Implement ML fraud detection pipeline and interactive graph visualiser

app.py

@@ -1,7 +1,11 @@
 import streamlit as st
+import os
 from src.data_loader import get_processed_data
 import src.graph_builder as gb
 from src.detectors.alert_engine import get_all_alerts
+from src.ml.features import engineer_features
+from src.ml.trainer import train_model, MODEL_PATH
+from src.ml.predictor import load_model, score_account
 from collections import Counter
 
 # Constants
@@ -90,6 +94,49 @@ def main():
         else:
             st.warning(f"⚠️ Only {len(unique_typologies)} typologies detected — expected >= 3.")
 
+        # --- ML Pipeline ---
+        st.subheader("ML Fraud Classifier")
+
+        # Gather cycle accounts from alerts
+        cycle_accounts = {a['account'] for a in alerts if 'RoundTripping' in a.get('typology', '')}
+
+        if not os.path.exists(MODEL_PATH):
+            with st.spinner("Engineering features..."):
+                feature_df = engineer_features(
+                    transactions_df, G, pagerank_scores,
+                    louvain_partition, cycle_accounts,
+                )
+            with st.spinner("Training XGBoost model..."):
+                model, scaler, metrics = train_model(feature_df)
+            st.write("**Model trained!** Metrics:")
+            st.write(f"AUC-ROC: {metrics['auc_roc']:.4f}")
+            st.write(f"F1: {metrics['f1']:.4f}")
+            st.write(f"Precision: {metrics['precision']:.4f}")
+            st.write(f"Recall: {metrics['recall']:.4f}")
+            if metrics['auc_roc'] >= 0.85:
+                st.success(f"✅ AUC-ROC = {metrics['auc_roc']:.4f} — above 0.85 threshold")
+            else:
+                st.warning(f"⚠️ AUC-ROC = {metrics['auc_roc']:.4f} — below 0.85 target")
+        else:
+            with st.spinner("Engineering features..."):
+                feature_df = engineer_features(
+                    transactions_df, G, pagerank_scores,
+                    louvain_partition, cycle_accounts,
+                )
+            st.write("Model already trained. Loading from disk.")
+
+        bundle = load_model()
+
+        # Score 3 sample accounts
+        st.subheader("Sample Account Scoring")
+        sample_accounts = feature_df['account'].head(3).tolist()
+        for acct in sample_accounts:
+            result = score_account(acct, feature_df, bundle)
+            st.write(
+                f"Account **{acct}**: Risk Score = {result['risk_score']}, "
+                f"Fraud Probability = {result['fraud_probability']:.4f}"
+            )
+
 
     except FileNotFoundError as e:
         st.error(f"Data file not found. Please ensure the raw data is placed at `data/raw/HI_Small_Trans.csv`. Details: {e}")

src/ml/explainer.py

@@ -0,0 +1,69 @@
+"""
+SHAP Explainer Module
+Provides interpretable explanations for individual fraud predictions.
+"""
+import shap
+import pandas as pd
+
+# Plain-English descriptions for every feature
+FEATURE_DESCRIPTIONS = {
+    'tx_count_total': 'Total number of outgoing transactions',
+    'tx_count_7d': 'Transaction count in the last 7 days',
+    'amount_sent_total': 'Total amount of funds sent',
+    'amount_sent_7d': 'Amount of funds sent in the last 7 days',
+    'amount_received_total': 'Total amount of funds received',
+    'amount_received_7d': 'Amount of funds received in the last 7 days',
+    'forward_ratio': 'Percentage of received funds immediately forwarded',
+    'avg_tx_amount': 'Average transaction amount sent',
+    'amount_std': 'Consistency of transaction amounts',
+    'in_out_ratio': 'Ratio of received to sent funds',
+    'pagerank_score': 'Network influence of this account',
+    'in_degree': 'Number of accounts sending funds to this account',
+    'out_degree': 'Number of accounts receiving funds from this account',
+    'fan_in_ratio': 'Concentration of incoming vs outgoing connections',
+    'community_id': 'Louvain community cluster assignment',
+    'is_in_cycle': 'Account detected in a circular transaction loop',
+    'account_age_days': 'Age of the account based on transaction history',
+    'currency_diversity': 'Number of distinct currencies used',
+    'channel_diversity': 'Number of distinct payment channels used',
+    'bank_diversity': 'Number of distinct destination banks used',
+}
+
+TOP_N_FEATURES = 5
+
+
+def explain_prediction(
+    account_id: str,
+    feature_df: pd.DataFrame,
+    bundle: dict,
+) -> list[dict]:
+    """
+    Generate SHAP-based explanations for a single account's fraud prediction.
+
+    Returns:
+        List of top-N feature explanation dicts sorted by |SHAP value|.
+    """
+    row = feature_df[feature_df['account'] == account_id]
+
+    if row.empty:
+        return []
+
+    feature_cols = bundle['feature_cols']
+    scaled_row = bundle['scaler'].transform(row[feature_cols])
+
+    explainer = shap.TreeExplainer(bundle['model'])
+    shap_values = explainer.shap_values(scaled_row)
+
+    results = []
+    for i, col in enumerate(feature_cols):
+        sv = float(shap_values[0][i])
+        results.append({
+            'feature_name': col,
+            'shap_value': sv,
+            'feature_value': float(row[col].values[0]),
+            'direction': 'increases risk' if sv > 0 else 'decreases risk',
+            'description': FEATURE_DESCRIPTIONS.get(col, col),
+        })
+
+    results.sort(key=lambda x: abs(x['shap_value']), reverse=True)
+    return results[:TOP_N_FEATURES]

