服装设计风格决策模型的研究与实现
|
摘要
服装设计风格作为一个抽象概念,并没有确切的定义和准确的度量方法。由于人的偏好、心理以及当时的情绪状态等多方面因素的影响,个人对服装款式设计风格的观点都有着自身独特的见解。针对款式风格的抽象性和主观性问题,建立满足用户个性化需求的风格决策推荐系统是必要的。构建个性化风格决策模型是完成整个系统的重要步骤。本文主要解决了个性化服装款式设计风格决策推荐系统中如下问题：
1、根据服装学相关资料,服装风格可分为多种类别,由于现有专家的服装风格评价样本数量有限,而支持向量机在解决小样本、非线性等问题都体现了特有优势。本文选取支持向量机建立专家评判分类决策模型,有助于将整个数据库中的服装款式按照专家信息进行分类。通过专家分类模型可为用户推荐符合专家评判的服装。
2、服装风格是一个模糊而又难以量化的概念,因此系统通过专家分类模型推荐给用户的服装并不能满足用户的个性化需求。本文采用交互式遗传算法偏好模型满足用户个性化需求,首先系统通过和用户交互获得用户对样本服装的评价信息,同时对服装部件编码,经过遗传算法建立服装风格个性化偏好模型,最终通过该模型为用户推荐可供参考的个性化服装。
3、用户评价服装款式设计风格的过程,其实是一种心理决策过程。本文采用多备择决策场理论所建立的决策模型不仅可以体现决策者判断服装风格的心理变化,更能够体现非理性效应对服装风格强弱度判断的影响。决策场理论决策模型,有效地解决小范围内用户对服装风格的心理评判。
4、三种决策模型分别解决了系统实现过程中的不同问题,本文结合三种模型,最终设计并实现了个性化服装风格决策推荐系统。
Fashion style as an abstract concept is no precise definition and accurate measurement method. Because of people's preferences, mental and emotional state at the time, and many other factors, people all have their own unique perspectives on fashion style. Due to subjectivity and abstract of style, to meet the needs of individual users to establish the style of decision-making recommendation system is necessary. Construction of personal style decision-making model is an important step to complete the system. About the personalized fashion style decision-making system, the paper mainly solves the following questions:
1. According to information of clothing, fashion style can be divided into several categories; as the existing expert evaluation of the limited sample size, and support vector machine to solve the small sample, non-linear and other issues are reflected unique advantages. So selecting support vector machine to establish expert evaluation classification decision model can help classify the entire clothing database according to experts'information. By experts'classification model, system can recommend clothing which meets the evaluation of experts for users.
2. Fashion style is a vague concept and difficult to quantify, so the clothes recommended by experts'classification model does not meet the user's individual requirements. In this paper, we use Preference Model of Interactive Genetic Algorithm to the needs of individual users. First, we obtain users assessment information of clothing samples and coding parts of the clothing. Then we can use genetic algorithm to establish personalized fashion style decision model. Finally by this model system can recommended personal reference costume for users.
3. The process that users evaluate Fashion style, in fact, is a psychological decision-making process. In this paper, the decision model of Multi-alternative decision field reflects the changes that decision makers determine the fashion style and the Irrational Effect. The decision model of Multi-alternative decision field can effectively solve the users' psychology evaluation of fashion styles in the small scale.
4. Three decision-making methods solve the system different issues in the implementation. Ultimately, we integrate three models, design and achieve the personalized fashion style decision-making system.
