该楼层疑似违规已被系统折叠 隐藏此楼查看此楼
从网上找到的编辑距离计算相似度的算法:
function [V,m,n] = EditDist(string1,string2)
% Edit Distance is a standard Dynamic Programming problem. Given two strings s1 and s2, the edit distance between s1 and s2 is the minimum number of operations required to convert string s1 to s2. The following operations are typically used:
% Replacing one character of string by another character.
% Deleting a character from string
% Adding a character to string
% Example:
% s1='article'
% s2='ardipo'
% EditDistance(s1,s2)
% > 4
% you need to do 4 actions to convert s1 to s2
% replace(t,d) , replace(c,p) , replace(l,o) , delete(e)
% using the other output, you can see the matrix solution to this problem
%
%
% by : Reza Ahmadzadeh (seyedreza_ahmadzadeh@yahoo.com - reza.ahmadzadeh@iit.it)
% 14-11-2012
m=length(string1);
n=length(string2);
v=zeros(m+1,n+1);
for i=1:1:m
v(i+1,1)=i;
end
for j=1:1:n
v(1,j+1)=j;
end
for i=1:m
for j=1:n
if (string1(i) == string2(j))
v(i+1,j+1)=v(i,j);
else
v(i+1,j+1)=1+min(min(v(i+1,j),v(i,j+1)),v(i,j));
end
end
end
V=v(m+1,n+1);
end
然后根据最小编辑距离,求两个字符串的相似度
[mindist m n]=EditDist(final_code,final_code2);
1-mindist/max(m,n )
这个算法如何修改可以实现字符串组的遍历并且输出成矩阵呢?