Mobile automatic question and answer (QA) system ismore and more popular as a mobile value-addedbusiness. But the current mobile QA systems always havelow query precision and recall ratio. In this paper, a newalgorithm by pattern matching in p2p mobile QA systemanswer extracting process is designed. It expresses thedocument as triad and extracts the question and answersentence from the triad based documents. Simulationexperimental analysis results show that under idealconditions the system’s average responding time is lessthan 58ms and precision is above 66%, and the recallratio is upper than 52% when there are no more than3000 simultaneous query nodes, and so the completenessand feasibility of the system’s design is proved.
Index Terms:
automatic question and answer, mobile, pattern match, answer extract, p2p
Citation:
Lei Wang, Shulin Chen, "A Pattern Matching Algorithm in P2P Mobile QA System Answer Extracting Process," isip, pp.214-218, 2008 International Symposiums on Information Processing, 2008