3 results about "Affix" patented technology

The invention discloses a foreign language word card, which is characterized in that a picture is arranged on the front side of the card and expresses characteristic elements or related attributes of a word; the core method of five-star networked words (FSNW: Five Star Networked Word) introduced in the invention is arranged on the back side of the card. The FSNW comprises the word itself and five most meaningful contexts thereof, namely definition, synonym, antonym, hyponym and affix. The display form of the networked words included in the FSNW can be changed according to the selected difficulty level, for example, an English word beader, and when the difficulty level is low, the display form is beader; and when the difficulty level is middle, the display form is b.... r, namely, each middle letter is replaced by a point (.). The FSNW card can be used like a word guessing game, that is, after the card is read, a learner needs to spell the networked word; the FSNW helps a learner to get rid of learning words in an isolated context, helps the learner to fully understand the words, and can easily retrieve the words when the words are forgotten.

The application discloses a text language automatic detection method and system based on syllable and affix features, relates to the field of natural language processing, and comprises the following steps: text preprocessing and basic language unit extraction are performed on a text to be detected to obtain a text syllable sequence and an affix list; based on the affix list, a multidimensional pragmatic scene distribution vector in the context of the text syllable sequence is extracted, and core connectivity in a pre-constructed morphological paradigm knowledge graph is acquired; the function load value of the affix is matched from an affix function load database, the affix function load value is normalized, and a weighted affix feature vector is generated; syntax dependency relationship analysis and preliminary structure construction are performed on the text to be detected, and a dependency syntax tree of the text to be detected is generated. The application improves the detection accuracy and result interpretability of highly similar languages and mixed code texts by explicitly modeling and quantitatively evaluating deep morphological-syntactic rule features of languages.

The invention discloses a multi-source data-based staphylographic large model pre-training method and system, and the method comprises the steps: collecting multi-source staphylographic texts of news, social contact, government affairs, academic and the like, and extracting the vocabulary coverage and language style features, thereby achieving the layering of corpus quality and the recognition of variants; analyzing a flection change structure of the staphylococcus by adopting a word form reduction and affix separation technology, and generating a training sample unit carrying a form label; high-value mask positions are identified based on morphological complexity, word-level, phrase-level and sentence-level mask tasks are configured, and learning of grammatical relationships such as subject-model consistency and name-form consistency is enhanced in combination with a morphological consistency constraint enhancement model; a course learning strategy is adopted to progressively execute training tasks according to difficulty, finally, optimal parameters are screened through multi-dimensional evaluation to output a staphylography pre-training model, and a high-quality language representation basis can be provided for downstream tasks such as staphylography text classification, named entity recognition and machine translation.