Fitness Watch Data Analysis involves analyzing the data collected by fitness wearables or smartwatches to gain insights into users’ health and activity patterns. These devices track metrics like steps taken, energy burned, walking speed, and more. So, if you want to learn how to analyze the data of a fitness watch, this article is for you. In this article, I’ll take you through the task of Fitness Watch Data Analysis using Python.
Fitness Watch Data Analysis: Process We Can Follow
Fitness Watch Data Analysis is a crucial tool for businesses in the health and wellness domain. By analyzing user data from fitness wearables, companies can understand user behaviour, offer personalized solutions, and contribute to improving users’ overall health and well-being.
Below is the process we can follow while working on the problem of Fitness Watch Data Analysis:
- Collect data from fitness watches, ensuring it’s accurate and reliable.
- Perform EDA to gain initial insights into the data.
- Create new features from the raw data that might provide more meaningful insights.
- Create visual representations of the data to communicate insights effectively.
- Segment user’s activity based on time intervals or the level of fitness metrics and analyze their performance.
So, the process starts with collecting data from a fitness watch. Every fitness watch works with an app on your smartphone. You can collect data from that app on your smartphone. For example, in my case, I collected my fitness watch’s data from Apple’s Health app. If you will also collect fitness data from any app, it will not be in a format to be used for analysis. So, I have already made a clean version of the data available online. You can find the dataset from here.
Fitness Watch Data Analysis using Python
Now let’s get started with the task of Fitness Watch Data Analysis by importing the necessary Python libraries and the dataset:
import pandas as pd
import plotly.io as pio
import plotly.graph_objects as go
pio.templates.default = "plotly_white"
import plotly.express as px
data = pd.read_csv("Apple-Fitness-Data.csv")
print(data.head())Date Time Step Count Distance Energy Burned \ 0 2023-03-21 16:01:23 46 0.02543 14.620 1 2023-03-21 16:18:37 645 0.40041 14.722 2 2023-03-21 16:31:38 14 0.00996 14.603 3 2023-03-21 16:45:37 13 0.00901 14.811 4 2023-03-21 17:10:30 17 0.00904 15.153 Flights Climbed Walking Double Support Percentage Walking Speed 0 3 0.304 3.060 1 3 0.309 3.852 2 4 0.278 3.996 3 3 0.278 5.040 4 3 0.281 5.184
Let’s have a look if this data contains any null values or not:
print(data.isnull().sum())
Date 0 Time 0 Step Count 0 Distance 0 Energy Burned 0 Flights Climbed 0 Walking Double Support Percentage 0 Walking Speed 0 dtype: int64
So, the data doesn’t have any null values. Let’s move further by analyzing my step count over time:
# Step Count Over Time
fig1 = px.line(data, x="Time",
y="Step Count",
title="Step Count Over Time")
fig1.show()
Now, let’s have a look at the distance covered over time:
# Distance Covered Over Time
fig2 = px.line(data, x="Time",
y="Distance",
title="Distance Covered Over Time")
fig2.show()
Now, let’s have a look at my energy burned over time:
# Energy Burned Over Time
fig3 = px.line(data, x="Time",
y="Energy Burned",
title="Energy Burned Over Time")
fig3.show()
Now, let’s have a look at my walking speed over time:
# Walking Speed Over Time
fig4 = px.line(data, x="Time",
y="Walking Speed",
title="Walking Speed Over Time")
fig4.show()
Now, let’s calculate and look at the average step counts per day:
# Calculate Average Step Count per Day
average_step_count_per_day = data.groupby("Date")["Step Count"].mean().reset_index()
fig5 = px.bar(average_step_count_per_day, x="Date",
y="Step Count",
title="Average Step Count per Day")
fig5.update_xaxes(type='category')
fig5.show()
Now, let’s have a look at my walking efficiency over time:
# Calculate Walking Efficiency
data["Walking Efficiency"] = data["Distance"] / data["Step Count"]
fig6 = px.line(data, x="Time",
y="Walking Efficiency",
title="Walking Efficiency Over Time")
fig6.show()
Now, let’s have a look at the step count and walking speed variations by time intervals:
# Create Time Intervals
time_intervals = pd.cut(pd.to_datetime(data["Time"]).dt.hour,
bins=[0, 12, 18, 24],
labels=["Morning", "Afternoon", "Evening"],
right=False)
data["Time Interval"] = time_intervals
# Variations in Step Count and Walking Speed by Time Interval
fig7 = px.scatter(data, x="Step Count",
y="Walking Speed",
color="Time Interval",
title="Step Count and Walking Speed Variations by Time Interval",
trendline='ols')
fig7.show()
Now, let’s compare the daily average of all the health and fitness metrics:
# Reshape data for treemap
daily_avg_metrics = data.groupby("Date").mean().reset_index()
daily_avg_metrics_melted = daily_avg_metrics.melt(id_vars=["Date"],
value_vars=["Step Count", "Distance",
"Energy Burned", "Flights Climbed",
"Walking Double Support Percentage",
"Walking Speed"])
# Treemap of Daily Averages for Different Metrics Over Several Weeks
fig = px.treemap(daily_avg_metrics_melted,
path=["variable"],
values="value",
color="variable",
hover_data=["value"],
title="Daily Averages for Different Metrics")
fig.show()
The above graph represents each health and fitness metric as a rectangular tile. The size of each tile corresponds to the value of the metric and the colour of the tiles represents the metric itself. Hover data displays the exact average value for each metric when interacting with the visualization.
The Step Count metric dominates the visualization due to its generally higher numerical values compared to other metrics, making it difficult to visualize variations in the other metrics effectively. As the value of step count is higher than the value of all other metrics, let’s have a look at this visualization again without step counts:
# Select metrics excluding Step Count
metrics_to_visualize = ["Distance", "Energy Burned", "Flights Climbed",
"Walking Double Support Percentage", "Walking Speed"]
# Reshape data for treemap
daily_avg_metrics_melted = daily_avg_metrics.melt(id_vars=["Date"], value_vars=metrics_to_visualize)
fig = px.treemap(daily_avg_metrics_melted,
path=["variable"],
values="value",
color="variable",
hover_data=["value"],
title="Daily Averages for Different Metrics (Excluding Step Count)")
fig.show()
So, this is how you can analyze and work with fitness data using Python.
Summary
So this is how to perform Fitness Data Analysis using Python. Fitness Watch Data Analysis is a crucial tool for businesses in the health and wellness domain. By analyzing user data from fitness wearables, companies can understand user behaviour, offer personalized solutions, and contribute to improving users’ overall health and well-being. I hope you liked this article on Fitness Watch Data Analysis using Python. Feel free to ask valuable questions in the comments section below.
