Abstract
With the fast development of industrialization and urbanization, heavy metal pollution caused by activities of people is becoming more and more serious, which is a vast challenge for protection arable land and food safety. To address this challenge, a hierarchical clustering algorithm is proposed for the source apportionment of heavy metals in soils combined with the domain knowledge of agricultural experts. In detail, a hierarchical clustering using a new similarity computing method is proposed for the analysis of land heavy metal data. And a heavy metal source table is established according to the expertise of land experts, which can make it easy to find the source of heavy metal pollution based on the clustering results. Furthermore, the practical feasibility of corresponding methods is validated by actual data. The results indicate that the proposed method can be a potential approach to protecting the land from heavy metal pollution.