高相似度英语词语自主选取系统设计

发布时间：2018-07-04 12:07

本文选题：高相似度英语词语 + 数据挖掘　；参考：《现代电子技术》2017年23期

【摘要】：针对传统的基于支持向量机的高相似度英语词语自主选取系统一直存在选取效果差、精度低的问题,提出一种基于数据挖掘的高相似度英语词语自主选取系统设计方法。首先根据英语词语的相似度概念,计算出两个英语词语义项的最短路径与其距离最近的公共父节点之间的深度。利用数据挖掘法将英语词语文本特征选择转换为一个多目标优化问题;然后以英语词语特征维数最少、分类正确率相对最高为选取标准,采用蚁群算法找到英语词语的最优特征子集;最后通过建立神经网络分类器完成高相似度英语词语自主选取系统设计。实验结果证明,所提方法可以精确地选取出高相似度英语词语,且选取时间较短,实用性广泛。
[Abstract]:Aiming at the problems of poor selection effect and low precision in the traditional high similarity English word autonomous selection system based on support vector machine, a design method of high similarity English word autonomous selection system based on data mining is proposed. Firstly, according to the concept of English word similarity, the depth between the shortest path of the meaning of two English words and the nearest common parent node is calculated. The method of data mining is used to transform the feature selection of English words into a multi-objective optimization problem, and then the feature dimension of English words is the least, and the classification accuracy is the highest. Ant colony algorithm is used to find the optimal feature subset of English words. Finally, a neural network classifier is established to design the autonomous selection system for English words with high similarity. The experimental results show that the proposed method can accurately select English words with high similarity, and the selection time is shorter, and the method has wide practicability.
【作者单位】：西安建筑科技大学;
【分类号】：TP391.1

【相似文献】