In recent years, the innovation of service products in the e-commerce Red Sea area is still facing challenges due to their own content heterogeneity and indivisibility. Identifying innovation based on the characteristics of service components can help companies improve product competitiveness in increasingly intense market competition. In this paper, we present a new analytic framework to identify service innovation of tour routes that tackles the challenges of product similarity measurement. A knowledge graph-based method of service products is developed to obtain the dynamic weight of each dimension, and aiming to calculate the importance of graph network nodes through entity definition and attribute extraction after dimension division. Furthermore, we construct a small-scale knowledge graph of tour routes in the Ctrip. The results show the effectiveness of the proposed approach in terms of accurately identifying the level of service innovation, and provide a benchmark for exploring the difference in product unstructured attributes.