Crowd sourced and Entity Resolution has recently attracted significant attentions because it can harness the wisdom of crowd to improve the quality of Entity Resolution. Entity Resolution can be defined as the process of identifying, matching, verifying accuracy and merging metadata that correspond to the same entities from several databases. Two main issues have been identified for crowd sourced Entity Resolution: data, relation harvesting and integration, and named Entity Resolution. In this paper, we address the issue of data and metadata integration from multi-sources. We propose a new semantic approach of data integration, called SMESE Trusted Smart Harvesting Algorithm based on Semantic Relationship and Social Network (SMESE-TSHA). SMESE-TSHA is based on efficient Semantic Harvesting Strategies (SHS)addresses the problem of performing Entity Resolution (MLM-TSHA) using trusted and ranked sources.SHS addresses the problem of semantic harvesting based on authority file sources, sources classification model and the data graph model nodes exploration patterns while MLM-TSHA addresses the problem of performing Entity Resolution on RDF graphs containing multiple types of nodes. We experimentally evaluate our SMES-TSHA approach on large real datasets and compare the performance results with existing approaches. Our experimental results show our proposed models perform well on the Entity Resolution compared to the existing approaches, while also satisfying the running time restrictions.
