基于近邻传播算法的茶园土壤墒情传感器布局优化
|
摘要
针对节水灌溉的土壤墒情传感器布局问题,提出了基于近邻传播算法(affinity propagation,AP聚类算法)的优化布局策略。策略在保证茶园传感网络全覆盖的基础上,实时采集试验区各节点的土壤墒情数据,构建节点土壤含水率的相似度矩阵,并迭代计算各节点的吸引度和归属度值,得出聚类数和聚类中心。结果表明,采用AP聚类算法对试验区域传感器进行优化布局,优化了传感器数量和位置,使传感器数量由25个减少到2个。在试验区随机采集土壤相对含水率,经验证,聚类中心节点的土壤相对含水率与试验区平均值相近,相对偏差近为0.76%,表明聚类中心节点的土壤墒情数据具有代表性。该方法有效降低了数据的冗余度,节约了系统成本。
Aiming at the layout problem of soil moisture sensors for water-saving irrigation, we proposed an optimal layout strategy of soil moisture sensors based on affinity propagation(AP) clustering algorithm. The soil moisture of tea plantation was as the research object. The tea plantation had 84-m width and 190-m length. Following the conventional method, 25 sensor nodes were evenly arranged in rectangular mode in tea plantation experimental area in order to guarantee full coverage of tea plantation sensor network. Soil moisture data of each sensor node in the test area was collected in real time for 3 days. The optimization of sensors was conducted based on soil water content and relative water content by AP clustering algorithm. Different clustering parameters were selected. The AP clustering algorithm was used to construct similarity matrix of node soil water content, to iteratively calculate the responsibility and availability of each node, and to form the clustering number and clustering center. When the clustering parameters were 10, 15, 20 and 25 times of preference, the AP clustering algorithm was used to calculate the soil moisture data in the experimental area for 3 days, the stable and consistent clustering results were obtained. Results showed that soil water content in the tested plantation presented an increasing trend from southwest to northeast and the largest difference of relative water content was 15%. The change is related to the topography of the tested area. For AP clustering, the maximum iterative times was designed as 1 000. Based on the results, the clustering result in the 3 days was 2. The number of sensors optimized by AP clustering algorithm was reduced from 25 to 2. The class mean of the relative water content of the soil in the experimental area was calculated, and compared with the relative water content of soil in the collection points of the cluster center, and the relative bias between them was less than 5%. The relative water content of the collection points in the cluster center was close to the average value of the experimental area, which indicated that the data collected by the cluster center can represent soil moisture situation in the experimental tea plantation. In order to verify the validity of this method, soil moisture data were collected randomly at 13 locations in the experimental area on January 2019. Results showed that the soil average relative water content of tea plantation in the experimental area based on 13 sampling points was 32.7%, the relative water content of soil in the cluster center based on 2 sensors was respectively 27.9% and 37% with an average in the cluster center of 32.45%. Compared with the average relative moisture in the experimental area, the relative bias was only 0.76%. It means that the AP clustering algorithm can optimize the distribution of soil moisture sensors in the experimental tea plantation. The relative soil moisture collected by the cluster center could reflect the overall situation of soil moisture in the tea plantation as long as using only 2 sensors arranged in the cluster center node determined by the optimization calculation. Thus, the AP clustering algorithm is suggested to use in optimization of the sensor layout, which can reduce the redundancy of data and accordingly realize cost saving in agricultural production system.
