Zynoox-IA/latam-slang-telemetry

Dataset Details

Dataset Sources

• Repository: https://huggingface.co/datasets/Zynoox-IA/latam-slang-telemetry

Uses

Direct Use

Ideal for fine-tuning LLMs, conversational agents, virtual assistants, or roleplay characters to sound organic and native in Latin American centennial spaces.

Out-of-Scope Use

This dataset should not be used for automated profiling of specific users, as all telemetry has been strictly anonymized and aggregated at a token-frequency level.

Dataset Structure

Each entry in the .jsonl file follows this production schema:

{
  "context": "discord_community_chat",
  "token_learned": "wenaas",
  "total_user_hits": 408,
  "timestamp_extracted": "2026-07-06"
}

Fields Matrix:

• context: (String) The origin environment or data stream.
• token_learned: (String) The specific linguistic token, slang, variation, or emoji captured.
• total_user_hits: (Integer) Cumulative frequency of use within the monitored environment (indicates token velocity/strength).
• timestamp_extracted: (String) Extraction and log dump date (YYYY-MM-DD).

Dataset Creation

Curation Rationale

To map the organic evolution of internet culture, centennial slang, gaming terminology, and niche community subcultures in Latin America by filtering raw chat streams into actionable frequency matrices. # Bias, Risks, and Limitations The dataset reflects informal internet language and gaming subcultures. It contains colloquial expressions, deliberate misspellings, and platform-specific emojis that may not align with standard or formal grammatical Spanish. # Dataset Card Contact Contact directly via Hugging Face Organization or open an Issue in this repository for full production dumps, commercial licensing, or custom data pipelines.

Detalles del Dataset (Español)

Descripción del Proyecto

Este dataset contiene telemetría lingüística automatizada, anonimizada y en tiempo real capturada de comunidades digitales activas en América Latina. Mapea la evolución orgánica de la cultura de internet, jergas centennials, terminología de videojuegos y subculturas de nicho.

• Propietario: Gustavo Enrique Ramos Ruiz / Zynoox IA
• Formato de Datos: JSON Lines (.jsonl)
• Frecuencia de Actualización: Automatizado cada 24 horas
• Contexto: Salas de chat en línea e interacciones comunitarias

Casos de Uso y Aplicaciones

1. Fine-Tuning de LLMs: Entrenar agentes conversacionales o modelos de lenguaje para usar modismos nativos de Latinoamérica.
2. Moderación de Contenido Avanzada: Identificar variaciones ortográficas de bypass y vocabulario emergente.
3. Investigación Lingüística: Análisis del ciclo de vida de los memes en la demografía juvenil.

# Dataset Card Contact

For technical inquiries, commercial licensing, custom data pipelines, or data partnerships, please contact:

- **Email:** [zynooxconnect@outlook.com]