Method for extracting event sentence pattern from Chinese sentence
A sentence and syntax technology, applied in the field of event sentence pattern extraction in Chinese sentences, can solve the problems of immaturity and reduction in analysis, and achieve the effect of overcoming immaturity and reducing requirements
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
specific example
[0242] The lower syntactic block sequence is: [P-vp eat / v-][O-np rice / n-]of / uJDE[H-np bowl / n-].
[0243] This string of syntactic blocks can be combined to form a large object block [O-np eat / v rice / n's / uJDE bowl / n-].
[0244] So the fusion rule is:
[0245] P O u J DE H → O 4 0.85.
[0246] It can be seen that the length of this rule is 4 (including the function word "uJDE" in the middle of the block), and the reliability is 0.85, that is, if there are 100 function block sequences in the format of "P O uJDE H", 85 of them are Sequences are merged into "O".
[0247] (4) System initialization
[0248] (4.1) Set data structure
[0249] Set the following data structures: word Verb, syntax block Block, each input sentence EveryRecord, all input sentences vector input, rule table vector for syntactic block fusion rules, where:
[0250] I. word Verb, information such as word and its part of speech in the reservation sentence, forms following format: {}, wherein:
[0251] ...
Embodiment
[0458] The event sentence pattern extraction method that the present invention proposes needs to complete the following steps successively:
[0459] (1) Data preparation
[0460] Data preparation is to complete three parts of operations, one is to call the part-of-speech mapping and extension module to convert the part-of-speech of each word in the input sentence into an extended TCT part-of-speech tag; the other is to call the syntax block analysis module to Sentences are analyzed as syntactic blocks; the third is to call the event knowledge base retrieval module, and use the target verb information and meaning item information in the input to obtain the complete meaning item description of the verb and the corresponding relationship between general roles and special roles in the verb frame.
[0461] The following example illustrates preprocessing:
[0462] Enter a sentence:
[0463] Ministry of Finance / n , / , Ministry of Agriculture / n , / , Ministry of Chemical In...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com