Browsing by Author "Lee, Shin-Jye"

Now showing 1 - 11 of 11

A Boosting Regression-Based Method to Evaluate the Vital Essence in Semiconductor Industry Performance
(2020-01-01)
Hsu, Ping-Yu
;
Yeh, I-Wen
;
Tseng, Ching-Hsun
;
Lee, Shin-Jye
;
科技管理研究所
In accordance with the statistical analysis, the industrial performance is usually related to research and development (R&D) intensity, and this factor indeed plausibly brings the biggest profit with patents and supporting products to the development of semiconductor industry. How to evaluate the completive performance of modern industries is an increasing issue, especially for the semiconductor industries in these decades. However, almost every traditional statistical model is deterred by the hypothesis of population and independent correlation among each feature, and this makes the result of typical regression model potentially lose reliability. To avoid this weakness, this article therefore applies a gradient boosting based method - XGBoost to evaluate the feature importance of semiconductor industries. In the simulation experiments, different findings revel certain information, apart from R&D intensity, actually sway the gross net value in the annual financial announcement of semiconductor industries. Moreover, this article proposes another concept to evaluate the essential factor contributing the development of semiconductor industries. Instead of only focusing on the effect of R&D intensity, this article also predicts the future growth rate (GR) of net value by applying the greedy search of XGBoost Regression.
A Deep Learning-Based Strategy to the Energy Management-Advice for Time-of-Use Rate of Household Electricity Consumption
(2020-01-01)
Wu, Lu-Xian
;
Lee, Shin-Jye
;
科技管理研究所
;
Institute of Management of Technology
With the high industrialization at a rapid pace, the demand for energy increases exponentially, but it is difficult to meet the balance of demand and supply. Therefore, how to effectively meet the balance of demand and supply intelligently has become a popular issue in this century. In recent years, Taiwan Power Company (Taipower), the biggest electric company in Taiwan, is committed to the construction of Advanced Metering Infrastructure (AMI), which provides communication channels and enables demand-side users to participate in load dispatch. On the other hand, the construction of AMI is expected to generate a tremendous number of valuable data on electricity consumption, but it is not easy to convert these data into effective information by the conventional quantitative methods. In as much as the rapid progression of AI technology in the industrial field, the application of AI technology in the technology management has become an increasing issue as an interdisciplinary study. To address this task, this work applies the recurrent neural network based on deep learning to predict low-voltage usage shortly by the electricity information of low-voltage user and meteorological data. After many vicissitudes, the electricity consumption per hour can be predicted and a sound energy arrangement can be therefore planned. Through introducing the proposed model, Taipower Company will have an effective capability that schedules power, reduces unnecessary backup power, and provides time-consuming electricity prices for industrial enterprises accurately among high usage of Taiwan industries.
A dimension-reduction based multilayer perception method for supporting the medical decision making
(2020-03-01)
Lee, Shin-Jye
;
Tseng, Ching-Hsun
;
Lin, G. T-R
;
Yang, Yun
;
Yang, Po
Due to the rapid development of Medical IoT recently, how to effectively apply these huge amounts of IoT data to enhance the reliability of the clinical decision making has become an increasing issue in the medical field. These data usually comprise high-complicated features with tremendous volume, and it implies that the simple inference models may less powerful to be practiced. In deep learning, multilayer perceptron (MLP) is a kind of feed-forward artificial neural network, and it is one of the high-performance methods about stochastic scheme, fitness approximation, and regression analysis. To process these high uncertain data, the proposed work based on MLP structure in particular integrates the boosting scheme and dimension-reduction process. In this proposed work, the advanced ReLU-based activation function is used. Also, the weight initialization is applied to improve the stable prediction and convergence. After the improved dimension-reduction process is introduced, the proposed method can effectively learn the hidden information from the reformative data and the precise labels also can be recognized by stacking a small amount of neural network layers with paying few extra cost. The proposed work shows a possible path of embedding dimension reduction in deep learning structure with minor price. In addition to the prediction issue, the proposed method can also be applied to assess risk and forecast trend among different information systems. (C) 2019 Elsevier B.V. All rights reserved.
