{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Cybersecurity news\n",
"Interactive exploratory notebook for trend analysis and marketing-ready visuals.\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"import os\n",
"import pandas as pd\n",
"import numpy as np\n",
"import plotly.express as px\n",
"import plotly.graph_objects as go\n",
"articles = pd.read_csv('cybersecurity-news-en-title-3000.csv')\n",
"weekly_path = 'weekly_counts.csv'\n",
"peaks_path = 'weekly_peak_headlines.csv'\n",
"weekly = pd.read_csv(weekly_path) if os.path.exists(weekly_path) else pd.DataFrame()\n",
"peaks = pd.read_csv(peaks_path) if os.path.exists(peaks_path) else pd.DataFrame()\n",
"articles.head()\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"articles['pub_date'] = pd.to_datetime(articles['pub_date'], errors='coerce', utc=True)\n",
"articles = articles.dropna(subset=['pub_date']).copy()\n",
"articles['week'] = articles['pub_date'].dt.to_period('W').dt.start_time\n",
"articles.shape\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"keyword_columns = [\"mentions_ransomware\", \"mentions_ddos\", \"mentions_data_breach\", \"mentions_vulnerability\", \"mentions_hacking\", \"mentions_cyber_attack\"]\n",
"if not weekly.empty and keyword_columns:\n",
" weekly['week'] = pd.to_datetime(weekly['week'])\n",
" long_df = weekly.melt(\n",
" id_vars=['week'],\n",
" value_vars=keyword_columns,\n",
" var_name='mention_flag',\n",
" value_name='count'\n",
" )\n",
" long_df['count'] = pd.to_numeric(long_df['count'], errors='coerce').fillna(0).astype(int)\n",
"\n",
" long_df['week_total'] = long_df.groupby('week')['count'].transform('sum')\n",
" long_df['share_pct'] = np.where(long_df['week_total'] > 0, (long_df['count'] / long_df['week_total']) * 100, 0.0)\n",
" long_df = long_df.sort_values(['mention_flag', 'week']).reset_index(drop=True)\n",
" long_df['wow_pct'] = (\n",
" long_df.groupby('mention_flag')['count']\n",
" .pct_change()\n",
" .replace([np.inf, -np.inf], np.nan)\n",
" .fillna(0.0) * 100\n",
" )\n",
"\n",
" weekly_rep_parts = []\n",
" for flag in keyword_columns:\n",
" if flag not in articles.columns:\n",
" continue\n",
" rep = (\n",
" articles[articles[flag].astype(str).str.lower().isin(['true', '1'])]\n",
" .sort_values('pub_date')\n",
" .groupby('week', as_index=False)\n",
" .first()[['week', 'title', 'source_title']]\n",
" )\n",
" rep['mention_flag'] = flag\n",
" weekly_rep_parts.append(rep)\n",
"\n",
" if weekly_rep_parts:\n",
" weekly_rep = pd.concat(weekly_rep_parts, ignore_index=True)\n",
" long_df = long_df.merge(weekly_rep, on=['week', 'mention_flag'], how='left')\n",
" else:\n",
" long_df['title'] = ''\n",
" long_df['source_title'] = ''\n",
"\n",
" long_df['title'] = long_df['title'].fillna('')\n",
" long_df['source_title'] = long_df['source_title'].fillna('')\n",
" long_df['headline_short'] = np.where(\n",
" long_df['title'].str.len() > 110,\n",
" long_df['title'].str.slice(0, 107) + '...',\n",
" long_df['title']\n",
" )\n",
"\n",
" fig = px.line(\n",
" long_df,\n",
" x='week',\n",
" y='count',\n",
" color='mention_flag',\n",
" markers=True,\n",
" custom_data=[\n",
" 'share_pct',\n",
" 'wow_pct',\n",
" 'headline_short',\n",
" 'source_title',\n",
" ],\n",
" title='Weekly Mention Counts (Interactive)'\n",
" )\n",
"\n",
" fig.update_traces(\n",
" hovertemplate=(\n",
" 'Mention: %{fullData.name}
'\n",
" 'Week: %{x|%Y-%m-%d}
'\n",
" 'Count: %{y:,}
'\n",
" 'Share: %{customdata[0]:.1f}%
'\n",
" 'Week-over-week change: %{customdata[1]:+.1f}%
'\n",
" 'Example headline: %{customdata[2]}
'\n",
" 'Source: %{customdata[3]}'\n",
" ''\n",
" ),\n",
" line={'width': 2.6},\n",
" marker={'size': 6},\n",
" )\n",
"\n",
" if not peaks.empty:\n",
" peaks['peak_week'] = pd.to_datetime(peaks['peak_week'])\n",
" peaks['headline_short'] = np.where(\n",
" peaks['representative_headline'].fillna('').str.len() > 110,\n",
" peaks['representative_headline'].fillna('').str.slice(0, 107) + '...',\n",
" peaks['representative_headline'].fillna('')\n",
" )\n",
" custom = peaks[['headline_short', 'source_title']].fillna('').to_numpy()\n",
" fig.add_trace(go.Scatter(\n",
" x=peaks['peak_week'],\n",
" y=peaks['peak_count'],\n",
" mode='markers',\n",
" marker=dict(size=10, symbol='diamond', line=dict(width=1)),\n",
" name='Peak week',\n",
" customdata=custom,\n",
" text=peaks['mention_flag'],\n",
" hovertemplate=(\n",
" 'Mention: %{text}
'\n",
" 'Week: %{x|%Y-%m-%d}
'\n",
" 'Count: %{y}
'\n",
" 'Example headline: %{customdata[0]}
'\n",
" 'Source: %{customdata[1]}'\n",
" ''\n",
" )\n",
" ))\n",
"\n",
" fig.update_layout(\n",
" template='plotly_white',\n",
" hovermode='closest',\n",
" legend_title_text='',\n",
" xaxis_title='Week',\n",
" yaxis_title='Article count',\n",
" font=dict(size=13),\n",
" margin=dict(l=30, r=30, t=70, b=30),\n",
" hoverlabel=dict(bgcolor='white', font_size=12),\n",
" )\n",
" fig.update_xaxes(showgrid=True, gridcolor='#EFEFEF')\n",
" fig.update_yaxes(showgrid=True, gridcolor='#EFEFEF')\n",
" fig.show()\n",
"else:\n",
" print('No weekly data available for interactive mention chart.')\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"top_sources = (\n",
" articles['source_title']\n",
" .value_counts()\n",
" .head(20)\n",
" .rename_axis('source_title')\n",
" .reset_index(name='articles')\n",
")\n",
"fig = px.bar(\n",
" top_sources,\n",
" x='articles',\n",
" y='source_title',\n",
" orientation='h',\n",
" title='Top 20 Sources by Article Count (Interactive)',\n",
" color='articles',\n",
" color_continuous_scale='Blues'\n",
")\n",
"fig.update_layout(\n",
" template='plotly_white',\n",
" yaxis={'categoryorder': 'total ascending'},\n",
" coloraxis_showscale=False,\n",
" margin=dict(l=30, r=20, t=60, b=20),\n",
")\n",
"fig.show()\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"if not peaks.empty:\n",
" peaks[['mention_flag', 'peak_week', 'peak_count', 'representative_headline', 'source_title', 'article_link']]\n",
"else:\n",
" print('No peak headline rows generated for this dataset.')\n"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"name": "python",
"version": "3"
}
},
"nbformat": 4,
"nbformat_minor": 5
}