{"title":"Shallow parsing with Hidden Markov Support Vector Machines","authors":"Shixi Fan, Lidan Chen, Xuan Wang, Buzhou Tang","doi":"10.1109/ICMLC.2014.7009716","DOIUrl":"https://doi.org/10.1109/ICMLC.2014.7009716","url":null,"abstract":"Shallow parsing system, providing natural part syntactic information statement, to meet a lot of language information processing requirements, has received much attention recent years. Hidden Markov Support Vector Machines (HM-SVMs) for sequence labeling offer advantages over both generative models like HMMs and classifying models like SVMs which give labeling result for each positionseparately. We show how to train a HM-SVM model to achieve good performance on the data set of CoNLL2000 share task. The HM-SVMs yields an F-score of 95.51% which is better than any system result of ConLL2000 share task.","PeriodicalId":335296,"journal":{"name":"2014 International Conference on Machine Learning and Cybernetics","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122069316","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"AN indicator-based selection multi-objective evolutionary algorithm with preference for multi-class ensemble","authors":"Jingjing Cao, S. Kwong, Ran Wang, Ke Li","doi":"10.1109/ICMLC.2014.7009108","DOIUrl":"https://doi.org/10.1109/ICMLC.2014.7009108","url":null,"abstract":"One of the most difficult components for multi-class classification system is to find an appropriate error-correcting output codes (ECOC) matrix, which is used to decompose the multi-class problem into several binary class problems. In this paper, an indicator based multi-objective evolutionary algorithm with preference involved is designed to search the high-quality ECOC matrix. Specifically, the Harrington's one-sided desirability function is integrated into an indicator-based evolutionary algorithm (IBEA), which aims to approximate the relevant regions of pareto front (PF) according to the preference of the decision maker. Simulation results show that the proposed approach has better classification performance than compared multi-class based algorithms.","PeriodicalId":335296,"journal":{"name":"2014 International Conference on Machine Learning and Cybernetics","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129824012","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Latent semantic KNN algorithm for multi-label learning","authors":"Zijie Chen, Z. Hao","doi":"10.1109/ICMLC.2014.7009129","DOIUrl":"https://doi.org/10.1109/ICMLC.2014.7009129","url":null,"abstract":"Exploiting label structures or label correlations is an important issue in multi-label learning, because taking into account such structures when learning can lead to improved predictive performance and time complexity. In this paper, a multi-label lazy learning approach based on k-nearest neighbor and latent semantics is presented, which is called LsKNN. Firstly, latent semantic analysis is applied to discover some semantic correlations between instances and class labels and the semantic features of each training sample are obtained. Then for each unseen instance, its k-nearest neighbors in the latent semantic subspace are identified and finally its proper label set is determined by resembling the votes of neighbors. Meanwhile, a support vector machine based pruning strategy called SVM-LsKNN, is proposed to deal with the slow testing of LsKNN. Experiments on three multi-label sets show that LsKNN needs no training, but can achieve at least comparable performance with some state-of-art multi-label learning algorithms. Extra experiments also verify the testing efficiency of the pruning technique.","PeriodicalId":335296,"journal":{"name":"2014 International Conference on Machine Learning and Cybernetics","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130547659","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fu-Hsiang Chan, Duan-Yu Chen, J. Hsieh, Chi-Hung Chuang
{"title":"Wrinkle of fingers based robust person identification","authors":"Fu-Hsiang Chan, Duan-Yu Chen, J. Hsieh, Chi-Hung Chuang","doi":"10.1109/ICMLC.2014.7009724","DOIUrl":"https://doi.org/10.1109/ICMLC.2014.7009724","url":null,"abstract":"This paper proposed a novel biometric identification system through 2D fingers' geometry measurements. First, the right middle finger and index finger are captured using a CCD camera. Then the fingers are segmented out from background based on skin colors. Splitting out two fingers based on their contours, multiple features such as the length, mean width, finger shape vector and wrinkle texture are computed and consequently are used for person identification. Experiments show that our proposed method can perform well in real time with the recognition rate being up to 97%.","PeriodicalId":335296,"journal":{"name":"2014 International Conference on Machine Learning and Cybernetics","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132395696","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A weak order in special type fuzzy ellipsoid number space and its application in ranking uncertain multi-channel information","authors":"Guixiang Wang, Fang Yuan","doi":"10.1109/ICMLC.2014.7009086","DOIUrl":"https://doi.org/10.1109/ICMLC.2014.7009086","url":null,"abstract":"In this paper, a binary relation on fuzzy ellipsoid number space is defined. Its properties are investigated and shown to be a weak order on the space. Then, specific calculation formulas of the weak order are obtained for some specific fuzzy ellipsoid number spaces. A practical example is given to demonstrate the application of the weak order in ranking uncertain or imprecise multidimensional digital information.","PeriodicalId":335296,"journal":{"name":"2014 International Conference on Machine Learning and