| 摘要 |
A word segmentation method and device. The method comprises the steps of: extracting text information about a webpage in a search resource (101); conducting word segmentation processing on the text information using a feature entry in a word segmentation dictionary, to obtain one or more candidate segmented words (102); when ambiguity occurs during the word segmentation processing, counting the word frequency number of the candidate segmented words characterizing context in the webpage (103); adjusting the weight of the feature entry in the word segmentation dictionary according to the word frequency number (104); and conducting word segmentation processing on the text information according to the feature entry with adjusted weight in the word segmentation dictionary, to determine a target candidate segmented word (105). The present invention assists word segmentation processing based on context, and fully takes account of the characteristics of natural language, thereby effectively reducing the influence of ambiguity on word segmentation processing, and improving the accuracy rate of word segmentation. |
