What is the consequence of using unfiltered data for model training?

Using unfiltered data for model training may result in biased and inaccurate outcomes due to several factors. Unfiltered data often contains noise, outliers, and irrelevant information that can skew the model's learning process. Moreover, if the data set includes historical biases or reflects systemic issues, the model can learn and perpetuate these biases, leading to unfair or incorrect predictions.

By utilizing unfiltered data, the model might capture misleading trends rather than the true relationships within the data, ultimately impairing its performance. Filtering and preprocessing data help ensure that only relevant and clean data points are used in training, which is essential for building robust predictive models. Therefore, the consequence of using unfiltered data can undermine the integrity and effectiveness of the model, making it critical to apply proper data curation techniques before training.