摘要
为解决用户搜索意图识别问题,提出了将改进型PSO-LSSVM算法用于搜索引擎分类意图识别,在传统的搜索引擎架构上加入了用户搜索意图识别模块以及所需的语料特征库。对PSO算法的局部搜索与全局搜索两个方面进行了改进,用于用户搜索意图识别处理器的参数设置;对搜索意图识别器训练集的获取和处理等问题做了相关讨论。通过实验对该方法以及理论架构进行验证,验证结果表明,相比普通的搜索引擎,基于该架构的搜索引擎在一定程度上较好地预测出用户的搜索意图,对用户的搜索意图识别问题研究具有启示意义。
To address the problem of users＇ search intention recognition, two methods are presented： an improved PSO-LSSVM algorithm is used to solve the problem and the recognition model of users＇ intention together with the feature of corpus needed is added to the traditional search engine architecture. The local and global searches of PSO algorithm are improved to set the process of users＇ search intention parameters. Some related discussion on obtaining and processing the training sets is made. An experiment is conducted to validate the improved PSO-LSSVM and architecture of the search engine＇s theory. The results show that the search engine based on this architecture, compared with normal search engines, according to the experiment result, has better effects on predicting users＇ search intention, and has enlightening significance to the future work in search engine.
出处
《计算机工程与设计》
CSCD
北大核心
2014年第4期1285-1292,共8页
Computer Engineering and Design
基金
贵州省工业攻关基金项目(黔科合GY字[2008]3009)
贵州省科学技术基金项目(黔科合J字[2011]2213)
贵州师范大学2012年度自然科学类学生科研基金重点基金项目(201219)
关键词
搜索引擎
粒子群优化算法
最小二乘支持向量机
搜索引擎架构
搜索意图
search engine
particle swarm optimization
least squares support vector machine
architecture of search engine
search intention
