學術出版;;Publications

Permanent URI for this communityhdl:11536/1

收集所有交通大學研究人員與學生個人或團體所發表之各項學術著作，涵蓋期刊論文、教師專書、學位論文等項。;;Collections of Publications of NCTU Researchers and Students Like Journal Papers and Thesis.

Browse

🏛️國立陽明交通大學機構典藏

Search Results

Now showing 1 - 10 of 15

Learning concepts by arranging appropriate training order
(2001-08-01)
Hsu, YT
;
Hong, TP
;
Tseng, SS
;
資訊工程學系
;
Department of Computer Science
Machine learning has been proven useful for solving the bottlenecks in building expert systems. Noise in the training instances will, however, confuse a learning mechanism. Two main steps are adopted here to solve this problem. The first step is to appropriately arrange the training order of the instances. It is well known from Psychology that different orders of presentation of the same set of training instances to a human may cause different learning results. This idea is used here for machine learning and an order arrangement scheme is proposed. The second step is to modify a conventional noise-free learning algorithm, thus making it suitable for noisy environment. The generalized version space learning algorithm is then adopted to process the training instances for deriving good concepts. Finally, experiments on the Iris Flower problem show that the new scheme can produce a good training order, allowing the generalized version space algorithm to have a satisfactory learning result.
Bayesian classification for data from the same unknown class
(2002-04-01)
Huang, HJ
;
Hsu, CN
;
資訊工程學系
;
Department of Computer Science
In this paper, we address the problem of how to classify a set of query vectors that belong to the same unknown class. Sets of data known to be sampled from the same class are naturally available in many application domains, such as speaker recognition. We refer to these sets as homologous sets. We show how to take advantage of homologous sets in classification to obtain improved accuracy over classifying each query vector individually. Our method, called homologous naive Bayes (HNB), is based on the naive Bayes classifier, a simple algorithm shown to be effective in many application domains. HNB uses a modified classification procedure that classifies multiple instances as a single unit. Compared with a voting method and several other variants of naive Bayes classification, HNB significantly outperforms these methods in a variety of test data sets, even when the number of query vectors in the homologous sets is small. We also report a successful application of HNB to speaker recognition. Experimental results show that HNB can achieve classification accuracy comparable to the Gaussian mixture model (GMM), the most widely used speaker recognition approach, while using less time for both training and classification.
Fuzzy inductive learning strategies
(2003-03-01)
Wang, CH
;
Tsai, CR
;
Hong, TP
;
Tseng, SS
;
資訊工程學系
In real applications, data provided to a learning system usually contain linguistic information which greatly influences concept descriptions derived by conventional inductive learning methods. Design of learning methods for working with vague data is thus very important. In this paper, we apply fuzzy set concepts to machine learning to solve this problem. A fuzzy learning algorithm based on the AQR learning strategy is proposed to manage linguistic information. The proposed learning algorithm generates fuzzy linguistic rules from "soft" instances. Experiments on the Sports and the Iris Flower classification problems are presented to compare the accuracy of the proposed algorithm with those of some other learning algorithms. Experimental results show that the rules derived from our approach are simpler and yield higher accuracy than those from some other learning algorithms.
Learning effective classifiers with Z-value measure based on genetic programming
(2004-10-01)
Chien, BC
;
Lin, JY
;
Yang, WP
;
資訊工程學系
;
Department of Computer Science
This paper presents a learning scheme for data classification based on genetic programming. The proposed learning approach consists of an adaptive incremental learning strategy and distance-based fitness functions for generating the discriminant functions using genetic programming. To classify data using the discriminant functions effectively, the mechanism called Z-value measure is developed. Based on the Z-value measure, we give two classification algorithms to resolve ambiguity among the discriminant functions. The experiments show that the proposed approach has less training time than previous GP learning methods. The learned classifiers also have high accuracy of classification in comparison with the previous classifiers. (C) 2004 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.
Learning search keywords for construction procurement
(2005-01-01)
Dzeng, RJ
;
Chang, SY
;
土木工程學系
;
Department of Civil Engineering
Seeking information from websites has become an essential part of a contractor's procurement undertaking. as more and more procurement websites become available on the Internet, Websites host extremely lame amounts of information: a keyword search, therefore, is often more efficient than browsing via an index. However. in order to find the desired information. it may be necessary to enter keywords using a trial-and-error process. This research recognizes that professional procurement experience can help users search website information more effectively, by using fewer keywords. and so proposes a learning model and suggestion model that can capture such experience. thus guiding inexperienced users in their search. Experiments, evaluating the performance of the system, were also conducted. (C) 2004 Elsevier B.V. All rights reserved.
Wireless and Wearable EEG System for Evaluating Driver Vigilance
(2014-04-01)
Lin, Chin-Teng
;
Chuang, Chun-Hsiang
;
Huang, Chih-Sheng
;
Tsai, Shu-Fang
;
Lu, Shao-Wei
