Naive Bayes (simple) Class >50K: P(C) = 0.24893913 Attribute age Mean: 43.95911028 Standard Deviation: 10.26963284 Attribute workclass Private 0.64888238 Self-emp-not-inc 0.09513039 Self-emp-inc 0.07996275 Federal-gov 0.04869611 Local-gov 0.08116019 State-gov 0.04590208 Without-pay 0.00013305 Never-worked 0.00013305 Attribute fnlwgt Mean: 188149.96217368 Standard Deviation: 102821.73711374 Attribute education Bachelors 0.28269537 Some-college 0.17769803 11th 0.00797448 HS-grad 0.21504519 Prof-school 0.05409357 Assoc-acdm 0.03415736 Assoc-voc 0.04585327 9th 0.00345561 7th-8th 0.00478469 12th 0.00398724 Masters 0.12214248 1st-4th 0.00093036 10th 0.00797448 Doctorate 0.03734716 5th-6th 0.0017278 Preschool 0.00013291 Attribute education-num Mean: 11.60641982 Standard Deviation: 2.36842295 Attribute marital-status Married-civ-spouse 0.85163007 Divorced 0.06027944 Never-married 0.06267465 Separated 0.0089155 Widowed 0.01077844 Married-spouse-absent 0.00425815 Married-AF-spouse 0.00146374 Attribute occupation Tech-support 0.0370912 Craft-repair 0.12084552 Other-service 0.01768147 Sales 0.12908801 Exec-managerial 0.25764424 Prof-specialty 0.24089338 Handlers-cleaners 0.01116724 Machine-op-inspct 0.03270407 Adm-clerical 0.06633874 Farming-fishing 0.01542143 Transport-moving 0.04254188 Priv-house-serv 0.00026589 Protective-serv 0.02805105 Armed-Forces 0.00026589 Attribute relationship Wife 0.09249401 Own-child 0.00865052 Husband 0.75592228 Not-in-family 0.10966196 Other-relative 0.00479106 Unmarried 0.02848017 Attribute race White 0.91042194 Asian-Pac-Islander 0.03314255 Amer-Indian-Eskimo 0.00465859 Other 0.00292826 Black 0.04884866 Attribute sex Female 0.1482024 Male 0.8517976 Attribute capital-gain Mean: 3937.6798082 Standard Deviation: 14386.06001898 Attribute capital-loss Mean: 193.75066596 Standard Deviation: 592.82558964 Attribute hours-per-week Mean: 45.70657965 Standard Deviation: 10.73698663 Attribute native-country United-States 0.92674526 Cambodia 0.00105974 England 0.0041065 Puerto-Rico 0.00172208 Canada 0.00490131 Germany 0.00596105 Outlying-US(Guam-USVI-etc) 0.00013247 India 0.00543118 Japan 0.00317923 Greece 0.00119221 South 0.00198702 China 0.00278183 Cuba 0.00344416 Iran 0.00251689 Honduras 0.00026494 Philippines 0.00808054 Italy 0.0033117 Poland 0.00158961 Jamaica 0.00145715 Vietnam 0.00079481 Mexico 0.00450391 Portugal 0.00066234 Ireland 0.00079481 France 0.00172208 Dominican-Republic 0.0003974 Laos 0.0003974 Ecuador 0.00066234 Taiwan 0.00264936 Haiti 0.00066234 Columbia 0.0003974 Hungary 0.00052987 Guatemala 0.00052987 Nicaragua 0.0003974 Scotland 0.0003974 Thailand 0.00052987 Yugoslavia 0.00092728 El-Salvador 0.00132468 Trinadad&Tobago 0.0003974 Peru 0.0003974 Hong 0.00092728 Holand-Netherlands 0.00013247 Class <=50K: P(C) = 0.75106087 Attribute age Mean: 36.60806039 Standard Deviation: 13.46463126 Attribute workclass Private 0.76829053 Self-emp-not-inc 0.07881034 Self-emp-inc 0.0209602 Federal-gov 0.02554938 Local-gov 0.0643809 State-gov 0.04130262 Without-pay 0.0006619 Never-worked 0.00004413 Attribute fnlwgt Mean: 190338.64672905 Standard Deviation: 