Cognitive,collaborative, conceptual and creative — Four characteristics of the next generation of knowledge-based CAD systems: A study in biologically inspired design

We envision that the next generation of knowledge-based CAD systems will be characterized by four features: they will be based on cognitive accounts of design, and they will support collaborative design, conceptual design, and creative design. In this paper, we first analyze these four dimensions of CAD. We then report on a study in the design, development and deployment of a knowledge-based CAD system for supporting biologically inspired design that illustrates these four characteristics. This system, called DANE for Design by Analogy to Nature Engine, provides access to functional models of biological systems. Initial results from in situ deployment of DANE in a senior-level interdisciplinary class on biologically inspired design indicates its usefulness in helping designers conceptualize design of complex systems, thus promising enough to motivate continued work on knowledge-based CAD for biologically inspired design. More importantly from our perspective, DANE illustrates how cognitive studies of design can inform the development of CAD systems for collaborative, conceptual, and creative design, help assess their use in practice, and provide new insights into human interaction with knowledge-based CAD systems.     

Classification capacity of a modular neural network implementing neurally inspired architecture and training rules

A three-layer neural network (NN) with novel adaptive architecture has been developed. The hidden layer of the network consists of slabs of single neuron models, where neurons within a slab-but not between slabs- have the same type of activation function. The network activation functions in all three layers have adaptable parameters. The network was trained using a biologically inspired, guided-annealing learning rule on a variety of medical data. Good training/testing classification performance was obtained on all data sets tested. The performance achieved was comparable to that of SVM classifiers. It was shown that the adaptive network architecture, inspired from the modular organization often encountered in the mammalian cerebral cortex, can benefit classification performance.     

Predicting Drivers of Mobile Entertainment Adoption: A Two-Stage SEM-Artificial-Neural-Network Analysis

This study aims to understand users’ motivations to adopt mobile entertainment (m-entertainment). Extending the Technology Acceptance Model (TAM), this study examined the effects of trust, perceived financial cost (PFC), and quality of the service on consumers’ decision in adopting the m-entertainment. Survey data were collected from 524 mobile users and analyzed using both structural equation modeling (SEM) and neural network (NN) . The result showed that perceived usefulness (PU), perceived ease of use (PEOU), and quality of service (QS) are important predictors of m-entertainment adoption. The study contributes to the existing literature by extending the TAM model as well as examining m-entertainment, an important and emerging business model in mobile commerce. A new analytical approach using both SEM and NN was also employed in this study.     

Systemic banking crisis early warning systems using dynamic Bayesian networks

For decades, the literature on banking crisis early-warning systems has been dominated by two methods, namely, the signal extraction and the logit model methods. However, these methods, do not model the dynamics of the systemic banking system. In this study, dynamic Bayesian networks are applied as systemic banking crisis early-warning systems. In particular, the hidden Markov model, the switching linear dynamic system and the naïve Bayes switching linear dynamic system models are considered. These dynamic Bayesian networks provide the means to model system dynamics using the Markovian framework. Given the dynamics, the probability of an impending crisis can be calculated. A unique approach to measuring the ability of a model to predict a crisis is utilised. The results indicate that the dynamic Bayesian network models can provide precise early-warnings compared with the signal extraction and the logit methods.     

感知认知技术在金融风险预警中的应用研究

金融风险全球溢出效应与国内金融业态创新发展中的伴生风险相叠加，使得我国所面临的国际金融及内生性金融风险形势非常严峻。针对传统风险预警技术因缺乏有效、及时的关键因子导致实践中对金融风险预警难度极大的技术难题，本文重点总结了如何利用感知认知技术从海量非结构化信息提取有效、及时的金融风险预警关键因子，并在回顾现有金融风险预警模型研究现状的基础上，对相关技术难点和未来研究趋势进行总结和展望。本文研究内容可为我国研发自主可控的金融风险预警技术提供必要参考。     

Minimum time ship maneuvering method using neural network and nonlinear model predictive compensator

