Assignment 2: Data Mining Madness

Kelompok 10:
1. Fathatia Cholida Syifa (5003231070)
2. Suhanur Haliza (5003231078)
3. Diazty Ifta Padvani (5003231171)
4. Pingky Oktania Tata Sefina (5003231184)

library(dplyr)
library(lubridate)
library(ggplot2)
library(tidyr)
library(scales)
library(forecast)

Filtering & Cleaning Data

WiFi Data

wifi=read.csv('C:/Users/Lenovo/Downloads/wifi.csv')
head(wifi)

colnames(wifi)

[1] "time"                       "Event.Time"                 "Associated.Client.Count"    "Authenticated.Client.Count"
[5] "Uni"                        "Building"                   "Floor"                     

str(wifi)

'data.frame':   1883844 obs. of  7 variables:
 $ time                      : chr  "2020-02-01 00:02:12" "2020-02-01 00:02:12" "2020-02-01 00:02:12" "2020-02-01 00:02:12" ...
 $ Event.Time                : chr  "Sat Feb 01 00:02:12 UTC 2020" "Sat Feb 01 00:02:12 UTC 2020" "Sat Feb 01 00:02:12 UTC 2020" "Sat Feb 01 00:02:12 UTC 2020" ...
 $ Associated.Client.Count   : int  184 6 18 23 45 16 32 6 73 5 ...
 $ Authenticated.Client.Count: int  182 6 18 23 45 16 32 6 73 5 ...
 $ Uni                       : chr  "Lancaster University " "Lancaster University " "Lancaster University " "Lancaster University " ...
 $ Building                  : chr  " Graduate College " " Management School " " SW hse 32-33 " " SW hse 29 " ...
 $ Floor                     : chr  " A Floor" " C Floor" " D Floor" " B floor" ...

wifi$time=as.POSIXct(wifi$time)
str(wifi)

'data.frame':   1883844 obs. of  7 variables:
 $ time                      : POSIXct, format: "2020-02-01 00:02:12" "2020-02-01 00:02:12" "2020-02-01 00:02:12" "2020-02-01 00:02:12" ...
 $ Event.Time                : chr  "Sat Feb 01 00:02:12 UTC 2020" "Sat Feb 01 00:02:12 UTC 2020" "Sat Feb 01 00:02:12 UTC 2020" "Sat Feb 01 00:02:12 UTC 2020" ...
 $ Associated.Client.Count   : int  184 6 18 23 45 16 32 6 73 5 ...
 $ Authenticated.Client.Count: int  182 6 18 23 45 16 32 6 73 5 ...
 $ Uni                       : chr  "Lancaster University " "Lancaster University " "Lancaster University " "Lancaster University " ...
 $ Building                  : chr  " Graduate College " " Management School " " SW hse 32-33 " " SW hse 29 " ...
 $ Floor                     : chr  " A Floor" " C Floor" " D Floor" " B floor" ...

unique(wifi$Building)

 [1] " Graduate College "                 " Management School "                " SW hse 32-33 "                    
 [4] " SW hse 29 "                        " Furness outer "                    " Slaidburn House (LUSU) "          
 [7] " SW hse 34 "                        " Bowland hall "                     " Fylde "                           
[10] " SW hse 55-56 "                     " SW hse 40-42 "                     " Bowland Twr (Old Bowland Annexe) "
[13] " Pendle "                           " Faraday and cTAP "                 " Institute for Advanced Studies "  
[16] " University House "                 " SW hse 53-54 "                     " Engineering "                     
[19] " Field Station "                    " SW hse 36 "                        " Infolab "                         
[22] " SW hse 21-23 "                     " LICA "                             " FU Hse 71-74 "                    
[25] " Grizedale "                        " Charles Carter "                   " Furness "                         
[28] " SW hse 43-45 "                     " SW hse 12-16 "                     " Bowland Main "                    
[31] " SW hse 35 "                        " Human Resources "                  " County "                          
[34] " FY Hse 65-70 "                     " Barker House Farm "                " SW hse 27-28 "                    
[37] " Bowland Annexe "                   " John Creed "                       " grize-res "                       
[40] " SW hse 24-26 "                     " SW hse 20 "                        " Ruskin Library "                  
[43] " County South/Cartmel "             " Library "                          " Conference Centre "               
[46] " Postgrad Stats (PSC) "             " Bowland North "                    " George Fox "                      
[49] " Hse 75 77 "                        " Bailrigg House "                   " Sports Centre "                   
[52] " SW hse 30-31 "                     " Bowland Ash "                      " Alex Square "                     
[55] " LEC "                              " MDC "                              " ISS Building "                    
[58] " Chaplaincy Centre "                " CETAD "                            " SW hse 50-52 "                    
[61] " SW hse 17-19 "                     " Central Workshops "                " SW hse 46-49 "                    
[64] " SW hse 39 "                        " Science and technology "           " Whewell "                         
[67] " SW hse 37-38 "                     " Lonsdale College (SW) "            " Physics "                         
[70] " County Field "                     " Great Hall "                       " SW hse-158-179 "                  
[73] " Preschool "                        " Reception "                        " Hazelrigg "                       
[76] " Energy Centre "                   

