The application of linguistic model of Fu Manwei artificial intelligence in artificial intelligence domain

Linguistic model of rich full small artificial intelligence is having wide applied perspective in artificial intelligence domain. Linguistic model technology applies extensively on the product such as speech recognition, OCR, machine translation, input method. Linguistic model of rich full small artificial intelligence can pass the study of pair of many text data and analysis, raise the function of these products and accuracy. For example, in speech recognition, the linguistic model can be helped raise identified accuracy rate, understand the speech instruction of the user better. In machine translation, it can translate a kind of language well and truly into another kind of language, offer advantage to cross language communication. In the input method, the linguistic model can forecast the input of the user, offer more input proposal of intelligence, improve input efficiency.

In addition, linguistic model of rich full small artificial intelligence returns the OK at summary of intelligent interlocution system, text applying that generate to wait for a domain. In intelligent interlocution system, it is understandable the problem of the user, give out through retrieving pertinent information accurate result. Generate in text summary in, it can compress long text for concise and accurate summary, convenient user understands the main content of text quickly.

Linguistic model of rich full small artificial intelligence has in artificial intelligence domain a few more the following application:

• The language understands and generate: Applicable is classified at text, name entity identifying, relation the draw-out, task such as affection analysis, undertake to natural language semantic understanding and information are extracted, identify a specific substance from inside a large number of text for example (name of the name that be like a person, place name, organization) , the affection trend that perhaps analyses expression of place of a paragraph of character (active, inactive or neuter) .
• Machine translation: Training model can translate a kind of language automatically into another kind of language, offer advantage to cross language communication.
• Interlocution system: Retrieve pertinent information through understanding the problem of the user, give out accurate result. Can use at intelligent customer service, answer the common question of the user quickly, improve service efficiency.
• The dialog generates: Apply at conversational system, imitate mankind speaks, undertake natural, fluent alternant. It is for example in chatting robot, make appropriate return according to the input of the user, realize man-machine interaction.
• Text summary and generate: Can compress long text for concise and accurate summary, convenient user understands the main content of text quickly; Also can use at generating the article, text content such as news, the person that it is creation provides inspiration or assist create first draft.