src/ml/features.py

@@ -0,0 +1,114 @@
+"""
+Feature Engineering Module
+Builds a per-account feature DataFrame for the ML fraud classifier.
+"""
+import pandas as pd
+import numpy as np
+import networkx as nx
+
+# Constants
+RECENT_WINDOW_DAYS = 7
+
+
+def engineer_features(
+    df: pd.DataFrame,
+    G: nx.DiGraph,
+    pagerank_scores: dict,
+    louvain_partition: dict,
+    cycle_accounts: set,
+) -> pd.DataFrame:
+    """
+    Build a feature DataFrame with one row per unique account.
+
+    Combines velocity, ratio, graph, and profile features with a fraud_flag target.
+
+    Returns:
+        pd.DataFrame with all features and fraud_flag column.
+    """
+    df = df.copy()
+    df['timestamp'] = pd.to_datetime(df['timestamp'])
+    max_date = df['timestamp'].max()
+    cutoff_7d = max_date - pd.Timedelta(days=RECENT_WINDOW_DAYS)
+
+    fraud_sources = set(df[df['is_laundering'] == 1]['source'].values)
+
+    # --- All unique accounts ---
+    all_accounts = set(df['source'].unique()) | set(df['target'].unique())
+
+    # --- Pre-aggregate sent stats ---
+    sent_all = df.groupby('source').agg(
+        tx_count_total=('amount', 'count'),
+        amount_sent_total=('amount', 'sum'),
+    )
+    sent_7d = df[df['timestamp'] >= cutoff_7d].groupby('source').agg(
+        tx_count_7d=('amount', 'count'),
+        amount_sent_7d=('amount', 'sum'),
+    )
+
+    # --- Pre-aggregate received stats ---
+    recv_all = df.groupby('target').agg(
+        amount_received_total=('amount', 'sum'),
+    )
+    recv_7d = df[df['timestamp'] >= cutoff_7d].groupby('target').agg(
+        amount_received_7d=('amount', 'sum'),
+    )
+
+    # --- Amount std: combine sent + received per account ---
+    sent_amounts = df[['source', 'amount']].rename(columns={'source': 'account'})
+    recv_amounts = df[['target', 'amount']].rename(columns={'target': 'account'})
+    combined_amounts = pd.concat([sent_amounts, recv_amounts], ignore_index=True)
+    amount_std = combined_amounts.groupby('account')['amount'].std().rename('amount_std')
+
+    # --- Profile features ---
+    first_tx = df.groupby('source')['timestamp'].min().rename('first_tx')
+    last_tx = df.groupby('source')['timestamp'].max().rename('last_tx')
+
+    currency_div = df.groupby('source')['Payment Currency'].nunique().rename('currency_diversity') \
+        if 'Payment Currency' in df.columns else pd.Series(dtype=float, name='currency_diversity')
+    channel_div = df.groupby('source')['payment_type'].nunique().rename('channel_diversity')
+    bank_div = df.groupby('source')['target_bank'].nunique().rename('bank_diversity')
+
+    # --- Build DataFrame ---
+    features = pd.DataFrame(index=list(all_accounts))
+    features.index.name = 'account'
+
+    # Velocity
+    features = features.join(sent_all)
+    features = features.join(sent_7d)
+    features = features.join(recv_all)
+    features = features.join(recv_7d)
+
+    # Ratio
+    features['forward_ratio'] = features['amount_sent_total'] / (features['amount_received_total'] + 1)
+    features['avg_tx_amount'] = features['amount_sent_total'] / (features['tx_count_total'] + 1)
+    features = features.join(amount_std)
+    features['in_out_ratio'] = features['amount_received_total'] / (features['amount_sent_total'] + 1)
+
+    # Graph
+    features['pagerank_score'] = features.index.map(lambda a: pagerank_scores.get(a, 0))
+    features['in_degree'] = features.index.map(lambda a: G.in_degree(a) if a in G else 0)
+    features['out_degree'] = features.index.map(lambda a: G.out_degree(a) if a in G else 0)
+    features['fan_in_ratio'] = features['in_degree'] / (features['in_degree'] + features['out_degree'] + 1)
+    features['community_id'] = features.index.map(lambda a: louvain_partition.get(a, -1))
+    features['is_in_cycle'] = features.index.map(lambda a: 1 if a in cycle_accounts else 0)
+
+    # Profile
+    features = features.join(first_tx)
+    features = features.join(last_tx)
+    features['account_age_days'] = (features['last_tx'] - features['first_tx']).dt.days
+    features = features.drop(columns=['first_tx', 'last_tx'], errors='ignore')
+    features = features.join(currency_div)
+    features = features.join(channel_div)
+    features = features.join(bank_div)
+
+    # Target
+    features['fraud_flag'] = features.index.map(lambda a: 1 if a in fraud_sources else 0)
+
+    # Clean infinities and NaNs
+    features = features.replace([np.inf, -np.inf], 0)
+    features = features.fillna(0)
+
+    # Reset index so 'account' becomes a column
+    features = features.reset_index()
+
+    return features

src/ml/predictor.py

@@ -0,0 +1,41 @@
+"""
+Predictor Module
+Loads the trained model and scores individual accounts.
+"""
+import pickle
+import streamlit as st
+import pandas as pd
+
+MODEL_PATH = 'models/xgb_fraud_model.pkl'
+
+
+@st.cache_resource
+def load_model() -> dict:
+    """
+    Load the trained model bundle from disk.
+
+    Returns:
+        Dict with 'model', 'scaler', and 'feature_cols'.
+    """
+    with open(MODEL_PATH, 'rb') as f:
+        bundle = pickle.load(f)
+    return bundle
+
+
+def score_account(account_id: str, feature_df: pd.DataFrame, bundle: dict) -> dict:
+    """
+    Score a single account's fraud risk using the trained model.
+
+    Returns:
+        Dict with 'risk_score' (0-99) and 'fraud_probability' (0.0-1.0).
+    """
+    row = feature_df[feature_df['account'] == account_id]
+
+    if row.empty:
+        return {'risk_score': 0, 'fraud_probability': 0.0}
+
+    scaled = bundle['scaler'].transform(row[bundle['feature_cols']])
+    proba = bundle['model'].predict_proba(scaled)[0][1]
+    risk_score = int(proba * 99)
+
+    return {'risk_score': risk_score, 'fraud_probability': float(proba)}

