updated

notebooks/data_exploration.ipynb

@@ -0,0 +1,513 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# IFVI Value Factors Data Exploration\n",
+    "\n",
+    "This notebook provides a starting point for exploring and visualizing the IFVI Value Factors dataset. The dataset contains environmental value factors organized by region, impact type, and policy domain.\n",
+    "\n",
+    "## Dataset Overview\n",
+    "\n",
+    "The IFVI Value Factors dataset is organized in multiple ways:\n",
+    "- By region (continental regions, economic zones, development status)\n",
+    "- By impact type (health, ecosystem, economic, social impacts)\n",
+    "- By policy domain (climate, air quality, land use, waste management, water resources)\n",
+    "\n",
+    "Data is available in multiple formats: JSON, CSV, and Parquet."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Setup and Dependencies\n",
+    "\n",
+    "First, let's import the necessary libraries for data exploration and visualization."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Import standard data analysis libraries\n",
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
+    "import seaborn as sns\n",
+    "import json\n",
+    "import os\n",
+    "from pathlib import Path\n",
+    "import glob\n",
+    "\n",
+    "# Set up plotting\n",
+    "plt.style.use('ggplot')\n",
+    "sns.set(style=\"whitegrid\")\n",
+    "%matplotlib inline\n",
+    "plt.rcParams['figure.figsize'] = (12, 8)\n",
+    "\n",
+    "# Display settings for pandas\n",
+    "pd.set_option('display.max_columns', None)\n",
+    "pd.set_option('display.max_rows', 100)\n",
+    "pd.set_option('display.width', 1000)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Data Loading Functions\n",
+    "\n",
+    "Let's define some helper functions to load data from different sources and formats."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Define the base data directory\n",
+    "DATA_DIR = Path('../data')\n",
+    "\n",
+    "def load_json_file(file_path):\n",
+    "    \"\"\"Load a JSON file into a Python dictionary.\"\"\"\n",
+    "    try:\n",
+    "        with open(file_path, 'r') as f:\n",
+    "            return json.load(f)\n",
+    "    except Exception as e:\n",
+    "        print(f\"Error loading {file_path}: {e}\")\n",
+    "        return None\n",
+    "\n",
+    "def load_csv_file(file_path):\n",
+    "    \"\"\"Load a CSV file into a pandas DataFrame.\"\"\"\n",
+    "    try:\n",
+    "        return pd.read_csv(file_path)\n",
+    "    except Exception as e:\n",
+    "        print(f\"Error loading {file_path}: {e}\")\n",
+    "        return None\n",
+    "    \n",
+    "def load_parquet_file(file_path):\n",
+    "    \"\"\"Load a Parquet file into a pandas DataFrame.\"\"\"\n",
+    "    try:\n",
+    "        return pd.read_parquet(file_path)\n",
+    "    except Exception as e:\n",
+    "        print(f\"Error loading {file_path}: {e}\")\n",
+    "        return None\n",
+    "    \n",
+    "def get_available_regions():\n",
+    "    \"\"\"Get a list of available regions in the dataset.\"\"\"\n",
+    "    continental_dir = DATA_DIR / 'by-region' / 'continental'\n",
+    "    if continental_dir.exists():\n",
+    "        return [d.name for d in continental_dir.iterdir() if d.is_dir()]\n",
+    "    return []\n",
+    "\n",
+    "def get_available_impact_types():\n",
+    "    \"\"\"Get a list of available impact types in the dataset.\"\"\"\n",
+    "    impact_dir = DATA_DIR / 'by-impact-type'\n",
+    "    if impact_dir.exists():\n",
+    "        return [d.name for d in impact_dir.iterdir() if d.is_dir()]\n",
+    "    return []"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Exploring the Dataset Structure\n",
+    "\n",
+    "Let's explore the structure of the dataset to understand what's available."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Check available regions\n",
+    "regions = get_available_regions()\n",
+    "print(f\"Available regions: {regions}\")\n",
+    "\n",
+    "# Check available impact types\n",