Brain activity associated with attention sustained on the task of safe driving has received considerable attention recently in many neurophysiological studies. Those investigations have also accurately estimated shifts in drivers\' levels of arousal, fatigue, and vigilance, as evidenced by variations in their task performance, by evaluating electroencephalographic (EEG) changes. However, monitoring the neurophysiological activities of automobile drivers poses a major measurement challenge when using a laboratory-oriented biosensor technology. This work presents a novel dry EEG sensor based mobile wireless EEG system (referred to herein as Mindo) to monitor in real time a driver\'s vigilance status in order to link the fluctuation of driving performance with changes in brain activities. The proposed Mindo system incorporates the use of a wireless and wearable EEG device to record EEG signals from hairy regions of the driver conveniently. Additionally, the proposed system can process EEG recordings and translate them into the vigilance level. The study compares the system performance between different regression models. Moreover, the proposed system is implemented using JAVA programming language as a mobile application for online analysis. A case study involving 15 study participants assigned a 90 min sustained-attention driving task in an immersive virtual driving environment demonstrates the reliability of the proposed system. Consistent with previous studies, power spectral analysis results confirm that the EEG activities correlate well with the variations in vigilance. Furthermore, the proposed system demonstrated the feasibility of predicting the driver\'s vigilance in real time.
A 48.6-to-105.2 mu W Machine Learning Assisted Cardiac Sensor SoC for Mobile Healthcare Applications
(2014-04-01)
Hsu, Shu-Yu
;
Ho, Yingchieh
;
Chang, Po-Yao
;
Su, Chauchin
;
Lee, Chen-Yi
A machine- learning ( ML) assisted cardiac sensor SoC ( CS- SoC) is designed for mobile healthcare applications. The heterogeneous architecture realizes the cardiac signal acquisition, filtering with versatile feature extractions and classifications, and enables the higher order analysis over traditional DSPs. Besides, the asynchronous architecture with dynamic standby controller further suppresses the system active duty and the leakage power dissipation. The proposed chip is fabricated in a 90- nm standard CMOS technology and operates at 0.5 V- 1.0 V ( 0.7 V- 1.0 V for SRAM and I/ O interface). Examined with healthcare monitoring applications, the CS- SoC dissipates 48.6/ 105.2 mu W for real- time syndrome detections of ECG- based arrhythmia/ VCG- based myocardial infarction with 95.8/ 99% detection accuracy, respectively.
STATISTICAL PREDICTION OF EMOTIONAL STATES BY PHYSIOLOGICAL SIGNALS WITH MANOVA AND MACHINE LEARNING
(2012-06-01)
Chueh, Tung-Hung
;
Chen, Tai-Been
;
Lu, Henry Horng-Shing
;
Ju, Shan-Shan
;
Tao, Teh-Ho
For the importance of communication between human and machine interface, it would be valuable to develop an implement which has the ability to recognize emotional states. In this paper, we proposed an approach which can deal with the daily dependence and personal dependence in the data of multiple subjects and samples. 30 features were extracted from the physiological signals of subject for three states of emotion. The physiological signals measured were: electrocardiogram (ECG), skin temperature (SKT) and galvanic skin response (GSR). After removing the daily dependence and personal dependence by the statistical technique of MANOVA, six machine learning methods including Bayesian network learning, naive Bayesian classification, SVM, decision tree of C4.5, Logistic model and K-nearest-neighbor (KNN) were implemented to differentiate the emotional states. The results showed that Logistic model gives the best classification accuracy and the statistical technique of MANOVA can significantly improve the performance of all six machine learning methods in emotion recognition system.
Machine learning with automatic feature selection for multi-class protein fold classification
(2005-07-01)
Huang, CD
;
Liang, SF
;
Lin, CT
;
Wu, RC
;
電控工程研究所
In machine learning, both the properly used networks and the selected features are important factors which should be considered carefully. These two factors will influence the result, whether for better or worse. In bioinformatics, the amount of features may be very large to make machine learning possible. In this study we introduce the idea of feature selection in the problem of bioinformatics. We use neural networks to complete our task where each input node is associated with a gate. At the beginning of the training, all gates are almost closed, and, at this time, no features are allowed to enter the network. During the training phase, gates are either opened or closed, depending oil the requirements. After the selection training phase has completed, gates corresponding to the helpful features are completely opened while gates Corresponding to the useless features are closed more tightly. Some gates may be partially open, depending oil the importance of the corresponding features. So, the network can not only select features in an online manner during learning, but it also does some feature extraction. We combine feature selection with our novel hierarchical machine learning architecture and apply it to multi-class protein fold classification. At the first level the network classifies the data into four major folds: all alpha, all beta, alpha + beta and alpha/beta. In the next level, we have another set of networks which further classifies the data into twenty-seven folds. This approach helps achieve the following. The gating network is found to reduce the number of features drastically. It is interesting to observe that, for the first level using just 50 features selected by the gating network, we can get a test accuracy comparable to that using 125 features in neural classifiers. The process also helps us get a better insight into the folding process. For example, tracking the evolution of different gates, we call find which characteristics (features) of the data are more important for the folding process. Eventually, it reduces the computation time. The use of the hierarchical architecture helps LIS get a better performance also.
Online mining maximal frequent structures in continuous landmark melody streams
(2005-08-01)
Li, HF
;
Lee, SY
;
Shan, MK
;
資訊工程學系
;
Department of Computer Science
In this paper, we address the problem of online mining maximal frequent structures (Type I & II melody structures) in unbounded, continuous landmark melody streams. An efficient algorithm, called MMSLMS (Maximal Melody Structures of Landmark Melody Streams), is developed for online incremental mining of maximal frequent melody substructures in one scan of the continuous melody streams. In MMSLMS, a space-efficient scheme, called CMB (Chord-set Memory Border), is proposed to constrain the upper-bound of space requirement of maximal frequent melody structures in such a streaming environment. Theoretical analysis and experimental study show that our algorithm is efficient and scalable for mining the set of all maximal melody structures in a landmark melody stream. (c) 2005 Elsevier B.V. All rights reserved.

學術出版;;Publications

Browse

Filters

Settings

Sort By

Results per page

🏛️國立陽明交通大學機構典藏

Search Results