聚类分析作业.docx

聚类分析作业.docx

《聚类分析作业.docx》由会员分享，可在线阅读，更多相关《聚类分析作业.docx（34页珍藏版）》请在冰豆网上搜索。

聚类分析作业聚类分析作业1根据调查得到某地42所学校的数据如下：

单位占地面积（m2）建筑面积（m2）教师总数学生总数学校12088562.0542434学校210344.84755761279学校32700410056820学校43967375167990学校55850.246173.25781241学校61803.265224.99721180学校72165239147671学校8383838151041400学校92268801156800学校1098093000901597学校112996388961745学校1228863372651722学校1332000180002002000学校1411842111061432006学校157610531069790学校1620518308671882658学校177860831377852学校18557055791031350学校191960086001081490学校2013351.4787841902580学校219534124461421800学校2242080195001872609学校232000210062332学校2479287024961267学校2526000150001602000学校2612371121731281634学校2785616556102714学校2811842111061632904学校29188508332851196学校3025244105351251400学校311000085002003700学校321503795231331500学校33657978571432285学校3430094250271752623学校354238.738522.381301200学校361084683772002300学校3727282.2521614.042002400学校3840000130001511800学校3953333.3160002242132学校406000050000360200学校41100000300002001100学校42173333600004202552试将这些学校分类。

解：

运用方法：

将所有学校分为3类：

第一类：

（3,11,15,4,9,17,27,1,23,7,19,30,32,21,26,14,10,12,5,6,2,8,18,24,29,36,33,20,34,37,22,16,25,38,13，39,31）第二类：

（40,41）第三类：

（42）216种饮料的热量、咖啡因、钠及价格四种变量数据如下表：

饮料编号热量咖啡因钠价格1207.203.3015.502.80236.805.9012.903.30372.207.308.202.40436.70.4010.504.005121.704.109.203.50689.104.0010.203.307146.704.309.701.80857.602.2013.602.10995.90.008.501.3010199.0.0010.603.501149.808.006.303.701216.604.706.301.501338.503.707.702.0014.004.2013.102.2015118.804.707.204.1016107.00.008.304.20试将这些饮料分类。

解：

所有饮料分为4类：

（5，6，7，3，2，4，11）（8,14,12,13,9）

（1）（15,16,10）320种啤酒的成分和价格数据如下表：

beernamecaloriesodiumalcoholcostBudweiser144.0019.004.70.43Schlitz181.0019.004.90.43Ionenbrau157.0015.004.90.48Kronensourc170.007.005.20.73Heineken152.0011.005.00.77Old-milnaukee145.0023.004.60.26Aucsberger175.0024.005.50.40Strchs-bohemi149.0027.004.70.42Miller-lite99.0010.004.30.43Sudeiser-lich113.006.003.70.44Coors140.0016.004.60.44Coorslicht102.0015.004.10.46Michelos-lich135.0011.004.20.50Secrs150.0019.004.70.76Kkirin149.006.005.00.79Pabst-extra-l68.0015.002.30.36Hamms136.0019.004.40.43Heilemans-old144.0024.004.90.43Olympia-gold-72.006.002.90.46Schlite-light97.007.004.20.47试将这些啤酒分类。

