Chinese named entity recognition method and system
A technology for named entity recognition and Chinese, which is applied in instruments, biological neural network models, electrical digital data processing, etc., can solve problems such as complex Chinese structures, and achieve the effect of improving accuracy and accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0036] Such as figure 1 As shown, this embodiment provides a Chinese named entity recognition method, and this embodiment uses the method applied to a server for illustration. It can be understood that this method can also be applied to terminals, and can also be applied to terminals and servers and system, and through the interaction between the terminal and the server. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or it can provide cloud services, cloud database, cloud computing, cloud function, cloud storage, network server, cloud communication, intermediate Cloud servers for basic cloud computing services such as software services, domain name services, security service CDN, and big data and artificial intelligence platforms. The terminal may be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited thereto. The t...
Embodiment approach
[0052] As one or more implementations, the determining the vector representation of the n-gram feature according to the extracted n-gram feature specifically includes: generating the feature vector of each character and the different byte length fragments corresponding to the beginning of the character The vectors are concatenated to obtain a vector representation of the n-gram features corresponding to each character.
[0053] As one or more implementation manners, after obtaining the vector representation of the n-gram feature includes: performing quantitative coding on the vector representation of the n-gram feature.
[0054] Specifically, in the second step, use the n-gram language model to extract byte segments with lengths of 2, 3, 4, and 5. Given a sequence, the 2-gram, 3-gram, 4-gram, 5-gram of the sequence such as image 3 shown.
[0055] Each byte fragment is called a gram, which counts the frequency of occurrence of all grams, and filters them according to a prese...
Embodiment 2
[0060] This embodiment provides a Chinese named entity recognition system.
[0061] A Chinese named entity recognition system, comprising:
[0062] A segmentation module, which is configured to: obtain the segment to be processed, and segment the processed segment by character;
[0063] The gated graph neural network module is configured to: obtain the node information corresponding to each character in the segment by using the gated graph neural network based on the segment to be processed;
[0064] A feature extraction module, which is configured to: extract an n-gram feature from the sentence to be processed, and determine a vector representation of the n-gram feature according to the extracted n-gram feature;
[0065] The recognition module is configured to: after splicing the node information and the vector representation of the n-gram feature, through a two-way long-short-term memory network, obtain a Chinese named entity recognition result.
[0066] What needs to be e...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More