1、根据服装学相关资料,服装风格可分为多种类别,由于现有专家的服装风格评价样本数量有限,而支持向量机在解决小样本、非线性等问题都体现了特有优势。本文选取支持向量机建立专家评判分类决策模型,有助于将整个数据库中的服装款式按照专家信息进行分类。通过专家分类模型可为用户推荐符合专家评判的服装。
2、服装风格是一个模糊而又难以量化的概念,因此系统通过专家分类模型推荐给用户的服装并不能满足用户的个性化需求。本文采用交互式遗传算法偏好模型满足用户个性化需求,首先系统通过和用户交互获得用户对样本服装的评价信息,同时对服装部件编码,经过遗传算法建立服装风格个性化偏好模型,最终通过该模型为用户推荐可供参考的个性化服装。
3、用户评价服装款式设计风格的过程,其实是一种心理决策过程。本文采用多备择决策场理论所建立的决策模型不仅可以体现决策者判断服装风格的心理变化,更能够体现非理性效应对服装风格强弱度判断的影响。决策场理论决策模型,有效地解决小范围内用户对服装风格的心理评判。
4、三种决策模型分别解决了系统实现过程中的不同问题,本文结合三种模型,最终设计并实现了个性化服装风格决策推荐系统。
Fashion style as an abstract concept is no precise definition and accurate measurement method. Because of people's preferences, mental and emotional state at the time, and many other factors, people all have their own unique perspectives on fashion style. Due to subjectivity and abstract of style, to meet the needs of individual users to establish the style of decision-making recommendation system is necessary. Construction of personal style decision-making model is an important step to complete the system. About the personalized fashion style decision-making system, the paper mainly solves the following questions:
1. According to information of clothing, fashion style can be divided into several categories; as the existing expert evaluation of the limited sample size, and support vector machine to solve the small sample, non-linear and other issues are reflected unique advantages. So selecting support vector machine to establish expert evaluation classification decision model can help classify the entire clothing database according to experts'information. By experts'classification model, system can recommend clothing which meets the evaluation of experts for users.
2. Fashion style is a vague concept and difficult to quantify, so the clothes recommended by experts'classification model does not meet the user's individual requirements. In this paper, we use Preference Model of Interactive Genetic Algorithm to the needs of individual users. First, we obtain users assessment information of clothing samples and coding parts of the clothing. Then we can use genetic algorithm to establish personalized fashion style decision model. Finally by this model system can recommended personal reference costume for users.
3. The process that users evaluate Fashion style, in fact, is a psychological decision-making process. In this paper, the decision model of Multi-alternative decision field reflects the changes that decision makers determine the fashion style and the Irrational Effect. The decision model of Multi-alternative decision field can effectively solve the users' psychology evaluation of fashion styles in the small scale.
4. Three decision-making methods solve the system different issues in the implementation. Ultimately, we integrate three models, design and achieve the personalized fashion style decision-making system.
引文
[1]冯玲玲.略谈服装设计风格[A].吉林工程技术师范学院学报.2008,12.
[2]刘晓刚.品牌服装设计[M].中国纺织大学出版社,2001.
[3]张海花.服装风格与传播研究[D],北京服装学院,2009.
[4]冯利、刘晓刚.服装风格的量化方法初探[J]：东华大学学报。2004.2,p57-60.
[5]邓苏、黄宏斌.决策支持系统[M],电子工业出版社.
[6]魏现梅.决策支持系统中的模型选择研究[D].大连交通大学,2010.
[7]Simon Haykin著,叶世伟等译.神经网络原理[M].北京：机械工业出版社,2004：353-432.
[8]蚁平、曹先彬.基于交互式遗传算法的个性化建筑物外观设计[J].计算机仿真,2006,23(5)：156-159.
[9]钟斌,心理学中决策模型的发展,科技信息,2008(8)：P289.
[10]徐光国、张进辅.决策场理论简介,中国学术期刊出版,1995.
[11]Roe, R, Busemeyer, JR, Townsend, JT. in press. Multi-alternative decision field theory:a dynamic connectional model of decision-making[J]. Psychological Review,2001,108(2):370-392.
[12]张兴旺,基于SVM的分类挖掘算法及其应用[J],大庆石油学院,2007,03.
[13]阎威武、邵惠鹤.支持向量机和最小二乘支持向量机的比较以应用研究.控制与决策,2003,(3)：378-381.
[14]John Shawe-Taylor,Nello Cristianini. Support Vector Machines and other kernel-based learning methods[m].Cambridge:Cambridge University Press.2000.
[15]边肇棋、张学工等.模式识别[M].北京：清华大学出版社,2000：234-343.