wifi=wifi[wifi$Building == " Library ",]
head(wifi)

wifi=wifi[,c('time','Associated.Client.Count')]
head(wifi)

wifi=wifi %>%
  mutate(time = floor_date(time, "10 minutes")) %>%
  group_by(time) %>%
  summarise(
    occupancy = mean(`Associated.Client.Count`, na.rm = TRUE),
    .groups = "drop"
  )
head(wifi)

colSums(is.na(wifi))

     time occupancy 
        0         0 

sum(duplicated(wifi))

[1] 0

dim(wifi)

[1] 3683    2

Library Energy Data

library1=read.csv('C:/Users/Lenovo/Downloads/library1.csv')
library2=read.csv('C:/Users/Lenovo/Downloads/library2.csv')
library3=read.csv('C:/Users/Lenovo/Downloads/library3.csv')

head(library1)

str(library1)

'data.frame':   18864 obs. of  6 variables:
 $ ts        : chr  "2020-01-01 00:00:00" "2020-01-01 00:10:00" "2020-01-01 00:20:00" "2020-01-01 00:30:00" ...
 $ name      : chr  "MC065-L01/M9R2048" "MC065-L01/M9R2048" "MC065-L01/M9R2048" "MC065-L01/M9R2048" ...
 $ reading   : num  1489442 1489449 1489456 1489464 1489471 ...
 $ units     : chr  "KWh" "KWh" "KWh" "KWh" ...
 $ cumulative: num  1489442 1489449 1489456 1489464 1489471 ...
 $ rate      : num  NA 7 7 8 7 8 7 8 7 8 ...

library1$ts=as.POSIXct(library1$ts)
library2$ts=as.POSIXct(library2$ts)
library3$ts=as.POSIXct(library3$ts)

str(library1)

'data.frame':   18864 obs. of  6 variables:
 $ ts        : POSIXct, format: "2020-01-01 00:00:00" "2020-01-01 00:10:00" "2020-01-01 00:20:00" "2020-01-01 00:30:00" ...
 $ name      : chr  "MC065-L01/M9R2048" "MC065-L01/M9R2048" "MC065-L01/M9R2048" "MC065-L01/M9R2048" ...
 $ reading   : num  1489442 1489449 1489456 1489464 1489471 ...
 $ units     : chr  "KWh" "KWh" "KWh" "KWh" ...
 $ cumulative: num  1489442 1489449 1489456 1489464 1489471 ...
 $ rate      : num  NA 7 7 8 7 8 7 8 7 8 ...

