基于TextRank的单文本关键字提取算法

A Single Text Keyword Extraction Algorithm Based on TextRank

TextRank算法在搜索关键字提取中应用十分广泛,针对TextRank算法因考虑句子单词顺序而导致权值范围过大的算法缺陷,提出了忽视句子单词的顺序,仅考虑某文本中的某一句的共现关系,进行频率、权值分析.结果表明,该种方式能够有效地提高查询效率.

In view of the TextRank algorithm being widely used in the search keyword extraction,and aiming at the algorithm defect only for sentence word order causing too large weight range,neglect of sentence of word order only the consideration of a text in a word co-occurrence relationship was proposed to analyze frequency and weight,the result shows that this method can be effective to improve the query efficiency.