论文部分内容阅读
为了提高情报分发的效率,解决雷达组网上信息过载的问题,提出了一种基于朴素贝叶斯分类算法的雷达情报按需分发技术。利用层次向量空间构建用户兴趣空间,对情报用户的历史情报和定制信息,通过朴素贝叶斯分类算法挖掘用户兴趣,建立用户兴趣模型;通过实时情报与用户兴趣模型的匹配,将情报用户感兴趣的情报推送给用户,过滤其不感兴趣的情报,从而实现雷达情报的按需分发。仿真实验将该技术与基于TF-IDF分类算法的情报分发技术做了准确率与覆盖率的对比实验,结果表明,该方法的准确率优于利用TF-IDF分类算法的情报分发技术,能够较好地实现雷达情报的按需分发。
In order to improve the efficiency of intelligence distribution and solve the problem of information overload in radar network, a technology of on-demand radar information distribution based on naive Bayesian classification algorithm is proposed. The user space of interest is constructed by using the hierarchical vector space. The user’s historical intelligence and customized information are obtained by using the naive Bayesian classification algorithm to mine the user’s interest and establish the user’s interest model. The intelligence users are also interested in the match between the real-time intelligence and the user’s interest model Of the intelligence pushed to the user, filtering the information they are not interested in, so as to achieve on-demand radar intelligence distribution. The simulation experiment compares the accuracy of the technology with the intelligence distribution technology based on the TF-IDF classification algorithm. The results show that the accuracy of the method is better than the intelligence distribution technology using the TF-IDF classification algorithm, which can compare Achieve on-demand distribution of radar intelligence.