src/ml/trainer.py

@@ -0,0 +1,123 @@
+"""
+Model Trainer Module
+Trains an XGBoost classifier for fraud detection.
+"""
+import os
+import pickle
+import xgboost as xgb
+import pandas as pd
+import numpy as np
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+from sklearn.metrics import (
+    roc_auc_score, f1_score, precision_score,
+    recall_score, confusion_matrix, roc_curve,
+)
+
+# Constants
+TEST_SIZE = 0.2
+RANDOM_STATE = 42
+N_ESTIMATORS = 300
+MAX_DEPTH = 6
+LEARNING_RATE = 0.05
+SUBSAMPLE = 0.8
+COLSAMPLE = 0.8
+EARLY_STOPPING = 20
+MODEL_PATH = 'models/xgb_fraud_model.pkl'
+
+FEATURE_COLS = [
+    'tx_count_total',
+    'tx_count_7d',
+    'amount_sent_total',
+    'amount_sent_7d',
+    'amount_received_total',
+    'amount_received_7d',
+    'forward_ratio',
+    'avg_tx_amount',
+    'amount_std',
+    'in_out_ratio',
+    'pagerank_score',
+    'in_degree',
+    'out_degree',
+    'fan_in_ratio',
+    'community_id',
+    'is_in_cycle',
+    'account_age_days',
+    'currency_diversity',
+    'channel_diversity',
+    'bank_diversity',
+]
+
+
+def train_model(feature_df: pd.DataFrame) -> tuple:
+    """
+    Train an XGBoost fraud classifier on the feature DataFrame.
+
+    Returns:
+        Tuple of (model, scaler, metrics_dict).
+    """
+    X = feature_df[FEATURE_COLS].copy()
+    y = feature_df['fraud_flag'].copy()
+
+    positive_count = (y == 1).sum()
+    negative_count = (y == 0).sum()
+    scale_pos_weight = negative_count / max(positive_count, 1)
+
+    X_train, X_test, y_train, y_test = train_test_split(
+        X, y,
+        test_size=TEST_SIZE,
+        random_state=RANDOM_STATE,
+        stratify=y,
+    )
+
+    X_val = X_test
+
+    scaler = StandardScaler()
+    X_train_scaled = scaler.fit_transform(X_train)
+    X_test_scaled = scaler.transform(X_test)
+
+    model = xgb.XGBClassifier(
+        n_estimators=N_ESTIMATORS,
+        max_depth=MAX_DEPTH,
+        learning_rate=LEARNING_RATE,
+        subsample=SUBSAMPLE,
+        colsample_bytree=COLSAMPLE,
+        scale_pos_weight=scale_pos_weight,
+        eval_metric='auc',
+        random_state=RANDOM_STATE,
+        tree_method='hist',
+    )
+
+    model.fit(
+        X_train_scaled, y_train,
+        eval_set=[(scaler.transform(X_val), y_test)],
+        verbose=False,
+    )
+
+    # Predictions
+    y_pred = model.predict(X_test_scaled)
+    y_proba = model.predict_proba(X_test_scaled)[:, 1]
+
+    fpr, tpr, thresholds = roc_curve(y_test, y_proba)
+
+    metrics = {
+        'auc_roc': roc_auc_score(y_test, y_proba),
+        'f1': f1_score(y_test, y_pred),
+        'precision': precision_score(y_test, y_pred),
+        'recall': recall_score(y_test, y_pred),
+        'confusion_matrix': confusion_matrix(y_test, y_pred),
+        'fpr': fpr,
+        'tpr': tpr,
+    }
+
+    # Save bundle
+    bundle = {
+        'model': model,
+        'scaler': scaler,
+        'feature_cols': FEATURE_COLS,
+    }
+    os.makedirs(os.path.dirname(MODEL_PATH), exist_ok=True)
+    with open(MODEL_PATH, 'wb') as f:
+        pickle.dump(bundle, f)
+
+    return model, scaler, metrics

