摘要 |
PURPOSE: A method for automatically extracting a keyword of a digital document is provided to automatically extract the keyword from the digital document at fast, and to be applied to the real environment and any language. CONSTITUTION: The method comprises the steps of extracting en element when the frequency of the element is greater than a threshold value and using the separator as a reference mark, converting the digital document into a list data structure including at least one list element, setting the joining region to a vacant state, and extracting the list element successively. In case that the extracted element is the separator, the extracted element is ignored. When the frequency of the extracted list element and the frequency of the next list element nearly arranged to the extracted element are greater than the threshold value, the new element is obtained by performing a join process. The joining region is converted into the list data structure and the previous steps are repeated until a stop condition is satisfied. The list element is displayed according to the display condition giving to the keyword search.
|