A new similarity/distance measure between intuitionistic fuzzy sets based on the transformed isosceles triangles and its applications to pattern recognition
(2019-02-01)
Jiang, Qian
;
Jin, Xin
;
Lee, Shin-Jye
;
Yao, Shaowen
;
科技管理研究所
Intuitionistic fuzzy set (IFS) is a representative model of fuzzy theory, and it is widely used to deal with fuzzy and uncertain problems in many practical applications. In IFS theory, the calculation of the distance and similarity between IFSs are significant techniques that are effective measurement methods for distinguishing the similarity degree between IFSs, and the two measures can be used in pattern recognition, decision-making and so on. In this paper, a novel similarity/distance measure between IFSs is proposed according to the intersections of the transformed isosceles triangles from IFSs, and the isosceles triangles are placed in a square area; furthermore, the properties of the proposed measure are also analyzed to prove whether the measure satisfies the definition of the similarity/distance measure for IFSs or not. The numerical experiments are implemented to test the validity of the proposed measure; in addition, several pattern recognition and clustering problems are also employed to further demonstrate its effectiveness. The experimental results show that the proposed measure is an accurate and superior measure that can avoid the shortcomings of most existing measures. Finally, the directions for future research of this work are also represented in the conclusion. (C) 2018 Elsevier Ltd. All rights reserved.
A Novel Mutual Fractional Grey Bernoulli Model With Differential Evolution Algorithm and Its Application in Education Investment Forecasting in China
(2020-01-01)
Xie, Wanli
;
Pu, Bin
;
Pei, Chunying
;
Lee, Shin-Jye
;
Kang, Yan
Nowadays, the nonlinear grey Bernoulli model [NGBM(1,1)] has been successfully applied to various fields. Its main advantage is that the power exponent can better reflect the non-linear characteristics of the original data. However, the parameters of the model (i.e., the order of accumulation, coefficient of background value, and power index) must be optimized to fit the development law of the system. In this study, a fractional non-linear grey Bernoulli model [MFNGBM (1,1)] is proposed to reduce the perturbation limit of the classical NGBM and further improve the accuracy of the model, which uses mutual fractional operators and a new optimization scheme with a differential evolution (DE) algorithm for forecasting education investment. In the scheme, the power exponent of the Bernoulli differential equation, coefficient of background, and cumulative order of the original sequence are taken as decision variables, and their optimal parameters obtained by iteratively adjusting fitness functions. The experimental evaluation is conducted on two types of open-source data, and the results show that the proposed method can be very competitive with the popular baselines. Finally, MFNGBM(1,1) is used to predict China & x2019;s education investment in 2020 & x2013;2025.
A social recommendation method based on the integration of social relationship and product popularity
(2019-01-01)
Lai, Chin-Hui
;
Lee, Shin-Jye
;
Huang, Hung-Ling
;
科技管理研究所
;
Institute of Management of Technology
Web 2.0 technology fosters the flourishing growth and development of social networks. More and more people are participating in the activities on social networks to interact and share information with each other. Thus, consumers are often making their purchasing decisions based on information from the Internet such as reviews, ratings, and comments on products, especially from their trusted friends. However, a great amount of available information may cause the problem of information overload for consumers. In seeking to attain a good recommendation performance by taking the high-potential factors into account as far as possible, this paper proposes a novel social recommendation method on the basis of the integration of interactions, trust relationships and product popularity to predict user preferences, and recommend relevant products in social networks. In addition, the proposed method mainly focuses on analyzing user interactions to infer their latent interactions in accordance with the user ratings and corresponding reviews. Additionally, users may be affected by the popularity of products, so this factor has also been taken into consideration in this work. The experimental results show that the proposed recommendation method has a better recommendation performance in comparisons to other methods because the proposed method can accurately analyze user preferences and further recommend high-potential products to target users in social networks to support their purchase decision making. Furthermore, the proposed method can not only reduce the time and effort users spend on querying information, but also positively relieve the problem of information overload.
Air pollution forecasting based on attention-based LSTM neural network and ensemble learning
(1970-01-01)
Liu, Duen-Ren
;
Lee, Shin-Jye
;
Huang, Yang
;
Chiu, Chien-Ju
;
科技管理研究所
With air pollution having become a global concern, scientists are committed to working on its amelioration. In the field of air pollution prediction, there have been good results in experimental research so far, but few studies have integrated weather forecast information and the properties of air pollution drift. In this work, we propose a novel wind-sensitive attention mechanism with a long short-term memory (LSTM) neural network model to predict the air pollution - PM2.5 concentrations by considering the influence of wind direction and speed on the changes of spatial-temporal PM2.5 concentrations in neighbouring areas. Preliminary predictions for PM2.5 are then made by an LSTM neural network regarding neighbouring pollution; these predictions are "paid attention to" and we finally apply an ensemble learning method based on eXtreme Gradient Boosting (XGBoost) to combine the preliminary predictions with weather forecasting to make second phase predictions of PM2.5. The experiment is conducted using PM2.5 data and weather forecast data. Our results illustrate that the proposed method is superior to other methods in predicting PM2.5 concentrations, including multi-layer perceptron, support vector regression, LSTM neural network, and extreme gradient boosting algorithm.
