Once you have base words, apply rules that reflect real user behavior:

| Mistake | Why it fails | Solution | | :--- | :--- | :--- | | | "Rapariga" means girl in PT; in Brazil, it is offensive slang. Users avoid it. | Separate wordlists for PT-PT and PT-BR. | | Ignoringão õe | The nasal diphthongs are extremely common (mão, coração, pão). | Generate numeric replacements: p4o , c0r4c40 . | | Forgetting compound words | English uses spaces (birthday cake). Portuguese uses hyphens or merging (beija-flor). | Use sed 's/ /-/g' to create hyphen variants. |

Run the following command to extract unique Portuguese words from your corpus:

: Localized lists emerged to capture the specific nuances of Brazilian Portuguese vs. European Portuguese, ensuring that common regionalisms were included. The Evolution: Contextual Permutations