colSums(is.na(library1))

        ts       name    reading      units cumulative       rate 
         0          0       3041          0       3041       3047 

colSums(is.na(library2))

        ts       name    reading      units cumulative       rate 
         0          0       3041          0       3041       3047 

colSums(is.na(library3))

        ts       name    reading      units cumulative       rate 
         0          0       3041          0       3041       3047 

library1$rate[is.na(library1$rate)]=mean(library1$rate[1:144],na.rm = TRUE)
library2$rate[is.na(library2$rate)]=mean(library2$rate[1:144],na.rm = TRUE)
library3$rate[is.na(library3$rate)]=mean(library3$rate[1:144],na.rm = TRUE)

library_all=merge(library1,library2,by='ts',all = TRUE)
library_all=merge(library_all,library3,by='ts',all = TRUE)
head(library_all)

library_all$energy_consumption<-library_all$rate.x + library_all$rate.y + library_all$rate
head(library_all)

names(library_all)[names(library_all)=='ts']='time'

sum(duplicated(library_all))

[1] 0

Data Integration

df=merge(wifi,library_all,by='time',all.x = TRUE)
df=df[,c('time','occupancy','energy_consumption')]
head(df)

dim(df)

[1] 3683    3

Visualization

Time Series Plot: Occupancy & Energy Consumption

ggplot(df, aes(x = time)) +
  geom_line(aes(y = occupancy, color = "Occupancy")) +
  geom_line(aes(y = energy_consumption, color = "Energy Consumption")) +
  scale_color_manual(values = c("Occupancy" = "blue", "Energy Consumption" = "red")) +
  labs(title = "Time Series of Occupancy and Energy Consumption",
       x = "Time",
       y = "Value",
       color = "Legend") +
  theme_minimal() +
  theme(legend.position = "bottom",plot.title = element_text(hjust = 0.5))

Scatter Plot: Occupancy vs Energy Consumption

ggplot(df, aes(x = occupancy, y = energy_consumption)) +
  geom_point(alpha = 0.5, color = "blue") +
  labs(title = "Scatter Plot: Occupancy vs Energy Consumption",
       x = "Occupancy (Number of Clients)",
       y = "Energy Consumption (kWh)") +
  theme_minimal() +
  theme(plot.title = element_text(hjust = 0.5))

Daily Profiles (24h)

Occupancy

df = df %>%
  mutate(
    date = as.Date(time),
    hour = hour(time) + minute(time)/60
  )
ggplot(df, aes(x = hour, y = occupancy, group = date)) +
  geom_line(alpha = 0.1, color = "blue") +
  geom_line(
    data = df %>% group_by(hour) %>% summarise(avg_occ = mean(occupancy, na.rm=TRUE)),
    aes(x = hour, y = avg_occ, group = 1),
    color = "blue", size = 1.2
  ) +
  labs(
    title = "Daily Profiles of Occupancy",
    x = "Hour of Day",
    y = "Occupancy"
  ) +
  scale_x_continuous(breaks = seq(0, 24, 1)) +
  theme_minimal() +
  theme(plot.title = element_text(hjust = 0.5))

df = df %>%
  mutate(
    date = as.Date(time),
    hour_decimal = hour(time) + minute(time)/60
  )
ggplot(df, aes(x = hour_decimal, y = occupancy, group = date)) +
  geom_line(alpha = 0.1, color = "blue") +
  geom_line(
    data = df %>% group_by(hour_decimal) %>% summarise(avg_occ = mean(occupancy, na.rm=TRUE)),
    aes(x = hour_decimal, y = avg_occ, group = 1),
    color = "blue", size = 1.2
  ) +
  labs(
    title = "Daily Profiles of Occupancy",
    x = "Time of Day",
    y = "Occupancy"
  ) +
  scale_x_continuous(
    breaks = seq(0, 24, by = 1/6),         
    labels = function(x) sprintf("%02d:%02d", floor(x), round((x %% 1) * 60))
  ) +
  theme_minimal() +
  theme(plot.title = element_text(hjust = 0.5,size=30),axis.text.x = element_text(angle = 90, hjust = 1, size = 15))