In this paper, a new minimum time ship maneuvering method using neural network (NN) and nonlinear model predictive compensator is proposed. In this proposed method the NN is used for interpolating the precomputed minimum time solution for real time situations and the nonlinear dynamical model of a ship is used for compensating the control error caused by some modeling errors, disturbances and so on. The introduction of the nonlinear model into the online control system is inspired by the idea that since the nonlinear dynamical model of a ship has been constructed for the off-line numerical computation of the optimal solutions, it could also be used to enhance online control performance. In order to investigate this method, simulation studies and actual sea tests were carried out using a training ship Shioji Maru.The results showed that the system gives approximate solutions in a short computing time and good tracking performance in the actual sea trials.     

Financial distress prediction based on OR-CBR in the principle of k-nearest neighbors

Financial distress prediction including bankruptcy prediction has called broad attention since 1960s. Various techniques have been employed in this area, ranging from statistical ones such as multiple discriminate analysis (MDA), Logit, etc. to machine learning ones like neural networks (NN), support vector machine (SVM), etc. Case-based reasoning (CBR), which is one of the key methodologies for problem-solving, has not won enough focus in financial distress prediction since 1996. In this study, outranking relations (OR), including strict difference, weak difference, and indifference, between cases on each feature are introduced to build up a new feature-based similarity measure mechanism in the principle of k-nearest neighbors. It is different from traditional distance-based similarity mechanisms and those based on NN, fuzzy set theory, decision tree (DT), etc. Accuracy of the CBR prediction method based on OR, which is called as OR-CBR, is determined directly by such four types of parameters as difference parameter, indifference parameter, veto parameter, and neighbor parameter. It is described in detail that what the model of OR-CBR is from various aspects such as its developed background, formalization of the specific model, and implementation of corresponding algorithm. With three year’s real-world data from Chinese listed companies, experimental results indicate that OR-CBR outperforms MDA, Logit, NN, SVM, DT, Basic CBR, and Grey CBR in financial distress prediction, under the assessment of leave-one-out cross-validation and the process of Max normalization. It means that OR-CBR may be a preferred model dealing with financial distress prediction in China.     

Predicting motivators of cloud computing adoption: A developing country perspective

Cloud computing is a recent and significant development in the domain of network applications with a new information technology perspective. This study attempts to develop a hybrid model to predict motivators influencing the adoption of cloud computing services by information technology (IT) professionals. The research proposes a new model by extending the Technology Acceptance Model (TAM) with three external constructs namely computer self-efficacy, trust, and job opportunity. One of the main contributions of this research is the introduction of a new construct, Job Opportunity (JO), for the first time in a technology adoption study. Data were collected from 101 IT professional and analyzed using multiple linear regression (MLR) and neural network (NN) modeling. Based on the RMSE values from the results of these models NN models were found to outperform the MLR model. The results obtained from MLR showed that computer self-efficacy, perceived usefulness, trust, perceived ease of use, and job opportunity. However, the NN models result showed that the best predictor of cloud computing adoption are job opportunity, trust, perceived usefulness, self-efficacy, and perceived ease of use. The findings of this study confirm the need to extend the fundamental TAM when studying a recent technology like cloud computing. This study will provide insights to IT service providers, government agencies, academicians, researchers and IT professionals.     

Predicting shareholder litigation on insider trading from financial text: An interpretable deep learning approach

The detrimental effects of insider trading on the financial markets and the economy are well documented. However, resource-constrained regulators face a great challenge in detecting insider trading and enforcing insider trading laws. We develop a text analytics framework that uses machine learning to predict ex-ante potentially opportunistic insider trading (using actual insider trading allegation by shareholders as the proxy) from corporate textual disclosures. Distinct from typical black-box neural network models, which have difficulty tracing a prediction back to key features, our approach combines the predictive power of deep learning with attention mechanisms to provide interpretability to the model. Further, our model utilizes representations from a business proximity network and incorporates the temporal variations of a firm’s financial disclosures. The empirical results offer new insights into insider trading and provide practical implications. Overall, we contribute to the literature by reconciling performance and interpretability in predictive analytics. Our study also informs the practice by proposing a new method for regulators to examine a large amount of text in order to monitor and predict financial misconduct.     

