简单实现根据Td-idf实现语句相似度

使用词频来计算两个语句的相似度，假设语句已经使用分词工具分好

public class Similar {

public static double getSimilarity(Vector<String> T1, Vector<String> T2) throws Exception {
int size = 0 , size2 = 0 ;
double sum=0;
if ( T1 != null && ( size = T1.size() ) > 0 && T2 != null && ( size2 = T2.size() ) > 0 ) {

Set<String> set=new HashSet<String>();
Map<String, Integer> s1 = new HashMap<String, Integer>();
Map<String, Integer> s2= new HashMap<String, Integer>();
//获取T1，T2所有的不重复单词
for (int i = 0; i < size; i++) {
set.add(T1.get(i));
}
for(int j=0;j<size2;j++){
set.add(T2.get(j));

}
//计算每个向量中的词语出现的次数
for (String string : set) {
int count=0;
int count1=0;
for (String a1 : T1) {
if (a1.equals(string)) {
count++;
}
}
for (String a2 : T2) {
if (a2.equals(string)) {
count1++;
}
}
s1.put(string, count);
s2.put(string, count1);
}

//计算百分比
int c1 = 0 , c2 = 0;
double sum1 = 0; //S1、S2
double b1=0,b2=0;
for (String string : set) {
c1=s1.get(string);
c2=s2.get(string);
sum1+=c1*c2;
b1+=c1*c1;
b2+=c2*c2;

}
sum=sum1/Math.sqrt(b1*b2);
}else {
System.out.println("参数有误");
}
return sum;

}
public static void main(String[] args) {
Vector<String> T1 = new Vector<String>() ;//对输入进行分词
Vector<String> T2 = new Vector<String>() ;//对输入进行分词

T1.add("i");
T1.add("kill");
T1.add("you");
T1.add("kill");
T1.add("you");
T1.add("fop");
T1.add("tip");
T1.add("cnce");

T2.add("kill");
T2.add("you");

try {
System.out.println(getSimilarity(T1, T2));
} catch (Exception e) {
// TODO Auto-generated catch block
e.printStackTrace();
}

}

}