[16]Y.A LeCun, L D Jackel,L Bottou, A Brunot, C Cortes, J S Denker, H Drucker, I Guyon, U A Muller, E sackinger, P Y Simard, V N Vapnik。Learning algorithms for classification:A comparism on handwritten digit recognition[J]. Neural Networks,1995:261-276.
[17]C J C Burges, B Scholkopf. Improving the accuracy and speed of support vector learning manchines. Advances in Neural Information Processing System[M].MIT Press,1997:375-381.
[18]袁立、穆志纯、刘磊明.基于核主元分析法和支持向量机的人耳识别.北京科技大学学报,2006,(9)：890-895.
[19]S Mukherjee, E Osuna, F.Girosi.Nonlinear prediction of chaotic time series using a support vector machine[J].New York:Neural Networks for Signal Processing,1997,7:511-520.
[20]k-R Muller, A J Smola, G Ratsch, B Scholkopf, J Kohlmorgen, V.N. Vapnik.Predicting time series with support vector machines[J].Berlin:Artcial Neural Networks-ICANN'97,1997:999-1004.
[21]A Smola, B.Scholkopf. A tutorial on support vector regression[J].Statistics and Computing,2004,14:199-222.
[22]Hu Jun, Huang Houkuan. An algorithm for text categorization with SVM.TENCON'02.Proceedings.2002 IEEE Region 10 Conference on Computers, Communications, Control and power Engineering,2002,1:47-50.
[23]苏金树、张博锋、徐昕.基于机器学习的文本分类研究进展[J].软件学报,2006,17(9)：1848-1859.
[24]Chang,C.-C. and C.-J.Lin(2001).LIBSVM:a bibrary for support vector machines Software available at http://www.csie.ntu.edu.tw/～cjlin/libsvm.
[25]Chih-Wei Hsu, Chih-Chung Chang, and Chih-Jen Lin.A Practical Guide to Support Vector Classification. http://www.csie.ntu.edu.tw/～cjlin/libsvm.
[26]纪哓燕、刘晓刚.服装款式设计零部件数字化问题[J].东华大学学报：自然科学版,2002,28(4)：P117-120.
[27]Ying Wang, Yan Chen, Zhi-ge Chen. The sensory research on the style of women's overcoats[J]. International Journal of Clothing Science and Technology,2008,20 (3):174-183.
[28]S. S. Keerthi, S. Sundararajan. K.-W. Chang, C.-J. Hsieh, and C.-J. Lin, A sequential dual method for large scale multi-class linear SVMs. KDD 2008.
[29]C.-J. Hsieh K.-W. Chang, C.-J. Lin, S. S. Keerthi, and S. Sundararajan. A Dual Coordinate Descent Method For Large-Scale Linear SVM. ICML 2008.
[30]H.-T. Lin and C.-J. Lin. A study on sigmoid kernels for SVM and the training of non-PSD Kernels by SMO-type methods. March 2003.
[31]陈国良、王煦法等.遗传算法及其应用.北京：人民邮电出版社,1996.28—40.
[32]王小平、曹立明.遗传算法——理论、应用与软件实现,西安交通大学出版社.
[33]R.Benayoun, J.de Montgolfier, J.Tergn and O.Laritchev. Linear programming with multiple objective functions:Step method (stem) [J].Mathematical Programming,1971 12,vol 1,366-375.
[34]Fukada Y,Sato K,Mitsukura Y,et al.The room design system of individual preference with IGA[C].Proceedings of the International Conference on Control,Automation and Systems.Seoul,Korea:IEEE Press,2007:2 158-2 161.
[35]Szeto K Y, Zhao S Y. Adaptive spatial allocation of resource for parallel genetic algorithm[J]. Studies in Computational Intelligence,2008,129(5):389-398.
[36]李博,交互式多目标遗传算法在调度知识库中的研究[D],大连交通大学,2009.
[37]魏兆旺、梁昌勇、陆青、张恩桥.基于径向基网络与交互式遗传算法的服装设计.计算机工程与设计,2008,29(2)
[38]陈雁、李栋高.服装颜色风格的客观评价[J].纺织学报,2003,24(1)：80-83.
[39]刘元风.服装设计学,北京：高等教育出版社,1997.