Aiming at the layout problem of soil moisture sensors for water-saving irrigation, we proposed an optimal layout strategy of soil moisture sensors based on affinity propagation(AP) clustering algorithm. The soil moisture of tea plantation was as the research object. The tea plantation had 84-m width and 190-m length. Following the conventional method, 25 sensor nodes were evenly arranged in rectangular mode in tea plantation experimental area in order to guarantee full coverage of tea plantation sensor network. Soil moisture data of each sensor node in the test area was collected in real time for 3 days. The optimization of sensors was conducted based on soil water content and relative water content by AP clustering algorithm. Different clustering parameters were selected. The AP clustering algorithm was used to construct similarity matrix of node soil water content, to iteratively calculate the responsibility and availability of each node, and to form the clustering number and clustering center. When the clustering parameters were 10, 15, 20 and 25 times of preference, the AP clustering algorithm was used to calculate the soil moisture data in the experimental area for 3 days, the stable and consistent clustering results were obtained. Results showed that soil water content in the tested plantation presented an increasing trend from southwest to northeast and the largest difference of relative water content was 15%. The change is related to the topography of the tested area. For AP clustering, the maximum iterative times was designed as 1 000. Based on the results, the clustering result in the 3 days was 2. The number of sensors optimized by AP clustering algorithm was reduced from 25 to 2. The class mean of the relative water content of the soil in the experimental area was calculated, and compared with the relative water content of soil in the collection points of the cluster center, and the relative bias between them was less than 5%. The relative water content of the collection points in the cluster center was close to the average value of the experimental area, which indicated that the data collected by the cluster center can represent soil moisture situation in the experimental tea plantation. In order to verify the validity of this method, soil moisture data were collected randomly at 13 locations in the experimental area on January 2019. Results showed that the soil average relative water content of tea plantation in the experimental area based on 13 sampling points was 32.7%, the relative water content of soil in the cluster center based on 2 sensors was respectively 27.9% and 37% with an average in the cluster center of 32.45%. Compared with the average relative moisture in the experimental area, the relative bias was only 0.76%. It means that the AP clustering algorithm can optimize the distribution of soil moisture sensors in the experimental tea plantation. The relative soil moisture collected by the cluster center could reflect the overall situation of soil moisture in the tea plantation as long as using only 2 sensors arranged in the cluster center node determined by the optimization calculation. Thus, the AP clustering algorithm is suggested to use in optimization of the sensor layout, which can reduce the redundancy of data and accordingly realize cost saving in agricultural production system.
引文
[1]Cheng Shunfeng,Michael H Azarian,Michael G Pecht.Sensor system for prognostics and health management[J].Sensors,2010,10(6):5774-5797.
[2]朱喜华,李颖晖,李宁,等.基于改进离散粒子群算法的传感器布局优化设计[J].电子学报,2013,41(10):2104-2108.Zhu Xihua,Li Yinhui,Li Ning,et al.Optimal sensor placement design based on improved discrete PSOalgorithm[J].Acta Electronica Sinica,2013,41(10):2104-2108.(in Chinese with English abstract)
[3]陈仲生,杨拥民,李聪,等.基于振动传递符号有向图的齿轮箱嵌入式传感器优化配置模型与算法[J].航空动力学报,2009,24(10):2384-2390.Chen Zhongsheng,Yang Yongmin,Li Cong,et al.Model and algorithm of optimal embedded sensor placement for gearboxes based on signed directed graph of vibration progagation[J].Journal of Aerospace Power,2009,24(10):2384-2390.(in Chinese with English abstract)
[4]王宏力,张忠泉,崔祥祥,等.基于改进PSO算法的实时故障监测诊断测试集优化[J].系统工程与电子技术,2011,33(4):958-962.Wang Hongli,Zhang Zhongquan,Cui Xiangxiang,et al.Test optimization of real-time monitoring and fault diagnosis system based on improved particle swarm optimization[J].Systems Engineering and Electonics,2011,33(4):958-962.(in Chinese with English abstract)
[5]杨帆,萧德云.故障检测的可靠性描述及传感器分布优化算法[J].应用科学学报,2006,24(2):125-130.Yang Fan,Xiao Deyun.Reliability description of fault detection and optimization algorithm of sensor location[J].Journal of Applied Sciences,2006,24(2):125-130.(in Chinese with English abstract)
[6]刘丽萍.无线传感器网络节能覆盖[D].杭州:浙江大学,2006.Liu Liping.Energy-efficient Coverage in Wireless Sensor Networks[D].Hangzhou:Zhejiang University,2006.(in Chinese with English abstract)
[7]Lin Frank Y S,Chiu P L.A simulated annealing algorithm for energy-efficient sensor network design[C]//International Symposium on Modeling&Optimization in Mobile.2005.
[8]Wang Y C,Hu C C,Tseng Y C.Efficient placement and dispatch of sensors in a wireless sensor network[J].IEEETransactions on Mobile Computing,2007,7(2):262-274.