src/visualiser/pyvis_graph.py

@@ -0,0 +1,183 @@
+"""
+PyVis Graph Visualiser
+Builds interactive network visualisations for account investigation.
+"""
+import json
+import tempfile
+import os
+import streamlit as st
+import streamlit.components.v1 as components
+import pandas as pd
+import networkx as nx
+from pyvis.network import Network
+
+# Constants
+NODE_SIZE_CENTER = 35
+NODE_SIZE_HIGH_RISK = 22
+NODE_SIZE_NORMAL = 14
+NODE_SIZE_HIGH_PAGERANK = 26
+EDGE_WIDTH_MAX = 8
+EDGE_WIDTH_SCALE = 1_000_000
+PAGERANK_TOP_PCT = 0.01
+PHYSICS_GRAVITY = -50
+PHYSICS_SPRING = 100
+PHYSICS_ITERATIONS = 150
+
+# Color palette
+COLOR_CENTER = '#FFD700'       # Gold for center node
+COLOR_FRAUD = '#FF4136'        # Red for confirmed fraud
+COLOR_HIGH_PR = '#FF851B'      # Orange for high PageRank
+COLOR_NORMAL = '#0074D9'       # Blue for normal
+COLOR_FRAUD_EDGE = '#FF4136'   # Red for fraudulent edges
+COLOR_NORMAL_EDGE = '#AAAAAA'  # Grey for normal edges
+
+PHYSICS_OPTIONS = json.dumps({
+    "nodes": {"borderWidth": 2, "shadow": True},
+    "edges": {
+        "smooth": {"type": "curvedCW", "roundness": 0.2},
+        "shadow": True,
+        "arrows": {"to": {"enabled": True, "scaleFactor": 0.8}},
+    },
+    "physics": {
+        "forceAtlas2Based": {
+            "gravitationalConstant": PHYSICS_GRAVITY,
+            "springLength": PHYSICS_SPRING,
+        },
+        "solver": "forceAtlas2Based",
+        "stabilization": {"iterations": PHYSICS_ITERATIONS},
+    },
+    "interaction": {"hover": True, "tooltipDelay": 100},
+})
+
+
+def build_pyvis_graph(
+    subgraph: nx.DiGraph,
+    df: pd.DataFrame,
+    center_node: str,
+    fraud_accounts: set,
+    pagerank_scores: dict,
+    louvain_partition: dict,
+) -> Network:
+    """
+    Build an interactive PyVis network from a NetworkX subgraph.
+
+    Nodes are sized and colored based on their role (center, fraud, high-PR, normal).
+    Edges are scaled by transaction amount and colored by fraud status.
+
+    Returns:
+        A pyvis.network.Network instance ready for rendering.
+    """
+    net = Network(
+        height='570px',
+        width='100%',
+        directed=True,
+        notebook=False,
+    )
+
+    net.set_options(PHYSICS_OPTIONS)
+
+    # Compute pagerank threshold for highlighting top nodes
+    all_pr = sorted(pagerank_scores.values(), reverse=True)
+    top_n = max(1, int(len(all_pr) * PAGERANK_TOP_PCT))
+    pagerank_threshold = all_pr[min(top_n, len(all_pr) - 1)]
+
+    # --- Add nodes ---
+    for node in subgraph.nodes():
+        node_data = subgraph.nodes[node]
+        is_center = (node == center_node)
+        is_fraud = node in fraud_accounts
+        is_high_pr = pagerank_scores.get(node, 0) >= pagerank_threshold
+
+        # Size and shape
+        if is_center:
+            size = NODE_SIZE_CENTER
+            shape = 'star'
+            color = COLOR_CENTER
+        elif is_high_pr:
+            size = NODE_SIZE_HIGH_PAGERANK
+            shape = 'diamond'
+            color = COLOR_HIGH_PR
+        elif is_fraud:
+            size = NODE_SIZE_HIGH_RISK
+            shape = 'dot'
+            color = COLOR_FRAUD
+        else:
+            size = NODE_SIZE_NORMAL
+            shape = 'dot'
+            color = COLOR_NORMAL
+
+        # Tooltip
+        total_sent = node_data.get('total_sent', 0)
+        total_received = node_data.get('total_received', 0)
+        count_sent = node_data.get('count_sent', 0)
+        community = louvain_partition.get(node, 'N/A')
+        status = 'FLAGGED' if is_fraud else 'Normal'
+
+        tooltip = (
+            f"Account: {node}\n"
+            f"Total Sent: {total_sent:,.0f}\n"
+            f"Total Received: {total_received:,.0f}\n"
+            f"Transactions: {count_sent}\n"
+            f"Community: {community}\n"
+            f"Status: {status}"
+        )
+
+        label = str(node)[:12]
+
+        net.add_node(
+            str(node),
+            label=label,
+            size=size,
+            shape=shape,
+            color=color,
+            title=tooltip,
+            borderWidth=2,
+            borderWidthSelected=4,
+        )
+
+    # --- Add edges ---
+    for src, tgt, data in subgraph.edges(data=True):
+        amount = data.get('amount', 0)
+        is_fraud_edge = data.get('is_laundering', 0) == 1
+
+        edge_width = min(amount / EDGE_WIDTH_SCALE, EDGE_WIDTH_MAX)
+        edge_width = max(edge_width, 0.5)
+
+        edge_color = COLOR_FRAUD_EDGE if is_fraud_edge else COLOR_NORMAL_EDGE
+
+        tooltip = (
+            f"Amount: {amount:,.0f}\n"
+            f"Channel: {data.get('payment_type', '')}\n"
+            f"Suspicious: {'Yes' if is_fraud_edge else 'No'}"
+        )
+
+        net.add_edge(
+            str(src),
+            str(tgt),
+            value=edge_width,
+            title=tooltip,
+            color=edge_color,
+            arrows='to',
+        )
+
+    return net
+
+
+def render_pyvis(net: Network) -> None:
+    """
+    Render a PyVis network inside a Streamlit app using an HTML component.
+    """
+    with tempfile.NamedTemporaryFile(delete=False, suffix='.html', mode='w') as f:
+        tmp_path = f.name
+    net.save_graph(tmp_path)
+    with open(tmp_path, 'r') as f:
+        html_content = f.read()
+    os.unlink(tmp_path)
+    components.html(html_content, height=580, scrolling=False)
+
+
+def save_pyvis_html(net: Network, path: str) -> None:
+    """
+    Save a PyVis network to an HTML file on disk.
+    """
+    net.save_graph(path)

