|
|
Для Герасимова 2. Разработка модели потребительских предпочтений на основе данных рекомендательной сети epinions com
9 кластеров
Рисунок 0.72 – Диаграмма центра кластера 1 для случая 9 кластеров
[1] "Points: 292"
Call:
lm(formula = overall
durability + eou + battery + lag + picture,
data = x)
Residuals:
Min 1Q Median 3Q Max
-3.6826 -0.3656 0.0113 0.4600 1.5064
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -1.64418 0.28403 -5.789 1.86e-08 ***
durability 0.54801 0.05289 10.361 < 2e-16 ***
eou 0.32352 0.05498 5.885 1.11e-08 ***
battery 0.09232 0.03691 2.501 0.0129 *
lag 0.09927 0.05341 1.859 0.0641 .
picture 0.36993 0.05880 6.292 1.18e-09 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.649 on 286 degrees of freedom
(96 observations deleted due to missingness)
Multiple R-squared: 0.6382, Adjusted R-squared: 0.6319
F-statistic: 100.9 on 5 and 286 DF, p-value: < 2.2e-16
Рисунок 0.73 – Диаграмма центра кластера 1 для случая 9 кластеров
[1] "Points: 109"
Call:
lm(formula = overall
durability + eou + battery + lag + picture,
data = x)
Residuals:
Min 1Q Median 3Q Max
-2.18814 -0.30385 0.06365 0.42838 1.30765
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.39362 0.31776 -1.239 0.218254
durability 0.42599 0.09131 4.665 9.3e-06 ***
eou 0.17095 0.07016 2.436 0.016546 *
battery 0.05523 0.07017 0.787 0.433042
lag 0.15312 0.07786 1.967 0.051930 .
picture 0.34525 0.08879 3.888 0.000179 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.649 on 103 degrees of freedom
(38 observations deleted due to missingness)
Multiple R-squared: 0.6975, Adjusted R-squared: 0.6828
F-statistic: 47.5 on 5 and 103 DF, p-value: < 2.2e-16
Рисунок 0.74 – Диаграмма центра кластера 3 для случая 9 кластеров
[1] "Points: 61"
Call:
lm(formula = overall
durability + eou + battery + lag + picture,
data = x)
Residuals:
Min 1Q Median 3Q Max
-1.84565 -0.30467 0.09569 0.47755 1.40900
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.86732 0.79474 -1.091 0.27989
durability 0.42746 0.16225 2.635 0.01092 *
eou 0.05182 0.11367 0.456 0.65031
battery 0.05122 0.10627 0.482 0.63173
lag 0.40036 0.12624 3.172 0.00248 **
picture 0.30354 0.13728 2.211 0.03121 *
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.7142 on 55 degrees of freedom
(25 observations deleted due to missingness)
Multiple R-squared: 0.4871, Adjusted R-squared: 0.4404
F-statistic: 10.45 on 5 and 55 DF, p-value: 4.426e-07
Рисунок 0.75 – Диаграмма центра кластера 4 для случая 9 кластеров
[1] "Points: 531"
Call:
lm(formula = overall
durability + eou + battery + lag + picture,
data = x)
Residuals:
Min 1Q Median 3Q Max
-3.1700 -0.4314 0.0949 0.4702 1.8507
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.67901 0.22825 -2.975 0.00307 **
durability 0.38567 0.04523 8.526 < 2e-16 ***
eou 0.12597 0.03912 3.220 0.00136 **
battery 0.15909 0.03397 4.684 3.59e-06 ***
lag 0.12532 0.03979 3.150 0.00173 **
picture 0.38440 0.03756 10.234 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.6865 on 525 degrees of freedom
(164 observations deleted due to missingness)
Multiple R-squared: 0.5183, Adjusted R-squared: 0.5137
F-statistic: 113 on 5 and 525 DF, p-value: < 2.2e-16
Рисунок 0.76 – Диаграмма центра кластера 5 для случая 9 кластеров
[1] "Points: 131"
Call:
lm(formula = overall
durability + eou + battery + lag + picture,
data = x)
Residuals:
Min 1Q Median 3Q Max
-1.83714 -0.41416 0.00331 0.45375 1.37421
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -1.20732 0.38564 -3.131 0.002171 **
durability 0.39497 0.07785 5.073 1.38e-06 ***
eou 0.24620 0.06838 3.600 0.000457 ***
battery 0.13768 0.06297 2.186 0.030645 *
