["International Journal of Consumer Studies, Volume 50, Issue 3, May 2026. ", "\nABSTRACT\nHouseholds account for over half of total food waste in Europe, with everyday food‐management practices playing a central role in its generation. Targeted interventions at the consumption stage are therefore critical for reducing food waste and achieving Sustainable Development Goal 12.3. This study applies machine‐learning techniques to identify the behavioral determinants of household food waste using survey data from 1536 Czech households. Among the tested models, Random Forest achieved the highest predictive accuracy (59.1%), enabling the classification of households based on waste‐related behaviors. Households with minimal food waste typically practiced regular and planned shopping, relied on shopping lists, monitored food storage, understood expiration labels, and were price‐sensitive. In contrast, factors such as purchase location, food appearance, or taste preferences showed no predictive value. These findings highlight that not all behavioral drivers contribute equally to household food waste. The results offer policy‐relevant insights into designing interventions: promoting informed shopping, targeting less price‐sensitive consumers, and improving storage knowledge are key leverage points. Beyond its methodological contribution, the study advances the conceptual understanding of household food waste by showing that waste behavior is structured primarily around planning capacity, perceived behavioral control, stock management, and routine food‐management practices, whereas several commonly assumed factors have limited predictive relevance.\n"]