步骤:
(1)读取3篇txt格式的英文文档;
(2)对每篇文档进行分词;
(3)利用正则表达式去掉逗号、句号等标点符号;
(4)计算文档频率;
(5)统计特征向量<?php
$_txts = array('1.txt','2.txt','3.txt');
$_len = count($_txts);
for ($i = 0;$i
$_contents[] = file_get_contents($_txts[$i]);
}
for ($i = 0;$i
//分词
$_words[] = explode(' ',trim($_contents[$i]));
foreach ($_words[$i] as $_key=>$_value){
$_value = trim($_value);
$_value = preg_replace('/[.|,|(|)|-|;]/','',$_value);
$_words[$i][$_key]=strtolower($_value);
}
//去重
$_words[$i]= array_unique($_words[$i]);
}
//合并
$_words_com = array_merge($_words[0],$_words[1],$_words[2]);
//文档频率
$_df = array_count_values($_words_com);
//特征向量
for ($i = 0;$i
$_vsm[$i] = $_df;
foreach($_vsm[$i] as $_key=>$_value){
$_vsm[$i][$_key] = 0;
}
for ($j=0;$j
if (in_array($_words[$i][$j],$_words_com)){
$_vsm[$i][($_words[$i][$j])] +=1;
}
}
}
for($i = 0;$i
echo '第'.($i+1).'篇文档的特征向量: ('. implode(",",$_vsm[$i]).')
';
}
print_r($_df);
?>