Incorporating topic and property for knowledge base synchronization

Open-domain knowledge bases have been widely used in many applications, and it is critical to maintain their freshness. Most existing studies update an open knowledge base by predicting the change frequencies of the entities and then updating those unstable ones. However, in the knowledge base, there are various entities and properties with complex structural information, and many entities are time-sensitive. In this work, we propose a novel topic-aware entity stability prediction framework which incorporates property and topic features of the entities to help efficiently update the knowledge base. To deal with the complex entity structure and various entity properties, we first build an entity property graph for each entity, with its property names as edges and property values as nodes. Then, with the constructed entity property graph, we analyze the topic information of the entities and propose a topic classifier via unsupervised clustering to further improve the accuracy of prediction. To tackle the time-sensitive challenge, we measure the monthly average update frequency of the entity, based on its revision history acquired from the source encyclopedia webpage, as the basis for labeling its stability. Finally, we formulate the prediction task as a binary classification problem and solve it with an entity stability predictor, wherein the topic information serves as strong supervision. Extensive experiments on collections of real-world entities have demonstrated the superior performance of our proposed method and also well shown the benefits of each new module in our framework.