[40]成果.个性化服装款式系统风格决策模型的研究及应用[D].东华大学.2010.
[41]周明、孙树栋.遗传算法原理及应用[M].北京：国防工业出版社,1999：46- 50.
[42]Gold J I, Shadlen M N. Rep resentation of a perceptual decision in developing oculomotor commands[J]. Nature,2000,404 (6776):390-394.
[43]Busemeyer, J. R., R. K. Jessup, Joseph G. Johnson, James T. Townsend. (2006). Building bridges between neural models and complex decision making behavior. Neural Networks 19(8):1047-1058.
[44]Busemeyer J R, Jessup R K, Johnson J G, et al Building bridges between neural models and complex decision making behavior [J]. Neural N etworks,2006,19(8): 1047-1058.
[45]Huber J, Payne J W, Puto C. Adding asymmetrically dominated alternatives: Violations of regularity and the similarity hypothesis Journal of Consumer Research,1982,9:90-98.
[46]Tversky A, Simonson I. Context-dependent Preferences. Management Science,1993,39(10):1179-1188.
[47]Tversky A. Elimination by aspects:A theory of choice. Psychological Review, 1972,79(4):281-299.
[2]刘晓刚.品牌服装设计[M].中国纺织大学出版社,2001.
[3]张海花.服装风格与传播研究[D],北京服装学院,2009.
[4]冯利、刘晓刚.服装风格的量化方法初探[J]：东华大学学报。2004.2,p57-60.
[5]邓苏、黄宏斌.决策支持系统[M],电子工业出版社.
[6]魏现梅.决策支持系统中的模型选择研究[D].大连交通大学,2010.
[7]Simon Haykin著,叶世伟等译.神经网络原理[M].北京：机械工业出版社,2004：353-432.
[8]蚁平、曹先彬.基于交互式遗传算法的个性化建筑物外观设计[J].计算机仿真,2006,23(5)：156-159.
[9]钟斌,心理学中决策模型的发展,科技信息,2008(8)：P289.
[10]徐光国、张进辅.决策场理论简介,中国学术期刊出版,1995.
[11]Roe, R, Busemeyer, JR, Townsend, JT. in press. Multi-alternative decision field theory:a dynamic connectional model of decision-making[J]. Psychological Review,2001,108(2):370-392.
[12]张兴旺,基于SVM的分类挖掘算法及其应用[J],大庆石油学院,2007,03.
[13]阎威武、邵惠鹤.支持向量机和最小二乘支持向量机的比较以应用研究.控制与决策,2003,(3)：378-381.
[14]John Shawe-Taylor,Nello Cristianini. Support Vector Machines and other kernel-based learning methods[m].Cambridge:Cambridge University Press.2000.
[15]边肇棋、张学工等.模式识别[M].北京：清华大学出版社,2000：234-343.
[16]Y.A LeCun, L D Jackel,L Bottou, A Brunot, C Cortes, J S Denker, H Drucker, I Guyon, U A Muller, E sackinger, P Y Simard, V N Vapnik。Learning algorithms for classification:A comparism on handwritten digit recognition[J]. Neural Networks,1995:261-276.
[17]C J C Burges, B Scholkopf. Improving the accuracy and speed of support vector learning manchines. Advances in Neural Information Processing System[M].MIT Press,1997:375-381.
[18]袁立、穆志纯、刘磊明.基于核主元分析法和支持向量机的人耳识别.北京科技大学学报,2006,(9)：890-895.
[19]S Mukherjee, E Osuna, F.Girosi.Nonlinear prediction of chaotic time series using a support vector machine[J].New York:Neural Networks for Signal Processing,1997,7:511-520.
[20]k-R Muller, A J Smola, G Ratsch, B Scholkopf, J Kohlmorgen, V.N. Vapnik.Predicting time series with support vector machines[J].Berlin:Artcial Neural Networks-ICANN'97,1997:999-1004.
[21]A Smola, B.Scholkopf. A tutorial on support vector regression[J].Statistics and Computing,2004,14:199-222.