test_ml.py

@@ -0,0 +1,69 @@
+"""
+Test the full ML pipeline: feature engineering, training, scoring, explainability.
+"""
+import warnings
+warnings.filterwarnings('ignore')
+import time
+
+from src.data_loader import get_processed_data
+import src.graph_builder as gb
+from src.ml.features import engineer_features
+from src.ml.trainer import train_model
+from src.ml.predictor import score_account
+from src.ml.explainer import explain_prediction
+import pickle
+
+print("Loading data...")
+df, nf = get_processed_data()
+print(f"Transactions: {len(df):,}, Node features: {len(nf):,}")
+
+print("\nBuilding graph...")
+G = gb.build_graph(df)
+G = gb.attach_node_features(G, nf)
+print(f"Graph: {G.number_of_nodes():,} nodes, {G.number_of_edges():,} edges")
+
+print("Computing PageRank...")
+pagerank_scores = gb.compute_pagerank(G)
+
+print("Computing Louvain...")
+t = time.time()
+louvain_partition = gb.compute_louvain(G)
+print(f"Louvain: {time.time()-t:.1f}s")
+
+# Use empty set for cycle_accounts (skip expensive cycle detection for test)
+cycle_accounts = set()
+
+print("\nEngineering features...")
+t = time.time()
+feature_df = engineer_features(df, G, pagerank_scores, louvain_partition, cycle_accounts)
+print(f"Features: {feature_df.shape} in {time.time()-t:.1f}s")
+print(f"Fraud flag distribution:\n{feature_df['fraud_flag'].value_counts()}")
+print(f"Feature columns: {list(feature_df.columns)}")
+
+print("\nTraining XGBoost model...")
+t = time.time()
+model, scaler, metrics = train_model(feature_df)
+print(f"Trained in {time.time()-t:.1f}s")
+print(f"\n--- METRICS ---")
+print(f"AUC-ROC   : {metrics['auc_roc']:.4f}  {'✅ >= 0.85' if metrics['auc_roc'] >= 0.85 else '❌ < 0.85'}")
+print(f"F1        : {metrics['f1']:.4f}")
+print(f"Precision : {metrics['precision']:.4f}")
+print(f"Recall    : {metrics['recall']:.4f}")
+print(f"Confusion Matrix:\n{metrics['confusion_matrix']}")
+
+print("\nLoading model from disk...")
+with open('models/xgb_fraud_model.pkl', 'rb') as f:
+    bundle = pickle.load(f)
+
+print("\nScoring 3 sample accounts...")
+sample_accounts = feature_df['account'].head(3).tolist()
+for acct in sample_accounts:
+    result = score_account(acct, feature_df, bundle)
+    print(f"  {acct}: risk_score={result['risk_score']}, fraud_prob={result['fraud_probability']:.4f}")
+
+print("\nSHAP explanation for first account...")
+explanation = explain_prediction(sample_accounts[0], feature_df, bundle)
+for e in explanation:
+    print(f"  {e['feature_name']:25s} SHAP={e['shap_value']:+.4f}  val={e['feature_value']:.2f}  ({e['direction']})")
+
+print("\n✅ ML pipeline fully verified.")

