Method and apparatus for learning Chinese new words

Abstract

This invention discloses a method for studying Chinese new words and a device, which uses a process module to process the searched engine daily record input via an input module, deletes single Chinese word and inquiry words containing non-Chinese composition and puts the left in order according to the inquiry sequence to set a threshold value and delete the inquiry words with sequence lower than the value. The word-dividing module divides the rest inquiry words taking the present words with the number smaller or equal to 4, if the number is greater than 4, then each time takes 4 words and one from the beginning till the last word, then divides the words based on the four-words method. The filter module arrays the new words according to the appeared frequency, sets new threshold value and deletes the new words lower than the threshold value and outputs the left by an output module.

Description

Technical field: [0001] The invention relates to a method and a device for learning new Chinese words. Background technique: [0002] Word segmentation is the premise and basis for effective Chinese information processing. Word segmentation technology has been widely used in search engines, information retrieval and other fields. The size of the dictionary is one of the most important factors affecting word segmentation accuracy. The dictionaries used for word segmentation are usually incomplete, and it is impossible to include all the entries whether it is a common general dictionary or a domain-oriented professional dictionary. Moreover, as a dynamic and open collection of Chinese vocabulary, a large number of new words will continue to emerge. Especially in the Web environment, various fashionable new vocabulary reflecting the characteristics of the times will appear from time to time. The updating speed of the manually maintained dictionary often lags behind the gene...

AI Technical Summary

Problems solved by technology

[0004] 1. The efficiency and accuracy of learning new words is low

Learning new words directly from the entire article requires complex lexical and syntactic analysis, which is relatively complicated to deal with. The complexity of new word learning algorithms is high. When faced with massive web page information, these methods are sometimes powerless

Moreover, due to the need for complex lexical and syntactic analysis, and there are a large number of ambiguity problems in the Chinese language, the accuracy of new word learning is also relatively low;

[0005] 2. The learning of new words is limited by the field

And when the application environment of new words has no clear domain characteristics, it is more difficult to use professional corpus to learn new words, so the application of this domain-oriented new word learning method is very limited;

[0006] 3. Too many subjective factors

The collection and arrangement of professional corpora often requires a lot of manpower and material resources, and is usually a time-consuming task

The sorted corpus samples are inevitably affected by human subjective factors and have deviations, which will eventually affect the accuracy of new word learning

Images