Energy Consumption

ggplot(df, aes(x = hour, y = energy_consumption, group = date)) +
  geom_line(alpha = 0.1, color = "red") +
  geom_line(
    data = df %>% group_by(hour) %>% summarise(avg_energy = mean(energy_consumption, na.rm=TRUE)),
    aes(x = hour, y = avg_energy, group = 1),
    color = "red", size = 1.2
  ) +
  labs(
    title = "Daily Profiles of Energy Consumption",
    x = "Hour of Day",
    y = "Energy Consumption"
  ) +
  scale_x_continuous(breaks = seq(0, 24, 1)) +
  theme_minimal()+
  theme(plot.title = element_text(hjust = 0.5))

ggplot(df, aes(x = hour_decimal, y = energy_consumption, group = date)) +
  geom_line(alpha = 0.1, color = "red") +
  geom_line(
    data = df %>% group_by(hour_decimal) %>% summarise(avg_energy = mean(energy_consumption, na.rm=TRUE)),
    aes(x = hour_decimal, y = avg_energy, group = 1),
    color = "red", size = 1.2
  ) +
  labs(
    title = "Daily Profiles of Energy Consumption",
    x = "Time of Day",
    y = "Energy Consumption"
  ) +
  scale_x_continuous(
    breaks = seq(0, 24, by = 1/6),        
    labels = function(x) sprintf("%02d:%02d", floor(x), round((x %% 1) * 60))
  ) +
  theme_minimal() +
  theme(plot.title = element_text(hjust = 0.5,size=30),axis.text.x = element_text(angle = 90, hjust = 1, size = 15))

Analysis

Peak Hour Occupancy

df_avg_hour <- df %>%
  group_by(hour_label = format(time, "%H:00")) %>% 
  summarise(mean_occupancy = base::mean(occupancy, na.rm = TRUE), .groups = "drop") %>%
arrange(hour_label)
peak_hours <- dplyr::arrange(df_avg_hour, desc(mean_occupancy))
head(peak_hours)

ggplot(df_avg_hour, aes(x = hour_label, y = mean_occupancy, group = 1)) +
  geom_line(color = "blue", linewidth = 1) +
  geom_point(color = "red", size = 3) +
  labs(
    title = "Rata-rata Occupancy per Jam",
    x = "Jam (00:00–23:00)",
    y = "Rata-rata Occupancy"
  ) +
  theme_minimal(base_size = 14) +
  theme(axis.text.x = element_text(angle = 45, hjust = 1),
        plot.title = element_text(hjust = 0.5))

Correlation

corr=cor(electricity$mean_client_count, electricity$total_electricity)
corr

[1] 0.8782117

Time Series Regression

df_reg <- df %>%
  arrange(time) %>%
  mutate(t = row_number(),
         hour = factor(format(time, "%H")))
model_tslm <- lm(energy_consumption ~ occupancy + t + hour, data = df_reg)
summary(model_tslm)

Call:
lm(formula = energy_consumption ~ occupancy + t + hour, data = df_reg)

Residuals:
    Min      1Q  Median      3Q     Max 
-55.416  -8.208   0.084   8.520  55.706 

Coefficients:
              Estimate Std. Error t value Pr(>|t|)    
(Intercept)  1.162e+02  1.097e+00 105.958  < 2e-16 ***
occupancy    1.763e-01  3.346e-03  52.688  < 2e-16 ***
t            9.812e-04  2.033e-04   4.825 1.46e-06 ***
hour01      -7.731e+00  1.463e+00  -5.286 1.32e-07 ***
hour02      -1.517e+01  1.463e+00 -10.367  < 2e-16 ***
hour03      -2.251e+01  1.465e+00 -15.368  < 2e-16 ***
hour04      -3.260e+01  1.465e+00 -22.263  < 2e-16 ***
hour05      -4.356e+01  1.465e+00 -29.738  < 2e-16 ***
hour06      -4.390e+01  1.466e+00 -29.952  < 2e-16 ***
hour07      -3.567e+01  1.465e+00 -24.346  < 2e-16 ***
hour08      -1.530e+01  1.462e+00 -10.468  < 2e-16 ***
hour09      -1.983e+00  1.468e+00  -1.351    0.177    
hour10       7.604e+00  1.538e+00   4.945 7.97e-07 ***
hour11       8.377e+00  1.637e+00   5.117 3.26e-07 ***
hour12       1.191e+01  1.700e+00   7.006 2.91e-12 ***
hour13       1.155e+01  1.758e+00   6.568 5.81e-11 ***
hour14       8.922e+00  1.819e+00   4.905 9.75e-07 ***
hour15       9.191e+00  1.845e+00   4.982 6.57e-07 ***
hour16       1.284e+01  1.796e+00   7.147 1.06e-12 ***
hour17       1.907e+01  1.700e+00  11.215  < 2e-16 ***
hour18       2.014e+01  1.611e+00  12.504  < 2e-16 ***
hour19       2.170e+01  1.562e+00  13.889  < 2e-16 ***
hour20       2.130e+01  1.530e+00  13.924  < 2e-16 ***
hour21       1.992e+01  1.509e+00  13.202  < 2e-16 ***
hour22       1.912e+01  1.490e+00  12.832  < 2e-16 ***
hour23       1.143e+01  1.479e+00   7.726 1.42e-14 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 12.91 on 3657 degrees of freedom
Multiple R-squared:  0.9133,    Adjusted R-squared:  0.9127 
F-statistic:  1541 on 25 and 3657 DF,  p-value: < 2.2e-16

