| 摘要 |
A computer-operated document retrieval system includes a lexicon of words contained in system documents, and a document look-up table that relates words by unique word numbers to the documents. A word look-up table identifies sets of words with common characteristics, specifically prefix value and word length, and a character look-up table identifies whether any word contains a specified character. A target set generator accesses the word look-up table to compose a target word set with characteristics corresponding to the search string. A refining module reduces the target set by selecting a set of characters from the search string, and accessing the character look-up table to identify which target words use the character set. The character look-up table is a boolean array with one bit elements that are processed in groups whose size corresponds to the maximum bit processing count of the computer, effectively culling non-matching words simultaneously. A string comparison module determines whether any word remaining in the target set matches the search string. The system quickly executes various searches, including prefix, exact match, wildcard, and fuzzy searches.
|