[9]赵伟霞,李久生,王珍,等.滴灌均匀性对土壤水分传感器埋设位置的影响[J].农业工程学报,2018,34(9):123-129.Zhao Weixia,Li Jiusheng,Wang Zhen,et al.Influence of drip irrigation uniformity on buried position of soil moisture sensor[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2018,34(9):123-129.(in Chinese with English abstract)
[10]蒋杰,方力,张鹤颖.无线传感器网络最小连通覆盖问题求解算法[J].软件学报2006,17(2):175-184.Jiang Jie,Fang Li,Zhang Heying.An algorithm for minimal connected cover set problem in wireless sensor networks.Journal of Software,2006,17(2):175-184.(in Chinese with English abstract)
[11]李飚,魏正英,张育斌,等.土壤墒情传感器布点算法研究[J].节水灌溉,2015(2):72-76.Li Biao,Wei Zhengying,Zhang Yubin,et al.Optimal placement algorithm of soil moisture sensors[J].Water-saving Irrigation,2015(2):72-76.(in Chinese with English abstract)
[12]姚月峰,满秀玲.毛乌素沙地不同林龄沙柳表层土壤水封空间异质性[J].水土保持学报,2007,21(1):112-115.Yao Yuefeng,Man Xiuling.Spatial Heterogeneity of top soil moisture in different stand age of salix psammophila in Mu US sandy land[J].Journal of Soil and Water Conservation,2007,21(1):112-115.(in Chinese with English abstract)
[13]黄弈龙,陈利顶,傅伯杰,等.黄土丘陵小流域土壤水分空间格局及其影响因素[J].自然资源学报,2005,20(4):483-492.Huang Yilong,Chen Liding,Fu Bojie,et al.Spatial pattern of soil moisture and its influencing factors in small watershed of loess hilly region[J].Journal of Natural Resources,2005,20(4):483-492.(in Chinese with English abstract)
[14]潘颜霞,王新平,苏延桂,等.不同植被类型沙地表层土壤水分变化特征[J].水土保持学报,2007,21(5):107-109.Pan Yanxia,Wang Xinping,Su Yangui,et al.Variability characteristics of surface soil moisture content in sand areas covered by different vegetation types[J].Journal of Soil and Water Conservation,2007,21(5):107-109.(in Chinese with English abstract)
[15]赵永宏,刘贤德,张学龙,等.祁连山区亚高山灌丛土壤含水量的空间分布与月份变化规律[J].自然资源学报,2016,31(4):672-681.Zhao Yonghong,Liu Xiande,Zhang Xuelong,et al.Spatial distribution and monthly variation of soil water content in subalpine shrubs in Qilian Mountains[J].Journal of Natural Resources,2016,31(4):672-681.(in Chinese with English abstract)
[16]许迪,Schmicl R.田间土壤特性的空间相关结构分析及其分布描述[J].灌溉排水,1996,15(4):16-20.Xu Di,Schmicl R.Spatial correlation structure analysis and distribution description of field soil properties[J].Irrigation and Drainage,1996,15(4):16-20(in Chinese with English abstract)
[17]刘自豪,张斌,祝宁,等.基于改进AP聚类算法的自学习应用层DDoS检测方法[J].计算机研究与发展,2018,55(6):1236-1246.Liu Zihao,Zhang Bin,Zhu Ning,et al.Adaptive app-ADDoSdetection method based on improved AP algorithm[J].Computer research and development,2018,55(6):1236-1246.(in Chinese with English abstract)
[18]唐丹,张正军.近邻传播聚类算法的优化[J].计算机应用,2017,37(z1):258-261.Tang Dan,Zhang Zhengjun.New Optimized affinity propagation clustering Algorithms[J].Computer Applications,2017,37(z1):258-261.(in Chinese with English abstract)
[19]Frey B J,Dueck D.Clustering by passing messages between data points[J].Science,2007,315(5814):972-976.
[20]董俊,王锁萍,熊范纶.可变相似性度量的近邻传播聚类[J].电子与信息学报,2010,32(3):509-514.Dong Jun,Wang Suoping,Xiong Fanlun.Affinity Propagation clustering based on variable-similarity measure[J].Journal of Electronics and Information,2010,32(3):509-514.(in Chinese with English abstract)
[21]王开军,张军英,李丹,等.自适应仿射传播聚类[J].自动化学报,2007,33(12):1242-1246.Wang Kaijun,Zhang Junying,Li Dan,et al.Adaptive affinity propagation clustering[J].Journal of Automation,2007,33(12):1242-1246.(in Chinese with English abstract)
[22]Wei Z,Wang Y,He S,et al.A novel intelligent method for bearing fault diagnosis based on affinity propagation clustering and adaptive feature selection[J].Knowledge-Based Systems,2017,116:1-12.
