed by International Civil Aviation Organization (ICAO), the airports AQIs in different scenarios are classified with consideration of the airport configuration, actual flight operations, aircraft performance, and related meteorological data. Taking the AQI classification results as sample data, several popular supervised learning methods are investigated for accurately predicting air quality in airports. The numerical tests implicate that the accuracy rate of prediction could reach more than 95% with only 0.022 sec; the proposed framework and the results could be used as the foundation for improving air quality impacts around airports. [ABSTRACT FROM AUTHOR]

Drivers consecutively direct their gaze to various areas to select relevant information from the traffic environment. The rate of crash risk increases with different off-road glance durations in different driving scenarios. This paper proposed an approach to identify current driving scenarios and predict driver’s eyes-off-road durations using Hidden Markov Model (HMM). A moving base driving simulator study with 26 participants driving in three driving scenarios (urban, rural, and motorway) was conducted. Three different fixed occlusion durations (0-s, 1-s, and 2-s) were applied to quantify eyes-off-road durations. Participants could initiate each occlusion for certain duration by pressing a microswitch on a finger. They were instructed to occlude their vision as often as possible while still driving safely. Drivers’ visual behavior and occlusion behavior were captured and analyzed based on manually frame by frame coding. Visual behaviors in terms of glance duration and glance location in time series were used as input to train HMMs. The results showed that current driving scenarios could be identified ideally using glance location sequences, the accuracy achieving up to 89.3%. And motorway was relatively distinguishable easily with over 90% accuracy. Moreover, HMM-based algorithms that fed up with both glance duration and glance location sequences resulted in a highest accuracy of 92.7% in driver’s eyes-off-road durations prediction. And higher accuracy achieved in longer eyes-off-road durations prediction. It indicates that time series of glance allocations could be used to predict driving behavior and indentify driving environment. The developed models in this study could contribute to the development of scenario sensitive visual inattention prewarning system. [ABSTRACT FROM AUTHOR]

A support vector machine is a new learning machine; it is based on the statistics learning theory and attracts the attention of all researchers. Recently, the support vector machines (SVMs) have been applied to the problem of financial early-warning prediction (Rose, 1999). The SVMs-based method has been compared with other statistical methods and has shown good results. But the parameters of the kernel function which influence the result and performance of support vector machines have not been decided. Based on genetic algorithms, this paper proposes a new scientific method to automatically select the parameters of SVMs for financial early-warning model. The results demonstrate that the method is a powerful and flexible way to solve financial early-warning problem. [ABSTRACT FROM AUTHOR]