[22]Hu Jun, Huang Houkuan. An algorithm for text categorization with SVM.TENCON'02.Proceedings.2002 IEEE Region 10 Conference on Computers, Communications, Control and power Engineering,2002,1:47-50.
[23]苏金树、张博锋、徐昕.基于机器学习的文本分类研究进展[J].软件学报,2006,17(9)：1848-1859.
[24]Chang,C.-C. and C.-J.Lin(2001).LIBSVM:a bibrary for support vector machines Software available at http://www.csie.ntu.edu.tw/～cjlin/libsvm.
[25]Chih-Wei Hsu, Chih-Chung Chang, and Chih-Jen Lin.A Practical Guide to Support Vector Classification. http://www.csie.ntu.edu.tw/～cjlin/libsvm.
[26]纪哓燕、刘晓刚.服装款式设计零部件数字化问题[J].东华大学学报：自然科学版,2002,28(4)：P117-120.
[27]Ying Wang, Yan Chen, Zhi-ge Chen. The sensory research on the style of women's overcoats[J]. International Journal of Clothing Science and Technology,2008,20 (3):174-183.
[28]S. S. Keerthi, S. Sundararajan. K.-W. Chang, C.-J. Hsieh, and C.-J. Lin, A sequential dual method for large scale multi-class linear SVMs. KDD 2008.
[29]C.-J. Hsieh K.-W. Chang, C.-J. Lin, S. S. Keerthi, and S. Sundararajan. A Dual Coordinate Descent Method For Large-Scale Linear SVM. ICML 2008.
[30]H.-T. Lin and C.-J. Lin. A study on sigmoid kernels for SVM and the training of non-PSD Kernels by SMO-type methods. March 2003.
[31]陈国良、王煦法等.遗传算法及其应用.北京：人民邮电出版社,1996.28—40.
[32]王小平、曹立明.遗传算法——理论、应用与软件实现,西安交通大学出版社.
[33]R.Benayoun, J.de Montgolfier, J.Tergn and O.Laritchev. Linear programming with multiple objective functions:Step method (stem) [J].Mathematical Programming,1971 12,vol 1,366-375.
[34]Fukada Y,Sato K,Mitsukura Y,et al.The room design system of individual preference with IGA[C].Proceedings of the International Conference on Control,Automation and Systems.Seoul,Korea:IEEE Press,2007:2 158-2 161.
[35]Szeto K Y, Zhao S Y. Adaptive spatial allocation of resource for parallel genetic algorithm[J]. Studies in Computational Intelligence,2008,129(5):389-398.
[36]李博,交互式多目标遗传算法在调度知识库中的研究[D],大连交通大学,2009.
[37]魏兆旺、梁昌勇、陆青、张恩桥.基于径向基网络与交互式遗传算法的服装设计.计算机工程与设计,2008,29(2)
[38]陈雁、李栋高.服装颜色风格的客观评价[J].纺织学报,2003,24(1)：80-83.
[39]刘元风.服装设计学,北京：高等教育出版社,1997.
[40]成果.个性化服装款式系统风格决策模型的研究及应用[D].东华大学.2010.
[41]周明、孙树栋.遗传算法原理及应用[M].北京：国防工业出版社,1999：46- 50.
[42]Gold J I, Shadlen M N. Rep resentation of a perceptual decision in developing oculomotor commands[J]. Nature,2000,404 (6776):390-394.
[43]Busemeyer, J. R., R. K. Jessup, Joseph G. Johnson, James T. Townsend. (2006). Building bridges between neural models and complex decision making behavior. Neural Networks 19(8):1047-1058.
[44]Busemeyer J R, Jessup R K, Johnson J G, et al Building bridges between neural models and complex decision making behavior [J]. Neural N etworks,2006,19(8): 1047-1058.
[45]Huber J, Payne J W, Puto C. Adding asymmetrically dominated alternatives: Violations of regularity and the similarity hypothesis Journal of Consumer Research,1982,9:90-98.
[46]Tversky A, Simonson I. Context-dependent Preferences. Management Science,1993,39(10):1179-1188.
[47]Tversky A. Elimination by aspects:A theory of choice. Psychological Review, 1972,79(4):281-299.