+    "impact_types = get_available_impact_types()\n",
+    "print(f\"Available impact types: {impact_types}\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Loading Aggregated Data\n",
+    "\n",
+    "Let's start by loading some of the aggregated data files to get an overview of the dataset."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Load composite value factors from CSV\n",
+    "composite_csv_path = DATA_DIR / 'aggregated' / 'composite_value_factors.csv'\n",
+    "if composite_csv_path.exists():\n",
+    "    composite_df = load_csv_file(composite_csv_path)\n",
+    "    if composite_df is not None:\n",
+    "        print(\"Composite Value Factors (CSV):\")\n",
+    "        display(composite_df.head())\n",
+    "else:\n",
+    "    print(f\"File not found: {composite_csv_path}\")\n",
+    "    \n",
+    "# Try loading some CSV files from the csv directory\n",
+    "csv_files = list(Path(DATA_DIR / 'csv' / 'by-methodology').glob('*.csv'))\n",
+    "if csv_files:\n",
+    "    print(f\"\\nFound {len(csv_files)} CSV files in by-methodology directory\")\n",
+    "    for file_path in csv_files[:3]:  # Show first 3 files\n",
+    "        print(f\"\\nLoading {file_path.name}:\")\n",
+    "        df = load_csv_file(file_path)\n",
+    "        if df is not None:\n",
+    "            display(df.head())"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Exploring Regional Data\n",
+    "\n",
+    "Let's explore the data for specific regions."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Function to get countries in a region\n",
+    "def get_countries_in_region(region):\n",
+    "    region_dir = DATA_DIR / 'by-region' / 'continental' / region\n",
+    "    if region_dir.exists():\n",
+    "        return [f.stem for f in region_dir.glob('*.json')]\n",
+    "    return []\n",
+    "\n",
+    "# Check countries in each region\n",
+    "for region in regions:\n",
+    "    countries = get_countries_in_region(region)\n",
+    "    print(f\"{region}: {len(countries)} countries\")\n",
+    "    if countries:\n",
+    "        print(f\"  Sample countries: {', '.join(countries[:5])}...\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Loading Country-Specific Data\n",
+    "\n",
+    "Let's load data for a specific country and explore it."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Choose a sample country (adjust as needed)\n",
+    "sample_region = regions[0] if regions else None\n",
+    "if sample_region:\n",
+    "    countries = get_countries_in_region(sample_region)\n",
+    "    sample_country = countries[0] if countries else None\n",
+    "    \n",
+    "    if sample_country:\n",
+    "        country_file = DATA_DIR / 'by-region' / 'continental' / sample_region / f\"{sample_country}.json\"\n",
+    "        print(f\"Loading data for {sample_country} in {sample_region}\")\n",
+    "        country_data = load_json_file(country_file)\n",
+    "        \n",
+    "        if country_data:\n",
+    "            # Display basic information about the country data\n",
+    "            print(f\"\\nData keys: {list(country_data.keys()) if isinstance(country_data, dict) else 'Not a dictionary'}\")\n",
+    "            \n",
+    "            # Convert to DataFrame if possible for easier exploration\n",
+    "            if isinstance(country_data, dict):\n",
+    "                try:\n",
+    "                    # This is a placeholder - adjust based on actual data structure\n",
+    "                    country_df = pd.json_normalize(country_data)\n",
+    "                    display(country_df.head())\n",
+    "                except Exception as e:\n",
+    "                    print(f\"Could not convert to DataFrame: {e}\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Exploring Impact Type Data\n",
+    "\n",
+    "Let's explore data organized by impact type."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Function to get files in an impact type directory\n",
+    "def get_files_in_impact_type(impact_type):\n",
+    "    impact_dir = DATA_DIR / 'by-impact-type' / impact_type\n",