解：

分为3类：

（1,17,11,3,2,7,8,18,6）（4,15,5,14）（9,20,10,12,13,16,19）450名学生参加10个测验项目的测试数据如下表：

学生编号X1X2X3X4X5X6X7X8X9X1018.25.73414622139922527.84.03614692230101913310.05.0371557242311211449.04.535143020261219958.04.030114317189121067.76.3331435223210191477.34.721143724199131288.76.0331457243412201498.24.72814291923101612108.24.53215332030121613117.23.5261164192081510127.75.726132925348128137.36.02812442419111410147.77.03114282417102013157.54.235134520149149167.74.23113311930111315177.74.03614502543918131810.54.53015512441122214197.56.83815592434102016207.75.22915652627101416217.25.519132019266125228.74.32211451736699237.24.7271349182391412248.35.325135218218134257.35.5221333142081412267.86.83213402634111816278.75.72515402126101215286.35.522155924105117298.55.815146021178137307.75.23013572325111613318.85.3251443192091512326.34.82714411825101913337.86.7271437223081416348.85.227146115217159356.33.5231517181491115368.25.7271445163091614378.86.72815732441101014388.03.8241230171491312397.06.32916412024132115407.77.22616431921111516417.06.5231525151791210428.24.52715262137121414439.77.03415532631111616447.53.722131719239128459.36.22814402142121714469.37.32715752226122413477.55.02916492130121914487.55.5231528212171212498.74.234143920348137508.75.2271365202691916试将学生分类。

解：

*HIERARCHICALCLUSTERANALYSIS*DendrogramusingWardMethodRescaledDistanceClusterCombineCASE0510152025LabelNum+-+-+-+-+-+Case99Case1010Case1616Case44Case33Case1818Case22Case1717Case11Case66Case1414Case88Case1919Case2626Case1313Case3030Case2020Case2727Case3131Case3636Case3333Case5050Case4242Case4545Case4949Case3939Case4747Case4040Case3737Case4343Case4646Case3838Case4444Case3535Case2323Case3232Case2525Case4141Case4848Case55Case1111Case1515Case2424Case3434Case2929Case2222Case77Case1212Case2121Case2828分为两类：

（9,10,16,4,3,18,2,17,1,6,14,8,19,26,13,30,30,20,27,31,36,33,50,42,45,49,39,47,40,37,43,46）（38,44,35,23,32,25,41,48,5,11,15,24,34,29,22,7,12,21,28）5下表列出了2007年我国31个省、市、自治区和直辖市的城镇居民家庭平均每人全年消费性支出的8个主要变量数据。

利用系统聚类法，对各地区进行聚类分析：

地区食品衣着居住家庭设备用品及服务医疗保健交通和通信教育文化娱乐服务杂项商品和服务北京4934.051512.881246.19981.131294.072328.512383.96649.66天津4249.311024.151417.45760.561163.981309.941639.83463.64河北2789.85975.94917.19546.75833.511010.51895.06266.16山西2600.371064.61991.77477.74640.221027.991054.05245.07内蒙古2824.891396.86941.79561.71719.131123.821245.09468.17辽宁3560.211017.651047.04439.28879.081033.361052.94400.16吉林2842.681127.091062.46407.35854.8873.88997.75394.29黑龙江2633.181021.45784.51355.67729.55746.03938.21310.67上海6125.451330.051412.1959.49857.113153.722653.67763.8江苏3928.71990.031020.09707.31689.371303.021699.26377.37浙江4892.581406.21168.08666.02859.062473.42158.32467.52安徽3384.38906.47850.24465.68554.44891.381169.99309.3福建4296.22940.721261.18645.4502.411606.91426.34375.98江西3192.61915.09728.76587.4385.91732.97973.38294.6山东3180.641238.341027.58661.03708.581333.631191.18325.64河南2707.441053.13795.39549.14626.55858.33936.55300.19湖北3455.981046.62856.97550.16525.32903.021120.29242.82湖南3243.881017.59869.59603.18668.53986.891285.24315.82广东5056.68814.571444.91853.18752.522966.081994.86454.09广西3398.09656.69803.04491.03542.07932.871050.04277.43海南3546.67452.85819.02519.99503.781401.89837.83210.85重庆3674.281171.15968.45706.77749.511118.791237.35264.01四川3580.14949.74690.27562.02511.781074.911031.81291.32贵州3122.46910.3718.65463.56354.52895.041035.96258.21云南3562.33859.65673.07280.62631.71034.71705.51174.23西藏3836.51880.1628.35271.29272.81866.33441.02335.66陕西3063.69910.29831.27513.08678.38866.761230.74332.84甘肃2824.42939.89768.28505.16564.25861.471058.66353.65青海2803.45898.54641.93484.71613.24785.27953.87331.38宁夏2760.74994.47910.68480.84645.98859.04863.36302.17新疆2760.691183.69736.99475.23598.78890.3896.79331.8试将这些地区分类解：