Image Classification Based on the Boost Convolutional Neural Network
(2018-01-01)
Lee, Shin-Jye
;
Chen, Tonglin
;
Yu, Lun
;
Lai, Chin-Hui
;
科技管理研究所
Convolutional neural networks (CNNs), which are composed of multiple processing layers to learn the representations of data with multiple abstract levels, are the most successful machine learning models in recent years. However, these models can have millions of parameters and many layers, which are difficult to train, and sometimes several days or weeks are required to tune the parameters. Within this paper, we present the usage of a trained deep convolutional neural network model to extract the features of the images, and then, used the AdaBoost algorithm to assemble the Softmax classifiers into recognizable images. This method resulted in a 3% increase of accuracy of the trained CNN models, and dramatically reduced the retraining time cost, and thus, it has good application prospects.
Multi-Sensor Image Fusion Based on Interval Type-2 Fuzzy Sets and Regional Features in Nonsubsampled Shearlet Transform Domain
(2018-03-15)
Jiang, Qian
;
Jin, Xin
;
Hou, Jingyu
;
Lee, Shin-Jye
;
Yao, Shaowen
Multi-scale geometric analysis, one of the most often-used multi-sensor image fusion (MSIF) techniques, can offer outstanding performance during extracting the features of source image. Interval type-2 fuzzy sets (Type-2 FS) have a good prospect in image fusion field, because it can effectively address the uncertain and fuzzy problem in image fusion for selecting the high-quality pixels or coefficients of source images. We try to extend the application fields of Type-2 FS and improve the performance of MSIF; therefore, this paper presents a hybrid method by combining the local spatial frequency (LSF) with interval Type-2 FS in nonsubsampled shearlet transform (NSST) domain. NSST is used to decompose source images, and interval Type-2 FS and LSF is employed to extract the regional features of source images; so it can extract and fuse the detailed features of different source images accurately. First, NSST is performed to decompose the source images into low frequency and high frequency sub-images. Second, LSF-based fusion rule is applied to fuse low frequency sub-images. Thirdly, a novel fusion process based on interval Type-2 FS is designed to fuse high frequency sub-images. At last, inverse NSST2 (INSST) is implemented to reconstruct the fused images. The experimental and contrastive results of different image sets show that the proposed method is an effective MSIF scheme, which can achieve better fusion effect than the existing representative methods.
The Application of Quality Function Deployment to Smartwatches The House of Quality for Improved Product Design
(2017-01-01)
Lee, An Wun
;
Lin, Grace T. R.
;
Kuo, Wen-Hsun
;
Lee, Shin-Jye
;
科技管理研究所
Consumers are confused about the benefits of using Smartwatches due to the various insignificant functionalities included and the unclear value propositions promoted by Smartwatch manufacturers. In order to improve the quality of products and to provide features that can satisfy consumers' needs, technicians must understand the expectations of consumers. This paper recommends the use of quality function deployment (QFD), a method that integrates consumers' dynamic requirements with product designs, which can be represented using a House of Quality (HoQ) matrix diagram. Here, the unmet and uncovered needs of consumers relating to the Smartwatch are analyzed extensively using a questionnairebased survey. The opinions of consumers and technicians are then examined, and a HoQ diagram is constructed to evaluate the relationship between the Smartwatch's design attributes and its technical attributes. Finally, the attributes that need to be improved are listed in a suggested order of priority.
Traffic sign recognition by combining global and local features based on semi-supervised classification
(2020-05-01)
He, Zhenli
;
Nan, Fengtao
;
Li, Xinfa
;
Lee, Shin-Jye
;
Yang, Yun
The legibility of traffic signs has been considered from the beginning of design, and traffic signs are easy to identify for humans. For computer systems, however, identifying traffic signs still poses a challenging problem. Both image-processing and machine-learning algorithms are constantly improving, aimed at better solving this problem. However, with a dramatic increase in the number of traffic signs, labelling a large amount of training data means high cost. Therefore, how to use a small number of labelled traffic sign data reasonably to build an efficient and high-quality traffic sign recognition (TSR) model in the Internet-of-things-based (IOT-based) transport system has been an urgent research goal. Here, the authors propose a novel semi-supervised learning approach combining global and local features for TSR in an IOT-based transport system. In their approach, histograms of oriented gradient, colour histograms (CH), and edge features (EF) are used to build different feature spaces. Meanwhile, on the unlabelled samples, a fusion feature space is found to alleviate the differences between different feature spaces. Extensive evaluations on a collection of signs from the German Traffic Sign Recognition Benchmark (GTSRB) dataset shows that the proposed approach outperforms the others and provides a potential solution for practical applications.