106571.34300494 Attribute education Bachelors 0.12876048 Some-college 0.23568593 11th 0.04367005 HS-grad 0.36277018 Prof-school 0.00604323 Assoc-acdm 0.0332157 Assoc-voc 0.04252316 9th 0.01901191 7th-8th 0.02307014 12th 0.01539479 Masters 0.03131892 1st-4th 0.00644023 10th 0.0336127 Doctorate 0.00423467 5th-6th 0.01221879 Preschool 0.00202911 Attribute education-num Mean: 9.62911627 Standard Deviation: 2.41359613 Attribute marital-status Married-civ-spouse 0.33833458 Divorced 0.16605622 Never-married 0.40849918 Separated 0.03856847 Widowed 0.03300825 Married-spouse-absent 0.01500375 Married-AF-spouse 0.00052954 Attribute occupation Tech-support 0.02801306 Craft-repair 0.13777131 Other-service 0.13591848 Sales 0.11536086 Exec-managerial 0.09070055 Prof-specialty 0.09828834 Handlers-cleaners 0.05593789 Machine-op-inspct 0.07596612 Adm-clerical 0.14222693 Farming-fishing 0.03860067 Transport-moving 0.05532028 Priv-house-serv 0.00630845 Protective-serv 0.01919005 Armed-Forces 0.00039704 Attribute relationship Wife 0.03146514 Own-child 0.19430715 Husband 0.2994263 Not-in-family 0.30467785 Other-relative 0.03773169 Unmarried 0.13239188 Attribute race White 0.84271151 Asian-Pac-Islander 0.02859791 Amer-Indian-Eskimo 0.01116554 Other 0.00931197 Black 0.10821307 Attribute sex Female 0.38272422 Male 0.61727578 Attribute capital-gain Mean: 148.89383773 Standard Deviation: 936.39227955 Attribute capital-loss Mean: 53.44800035 Standard Deviation: 310.27026297 Attribute hours-per-week Mean: 39.34859186 Standard Deviation: 11.95077414 Attribute native-country United-States 0.90372329 Cambodia 0.00052875 England 0.00251157 Puerto-Rico 0.00431813 Canada 0.0031725 Germany 0.00374532 Outlying-US(Guam-USVI-etc) 0.00066094 India 0.00268782 Japan 0.00163032 Greece 0.00096938 South 0.00255563 China 0.00215907 Cuba 0.00299625 Iran 0.00110156 Honduras 0.00052875 Philippines 0.00568407 Italy 0.00198282 Poland 0.00202688 Jamaica 0.00312844 Vietnam 0.00264375 Mexico 0.02546816 Portugal 0.00136594 Ireland 0.00088125 France 0.000705 Dominican-Republic 0.00290813 Laos 0.000705 Ecuador 0.0010575 Taiwan 0.0010575 Haiti 0.00171844 Columbia 0.00242344 Hungary 0.00048469 Guatemala 0.00268782 Nicaragua 0.00141 Scotland 0.00044063 Thailand 0.00066094 Yugoslavia 0.00048469 El-Salvador 0.00405376 Trinadad&Tobago 0.00074906 Peru 0.00127781 Hong 0.00061688 Holand-Netherlands 0.00008813 === Error on training data === Correctly Classified Instances 24956 82.7399 % Incorrectly Classified Instances 5206 17.2601 % Mean absolute error 0.1795 Root mean squared error 0.38 Relative absolute error 48.0074 % Root relative squared error 87.8843 % Total Number of Instances 30162 === Confusion Matrix === a b <-- classified as 3854 3654 | a = >50K 1552 21102 | b = <=50K === Error on test data === Correctly Classified Instances 12411 82.4104 % Incorrectly Classified Instances 2649 17.5896 % Mean absolute error 0.1818 Root mean squared error 0.3827 Relative absolute error 48.8345 % Root relative squared error 88.8846 % Total Number of Instances 15060 === Confusion Matrix === a b <-- classified as 1867 1833 | a = >50K 816 10544 | b = <=50K