Streamlitplotly.py

import streamlit as st

import pandas as pd

import matplotlib.pyplot as plt

from urllib.parse import urlparse

data = pd.read_csv("csvdata/dark-patterns.csv")

st.dataframe(data)

# Extracting domain from the 'Website Page' column

data['Domain'] = data['Website Page'].apply(lambda x: urlparse(x).netloc)

# Counting the frequency of each domain and sorting them from most to least occurrences

domain_counts = data['Domain'].value_counts().sort_values(ascending=False)

st.divider()

st.subheader("Domain Counts")

st.write(domain_counts)

fig, ax = plt.subplots()

# only show top 10

ax.bar(domain_counts.index[:10], domain_counts.values[:10])

plt.title('Distribution of Domains')

plt.xticks(rotation=45, ha='right')

plt.ylabel('Count')

plt.xlabel('Domain')

st.pyplot(fig)

st.divider()

st.subheader("Insights")

st.write("The following graphs show the distribution of the data collected.")

category_counts = data['Pattern Category'].value_counts()

type_counts = data['Pattern Type'].value_counts()

deceptive_counts = data['Deceptive?'].value_counts()

fig, ax = plt.subplots()

ax.bar(category_counts.index, category_counts.values)

plt.title('Distribution of Pattern Categories')

plt.xticks(rotation=45, ha='right')

plt.ylabel('Count')

plt.xlabel('Pattern Category')

st.pyplot(fig)

fig, ax = plt.subplots()

ax.bar(type_counts.index, type_counts.values)

plt.title('Distribution of Pattern Types')

plt.xticks(rotation=45, ha='right')

plt.ylabel('Count')

plt.xlabel('Pattern Type')

st.pyplot(fig)