+    "    if impact_dir.exists():\n",
+    "        return list(impact_dir.glob('**/*.*'))\n",
+    "    return []\n",
+    "\n",
+    "# Check files in each impact type\n",
+    "for impact_type in impact_types:\n",
+    "    files = get_files_in_impact_type(impact_type)\n",
+    "    print(f\"{impact_type}: {len(files)} files\")\n",
+    "    if files:\n",
+    "        print(f\"  Sample files: {', '.join(str(f.relative_to(DATA_DIR / 'by-impact-type')) for f in files[:3])}...\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Data Visualization Examples\n",
+    "\n",
+    "Let's create some example visualizations based on the available data."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Example 1: Bar chart comparing values across countries (placeholder)\n",
+    "# Replace with actual data loading and processing based on your exploration\n",
+    "def plot_country_comparison(region, value_factor, n_countries=10):\n",
+    "    \"\"\"Plot a comparison of a specific value factor across countries in a region.\"\"\"\n",
+    "    # This is a placeholder - replace with actual data loading logic\n",
+    "    countries = get_countries_in_region(region)[:n_countries]\n",
+    "    \n",
+    "    # Placeholder for data - replace with actual data\n",
+    "    np.random.seed(42)  # For reproducibility\n",
+    "    values = np.random.rand(len(countries)) * 100\n",
+    "    \n",
+    "    plt.figure(figsize=(12, 8))\n",
+    "    plt.bar(countries, values)\n",
+    "    plt.title(f\"{value_factor} Comparison Across {region} Countries\")\n",
+    "    plt.xlabel(\"Country\")\n",
+    "    plt.ylabel(f\"{value_factor} Value\")\n",
+    "    plt.xticks(rotation=45, ha='right')\n",
+    "    plt.tight_layout()\n",
+    "    plt.show()\n",
+    "\n",
+    "# Example visualization with placeholder data\n",
+    "if regions:\n",
+    "    plot_country_comparison(regions[0], \"CO2 Emissions Value Factor\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Example 2: Heatmap of correlations between different value factors (placeholder)\n",
+    "# Replace with actual data loading and processing\n",
+    "\n",
+    "# Placeholder for correlation data - replace with actual data\n",
+    "np.random.seed(42)  # For reproducibility\n",
+    "factor_names = ['CO2', 'PM2.5', 'NOx', 'SOx', 'Land Use', 'Water Consumption']\n",
+    "corr_matrix = np.random.rand(len(factor_names), len(factor_names))\n",
+    "# Make it symmetric for a valid correlation matrix\n",
+    "corr_matrix = (corr_matrix + corr_matrix.T) / 2\n",
+    "np.fill_diagonal(corr_matrix, 1)\n",
+    "\n",
+    "# Create a DataFrame\n",
+    "corr_df = pd.DataFrame(corr_matrix, index=factor_names, columns=factor_names)\n",
+    "\n",
+    "# Plot the heatmap\n",
+    "plt.figure(figsize=(10, 8))\n",
+    "sns.heatmap(corr_df, annot=True, cmap='coolwarm', vmin=-1, vmax=1)\n",
+    "plt.title('Correlation Between Different Value Factors (Example)')\n",
+    "plt.tight_layout()\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Geographic Visualization\n",
+    "\n",
+    "Let's create a map visualization to show value factors across different regions."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# For geographic visualizations, we'll need additional libraries\n",
+    "# Uncomment and run if needed\n",
+    "# !pip install geopandas matplotlib\n",
+    "\n",
+    "# Example code for geographic visualization\n",
+    "try:\n",
+    "    import geopandas as gpd\n",
+    "    \n",
+    "    # Load world map data\n",
+    "    world = gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))\n",
+    "    \n",
+    "    # Placeholder for value factor data - replace with actual data\n",
+    "    # Here we're just assigning random values to countries\n",
+    "    np.random.seed(42)  # For reproducibility\n",
+    "    world['value_factor'] = np.random.rand(len(world)) * 100\n",
+    "    \n",
+    "    # Create the plot\n",
+    "    fig, ax = plt.subplots(1, 1, figsize=(15, 10))\n",