选择方法如下：

聚类分析树状图如下：

将所有地区分为2类：

第一类：

（28,29,12,27,18,14,24,17,23,6,7,8,31,16,30,4,3,20,21,25,26,10,13,15,22,5）第二类：

（1,9,11,19,2）6在全国服装标准制定中，对某地区成年女子的14个部位尺寸（体型尺寸）进行了测量，根据测量数据计算得到14个部位尺寸之间的相关系数矩阵，如下表所示，：

试对14个变量进行聚类分析：

x1x2x3x4x5x6x7x8x9x10x11x12x131上体长12手臂长0.36613胸围0.2420.23314颈围0.280.1940.5915总肩宽0.360.3240.4760.43516前胸宽0.2820.2630.4830.470.45217后背宽0.2450.2650.540.4780.5350.66318前腰0.4480.3450.4520.4040.4310.3220.2661节高9后腰节高0.4860.3670.3650.3570.4290.2830.2870.82110总体长0.6480.6620.2160.3160.4290.2830.2630.5270.547111身高0.6790.6810.2430.3130.430.3020.2940.520.5580.957112下体长0.4860.6360.1740.2430.3750.290.2550.4030.4170.8570.582113腰围0.1330.1530.7320.4770.3390.3920.4460.2660.2410.0540.0990.055114臀围0.3760.2520.6760.5810.4410.4470.440.4240.3720.3630.3760.3210.627解：

*HIERARCHICALCLUSTERANALYSIS*DendrogramusingWardMethodRescaledDistanceClusterCombineCASE0510152025LabelNum+-+-+-+-+-+x99x1010x55x1111x1212x11x22x88x44x66x77x1313x33分为2类：

（9,10,5,11,12,1,2,8,4,6,）（7,13,3）7.下表列出了2006年我国31个省、市、自治区和直辖市的12个月的月平均气温数据。

数据来源：

中华人民共和国国家统计局网站，现利用聚类法，对各地区进行聚类分析。

主要城市平均气温（2006年）单位：

摄氏度城市1月2月3月4月5月6月7月8月9月10月11月12月北京-1.9-0.98.013.520.425.925.926.421.816.16.7-1.0天津-2.7-1.47.513.220.326.425.926.421.316.26.5-1.7石家庄-0.91.610.315.121.327.427.025.921.817.88.00.4太原-3.6-0.46.814.519.123.225.723.117.413.44.4-2.5呼和浩特-9.2-7.02.210.317.421.824.522.016.311.51.3-7.7沈阳-12.7-8.10.58.018.321.624.224.317.511.60.8-6.7长春-14.5-10.6-1.36.117.020.223.523.317.19.6-2.3-9.3哈尔滨-17.7-12.6-2.85.917.119.923.423.116.27.4-4.5-12.1上海5.75.611.116.620.825.629.430.223.922.115.78.2南京3.94.311.317.121.226.528.729.522.520.312.85.2杭州5.86.112.418.321.525.930.130.623.321.915.17.7合肥3.44.511.717.221.726.728.829.022.220.412.85.0福州12.512.514.019.422.326.529.429.025.924.419.814.1南昌6.66.512.719.322.726.030.030.024.322.115.08.1济南0.02.110.216.521.526.927.426.021.419.510.01.6郑州0.33.911.517.121.827.827.126.121.219.010.83.0武汉4.25.812.819.023.928.430.229.724.021.014.06.8长沙5.36.212.519.923.627.030.129.524.021.314.77.8广州15.817.317.923.625.327.829.829.427.026.421.916.0南宁14.314.317.523.925.227.628.027.225.725.620.414.0海口18.520.521.826.728.329.430.028.527.427.125.320.8重庆7.89.013.319.222.925.431.032.424.820.614.69.4温州5.87.512.117.921.624.026.926.620.919.013.36.9贵阳4.35.410.217.018.921.123.823.220.516.711.25.8昆明10.813.215.918.018.020.421.320.618.316.9