physical_activity_data_science_code.Rmd

---
title: "Physical Activity and Data Science"
author: "Daniel Fuller"
date: "06/05/2019"
output:
      html_document:
        keep_md: true
---

```{r setup, include=TRUE}
knitr::opts_chunk$set(echo = TRUE)
```

# Answer key for Physical Activity Data Science Workshop

## Get the libraries we need
```{r}
library(tidyverse)
library(ggmap)
```

## Accelerometer data section

### Read in the data
```{r}
accel_data <- read_csv("accel_data_example.csv")
```

### Show the first 6 rows
```{r}
head(accel_data)
```

### Accel Data description
* rowid: unique identifier for each row
* utcdate: the data to the second in utc time
* ts: second of measurement to 13 decimal places
* x: the x axis of acceleration measured in g units of gravity (1g = 9.81m/s^2)
* y: the x axis of acceleration measured in g 
* z: the x axis of acceleration measured in g 

### Exploring x,y,z

Calculate the mean and standard deviation for the x axis and create a new data frame. 
```{r}
accel_mean_x <- accel_data %>% 
                summarize(
                  x_mean = mean(x),
                  x_sd = sd(x)
                )

head(accel_mean_x)
```

Calculate the mean and standard deviation for all variables and create a new data frame. 
```{r}
accel_mean <- accel_data %>% 
                summarize(
                  x_mean = mean(x),
                  y_mean = mean(y),
                  z_mean = mean(z),
                  x_sd = sd(x),
                  y_sd = sd(y),
                  z_sd = sd(z),
                )
head(accel_mean)
```

### Create a plot of the histogram of X
```{r}
accel_hist_x <- ggplot(accel_data) + 
                  geom_histogram(aes(x))
plot(accel_hist_x)
```

### Create a plot of the desnity plot of X
```{r}
accel_dens_x <- ggplot(accel_data) + 
                  geom_density(aes(x), colour = "blue")
plot(accel_dens_x)
```

### Create a plot of the desnity plot of X, Y, and Z
```{r}
accel_hist <- ggplot(accel_data) + 
                  geom_density(aes(x), colour = "blue") + 
                  geom_density(aes(y), colour = "red") + 
                  geom_density(aes(z), colour = "purple")
plot(accel_hist)
```

### Create a plot of the x axis over time with time on the x axis and acceleration 
```{r}
accel_plot_x <- ggplot(accel_data) + 
                  geom_point(aes(x = utcdate, y = x), colour = "blue", alpha = 0.1)
plot(accel_plot_x)
```

Cool. Now we can add all the other axes and see them together. 
```{r}
accel_plot <- ggplot(accel_data) + 
                  geom_point(aes(x = utcdate, y = x), colour = "blue", alpha = 0.1) + 
                  geom_point(aes(x = utcdate, y = y), colour = "red", alpha = 0.1) + 
                  geom_point(aes(x = utcdate, y = z), colour = "purple", alpha = 0.1) 
plot(accel_plot)
```

## GPS data section

### Read in the data
```{r}
gps_data <- read_csv("gps_data_example.csv")
```

### Check out the data
```{r}
head(gps_data) 
```

### Missing data

I know there is missing data. We are going to remove all missing. If you had
```{r}
gps_data <- gps_data %>% 
              drop_na()
```

### GPS Data description
* ts: second of measurement to 13 decimal places
* utcdate: the data to the second in utc time
* lat: the latitude 
    * geographic coordinate that specifies the north–south position of a point on the Earth's surface.
* lon: the longitude
    * geographic coordinate that specifies the east–west position of a point on the Earth's surface.
* sat_used: number of satellites used to fix the lat and lon coordinates

### Calculate the mean and standard deviation of speed
```{r}
gps_mean_x <- gps_data %>% 
                summarize(
                  speed_mean = mean(speed),
                  speed_sd = sd(speed)
                )

head(gps_mean_x)
```

### Make a plot of the lat and lon data

Which axes should we have on the x and y axes
```{r}
gps_plot_1 <- ggplot(gps_data) + 
                  geom_point(aes(x = lon, y = lat))
plot(gps_plot_1)
```

### Connect with Google Maps using `ggmap` 
```{r}
avalon_basemap <- get_map(location = "St. John's, Newfoundland, Canada",
                     source = "google",
                     maptype = "roadmap", crop = FALSE,
                     zoom = 14)
plot(avalon_basemap)
```

### Combine ggmap with points
```{r}
maps_points <- ggmap(avalon_basemap) + 
                  geom_point(aes(x = lon, y = lat), data = gps_data, alpha = 0.2)

plot(maps_points)
```

### Colour GPS data by speed
```{r}
maps_speed <- ggmap(avalon_basemap) + 
                  geom_point(aes(x = lon, y = lat, colour = speed), data = gps_data, alpha = 0.2)

plot(maps_speed)
```