+    "    world.plot(column='value_factor', ax=ax, legend=True,\n",
+    "               legend_kwds={'label': \"Value Factor (Example)\",\n",
+    "                           'orientation': \"horizontal\"},\n",
+    "               cmap='YlOrRd')\n",
+    "    ax.set_title('Global Distribution of Value Factors (Example)', fontsize=15)\n",
+    "    plt.tight_layout()\n",
+    "    plt.show()\n",
+    "    \n",
+    "except ImportError:\n",
+    "    print(\"To create geographic visualizations, install geopandas:\")\n",
+    "    print(\"pip install geopandas matplotlib\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Time Series Analysis\n",
+    "\n",
+    "If the data contains time series information, we can analyze trends over time."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Placeholder for time series data - replace with actual data if available\n",
+    "# Generate example time series data\n",
+    "dates = pd.date_range(start='2015-01-01', end='2024-01-01', freq='M')\n",
+    "np.random.seed(42)  # For reproducibility\n",
+    "values = np.cumsum(np.random.randn(len(dates))) + 50  # Random walk with drift\n",
+    "\n",
+    "# Create a DataFrame\n",
+    "ts_df = pd.DataFrame({'Date': dates, 'Value Factor': values})\n",
+    "\n",
+    "# Plot the time series\n",
+    "plt.figure(figsize=(14, 7))\n",
+    "plt.plot(ts_df['Date'], ts_df['Value Factor'])\n",
+    "plt.title('Value Factor Trend Over Time (Example)', fontsize=15)\n",
+    "plt.xlabel('Date')\n",
+    "plt.ylabel('Value Factor')\n",
+    "plt.grid(True)\n",
+    "plt.tight_layout()\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Comparative Analysis\n",
+    "\n",
+    "Let's compare value factors across different categories or regions."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Placeholder for comparative data - replace with actual data\n",
+    "categories = ['Health Impacts', 'Ecosystem Impacts', 'Economic Impacts', 'Social Impacts']\n",
+    "regions_sample = ['Africa', 'Asia', 'Europe', 'North America', 'South America']\n",
+    "\n",
+    "# Generate random data for demonstration\n",
+    "np.random.seed(42)  # For reproducibility\n",
+    "data = np.random.rand(len(regions_sample), len(categories)) * 100\n",
+    "\n",
+    "# Create a DataFrame\n",
+    "comp_df = pd.DataFrame(data, index=regions_sample, columns=categories)\n",
+    "\n",
+    "# Plot the data\n",
+    "comp_df.plot(kind='bar', figsize=(14, 8))\n",
+    "plt.title('Value Factors by Impact Type Across Regions (Example)', fontsize=15)\n",
+    "plt.xlabel('Region')\n",
+    "plt.ylabel('Value Factor')\n",
+    "plt.legend(title='Impact Type')\n",
+    "plt.grid(axis='y')\n",
+    "plt.tight_layout()\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Next Steps\n",
+    "\n",
+    "Based on this initial exploration, here are some suggested next steps:\n",
+    "\n",
+    "1. **Deeper Data Exploration**:\n",
+    "   - Explore the structure of country-specific JSON files in detail\n",
+    "   - Analyze the relationships between different value factors\n",
+    "   - Compare value factors across different regions and impact types\n",
+    "\n",
+    "2. **Advanced Visualizations**:\n",
+    "   - Create interactive visualizations using libraries like Plotly\n",
+    "   - Develop choropleth maps to show global distribution of value factors\n",
+    "   - Create dashboards for comprehensive data exploration\n",
+    "\n",
+    "3. **Statistical Analysis**:\n",
+    "   - Perform correlation analysis between different value factors\n",
+    "   - Conduct cluster analysis to identify groups of countries with similar profiles\n",
+    "   - Develop predictive models based on the value factors\n",
+    "\n",
+    "4. **Policy Implications**:\n",
+    "   - Analyze how value factors relate to policy decisions\n",
+    "   - Compare value factors with actual policy implementations\n",
+    "   - Evaluate the economic implications of different value factors"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.10"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}