MonitorPassos/MonitorPassos.R at master · leonardo78rs/MonitorPassos · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
# Pesquisa Reproduzível: Analisando um monitor de passos de uma pessoa.
### Carregando e pré-processando os dados

library(data.table)
Originais <- fread("activity.csv")
Originais$date <- as.Date(Originais$date, "%Y-%m-%d")

summary(Originais)
head(Originais)
length(Originais$steps)

## Qual é o número total médio de passos dados por dia?

Passos <- Originais[, lapply(.SD, sum), by = date, .SDcols = "steps"]

par(lend = "square")

plot(Passos$date, Passos$steps,
     type = "h", lwd = 10, col = "black",
     xlab = "date", ylab = "number of steps",
     main = "number of steps per day")

lines(Passos$date, Passos$steps, type = "h", lwd = 7, col = "magenta")

mean(Passos$steps, na.rm = T);median(Passos$steps, na.rm = T)

## Qual é o padrão de atividade média diária?

Passos2 <- Originais[!is.na(Originais$steps)][,lapply(.SD, mean), by = interval, .SDcols = "steps"]

plot(Passos2$interval, Passos2$steps, type = "l",
     xlab = "interval", ylab = "steps",
     main = "steps by interval")

Passos2[which.max(Passos2$steps)]

## Inserindo valores ausentes

length(Originais$steps[is.na(Originais$steps)])

Originais$steps[is.na(Originais$steps)] <- with(Originais, ave(steps, interval,
                                                               FUN = function(x) median(x, na.rm = T)))[is.na(Originais$steps)]
#median substitute missing values
#Replot "Steps Per Day"
Passos3 <- Originais[, lapply(.SD, sum), by = date, .SDcols = "steps"]

par(lend = "square")

plot(Passos3$date, Passos3$steps,
     type = "h", lwd = 8, col = "black",
     xlab = "date", ylab = "number of steps",
     main = "number of steps per day")

lines(Passos3$date, Passos3$steps, type = "h", lwd = 7, col = "blue")

mean(Passos3$steps, na.rm = T);
median(Passos3$steps, na.rm = T)

library(chron)
library(plyr)
Originais$day <- revalue(as.factor(is.weekend(Originais$date)),
                         c("FALSE" = "WEEKDAY", "TRUE"="WEEKEND"))

library(reshape2)
PassosMerge <- suppressWarnings(melt(Originais, c("interval", "day")))
PassosSoma  <- dcast(PassosMerge, day + interval ~ variable, mean)[,1:3]

library(lattice)
xyplot(steps ~ interval | day, data = PassosSoma, layout = c(1,2), type = "l")