Waste sorting and recycling is an important part of achieving sustainable development, relying on the good environmental awareness of community residents. Residents’ environmental awareness determines the preference choice of garbage classification scheme, and the ambiguous preference of the group directly affects the consensus effect of the scheme. Social network analysis can effectively solve the decision-making consensus problem of community group garbage classification scheme, which makes the research on the large-group decision-making (LGDM) consensus problem for social network has important practical significance. The research proposes an interactive consensus model based on social network consensus: Firstly, according to the social relationship between decision makers, the social network formed by decision makers is clustered using methods. Second, the preference of the categorical subgroup is represented based on the probability distribution of hesitant fuzzy elements. Then calculate the consensus level. Finally, the unreasonable decision-making level is improved through the interactive feedback mechanism, and the decision-making consensus of the garbage classification scheme is gradually reached. Considering that the choice of garbage classification scheme is actually a large group decision-making problem, the advantages and reliability of social network consensus analysis in selecting garbage classification schemes are verified through the case analysis of four types of garbage classification scheme selection.