摘要
针对当前已有方法应用于搜索浏览设计资源占用量大、效率低的问题,研究提出了一种基于大宗日志数据、支持多个复杂项目搜索的粒子群优化算法,该方法采用非线性惯性权重,设计了两个卷积神经网络模块,构建粒子质点,利用该质点在伪逻辑数据库形成的粒子空间中移动到最佳位置,形成粒子群搜索结果,采用六维数据空间下的距离函数进行数据处理,测试了该算法下的搜索效率及搜索耦合度,具有较高的搜索精准度与敏感度。实验验证该方法对文本数据、图片、音频识别敏感度有所增强,用户体验得到提升。
Regarding the problems of large resource consumption and low efficiency in the application of existing methods in search and browsing design,a particle swarm optimization algorithm based on bulk log data and supporting multiple complex item searches is proposed.Two convolutional neural network modules are designed based on nonlinear inertial weight to construct particle particles,which are used to move to the best position in the particle space formed by the pseudo-logical database to form particle swarm search results,and the distance function in the six-dimensional data space is used for data processing.The search efficiency and search coupling degree of the algorithm are tested,showing that it has high search accuracy and sensitivity.Experiment verification shows that the method has enhanced sensitivity to text data,pictures and audio recognition,and improved user experience.
作者
梁光瑞
巩志强
王宁
高尚建
魏国
LIANG Guang-rui;GONG Zhi-qiang;WANG Ning;GAO Shang-jian;WEI Guo(CNOOC Energy Development Equipment Technology Co.,Ltd.,Tianjin 300452,China)
出处
《信息技术》
2022年第7期142-146,153,共6页
Information Technology
关键词
搜索引擎
粒子群算法
六维逻辑空间
卷积神经网络
异构化数据
search engine
particle swarm algorithm
six-dimensional logical space
convolutional neural network
heterogeneous data