Time Series Plot

df$fitted <- fitted(model_tslm)

ggplot(df, aes(x = time)) +
  geom_line(aes(y = energy_consumption, color = "Observed"), linewidth = 0.8, alpha = 0.7) +
  geom_line(aes(y = fitted, color = "Fitted"), linewidth = 1) +
  scale_color_manual(values = c("Observed" = "blue", "Fitted" = "red")) +
  labs(
    title = "Time Series Regression",
    x = "Time",
    y = "Energy Consumption",
    color = ""
  ) +
  theme_minimal(base_size = 14)+
  theme(legend.position = "bottom",plot.title = element_text(hjust = 0.5))

Weekend vs Weekday Comparison

# Separate data into weekdays and weekends
df$time <- as_datetime(df$time)
df$day_of_week <- wday(df$time, week_start = 1) - 1
df_weekday <- df %>% filter(day_of_week < 5)
df_weekend <- df %>% filter(day_of_week >= 5)

df_weekday$time_of_day <- hms::as_hms(df_weekday$time)
df_weekend$time_of_day <- hms::as_hms(df_weekend$time)

Time Series Plot for Weekdays

ggplot(df_weekday, aes(x = time)) +
  geom_line(aes(y = occupancy, color = "Occupancy")) +
  geom_line(aes(y = energy_consumption, color = "Energy Consumption")) +
  labs(title = "Time Series: Occupancy vs Energy Consumption (Weekdays)",
       x = "Time",
       y = "Value") +
  scale_color_manual(values = c("Occupancy" = "blue", "Energy Consumption" = "red")) +
  theme_minimal() +
  theme(legend.position = "bottom",plot.title = element_text(hjust = 0.5))

Time Series Plot for Weekends

ggplot(df_weekend, aes(x = time)) +
  geom_line(aes(y = occupancy, color = "Occupancy")) +
  geom_line(aes(y = energy_consumption, color = "Energy Consumption")) +
  labs(title = "Time Series: Occupancy vs Energy Consumption (Weekends)",
       x = "Time",
       y = "Value") +
  scale_color_manual(values = c("Occupancy" = "blue", "Energy Consumption" = "red")) +
  theme_minimal() +
  theme(legend.position = "bottom",plot.title = element_text(hjust = 0.5))

Time Series Plot for Weekdays and Weekends

ggplot(df, aes(x = time)) +
  geom_line(data = df_weekday, aes(y = occupancy, color = "Occupancy - Weekdays")) +
  geom_line(data = df_weekday, aes(y = energy_consumption, color = "Energy Consumption - Weekdays")) +
  geom_line(data = df_weekend, aes(y = occupancy, color = "Occupancy - Weekends"), linetype = "dashed") +
  geom_line(data = df_weekend, aes(y = energy_consumption, color = "Energy Consumption - Weekends"), linetype = "dashed") +
  labs(title = "Time Series: Occupancy vs Energy Consumption (Weekdays and Weekends)",
       x = "Time",
       y = "Value") +
  scale_color_manual(values = c("Occupancy - Weekdays" = "blue", "Energy Consumption - Weekdays" = "red",
                                "Occupancy - Weekends" = "cyan", "Energy Consumption - Weekends" = "magenta")) +
  theme_minimal() +
  theme(legend.position = "bottom", plot.title = element_text(hjust = 0.5))

