Unregistered word recognition processing strategy: to build a one-word word list
Strategy 2:
Step by step to identify the user input the length of the word is equal to 1 or greater than 4, whether as the preset of word segmentation dictionary or users have stated that exist in the word, whether contained in one word segmentation dictionary or user dictionary, extract the unknown words from possible add to the user input word dictionary, temporary records when user input described further identify network in the entry word for word, The user input word is added to the user dictionary and deleted from the user input word dictionary. By identifying the words entered by the user step by step, the possible unlogged words are added to the user dictionary, which enriches the user dictionary and improves the word segmentation effect when it is necessary to segment the words entered by the user based on the user dictionary.
Source: patents.google.com/patent/WO20…