Search

Machine learning approaches are increasingly being employed to forecast the key characteristics of strong ground motions, including the challenging classification of Pulse-Like (PL) ground motions. The PL ground motions are characterized by their impulsive nature and have the potential to cause significant damage to structures. The classification of PL ground motions continues to be a significant challenge due to the absence of consensus on their definition and categorization. This paper investigates the potential benefits of several Machine Learning Classifiers (MLCs) algorithms such as decision tree, random forest, logistic regression, naive Bayes, support vector machine, K-nearest neighbor, ensemble model, and artificial neural network models for predicting PL and Non-Pulse-Like (NPL) ground motions. In this regard, a dataset comprising 200 near-fault ground motions records compiled from active tectonic regions like Taiwan, Turkey, Iran, and Japan, was divided into 2 portions, with 75% used to train the model and the remaining 25% used for testing. Plots of performance metrics, confusion matrix, and receiver operating curve indicate that the ensemble classifier outperforms the other classifier with 86.2% accuracy and the lowest misclassification of PL and NPL ground motions. Additionally, the trained MLC has been compared with the existing ground motion classification models to further assess the accuracy of the different classifiers in the present study. [ABSTRACT FROM AUTHOR]

This research concentrates on segmenting credit card clients of Taiwan into optimal groups. Unsupervised Learning plays a significant role in dividing customers into similar groups based on several parameters. If customers are clustered in groups optimally, it leads towards the retrieval of better precision from machine learning models applied to customers associated with the clusters. Different machine learning algorithms (Linear discriminant analysis, logistic regression and random forest) were applied on the obtained clusters through K-means, hierarchical and HK Means clustering algorithm, and predictive accuracy is compared with the accuracy obtained via applying mentioned machine learning models on the whole dataset. In this paper, a novel approach of combining K Means and hierarchical clustering (HK Means) is used. In this approach, HK means clustering algorithms are applied on the factorial coordinates, obtained from multiple correspondence analyses for segmenting customers into optimal groups has been proposed. The accuracy of the clustering techniques is evaluated from the decomposition of inertia. The results demonstrate that the combination of K Means and hierarchical clustering proved to be optimal clustering techniques for customer segmentation which can be used further for applying Machine Learning techniques for credit risk analysis. [ABSTRACT FROM AUTHOR]

The wrapping layer's foaming process of the car steering wheel industry usually relies on the manual parameter setting by experienced engineers, but the reliability and validity are usually hard to control due to many foaming factors that affect the hardness of steering wheel wrapping layer (e.g., vulcanization time, mold temperature, material liquid temperature, material discharge pressure, humidity). This paper first proposes an intelligent process parameter recommendation system architecture, then develops a neural network-based hardness prediction (NNHP) model to predict the car steering wheel's wrapping hardness. Consequently, we combine NNHP and genetic algorithm (GA) to develop a foaming process parameter recommendation model (NNHP-GA) to effectively recommend the most suitable combinations of process parameters for achieving the desired wrapping hardness in an IoT-based intelligent manufacturing environment. An empirical study is applied in Taiwan's world-leading car steering wheel company. By analyzing the 16 important process sensor data installed in each casting machine, the five experimental results show that the NNHP-GA recommendation system can successfully predict the hardness of the steering wheel's wrapping layer and recommend appropriate process parameters (e.g., sensors S1-vulcanization time, S2-mold temperature, S5-A.P. pressure, S6-B.I. pressure, and "pressurization time") under a specific hardness target. The manufacturing company may employ the NNHP-GA model to build a process parameter recommendation system and increase the quality of process parameter setting. [ABSTRACT FROM AUTHOR]

Pairs trading is an effective statistical arbitrage strategy considering the spread of paired stocks in a stable cointegration relationship. Nevertheless, rapid market changes may break the relationship (namely structural break), which further leads to tremendous loss in intraday trading. In this paper, we design a two-phase pairs trading strategy optimization framework, namely structural break-aware pairs trading strategy (SAPT), by leveraging machine learning techniques. Phase one is a hybrid model extracting frequency- and time-domain features to detect structural breaks. Phase two optimizes pairs trading strategy by sensing important risks, including structural breaks and market-closing risks, with a novel reinforcement learning model. In addition, the transaction cost is factored in a cost-aware objective to avoid significant reduction of profitability. Through large-scale experiments in real Taiwan stock market datasets, SAPT outperforms the state-of-the-art strategies by at least 456% and 934% in terms of profit and Sortino ratio, respectively. [ABSTRACT FROM AUTHOR]

Missing values are common in the Internet of Things (IoT) environment for various reasons, including regular maintenance or malfunction. In time-series prediction in the IoT, missing values may have a relationship with the target labels, and their missing patterns result in informative missingness. Thus, missing values can be a barrier to achieving high accuracy of prediction and analysis in data mining in the IoT. Although several methods have been proposed to estimate values that are missing, few studies have investigated the comparison of interpolation methods using conventional and deep learning models. There has thus far been relatively little research into interpolation methods in the IoT environment. To address these problems, this paper presents the use of linear regression, support vector regression, artificial neural networks, and long short-term memory to make time-series predictions for missing values. Finally, a full comparison and analysis of interpolation methods are presented. We believe that these findings can be of value to future work in IoT applications. [ABSTRACT FROM AUTHOR]

The techniques pertaining to soft computing are the base for commencement, model and organization of intelligent systems in order to offer more perfect, economical and realistic solution which has minimal complexity levels. The concept of intelligent systems has for long supported objectives on sustainability, improvisation of efficiency and symbolized various kinds of activities like creation of jobs, earning profits, providing services and improvement in capacities in ICT. The applications based on this system are widespread in the technological market because of massive development in grid, cloud, mobile and big data applications and its corresponding connectivity advantages. ICT triggers data scientists to satisfy technical-based demands of intelligent systems around data analytics and big data applications. The major features of big data like that of reservation of designs on information and knowledge have offered the public undertaking an opportunity to enhance production levels, improved efficiency and its effectiveness. The main objective of this paper is to enhance the accuracy of predictive modelling using an optimized predictive modelling based on rough set (RS) and Elman neural network (ElNN). These advanced predictive models are designed on the basis of RS approach in initial stages and in later processes enhanced with the support of Elman-NN. RS has an excellent feature selection capability, and Elman-NN is the best at nonlinear system modelling. By integrating them, the proposed method can limit the input dimension and optimize the structure for ElNN modelling. This can reduce the mathematical computation complexity with the progress of predictive models. The experimental results indicate that through RS feature selection and the structure of Elman-NN, the predictive model can be simplified significantly with enhanced model performance. The predictive accuracy of data sets, namely air quality in Northern Taiwan, hazardous air pollutants, historical hourly weather data and US pollutants through optimization, is above 99%, and this model proves that the results of optimized predictive error are far better than those obtained by other neural networks like PCA-RBF, PCA-NN, FFNN-BP with PCA, MLR, FFNN-BP, ELM, SOM, RBF and ART2. [ABSTRACT FROM AUTHOR]