Scatter Plot for Weekdays

ggplot(df_weekday, aes(x = occupancy, y = energy_consumption)) +
  geom_point(alpha = 0.4, color = "blue") +
  labs(title = "Scatter Plot: Occupancy vs Energy (Weekdays)",
       x = "Occupancy (WiFi Clients)",
       y = "Energy Consumption") +
  theme_minimal() +
  theme(plot.title = element_text(hjust = 0.5))

Scatter Plot for Weekends

ggplot(df_weekend, aes(x = occupancy, y = energy_consumption)) +
  geom_point(alpha = 0.4, color = "red") +
  labs(title = "Scatter Plot: Occupancy vs Energy (Weekends)",
       x = "Occupancy (WiFi Clients)",
       y = "Energy Consumption") +
  theme_minimal() + 
  theme(plot.title = element_text(hjust = 0.5))

Scatter Plot for Weekdays and Weekends

ggplot(df, aes(x = occupancy, y = energy_consumption, color = factor(day_of_week < 5))) +
  geom_point(alpha = 0.4) +
  scale_color_manual(values = c("TRUE" = "blue", "FALSE" = "red"), labels = c("Weekdays", "Weekends")) +
  labs(title = "Scatter Plot: Occupancy vs Energy (Weekdays and Weekends)",
       x = "Occupancy (WiFi Clients)",
       y = "Energy Consumption",
       color = "Day Type") +
  theme_minimal() +
  theme(legend.position = "bottom",plot.title = element_text(hjust = 0.5))

Daily Profiles (24h) - Occupancy for Weekdays

df_weekday$hour <- hour(df_weekday$time) + minute(df_weekday$time)/60.0
ggplot(df_weekday, aes(x = hour, y = occupancy, group = date(time))) +
  geom_line(alpha = 0.2, color = "blue") +
  stat_summary(aes(group = 1), fun = mean, geom = "line", color = "blue", size = 1.5, linetype = "solid") +
  labs(title = "Daily Profiles of Occupancy (Weekdays)",
       x = "Time of Day",
       y = "Occupancy (WiFi Clients)") +
  scale_x_continuous(
    breaks = seq(0, 24, by = 10/60),   # interval 10 menit
    labels = function(x) sprintf("%02d:%02d", floor(x), round((x%%1)*60))
  ) +
  theme_minimal() +
  theme(plot.title = element_text(hjust = 0.5,size=30),axis.text.x = element_text(angle = 90, hjust = 1, size = 15))

Daily Profiles (24h) - Occupancy for Weekends

df_weekend$hour <- hour(df_weekend$time) + minute(df_weekend$time)/60.0
ggplot(df_weekend, aes(x = hour, y = occupancy, group = date(time))) +
  geom_line(alpha = 0.2, color = "blue") +
  stat_summary(aes(group = 1), fun = mean, geom = "line", color = "blue", size = 1.5, linetype = "solid") +
  labs(title = "Daily Profiles of Occupancy (Weekends)",
       x = "Time of Day",
       y = "Occupancy (WiFi Clients)") +
    scale_x_continuous(
    breaks = seq(0, 24, by = 10/60),
    labels = function(x) sprintf("%02d:%02d", floor(x), round((x%%1)*60))
  ) +
  theme_minimal() +
  theme(plot.title = element_text(hjust = 0.5,size=30),axis.text.x = element_text(angle = 90, hjust = 1, size = 15))

