基于改进哈夫曼编码的全文索引结构压缩算法

Compressed Format Full-Text Index Based on Improved Huffman Code and Its Implement

为解决全文索引的索引结构压缩问题,提出了文本的基于正规哈夫曼编码小波树形式,并将该结构与后缀数组结合,实现了基于正规哈夫曼编码的小波树和高效构造算法。实验结果表明,在不降低运行效率的前提下,存储空间得到有效的压缩,从而证明了改进方法的有效性。

To solve the problem of index structures compression of full-text indexes, we introduce the canonical Huffman code to encode the BWT （ Burrows-Wheeler Transform） of a text. In the end, we present an efficient construction algorithm for this index, which is on-line and linear. Experimental results show that, without reducing the efficiency, the effective storage space compression is gained, which improves the effectiveness of the method.