tests/graph_test_output.html

@@ -0,0 +1,127 @@
+<html>
+    <head>
+        <meta charset="utf-8">
+        
+            <script src="lib/bindings/utils.js"></script>
+            <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/vis-network/9.1.2/dist/dist/vis-network.min.css" integrity="sha512-WgxfT5LWjfszlPHXRmBWHkV2eceiWTOBvrKCNbdgDYTHrT2AeLCGbF4sZlZw3UMN3WtL0tGUoIAKsu8mllg/XA==" crossorigin="anonymous" referrerpolicy="no-referrer" />
+            <script src="https://cdnjs.cloudflare.com/ajax/libs/vis-network/9.1.2/dist/vis-network.min.js" integrity="sha512-LnvoEWDFrqGHlHmDD2101OrLcbsfkrzoSpvtSQtxK3RMnRV0eOkhhBN2dXHKRrUU8p2DGRTk35n4O8nWSVe1mQ==" crossorigin="anonymous" referrerpolicy="no-referrer"></script>
+            
+        
+<center>
+<h1></h1>
+</center>
+
+<!-- <link rel="stylesheet" href="../node_modules/vis/dist/vis.min.css" type="text/css" />
+<script type="text/javascript" src="../node_modules/vis/dist/vis.js"> </script>-->
+        <link
+          href="https://cdn.jsdelivr.net/npm/bootstrap@5.0.0-beta3/dist/css/bootstrap.min.css"
+          rel="stylesheet"
+          integrity="sha384-eOJMYsd53ii+scO/bJGFsiCZc+5NDVN2yr8+0RDqr0Ql0h+rP48ckxlpbzKgwra6"
+          crossorigin="anonymous"
+        />
+        <script
+          src="https://cdn.jsdelivr.net/npm/bootstrap@5.0.0-beta3/dist/js/bootstrap.bundle.min.js"
+          integrity="sha384-JEW9xMcG8R+pH31jmWH6WWP0WintQrMb4s7ZOdauHnUtxwoG2vI5DkLtS3qm9Ekf"
+          crossorigin="anonymous"
+        ></script>
+
+
+        <center>
+          <h1></h1>
+        </center>
+        <style type="text/css">
+
+             #mynetwork {
+                 width: 100%;
+                 height: 570px;
+                 background-color: #ffffff;
+                 border: 1px solid lightgray;
+                 position: relative;
+                 float: left;
+             }
+
+             
+
+             
+
+             
+        </style>
+    </head>
+
+
+    <body>
+        <div class="card" style="width: 100%">
+            
+            
+            <div id="mynetwork" class="card-body"></div>
+        </div>
+
+        
+        
+
+        <script type="text/javascript">
+
+              // initialize global variables.
+              var edges;
+              var nodes;
+              var allNodes;
+              var allEdges;
+              var nodeColors;
+              var originalNodes;
+              var network;
+              var container;
+              var options, data;
+              var filter = {
+                  item : '',
+                  property : '',
+                  value : []
+              };
+
+              
+
+              
+
+              // This method is responsible for drawing the graph, returns the drawn network
+              function drawGraph() {
+                  var container = document.getElementById('mynetwork');
+
+                  
+
+                  // parsing and collecting nodes and edges from the python
+                  nodes = new vis.DataSet([{"borderWidth": 2, "borderWidthSelected": 4, "color": "#FFD700", "id": "8101AEF90", "label": "8101AEF90", "shape": "star", "size": 35, "title": "Account: 8101AEF90\nTotal Sent: 30,734\nTotal Received: 1,965,570\nTransactions: 11.0\nCommunity: 2769\nStatus: FLAGGED"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "8105DEB80", "label": "8105DEB80", "shape": "dot", "size": 14, "title": "Account: 8105DEB80\nTotal Sent: 1,399\nTotal Received: 142,255\nTransactions: 2.0\nCommunity: 61635\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "810EAAC30", "label": "810EAAC30", "shape": "dot", "size": 14, "title": "Account: 810EAAC30\nTotal Sent: 65\nTotal Received: 2,572,429\nTransactions: 1.0\nCommunity: 41974\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "810518810", "label": "810518810", "shape": "dot", "size": 14, "title": "Account: 810518810\nTotal Sent: 6,630,785\nTotal Received: 316,695\nTransactions: 31.0\nCommunity: 302\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "8000D4D50", "label": "8000D4D50", "shape": "dot", "size": 14, "title": "Account: 8000D4D50\nTotal Sent: 9,756,656\nTotal Received: 755,453\nTransactions: 26.0\nCommunity: 15615\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "80D876800", "label": "80D876800", "shape": "dot", "size": 14, "title": "Account: 80D876800\nTotal Sent: 278,055\nTotal Received: 2,238,568\nTransactions: 34.0\nCommunity: 357\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#FF4136", "id": "80DF9EF10", "label": "80DF9EF10", "shape": "dot", "size": 22, "title": "Account: 80DF9EF10\nTotal Sent: 702,187\nTotal Received: 94,302\nTransactions: 17.0\nCommunity: 2769\nStatus: FLAGGED"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "810217F20", "label": "810217F20", "shape": "dot", "size": 14, "title": "Account: 810217F20\nTotal Sent: 0\nTotal Received: 4,076,387\nTransactions: 0.0\nCommunity: 302\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "80FFFF350", "label": "80FFFF350", "shape": "dot", "size": 14, "title": "Account: 80FFFF350\nTotal Sent: 2,397,697\nTotal Received: 16,532,279\nTransactions: 59.0\nCommunity: 2769\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "8000D2450", "label": "8000D2450", "shape": "dot", "size": 14, "title": "Account: 8000D2450\nTotal Sent: 19,950,045\nTotal Received: 49\nTransactions: 84.0\nCommunity: 61613\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "80FDF7EF0", "label": "80FDF7EF0", "shape": "dot", "size": 14, "title": "Account: 80FDF7EF0\nTotal Sent: 5,073,726\nTotal Received: 654,042\nTransactions: 44.0\nCommunity: 302\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "8005A7740", "label": "8005A7740", "shape": "dot", "size": 14, "title": "Account: 8005A7740\nTotal Sent: 21,627,411\nTotal Received: 1,623,296\nTransactions: 94.0\nCommunity: 364\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "811247290", "label": "811247290", "shape": "dot", "size": 14, "title": "Account: 811247290\nTotal Sent: 642,819\nTotal Received: 721,317\nTransactions: 4.0\nCommunity: 302\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "810F29690", "label": "810F29690", "shape": "dot", "size": 14, "title": "Account: 810F29690\nTotal Sent: 208,594\nTotal Received: 1,617,539\nTransactions: 19.0\nCommunity: 1895\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "8115CB540", "label": "8115CB540", "shape": "dot", "size": 14, "title": "Account: 8115CB540\nTotal Sent: 3,341,743\nTotal Received: 3,483,443\nTransactions: 2.0\nCommunity: 302\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "811279D10", "label": "811279D10", "shape": "dot", "size": 14, "title": "Account: 811279D10\nTotal Sent: 83,238\nTotal Received: 943,537\nTransactions: 8.0\nCommunity: 14817\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "811842A50", "label": "811842A50", "shape": "dot", "size": 14, "title": "Account: 811842A50\nTotal Sent: 5,830\nTotal Received: 18,576\nTransactions: 2.0\nCommunity: 61604\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "8106F24B0", "label": "8106F24B0", "shape": "dot", "size": 14, "title": "Account: 8106F24B0\nTotal Sent: 41,687,726\nTotal Received: 917,480\nTransactions: 15.0\nCommunity: 54823\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "810A1A5D0", "label": "810A1A5D0", "shape": "dot", "size": 14, "title": "Account: 810A1A5D0\nTotal Sent: 25,456\nTotal Received: 51,514\nTransactions: 1.0\nCommunity: 302\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "8100C70F0", "label": "8100C70F0", "shape": "dot", "size": 14, "title": "Account: 8100C70F0\nTotal Sent: 492,671\nTotal Received: 193,322\nTransactions: 38.0\nCommunity: 302\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "810BEA5B0", "label": "810BEA5B0", "shape": "dot", "size": 14, "title": "Account: 810BEA5B0\nTotal Sent: 435,216\nTotal Received: 7,052,896\nTransactions: 1.0\nCommunity: 2769\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "8114D3220", "label": "8114D3220", "shape": "dot", "size": 14, "title": "Account: 8114D3220\nTotal Sent: 69\nTotal Received: 217,291\nTransactions: 1.0\nCommunity: 1895\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "8105DE7A0", "label": "8105DE7A0", "shape": "dot", "size": 14, "title": "Account: 8105DE7A0\nTotal Sent: 271,174\nTotal Received: 113,753\nTransactions: 29.0\nCommunity: 61635\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "810F29870", "label": "810F29870", "shape": "dot", "size": 14, "title": "Account: 810F29870\nTotal Sent: 1,618,474\nTotal Received: 935\nTransactions: 3.0\nCommunity: 1895\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "81067F4C0", "label": "81067F4C0", "shape": "dot", "size": 14, "title": "Account: 81067F4C0\nTotal Sent: 863,027\nTotal Received: 796,781\nTransactions: 38.0\nCommunity: 302\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "804E53900", "label": "804E53900", "shape": "dot", "size": 14, "title": "Account: 804E53900\nTotal Sent: 386,927\nTotal Received: 13,704\nTransactions: 34.0\nCommunity: 673\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "810B4B330", "label": "810B4B330", "shape": "dot", "size": 14, "title": "Account: 810B4B330\nTotal Sent: 94,008\nTotal Received: 694,430\nTransactions: 2.0\nCommunity: 2769\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#0074D9", "id": "8108149D0", "label": "8108149D0", "shape": "dot", "size": 14, "title": "Account: 8108149D0\nTotal Sent: 5,273\nTotal Received: 18,751\nTransactions: 1.0\nCommunity: 2769\nStatus: Normal"}, {"borderWidth": 2, "borderWidthSelected": 4, "color": "#FF851B", "id": "1004289C0", "label": "1004289C0", "shape": "diamond", "size": 26, "title": "Account: 1004289C0\nTotal Sent: 10,517,190,141\nTotal Received: 313,581\nTransactions: 16794.0\nCommunity: 302\nStatus: FLAGGED"}]);
