Applying language modeling to session identification from database trace logs

作者:Xiangji Huang, Qingsong Yao, Aijun An

摘要

A database session is a sequence of requests presented to the database system by a user or an application to achieve a certain task. Session identification is an important step in discovering useful patterns from database trace logs. The discovered patterns can be used to improve the performance of database systems by prefetching predicted queries, rewriting the current query or conducting effective cache replacement.

论文关键词:Statistical language modeling, Session identification, Database trace logs

论文评审过程:

论文官网地址:https://doi.org/10.1007/s10115-006-0015-9