Comparing lazy and eager learning models for water level forecasting in river-reservoir basins of inundation regions

This study developed a methodology for formulating water level models to forecast river stages during typhoons, comparing various models by using lazy and eager learning approaches. Two lazy learning models were introduced: the locally weighted regression (LWR) and the k-nearest neighbor (kNN) models. Their efficacy was compared with that of three eager learning models, namely, the artificial neural network (ANN), support vector regression (SVR), and linear regression (REG). These models were employed to analyze the Tanshui River Basin in Taiwan. The data collected comprised 50 historical typhoon events and relevant hourly hydrological data from the river basin during 1996–2007. The forecasting horizon ranged from 1 h to 4 h. Various statistical measures were calculated, including the correlation coefficient, mean absolute error, and root mean square error. Moreover, significance, computation efficiency, and Akaike information criterion were evaluated. The results indicated that (a) among the eager learning models, ANN and SVR yielded more favorable results than REG (based on statistical analyses and significance tests). Although ANN, SVR, and REG were categorized as eager learning models, their predictive abilities varied according to various global learning optimizers. (b) Regarding the lazy learning models, LWR performed more favorably than kNN. Although LWR and kNN were categorized as lazy learning models, their predictive abilities were based on diverse local learning optimizers. (c) A comparison of eager and lazy learning models indicated that neither were effective or yielded favorable results, because the distinct approximators of models that can be categorized as either eager or lazy learning models caused the performance to be dependent on individual models.     

Predicting the financial crisis by Mahalanobis–Taguchi system – Examples of Taiwan’s electronic sector

In the past researches of financial crisis early-warning model, multiple regression, linear probability model, and multiple discriminate analysis are commonly adopted, all of which have generated good discrimination effects, with over 90% accuracy. Dr. Taguchi, well known for his robust design, has lately brought up a new method – Mahalanobis–Taguchi System (MTS), which is mainly used to conduct multivariate diagnoses and forecasts. This study attempts to use MTS to build up a financial crisis early-warning model for Taiwan’s companies. It chooses both in financial sound judgment and in financial trouble TSE- and OTC-listed electronic companies in 2005 as training set and uses both in financial sound judgment and in financial trouble TSE- and OTC-listed electronic companies in 2006 as testing set to verify the accuracy of this model. There are two phases in our research, in which we firstly use MTS, logistic regression and neural network to establish the financial crisis early-warning model, followed by a comparative analysis of average accuracy rate of financial prediction in the second phase. The result of experiment shows that the accuracy rate of financial crisis early-warning system established by MTS, logistic regression and neural network are 96.1%, 92.3%, and 96.1%, respectively, indicating that MTS provides greater application effect in predicting financial crisis.     

A real-valued genetic algorithm to optimize the parameters of support vector machine for predicting bankruptcy

Two parameters, C and σ, must be carefully predetermined in establishing an efficient support vector machine (SVM) model. Therefore, the purpose of this study is to develop a genetic-based SVM (GA-SVM) model that can automatically determine the optimal parameters, C and σ, of SVM with the highest predictive accuracy and generalization ability simultaneously. This paper pioneered on employing a real-valued genetic algorithm (GA) to optimize the parameters of SVM for predicting bankruptcy. Additionally, the proposed GA-SVM model was tested on the prediction of financial crisis in Taiwan to compare the accuracy of the proposed GA-SVM model with that of other models in multivariate statistics (DA, logit, and probit) and artificial intelligence (NN and SVM). Experimental results show that the GA-SVM model performs the best predictive accuracy, implying that integrating the RGA with traditional SVM model is very successful.     