+                  edges = new vis.DataSet([{"arrows": "to", "color": "#AAAAAA", "from": "8101AEF90", "title": "Amount: 1,245\nChannel: Cheque\nSuspicious: No", "to": "8105DEB80", "value": 0.5}, {"arrows": "to", "color": "#FF4136", "from": "8101AEF90", "title": "Amount: 18,288\nChannel: ACH\nSuspicious: Yes", "to": "80DF9EF10", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "8105DEB80", "title": "Amount: 32\nChannel: Reinvestment\nSuspicious: No", "to": "8105DEB80", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "810EAAC30", "title": "Amount: 65\nChannel: Reinvestment\nSuspicious: No", "to": "810EAAC30", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "810518810", "title": "Amount: 103,085\nChannel: Reinvestment\nSuspicious: No", "to": "810518810", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "8000D4D50", "title": "Amount: 754,950\nChannel: Reinvestment\nSuspicious: No", "to": "8000D4D50", "value": 0.75494959}, {"arrows": "to", "color": "#AAAAAA", "from": "8000D4D50", "title": "Amount: 27\nChannel: Wire\nSuspicious: No", "to": "80FDF7EF0", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "80D876800", "title": "Amount: 29\nChannel: Credit Card\nSuspicious: No", "to": "80DF9EF10", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "80DF9EF10", "title": "Amount: 42,597\nChannel: Reinvestment\nSuspicious: No", "to": "80DF9EF10", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "80DF9EF10", "title": "Amount: 18,754\nChannel: Credit Card\nSuspicious: No", "to": "810B4B330", "value": 0.5}, {"arrows": "to", "color": "#FF4136", "from": "80DF9EF10", "title": "Amount: 65,262\nChannel: ACH\nSuspicious: Yes", "to": "810BEA5B0", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "80FFFF350", "title": "Amount: 102,125\nChannel: ACH\nSuspicious: No", "to": "8101AEF90", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "80FFFF350", "title": "Amount: 423,324\nChannel: Reinvestment\nSuspicious: No", "to": "80FFFF350", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "80FFFF350", "title": "Amount: 388\nChannel: Credit Card\nSuspicious: No", "to": "810518810", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "80FFFF350", "title": "Amount: 1,291\nChannel: Credit Card\nSuspicious: No", "to": "8108149D0", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "8000D2450", "title": "Amount: 49\nChannel: Reinvestment\nSuspicious: No", "to": "8000D2450", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "8000D2450", "title": "Amount: 257\nChannel: Wire\nSuspicious: No", "to": "80FDF7EF0", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "80FDF7EF0", "title": "Amount: 170,137\nChannel: Cash\nSuspicious: No", "to": "81067F4C0", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "80FDF7EF0", "title": "Amount: 559,189\nChannel: Reinvestment\nSuspicious: No", "to": "80FDF7EF0", "value": 0.5591893}, {"arrows": "to", "color": "#AAAAAA", "from": "80FDF7EF0", "title": "Amount: 182,883\nChannel: Wire\nSuspicious: No", "to": "810217F20", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "80FDF7EF0", "title": "Amount: 4,364\nChannel: Credit Card\nSuspicious: No", "to": "8101AEF90", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "80FDF7EF0", "title": "Amount: 548\nChannel: Wire\nSuspicious: No", "to": "810A1A5D0", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "8005A7740", "title": "Amount: 1,609,724\nChannel: Reinvestment\nSuspicious: No", "to": "8005A7740", "value": 1.60972417}, {"arrows": "to", "color": "#AAAAAA", "from": "8005A7740", "title": "Amount: 2,556\nChannel: Credit Card\nSuspicious: No", "to": "80DF9EF10", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "811247290", "title": "Amount: 370,069\nChannel: Reinvestment\nSuspicious: No", "to": "811247290", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "810F29690", "title": "Amount: 608\nChannel: Credit Card\nSuspicious: No", "to": "8100C70F0", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "810F29690", "title": "Amount: 29,084\nChannel: Cheque\nSuspicious: No", "to": "8114D3220", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "810F29690", "title": "Amount: 116\nChannel: Credit Card\nSuspicious: No", "to": "8101AEF90", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "810F29690", "title": "Amount: 548\nChannel: Credit Card\nSuspicious: No", "to": "811842A50", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "810F29690", "title": "Amount: 732\nChannel: Credit Card\nSuspicious: No", "to": "811247290", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "810F29690", "title": "Amount: 5\nChannel: Credit Card\nSuspicious: No", "to": "811279D10", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "810F29690", "title": "Amount: 79\nChannel: Credit Card\nSuspicious: No", "to": "8106F24B0", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "810F29690", "title": "Amount: 440\nChannel: Credit Card\nSuspicious: No", "to": "810EAAC30", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "810F29690", "title": "Amount: 27\nChannel: Credit Card\nSuspicious: No", "to": "8115CB540", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "8115CB540", "title": "Amount: 65\nChannel: Reinvestment\nSuspicious: No", "to": "8115CB540", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "811279D10", "title": "Amount: 31\nChannel: Reinvestment\nSuspicious: No", "to": "811279D10", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "811842A50", "title": "Amount: 65\nChannel: Reinvestment\nSuspicious: No", "to": "811842A50", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "8106F24B0", "title": "Amount: 692,939\nChannel: Reinvestment\nSuspicious: No", "to": "8106F24B0", "value": 0.69293866}, {"arrows": "to", "color": "#AAAAAA", "from": "810A1A5D0", "title": "Amount: 25,456\nChannel: Reinvestment\nSuspicious: No", "to": "810A1A5D0", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "8100C70F0", "title": "Amount: 21,509\nChannel: Reinvestment\nSuspicious: No", "to": "8100C70F0", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "810BEA5B0", "title": "Amount: 435,216\nChannel: Reinvestment\nSuspicious: No", "to": "810BEA5B0", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "8114D3220", "title": "Amount: 69\nChannel: Reinvestment\nSuspicious: No", "to": "8114D3220", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "8105DE7A0", "title": "Amount: 128,411\nChannel: ACH\nSuspicious: No", "to": "8105DEB80", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "810F29870", "title": "Amount: 37\nChannel: Reinvestment\nSuspicious: No", "to": "810F29870", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "810F29870", "title": "Amount: 1,617,539\nChannel: Cheque\nSuspicious: No", "to": "810F29690", "value": 1.617539}, {"arrows": "to", "color": "#AAAAAA", "from": "81067F4C0", "title": "Amount: 358,264\nChannel: Reinvestment\nSuspicious: No", "to": "81067F4C0", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "804E53900", "title": "Amount: 1,254\nChannel: Cheque\nSuspicious: No", "to": "80DF9EF10", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "804E53900", "title": "Amount: 276\nChannel: Reinvestment\nSuspicious: No", "to": "804E53900", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "810B4B330", "title": "Amount: 13\nChannel: Reinvestment\nSuspicious: No", "to": "810B4B330", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "8108149D0", "title": "Amount: 5,273\nChannel: Reinvestment\nSuspicious: No", "to": "8108149D0", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "1004289C0", "title": "Amount: 24,315\nChannel: Cheque\nSuspicious: No", "to": "811247290", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "1004289C0", "title": "Amount: 872\nChannel: Cash\nSuspicious: No", "to": "80FFFF350", "value": 0.5}, {"arrows": "to", "color": "#AAAAAA", "from": "1004289C0", "title": "Amount: 13,381\nChannel: Credit Card\nSuspicious: No", "to": "8100C70F0", "value": 0.5}]);
+
+                  nodeColors = {};
+                  allNodes = nodes.get({ returnType: "Object" });
+                  for (nodeId in allNodes) {
+                    nodeColors[nodeId] = allNodes[nodeId].color;
+                  }
+                  allEdges = edges.get({ returnType: "Object" });
+                  // adding nodes and edges to the graph
+                  data = {nodes: nodes, edges: edges};
+
+                  var options = {"nodes": {"borderWidth": 2, "shadow": true}, "edges": {"smooth": {"type": "curvedCW", "roundness": 0.2}, "shadow": true, "arrows": {"to": {"enabled": true, "scaleFactor": 0.8}}}, "physics": {"forceAtlas2Based": {"gravitationalConstant": -50, "springLength": 100}, "solver": "forceAtlas2Based", "stabilization": {"iterations": 150}}, "interaction": {"hover": true, "tooltipDelay": 100}};
+
+                  
+
+
+                  
+
+                  network = new vis.Network(container, data, options);
+
+                  
+
+                  
+
+                  
+
+
+                  
+
+                  return network;
+
+              }
+              drawGraph();
+        </script>
+    </body>
+</html>

