基于改进TFN-AHP的微博用户属性特征向量提取算法研究
An improved TFN-AHP vector extraction algorithm for determining the microblog users' attributes
云南民族大学学报:自然科学版,2016,25(4):354-360

黎吾鑫 LWX

摘要


在对传统TFN-AHP算法进行研究的基础上,针对其中判断矩阵标度选择不合理、计算过程中会出现错误以及各属性评价指标权重差异小不易区分和排序的问题,提出了改进的TFN-AHP算法,该算法通过构造模糊精度矩阵和采用闭区间[0,1]的实数作为模糊判断矩阵标度值,避免了传统TFN-AHP算法中将某一属性特征权重武断判定为0的错误,同时使用可控迭代精度的迭代方法计算特征向量,使各属性权重间有较好的区分性,有利于各属性的重要性排序,并基于该算法提取了微博用户属性特征向量. This paper gives a detailed analysis of the traditional TFN-AHP algorithm, and points out its defects like the unreasonable choices of the scale value of the fuzzy judgment matrix, possible errors in the calculation process and the difficulty in distinguishing the eigenvectors due to the insignificant weight differences of the assessment indexes for the relevant attributes. This paper brings up an improved TFN-AHP algorithm by constructing a fuzzy precision matrix and using the real number within the closed interval [0,1] as the scale value of the fuzzy judgment matrix to avoid some errors like the judgment of some attribute as 0 in the traditional TFN-AHP algorithm. More importantly, by using the alterable accuracy iteration method to calculate the eigenvectors of the fuzzy judgment matrix, the results of the eigenvectors are distinguishable from each another which would help determine the degree of importance of different users’ attributes.

参考



全文: PDF      下载: 451      浏览: 163


counter for myspace
云南民族大学学报(自然科学版) 1991—2016 Copyright
地址:云南省昆明市一二.一大街134号 邮编:650031 全国邮发代号:64-47
电话:0871-65132114 传真:0871-65137493 Email:ynmzxyxb@163.com