大数据中用户特征信息优化定位检测仿真

Simulation and Optimization of User Characteristic Information in Large Data

对大数据中用户特征信息定位检测,能够有效提高大数据中用户搜索效率。对特征信息的定位检测,需要先给出极限学习机的目标函数,进而优化相关参数,完成用户特征信息定位检测。传统方法利用差分图像像素行、列的粗糙度特征,引入彩色梯度的方向信息,计算梯度幅值,但忽略了优化相关参数,导致检测精度偏低。提出基于捕鱼算法优化结合极限学习机的用户特征信息定位检测方法,给出单独应用极限学习机用户特征信息定位检测的目标函数,介绍了捕鱼算法中转移搜索、缩面搜索和增速搜索的应用条件和优化函数,通过捕鱼算法出发对极限学习机进行参数优化,设计优化流程。对用户特征信息挖掘进行归一化处理并制定权值比例,最贴近现实的用户特征信息定位检测结果。实验结果表明,所提方法在单步定位中用户特征信息挖掘深入,具有很高的使用价值。

A location detection method of user feature information is proposed based on fishing algorithm optimiza- tion integrated with extreme learning machine. Firstly, objective function of the location detection applying the ex- treme learning machine alone is provided. Then, application condition and optimization function of transfer search, shrinking face search and speed - up search in fishing algorithm are introduced. Parameter optimization is carried out for the extreme learning machine via the fishing algorithm to design optimization process. Finally, normalization pro- cessing is carried out for mining of the user feature information, and the location detection results of the user which is closest to reality is formulated. Simulation results show that the proposed method can mine the user feature informa- tion deeply in single step location.