16Fibrosis is a key component in the pathogenic mechanism of many diseases. These 17 diseases involving fibrosis may share common mechanisms, therapeutic targets and 18 therefore, common intervention strategies and medicines may be applicable for these 19 diseases. For this reason, deliberately introducing anti-fibrosis characteristics into 20 modelling may lead to more success in drug repositioning. In this study, anti-fibrosis 21 knowledge base was first built by collecting data from multiple resources. Both 22 structural and biological profiles were derived from the knowledge base and used for 23 constructing machine learning models including Structural Profile Prediction Model 24 (SPPM) and Biological Profile Prediction Model (BPPM). Three external public data 25 sets were employed for validation purpose and further exploration of potential 26 repositioning drugs in wider chemical space. The resulting SPPM and BPPM models 27 achieve area under the receiver operating characteristic curve (AUC) of 0.879 and 28 0.972 in the training set, and 0.814 and 0.874 in the testing set. Additionally, our 29 results also demonstrate that substantial amount of multi-targeting natural products 30 possess notable anti-fibrosis characteristics and might serve as encouraging candidates 31 in fibrosis treatment and drug repositioning. To leverage our methodology and 32 findings, we developed repositioning prediction platform, Drug Repositioning based 33 on Anti-Fibrosis Characteristic (Dr AFC) that is freely accessible via 34 https://www.biosino.org/drafc. 35 36