tests/test_graph_visual.py

@@ -0,0 +1,66 @@
+"""
+Test the PyVis graph visualiser.
+Loads a sample, builds graph, picks a fraud account, and renders to HTML.
+"""
+import warnings
+warnings.filterwarnings('ignore')
+
+from src.data_loader import get_processed_data
+import src.graph_builder as gb
+from src.visualiser.pyvis_graph import build_pyvis_graph, save_pyvis_html
+
+OUTPUT_PATH = 'tests/graph_test_output.html'
+
+print("Loading data...")
+df, nf = get_processed_data()
+
+# Use a 1000-row sample for faster testing
+sample_df = df.sample(n=min(1000, len(df)), random_state=42).copy()
+print(f"Sample: {len(sample_df)} rows")
+
+print("Building graph from sample...")
+G = gb.build_graph(sample_df)
+G = gb.attach_node_features(G, nf)
+print(f"Sample graph: {G.number_of_nodes()} nodes, {G.number_of_edges()} edges")
+
+print("Computing PageRank on sample graph...")
+pagerank_scores = gb.compute_pagerank(G)
+
+print("Computing Louvain on sample graph...")
+import community as community_louvain
+G_undirected = G.to_undirected()
+louvain_partition = community_louvain.best_partition(G_undirected, weight='amount')
+
+# Pick a confirmed fraud account
+fraud_accounts = set(df[df['is_laundering'] == 1]['source'].values)
+fraud_in_sample = fraud_accounts & set(G.nodes())
+if fraud_in_sample:
+    center_node = list(fraud_in_sample)[0]
+else:
+    center_node = list(G.nodes())[0]
+print(f"Center node: {center_node} (fraud={center_node in fraud_accounts})")
+
+# Extract subgraph
+sub_G = gb.get_subgraph(G, center_node, hops=2, max_nodes=60)
+print(f"Subgraph: {sub_G.number_of_nodes()} nodes, {sub_G.number_of_edges()} edges")
+
+# Build PyVis graph
+print("Building PyVis graph...")
+net = build_pyvis_graph(
+    subgraph=sub_G,
+    df=sample_df,
+    center_node=center_node,
+    fraud_accounts=fraud_accounts,
+    pagerank_scores=pagerank_scores,
+    louvain_partition=louvain_partition,
+)
+
+# Save HTML
+save_pyvis_html(net, OUTPUT_PATH)
+print(f"\n✅ Graph built successfully with {sub_G.number_of_nodes()} nodes and {sub_G.number_of_edges()} edges")
+print(f"HTML saved to: {OUTPUT_PATH}")
+
+# Verify file exists and has content
+import os
+size = os.path.getsize(OUTPUT_PATH)
+print(f"HTML file size: {size:,} bytes")