[23]孙劲光,赵欣.一种改进近邻传播聚类的图像分割算法[J].计算机工程与应用,2017,53(6):178-182.Sun Jinguang,Zhao Xin.Image segmentation algorithm based on improved affinity propagation clustering.Computer Engineering and Applications,2017,53(6):178-182.(in Chinese with English abstract)
[24]储岳中,刘恒,张学锋.基于迁移AP聚类与稀疏表示的遥感图像分类[J].计算机工程与设计,2018,39(2):547-550.Chu Yuezhong,Liu Heng,Zhang Xuefeng.Remote sensing image classification method based on tansfer affinity propagation and sparse representation[J].Computer Engineering and Design,2018,39(2):547-550.(in Chinese with English abstract)
[25]江颉,王卓芳,陈铁明,等.自适应AP聚类算法及其在入侵检测中的应用[J].通信学报,2015,36(11):118-126.Jiang Jie,Wang Zhuofang,Chen Tieming,et al.Adaptive APclustering algorithm and its application on intrusion detection[J].Journal of Communications,2015,36(11):118-126.(in Chinese with English abstract)
[26]赵延龙,滑楠.基于初始偏向度的AP算法聚类性能优化研究[J].计算机应用研究,2018(2):372-374.Zhao Yanlong,Hua Nan.Research on optimization of APalgorithm performance based on initial preference[J].Computer Application,2018(2):372-374.(in Chinese with English abstract)
[27]陈莹.基于AP聚类算法的图像分割技术研究[D].吉林:吉林建筑大学,2015.Chen Ying.Research on Image Segmentation Technology Based on AP Clustering Algorithm[D].Jilin:Jilin University of Architecture,2015.(in Chinese with English abstract)
[28]Yang C,Liu S,Bruzzone L,et al.A feature-metric-based affinity propagation technique for feature selection in hyperspectral image classification[J].IEEE Geoscience&Remote Sensing Letters,2013,10(5):1152-1156.
[29]Gao J,Ma Z,Qin Y,et al.Application of affinity propagation clustering algorithm in fault diagnosis of metro vehicle auxiliary inverter[J].Lecture Notes in Electrical Engineering,2014,288:3-9.
[30]刘晓勇,付辉.一种快速AP聚类算法[J].山东大学学报:工学版,2011,41(4):20-23.Liu Xiaoyong,Fu Hui.A fast affinity propagation clustering algorithm[J].Journal of Shandong University:Engineering Edition,2011,41(4):20-23.(in Chinese with English abstract)
[31]肖宇,于剑.基于近邻传播算法的半监督聚类[J].软件学报,2008,19(11):2803-2813.Xiao Yu,Yu Jian.Semi-supervised clustering based on affinity propagation algorithms[J].Journal of Software,2008,19(11):2803-2813.(in Chinese with English abstract)
[2]朱喜华,李颖晖,李宁,等.基于改进离散粒子群算法的传感器布局优化设计[J].电子学报,2013,41(10):2104-2108.Zhu Xihua,Li Yinhui,Li Ning,et al.Optimal sensor placement design based on improved discrete PSOalgorithm[J].Acta Electronica Sinica,2013,41(10):2104-2108.(in Chinese with English abstract)
[3]陈仲生,杨拥民,李聪,等.基于振动传递符号有向图的齿轮箱嵌入式传感器优化配置模型与算法[J].航空动力学报,2009,24(10):2384-2390.Chen Zhongsheng,Yang Yongmin,Li Cong,et al.Model and algorithm of optimal embedded sensor placement for gearboxes based on signed directed graph of vibration progagation[J].Journal of Aerospace Power,2009,24(10):2384-2390.(in Chinese with English abstract)
[4]王宏力,张忠泉,崔祥祥,等.基于改进PSO算法的实时故障监测诊断测试集优化[J].系统工程与电子技术,2011,33(4):958-962.Wang Hongli,Zhang Zhongquan,Cui Xiangxiang,et al.Test optimization of real-time monitoring and fault diagnosis system based on improved particle swarm optimization[J].Systems Engineering and Electonics,2011,33(4):958-962.(in Chinese with English abstract)
[5]杨帆,萧德云.故障检测的可靠性描述及传感器分布优化算法[J].应用科学学报,2006,24(2):125-130.Yang Fan,Xiao Deyun.Reliability description of fault detection and optimization algorithm of sensor location[J].Journal of Applied Sciences,2006,24(2):125-130.(in Chinese with English abstract)
[6]刘丽萍.无线传感器网络节能覆盖[D].杭州:浙江大学,2006.Liu Liping.Energy-efficient Coverage in Wireless Sensor Networks[D].Hangzhou:Zhejiang University,2006.(in Chinese with English abstract)
[7]Lin Frank Y S,Chiu P L.A simulated annealing algorithm for energy-efficient sensor network design[C]//International Symposium on Modeling&Optimization in Mobile.2005.