Cybernetics","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127982311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"The guaranteed cost control for a delay dependent nonlinear singular system","authors":"Lian-Qing Su, Xia Zhao","doi":"10.1109/ICMLC.2014.7009693","DOIUrl":"https://doi.org/10.1109/ICMLC.2014.7009693","url":null,"abstract":"The problem of the guaranteed cost control for a delay dependent nonlinear singular system is considered and to design a memory state feedback controller. Its disturbance meets conditions of Lipschitz, and bases on Lyapunov stability theory and LMI technique. We obtained a sufficient stabilization condition of closed loop system and expressions of guaranteed cost control. Finally, a numerical example is provided to prove the effectiveness of the proposed method.","PeriodicalId":335296,"journal":{"name":"2014 International Conference on Machine Learning and Cybernetics","volume":"157 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123323786","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Research on the fuzzy effect equilibrium value measurement method","authors":"Fachao Li, Meng Yang","doi":"10.1109/ICMLC.2014.7009697","DOIUrl":"https://doi.org/10.1109/ICMLC.2014.7009697","url":null,"abstract":"Fuzzy number, as the extension of quantity and set, is a common tool for describing uncertain information in practical problems. How to construct measure system reflecting the size characteristics of fuzzy number has important theoretical significance. In this paper, by analyzing the deficiency, and combining with the balance theory of particle system, we establish a measure method for fuzzy equilibrium value based on cut set (CFEV for short), and give the calculation methods for triangular fuzzy number and trapezoidal fuzzy number. Finally, we apply CFEV to a programming example. The result shows that CFEV has good interpretability and easy analysis, it can easily deal with the algebraic operation problems of fuzzy number. All these discussions can provide useful theoretical foundation for fuzzy decision making under complex environment.","PeriodicalId":335296,"journal":{"name":"2014 International Conference on Machine Learning and Cybernetics","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126417727","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Multivariable self-tuning PID controller based on wavelet fuzzy neural networks","authors":"Chi-Huang Lu, Pengcheng Liao, Yuan-Hai Charng, Chi-Ming Liu, Jheng-Yu Guo","doi":"10.1109/ICMLC.2014.7009704","DOIUrl":"https://doi.org/10.1109/ICMLC.2014.7009704","url":null,"abstract":"This paper presents a multivariable self-tuning proportional-integral-derivative (PID) controller based on wavelet fuzzy neural networks (WFNNs) for a class of nonlinear systems. A mathematic model using WFNN is constructed for the controlled nonlinear multivariable system, and the self-tuning PID controller is derived via a generalized predictive performance criterion. Numerical simulations exhibit that the proposed multivariable self-tuning PID control law gives satisfactory tracking and disturbance rejection performances.","PeriodicalId":335296,"journal":{"name":"2014 International Conference on Machine Learning and Cybernetics","volume":"174 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123009665","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Study on distribution network expansion planning improved algorithm based on QPSO for contain distributed power","authors":"Mingguang Zhang, Shi-Liang Wang","doi":"10.1109/ICMLC.2014.7009102","DOIUrl":"https://doi.org/10.1109/ICMLC.2014.7009102","url":null,"abstract":"When distributed power is connected to distribution network, it will produce all kinds of influence on distribution network. So we need to optimize all kinds of distributed power that access the distribution network. Mathematical model which contains the distributed power distribution network expansion planning is presented in this paper, takeing construction and maintenance cost minimum as objective function and applying binary coding quantum particle swarm algorithm to optimize type, location and capacity of distributed power and new or upgraded lines. To solve a large number of infeasible solutions of which is caused by distribution network planning, the related literatures are referred to repair infeasible solution, isolated chain and ring, converts the non-radial network to radial network. The example analysis shows that the planning schemes of power distribution network planning are more diversity and robustness.","PeriodicalId":335296,"journal":{"name":"2014 International Conference on Machine Learning and Cybernetics","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126299711","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Weighted grid Principal Component Analysis hashing","authors":"Xiancheng Zhou, Zhi-Qian Huang, Wing W. Y. Ng","doi":"10.1109/ICMLC.2014.7009117","DOIUrl":"https://doi.org/10.1109/ICMLC.2014.7009117","url":null,"abstract":"Principal Component Analysis (PCA) is one of the most widely used components of hashing. In this paper, we propose three PCA-based hashing methods to improve the performance of the Principal Component Hashing (PCH). Different principal components have different among of variances of data. In the PCH, each principal component corresponds to a hash function. Hence, the PCH treats each principal component to have the same importance which will lead to the loss of much information in constructing hashing table. To deal with this shortage, we propose the weighted PCH (WPCH), the grid PCH (GPCH) and the weighted grid PCH (WGPCH).","PeriodicalId":335296,"journal":{"name":"2014 International Conference on Machine Learning and Cybernetics","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126584319","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}