Tugas Business Intelligence
5006211074_Cliff Ganisson
2023-11-10
Dataset
## [1] "C:/Users/cliff/OneDrive/Documents/File Aktu ITS/Semester 5/Business Intelligence"## # A tibble: 7,043 × 16
## `Customer ID` `Tenure Months` Location `Device Class` `Games Product`
## <dbl> <dbl> <chr> <chr> <chr>
## 1 0 2 Jakarta Mid End Yes
## 2 1 2 Jakarta High End No
## 3 2 8 Jakarta High End No
## 4 3 28 Jakarta High End No
## 5 4 49 Jakarta High End No
## 6 5 10 Jakarta Mid End No
## 7 6 1 Jakarta Mid End No
## 8 7 1 Jakarta Low End No internet service
## 9 8 47 Jakarta High End No
## 10 9 1 Jakarta Mid End No
## # ℹ 7,033 more rows
## # ℹ 11 more variables: `Music Product` <chr>, `Education Product` <chr>,
## # `Call Center` <chr>, `Video Product` <chr>, `Use MyApp` <chr>,
## # `Payment Method` <chr>, `Monthly Purchase (Thou. IDR)` <dbl>,
## # `Churn Label` <chr>, Longitude <dbl>, Latitude <dbl>,
## # `CLTV (Predicted Thou. IDR)` <dbl>## Customer ID Tenure Months Location Device Class
## Min. : 0 Min. : 0.00 Length:7043 Length:7043
## 1st Qu.:1760 1st Qu.: 9.00 Class :character Class :character
## Median :3521 Median :29.00 Mode :character Mode :character
## Mean :3521 Mean :32.37
## 3rd Qu.:5282 3rd Qu.:55.00
## Max. :7042 Max. :72.00
## Games Product Music Product Education Product Call Center
## Length:7043 Length:7043 Length:7043 Length:7043
## Class :character Class :character Class :character Class :character
## Mode :character Mode :character Mode :character Mode :character
##
##
##
## Video Product Use MyApp Payment Method
## Length:7043 Length:7043 Length:7043
## Class :character Class :character Class :character
## Mode :character Mode :character Mode :character
##
##
##
## Monthly Purchase (Thou. IDR) Churn Label Longitude
## Min. : 23.73 Length:7043 Min. :106.8
## 1st Qu.: 46.15 Class :character 1st Qu.:106.8
## Median : 91.45 Mode :character Median :106.8
## Mean : 84.19 Mean :107.0
## 3rd Qu.:116.81 3rd Qu.:107.6
## Max. :154.38 Max. :107.6
## Latitude CLTV (Predicted Thou. IDR)
## Min. :-6.915 Min. :2604
## 1st Qu.:-6.915 1st Qu.:4510
## Median :-6.200 Median :5885
## Mean :-6.404 Mean :5720
## 3rd Qu.:-6.200 3rd Qu.:6995
## Max. :-6.200 Max. :8450Including Plots
Untuk menjelaskan secara lebih lanjut mengenai deskriptif data dapat dilihat pada visualisasi berikut
Analisis Korelasi
library(corrplot)## Warning: package 'corrplot' was built under R version 4.2.3## corrplot 0.92 loadednumdata<-data[,c(2,12,16)]
cormatrix<-cor(numdata)
corrplot(cormatrix,method="number")
Menggunakan analisis korelasi data numerik dapat dilihat bahwa variabel
numerik pada dataset memiliki hubunga satu sama lain yang relatif kecil
yaitu dibawah 0.5 dengan hubungan tertinggi antara data numerik adalah
antara CLTV dengan Tenure Months atau lamanya customer membeli paket
data atau kuota secara berapa bulan berturut-turut.
Regresi Linear
simpleregression<-lm(formula=`CLTV (Predicted Thou. IDR)`~`Tenure Months`,data=data)
summary(simpleregression)##
## Call:
## lm(formula = `CLTV (Predicted Thou. IDR)` ~ `Tenure Months`,
## data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3456.2 -1055.1 64.9 1145.8 2853.2
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4916.8068 27.8395 176.61 <2e-16 ***
## `Tenure Months` 24.8239 0.6852 36.23 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1412 on 7041 degrees of freedom
## Multiple R-squared: 0.1571, Adjusted R-squared: 0.157
## F-statistic: 1313 on 1 and 7041 DF, p-value: < 2.2e-16Berdasarkan hasil analisis regresi sederhana 1 variabel dengan variabel independen CLTV memberikan hasil kebaikan model 15.7%
multiregression<-lm(formula=`CLTV (Predicted Thou. IDR)`~`Tenure Months`+`Monthly Purchase (Thou. IDR)`,data=data)
summary(multiregression)##
## Call:
## lm(formula = `CLTV (Predicted Thou. IDR)` ~ `Tenure Months` +
## `Monthly Purchase (Thou. IDR)`, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3455.1 -1055.4 65.3 1145.6 2852.7
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.916e+03 4.220e+01 116.49 <2e-16 ***
## `Tenure Months` 2.482e+01 7.073e-01 35.09 <2e-16 ***
## `Monthly Purchase (Thou. IDR)` 1.777e-02 4.441e-01 0.04 0.968
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1412 on 7040 degrees of freedom
## Multiple R-squared: 0.1571, Adjusted R-squared: 0.1569
## F-statistic: 656.2 on 2 and 7040 DF, p-value: < 2.2e-16Berdasarkan hasil analisis regresi berganda 2 variabel dengan variabel independen CLTV memberikan hasil kebaikan model 15.69%.