On Instant Knowledge Evolution from Learning User Search Intent

In this paper, we explore a novel problem to identify timely new knowledge triples. In the literature, the need of knowledge enrichment has been recognized as the key to the success of semantic search. However, previous work of automatic knowledge extraction, such as Google Knowledge Vault, aim at identifying the unannotated knowledge triples from the full web-scale content in the offline execution. In our study, we show that most people demand the updated knowledge soon after the information is announced. However, the number of queries of such knowledge dramatically declines after a few days, meaning that the most people cannot obtain the precise knowledge from the execution of the offline knowledge enrichment when they inquire the answer of the timely events. To remedy this, we propose the SCKE framework to extract new knowledge triples which can be executed in the online scenario. We model the 'Query-Click' bipartite graph to extract the query correlation and to identify temporally coexistent entity pairs. Our experimental studies show that new triples can also be identified effectively and efficiently.