A Multi-Pattern Matching Algorithm on Multi-Language Mixed Texts for Content-Based Network Information Audit

Content-based network information audit systems have to process multi-language mixed texts usually. The characteristics of multi-pattern matching on multi- language mixed texts and how existing multi-pattern matching algorithms perform on multi-language mixed texts are analyzed. A novel multi-pattern matching algorithm based on the hash Trie tree is proposed, which expands the standard Trie structure, constructs the hash Trie matching machine with the ISN of characters. Theoretic analysis and experimental results demonstrate that the proposed algorithm efficiently solves the space cost expansion problem and processes multi-language mixed texts correctly and efficiently with lower time and space complexity, satisfied the requirement of content-based network information audit. Keywords: Multi-Pattern Matching; Multi-language Mixed; Hash; Trie;