The authors highlight the integration of Collaborative Filtering (CF) as a preprocessing step to address missing values and filter discrete data, enhancing the effectiveness of SOM training. This combined approach allows for deeper insights into car rejection reasons, including their temporal relationships and dependencies.