La media micro es un método estadístico comúnmente empleado en clasificación multiclase tasks within the field of Inteligencia Artificial and Aprendizaje Automático. It provides a way to assess the performance of a classification model by considering all classes together, rather than calculating metrics for each class separately. This approach is particularly useful when dealing with conjuntos de datos desequilibrados donde algunas clases pueden tener significativamente más muestras que otras.
In micro-averaging, the true positives, false positives, and false negatives are summed across all classes before calculating the overall precision, recall, or F1 score. This means that each instance contributes equally to the final metric, regardless of which class it belongs to. The formula for micro-average precision (P) and recall (R) can be expressed as:
- Precisión en media micro: P = (Suma de Verdaderos Positivos) / (Suma de Verdaderos Positivos + Suma de Falsos Positivos)
- Recall en media micro: R = (Suma de Verdaderos Positivos) / (Suma de Verdaderos Positivos + Suma de Falsos Negativos)
Micro-average is particularly advantageous in scenarios where the focus is on overall model performance rather than on individual class performance. This can help in providing a more holistic view of a model’s capabilities, especially in applications such as clasificación de imágenes, text categorization, and speech recognition.