Daily Profiles (24h) - Occupancy (Weekdays and Weekends)

df$hour <- hour(df$time) + minute(df$time)/60.0
ggplot(df, aes(x = hour, y = occupancy, group = date(time))) +
  geom_line(data = df_weekday, aes(color = "Weekdays"), alpha = 0.2) +
  stat_summary(data = df_weekday, aes(color = "Weekdays", group = 1), fun = mean, geom = "line", size = 1.5, linetype = "solid") +
  geom_line(data = df_weekend, aes(color = "Weekends"), alpha = 0.2, linetype = "dashed") +
  stat_summary(data = df_weekend, aes(color = "Weekends", group = 1), fun = mean, geom = "line", size = 1.5, linetype = "dashed") +
  labs(title = "Daily Profiles of Occupancy (Weekdays and Weekends)",
       x = "Time of Day",
       y = "Occupancy (WiFi Clients)") +
  scale_color_manual(values = c("Weekdays" = "red", "Weekends" = "blue")) +
  scale_x_continuous(
  breaks = seq(0, 24, by = 10/60),
  labels = function(x) sprintf("%02d:%02d", floor(x), round((x%%1)*60))
  ) +
  theme_minimal() +
  theme(legend.position = "bottom",plot.title = element_text(hjust = 0.5,size=30),
        axis.text.x = element_text(angle = 90, hjust = 1, size = 15))

Daily Profiles (24h) - Energy Consumption for Weekdays

ggplot(df_weekday, aes(x = hour, y = energy_consumption, group = date(time))) +
  geom_line(alpha = 0.2, color = "red") +
  stat_summary(aes(group = 1), fun = mean, geom = "line", color = "red", size = 1.5, linetype = "solid") +
  labs(title = "Daily Profiles of Energy Consumption (Weekdays)",
       x = "Time of Day",
       y = "Energy Consumption") +
  scale_x_continuous(
    breaks = seq(0, 24, by = 10/60),   # interval 10 menit
    labels = function(x) sprintf("%02d:%02d", floor(x), round((x%%1)*60))
  ) +
  theme_minimal() +
  theme(plot.title = element_text(hjust = 0.5,size=30),
        axis.text.x = element_text(angle = 90, hjust = 1, size = 15))

Daily Profiles (24h) - Energy Consumption for Weekends

ggplot(df_weekend, aes(x = hour, y = energy_consumption, group = date(time))) +
  geom_line(alpha = 0.2, color = "red") +
  stat_summary(aes(group = 1), fun = mean, geom = "line", color = "red", size = 1.5, linetype = "solid") +
  labs(title = "Daily Profiles of Energy Consumption (Weekends)",
       x = "Time of Day",
       y = "Energy Consumption") +
  scale_x_continuous(
    breaks = seq(0, 24, by = 10/60),   # interval 10 menit
    labels = function(x) sprintf("%02d:%02d", floor(x), round((x%%1)*60))
  ) +
  theme_minimal() +
  theme(plot.title = element_text(hjust = 0.5,size=30),
        axis.text.x = element_text(angle = 90, hjust = 1, size = 15))

Daily Profiles (24h) - Energy Consumption (Weekdays and Weekends)

ggplot(df, aes(x = hour, y = energy_consumption, group = date(time))) +
  geom_line(data = df_weekday, aes(color = "Weekdays"), alpha = 0.2) +
  stat_summary(data = df_weekday, aes(color = "Weekdays", group = 1), fun = mean, geom = "line", size = 1.5, linetype = "solid") +
  geom_line(data = df_weekend, aes(color = "Weekends"), alpha = 0.2, linetype = "dashed") +
  stat_summary(data = df_weekend, aes(color = "Weekends", group = 1), fun = mean, geom = "line", size = 1.5, linetype = "dashed") +
  labs(title = "Daily Profiles of Energy Consumption (Weekdays and Weekends)",
       x = "Time of Day",
       y = "Energy Consumption") +
  scale_color_manual(values = c("Weekdays" = "red", "Weekends" = "blue")) +
  scale_x_continuous(
    breaks = seq(0, 24, by = 10/60),   # interval 10 menit
    labels = function(x) sprintf("%02d:%02d", floor(x), round((x%%1)*60))
  ) +
  theme_minimal()+
  theme(legend.position = "bottom",plot.title = element_text(hjust = 0.5,size=30),
        axis.text.x = element_text(angle = 90, hjust = 1, size = 15))