preprocessing/step5_metadata_consistency_check.py

import pandas as pd

DATA_PATH = "dataset/processed/dataset_step2_cleaned.csv"

def main():
    df = pd.read_csv(DATA_PATH)

    # -------------------------------
    # Handle Label column safely
    # -------------------------------
    if "Label" not in df.columns:
        if "Label (0- HUMAN, 1-AI)" in df.columns:
            df = df.rename(columns={"Label (0- HUMAN, 1-AI)": "Label"})
        else:
            print("⚠️ Label column not found. Skipping label-based checks.")

    # -------------------------------
    # Normalize Language column
    # -------------------------------
    if "Language" in df.columns:
        df["Language"] = df["Language"].astype(str).str.strip().str.lower()

    print("\n🧩 METADATA CONSISTENCY CHECK\n")

    # -------------------------------
    # Language consistency
    # -------------------------------
    print("🌐 Language distribution:")
    if "Language" in df.columns:
        print(df["Language"].value_counts())
    else:
        print("  Language column not found")

    # -------------------------------
    # Label vs Source_Type consistency
    # -------------------------------
    print("\n🏷️ Label vs Source_Type consistency:")
    if "Label" in df.columns and "Source_Type" in df.columns:
        cross_tab = pd.crosstab(df["Label"], df["Source_Type"])
        print(cross_tab)

        print("\nExpected behavior:")
        print("  Label 0 (Human) → Source_Type should be 'human'")
        print("  Label 1 (AI)    → Source_Type should be 'ai'")

    else:
        print("  Required columns not found for consistency check")

    print("\nMetadata consistency check completed ✅")

if __name__ == "__main__":
    main()