[8]Wang Y C,Hu C C,Tseng Y C.Efficient placement and dispatch of sensors in a wireless sensor network[J].IEEETransactions on Mobile Computing,2007,7(2):262-274.
[9]赵伟霞,李久生,王珍,等.滴灌均匀性对土壤水分传感器埋设位置的影响[J].农业工程学报,2018,34(9):123-129.Zhao Weixia,Li Jiusheng,Wang Zhen,et al.Influence of drip irrigation uniformity on buried position of soil moisture sensor[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2018,34(9):123-129.(in Chinese with English abstract)
[10]蒋杰,方力,张鹤颖.无线传感器网络最小连通覆盖问题求解算法[J].软件学报2006,17(2):175-184.Jiang Jie,Fang Li,Zhang Heying.An algorithm for minimal connected cover set problem in wireless sensor networks.Journal of Software,2006,17(2):175-184.(in Chinese with English abstract)
[11]李飚,魏正英,张育斌,等.土壤墒情传感器布点算法研究[J].节水灌溉,2015(2):72-76.Li Biao,Wei Zhengying,Zhang Yubin,et al.Optimal placement algorithm of soil moisture sensors[J].Water-saving Irrigation,2015(2):72-76.(in Chinese with English abstract)
[12]姚月峰,满秀玲.毛乌素沙地不同林龄沙柳表层土壤水封空间异质性[J].水土保持学报,2007,21(1):112-115.Yao Yuefeng,Man Xiuling.Spatial Heterogeneity of top soil moisture in different stand age of salix psammophila in Mu US sandy land[J].Journal of Soil and Water Conservation,2007,21(1):112-115.(in Chinese with English abstract)
[13]黄弈龙,陈利顶,傅伯杰,等.黄土丘陵小流域土壤水分空间格局及其影响因素[J].自然资源学报,2005,20(4):483-492.Huang Yilong,Chen Liding,Fu Bojie,et al.Spatial pattern of soil moisture and its influencing factors in small watershed of loess hilly region[J].Journal of Natural Resources,2005,20(4):483-492.(in Chinese with English abstract)
[14]潘颜霞,王新平,苏延桂,等.不同植被类型沙地表层土壤水分变化特征[J].水土保持学报,2007,21(5):107-109.Pan Yanxia,Wang Xinping,Su Yangui,et al.Variability characteristics of surface soil moisture content in sand areas covered by different vegetation types[J].Journal of Soil and Water Conservation,2007,21(5):107-109.(in Chinese with English abstract)
[15]赵永宏,刘贤德,张学龙,等.祁连山区亚高山灌丛土壤含水量的空间分布与月份变化规律[J].自然资源学报,2016,31(4):672-681.Zhao Yonghong,Liu Xiande,Zhang Xuelong,et al.Spatial distribution and monthly variation of soil water content in subalpine shrubs in Qilian Mountains[J].Journal of Natural Resources,2016,31(4):672-681.(in Chinese with English abstract)
[16]许迪,Schmicl R.田间土壤特性的空间相关结构分析及其分布描述[J].灌溉排水,1996,15(4):16-20.Xu Di,Schmicl R.Spatial correlation structure analysis and distribution description of field soil properties[J].Irrigation and Drainage,1996,15(4):16-20(in Chinese with English abstract)
[17]刘自豪,张斌,祝宁,等.基于改进AP聚类算法的自学习应用层DDoS检测方法[J].计算机研究与发展,2018,55(6):1236-1246.Liu Zihao,Zhang Bin,Zhu Ning,et al.Adaptive app-ADDoSdetection method based on improved AP algorithm[J].Computer research and development,2018,55(6):1236-1246.(in Chinese with English abstract)
[18]唐丹,张正军.近邻传播聚类算法的优化[J].计算机应用,2017,37(z1):258-261.Tang Dan,Zhang Zhengjun.New Optimized affinity propagation clustering Algorithms[J].Computer Applications,2017,37(z1):258-261.(in Chinese with English abstract)
[19]Frey B J,Dueck D.Clustering by passing messages between data points[J].Science,2007,315(5814):972-976.