A T-cell algorithm for solving dynamic optimization problems

In this paper, a metaheuristic inspired on the T-Cell model of the immune system (i.e., an artificial immune system) is introduced. The proposed approach (called DTC, for Dynamic T-Cell) is used to solve dynamic optimization problems, and is validated using test problems taken from the specialized literature on dynamic optimization. Results are compared with respect to artificial immune approaches representative of the state-of-the-art in the area. Some statistical analyses are also performed, in order to determine the sensitivity of the proposed approach to its parameters.     

Neural network methods for forecasting turning points in economic time series: an asymmetric verification to business cycles

This paper examines the relevance of various financial and economic indicators in forecasting business cycle turning points using neural network (NN) models. A three-layer feed-forward neural network model is used to forecast turning points in the business cycle of China. The NN model uses 13 indicators of economic activity as inputs and produces the probability of a recession as its output. Different indicators are ranked in terms of their effectiveness of predicting recessions in China. Out-of-sample results show that some financial and economic indicators, such as steel output, M2, Pig iron yield, and the freight volume of the entire society are useful for predicting recession in China using neural networks. The asymmetry of business cycle can be verified using our NN method.     

基于神经网络集成学习算法的金融时间序列预测

本文在传统神经网络(NN)、循环神经网络(RNN)、长短时记忆网络(LSTM)与门控循环单元(GRU)等神经网络时间预测模型基础上, 进一步构建集成学习(EL)时间序列预测模型, 研究神经网络类模型、集成学习模型和传统时间序列模型在股票指数预测上的表现. 本文以16只A股和国际股票市场指数为样本, 比较模型在不同预测期间和不同国家和地区股票市场上的表现.本文主要结论如下: 第一, 神经网络类时间序列预测模型和神经网络集成学习时间序列预测模型在表现上显著稳健优于传统金融时间序列预测模型, 预测性能提高大约35%; 第二, 神经网络类模型和神经网络集成学习模型在中国和美国股票市场上的表现优于其他发达国家和地区的股票市场.     

Comparison of country risk models: hybrid neural networks,logit models,discriminant analysis and cluster techniques

This paper looks at the ability of a relatively new technique, hybrid ANN's, to predict country risk rating. These models are compared with traditional statistical techniques and conventional ANN models. The performance of hierarchical cluster analysis and another type of ANN, the self-organizing map were also investigated, as possible methods for making country risk analysis with visual effects. The results indicate that hybrid neural networks outperform all other models. This suggests that for researchers, policymakers and others interested in early warning systems, hybrid network may be a useful tool for country risk analysis.     

Q-高斯核支持向量机的财务危机预报

针对科学实践、经济生活等诸多领域数据分布相对复杂的分类问题,使用传统支持向量机(SVM)无法很好地刻画其变量间的相关性,从而影响分类性能。对于这一情况,提出使用经典高斯函数的参数推广形式--Q-高斯函数作为SVM的核函数构建财务危机预警模型。结合沪深股市A股制造业上市公司的财务数据分别建立T-2和T-3财务预警模型进行实证分析,采用显著性检验筛选出合适的财务指标并利用交叉验证方法确定模型参数。相比高斯核SVM财务危机预警模型,使用Q-高斯核SVM建立的T-2和T-3模型的预报准确率都提高了大约3%,而且成本较高的第Ⅰ类错误最多降低了14.29%。     

Bankruptcy prediction for credit risk using neural networks: Asurvey and new results

The prediction of corporate bankruptcies is an important and widely studied topic since it can have significant impact on bank lending decisions and profitability. This work presents two contributions. First we review the topic of bankruptcy prediction, with emphasis on neural-network (NN) models. Second, we develop an NN bankruptcy prediction model. Inspired by one of the traditional credit risk models developed by Merton (1974), we propose novel indicators for the NN system. We show that the use of these indicators in addition to traditional financial ratio indicators provides a significant improvement in the (out-of-sample) prediction accuracy (from 81.46% to 85.5% for a three-year-ahead forecast).