lag 0.01203 0.07005 0.172 0.863892
picture 0.54755 0.06993 7.830 1.80e-12 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.6626 on 125 degrees of freedom
(68 observations deleted due to missingness)
Multiple R-squared: 0.6841, Adjusted R-squared: 0.6715
F-statistic: 54.14 on 5 and 125 DF, p-value: < 2.2e-16
Рисунок 0.77 – Диаграмма центра кластера 6 для случая 9 кластеров
[1] "Points: 54"
Call:
lm(formula = overall
durability + eou + battery + lag + picture,
data = x)
Residuals:
Min 1Q Median 3Q Max
-1.20201 -0.31322 0.03151 0.35530 1.22153
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.66722 0.55820 -1.195 0.23784
durability 0.45886 0.11088 4.138 0.00014 ***
eou 0.23707 0.09986 2.374 0.02165 *
battery 0.17940 0.08914 2.013 0.04979 *
lag -0.02844 0.10292 -0.276 0.78351
picture 0.31115 0.09732 3.197 0.00246 **
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.5444 on 48 degrees of freedom
(40 observations deleted due to missingness)
Multiple R-squared: 0.6308, Adjusted R-squared: 0.5924
F-statistic: 16.41 on 5 and 48 DF, p-value: 2.031e-09
Рисунок 0.78 – Диаграмма центра кластера 7 для случая 9 кластеров
[1] "Points: 4"
Call:
lm(formula = overall
durability + eou + battery + lag + picture,
data = x)
Residuals:
ALL 4 residuals are 0: no residual degrees of freedom!
Coefficients: (2 not defined because of singularities)
Estimate Std. Error t value Pr(>|t|)
(Intercept) 11 NA NA NA
durability -4 NA NA NA
eou 1 NA NA NA
battery 2 NA NA NA
lag NA NA NA NA
picture NA NA NA NA
Residual standard error: NaN on 0 degrees of freedom
(3 observations deleted due to missingness)
Multiple R-squared: 1, Adjusted R-squared: NaN
F-statistic: NaN on 3 and 0 DF, p-value: NA
Рисунок 0.79 – Диаграмма центра кластера 8 для случая 9 кластеров
[1] "Points: 39"
Call:
lm(formula = overall
durability + eou + battery + lag + picture,
data = x)
Residuals:
Min 1Q Median 3Q Max
-1.21990 -0.23119 0.00792 0.26253 1.29068
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.83769 0.60675 -1.381 0.1767
durability 0.25423 0.17481 1.454 0.1553
eou 0.26912 0.13744 1.958 0.0587 .
battery 0.03395 0.08737 0.389 0.7001
lag 0.04569 0.12920 0.354 0.7259
picture 0.64567 0.10087 6.401 2.98e-07 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.5853 on 33 degrees of freedom
(24 observations deleted due to missingness)
Multiple R-squared: 0.789, Adjusted R-squared: 0.7571
F-statistic: 24.69 on 5 and 33 DF, p-value: 2.85e-10
Рисунок 0.80 – Диаграмма центра кластера 9 для случая 9 кластеров
[1] "Points: 38"
Call:
lm(formula = overall
durability + eou + battery + lag + picture,
data = x)
Residuals:
Min 1Q Median 3Q Max
-1.05039 -0.32361 -0.08176 0.26412 1.41148
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.20857 0.65349 0.319 0.751674
durability 0.07381 0.12453 0.593 0.557535
eou 0.28022 0.11554 2.425 0.021113 *
battery 0.04546 0.11826 0.384 0.703194
lag 0.15047 0.12411 1.212 0.234242
picture 0.43313 0.11609 3.731 0.000741 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.5776 on 32 degrees of freedom
(16 observations deleted due to missingness)
Multiple R-squared: 0.6762, Adjusted R-squared: 0.6256
F-statistic: 13.37 on 5 and 32 DF, p-value: 4.558e-07
Сравнение групповых предпочтений
Рисунок 0.81 – График предпочтений по автономной работе для 9 кластеров
Рисунок 0.82 – График предпочтений по надежности для 9 кластеров
Рисунок 0.83 – График предпочтений по простоте использования для 9 кластеров
Рисунок 0.84 – График предпочтений по скорости срабатывания для 9 кластеров
Рисунок 0.85 – График предпочтений по качеству снимка для 9 кластеров |
|
|
Скачать 2.87 Mb.