[20]董俊,王锁萍,熊范纶.可变相似性度量的近邻传播聚类[J].电子与信息学报,2010,32(3):509-514.Dong Jun,Wang Suoping,Xiong Fanlun.Affinity Propagation clustering based on variable-similarity measure[J].Journal of Electronics and Information,2010,32(3):509-514.(in Chinese with English abstract)
[21]王开军,张军英,李丹,等.自适应仿射传播聚类[J].自动化学报,2007,33(12):1242-1246.Wang Kaijun,Zhang Junying,Li Dan,et al.Adaptive affinity propagation clustering[J].Journal of Automation,2007,33(12):1242-1246.(in Chinese with English abstract)
[22]Wei Z,Wang Y,He S,et al.A novel intelligent method for bearing fault diagnosis based on affinity propagation clustering and adaptive feature selection[J].Knowledge-Based Systems,2017,116:1-12.
[23]孙劲光,赵欣.一种改进近邻传播聚类的图像分割算法[J].计算机工程与应用,2017,53(6):178-182.Sun Jinguang,Zhao Xin.Image segmentation algorithm based on improved affinity propagation clustering.Computer Engineering and Applications,2017,53(6):178-182.(in Chinese with English abstract)
[24]储岳中,刘恒,张学锋.基于迁移AP聚类与稀疏表示的遥感图像分类[J].计算机工程与设计,2018,39(2):547-550.Chu Yuezhong,Liu Heng,Zhang Xuefeng.Remote sensing image classification method based on tansfer affinity propagation and sparse representation[J].Computer Engineering and Design,2018,39(2):547-550.(in Chinese with English abstract)
[25]江颉,王卓芳,陈铁明,等.自适应AP聚类算法及其在入侵检测中的应用[J].通信学报,2015,36(11):118-126.Jiang Jie,Wang Zhuofang,Chen Tieming,et al.Adaptive APclustering algorithm and its application on intrusion detection[J].Journal of Communications,2015,36(11):118-126.(in Chinese with English abstract)
[26]赵延龙,滑楠.基于初始偏向度的AP算法聚类性能优化研究[J].计算机应用研究,2018(2):372-374.Zhao Yanlong,Hua Nan.Research on optimization of APalgorithm performance based on initial preference[J].Computer Application,2018(2):372-374.(in Chinese with English abstract)
[27]陈莹.基于AP聚类算法的图像分割技术研究[D].吉林:吉林建筑大学,2015.Chen Ying.Research on Image Segmentation Technology Based on AP Clustering Algorithm[D].Jilin:Jilin University of Architecture,2015.(in Chinese with English abstract)
[28]Yang C,Liu S,Bruzzone L,et al.A feature-metric-based affinity propagation technique for feature selection in hyperspectral image classification[J].IEEE Geoscience&Remote Sensing Letters,2013,10(5):1152-1156.
[29]Gao J,Ma Z,Qin Y,et al.Application of affinity propagation clustering algorithm in fault diagnosis of metro vehicle auxiliary inverter[J].Lecture Notes in Electrical Engineering,2014,288:3-9.
[30]刘晓勇,付辉.一种快速AP聚类算法[J].山东大学学报:工学版,2011,41(4):20-23.Liu Xiaoyong,Fu Hui.A fast affinity propagation clustering algorithm[J].Journal of Shandong University:Engineering Edition,2011,41(4):20-23.(in Chinese with English abstract)
[31]肖宇,于剑.基于近邻传播算法的半监督聚类[J].软件学报,2008,19(11):2803-2813.Xiao Yu,Yu Jian.Semi-supervised clustering based on affinity propagation algorithms[J].Journal of Software,2008,19(11):2803-2813.(in Chinese with English abstract)