| 摘要 |
PURPOSE:To efficiently compress text data irrespective of a data appearance pattern in the text data. CONSTITUTION:Text data inputted in SO1 is divided into words in SO2. Subsequently. in SO4, whether the divided words exist in a dictionary prepared in advance or not is retrieved, and whether they exist or not is decided in SO5. In the case a coincident key exists in the dictionary, its word in the text data is converted to an address allocated to its word in the dictionary and outputted in SO6. In the case the word does not exist in the dictionary, an unregistered word processing prepared separately being different from SO6 is executed in S07. The word existing in the dictionary referred to in SO4 can be compressed efficiently to short data by this method. |
