苏剑林分词，去掉》开头或《结尾的候选词

自然语言处理/短语挖掘与新词发现/苏剑林/main_sujianlin.py

@@ -30,6 +30,13 @@ def isChinese(word):
             return True
     return False
 
+notStartChar = ['》', '」', '】', ')', ']', '·', '・', '•']
+notEndChar = ['《', '「', '【', '(', '[', '·', '・', '•']
+def notStartEnd(word):
+    if word[0] not in notStartChar and word[len(work) - 1] not in notEndChar: 
+        return True
+    return False
+
 for m in range(2, max_sep+1):
     print(u'正在生成%s字词...'%m)
     t.append([])
@@ -44,7 +51,7 @@ for m in range(2, max_sep+1):
     qq = np.array(tt.index)
     qqfilter = []
     for word in qq:
-        qqfilter.append(isChinese(word))
+        qqfilter.append(isChinese(word) and notStartEnd(word))
     qq = qq[qqfilter] #非汉字过滤
     tt = tt[qq]
     #非汉字过滤 end