WO2019196236A1 - Semantic role analysis method, readable storage medium, terminal device and apparatus - Google Patents

Abstract

The present application belongs to the technical field of computers, and relates in particular to a semantic role analysis method, a computer-readable storage medium, a terminal device and an apparatus. The method comprises: in the process of part-of-speech analysis, one neural network model is used for forward part-of-speech analysis, and another neural network model is used for reverse part-of-speech analysis. In the process of semantic role analysis, one neural network model is used for forward semantic role analysis, and another neural network model is used for reverse semantic role analysis. Thus, an originally complex neural network model is divided into relatively simple neural network models, and then the output of each neural network model is comprehensively processed to obtain a result. Due to the simplification of the neural network model structure, the computational load is greatly reduced and the analysis efficiency is improved.

Description

Semantic role analysis method, readable storage medium, terminal device and device

This application claims priority to Chinese Patent Application No. 201101309685.6, entitled "Semantic Role Analysis Method, Computer Readable Storage Media, and Terminal Equipment", filed on April 9, 2018, the entire disclosure of which is hereby incorporated by reference. The content is incorporated herein by reference.

Technical field

The present application belongs to the field of computer technology, and in particular, to a semantic role analysis method, a computer readable storage medium, a terminal device and a device.

Background technique

At present, the mainstream semantic role analysis research mainly focuses on the use of various machine learning techniques, using multiple linguistic features to identify and classify semantic roles. The usual practice is to first use a neural network model to perform the participle of each participle. Determine, and then use a neural network model to determine the semantic role of each word segment. Because in the calculation process, the impact of the whole sentence on the word segmentation result needs to be considered in a single neural network model, the neural network model is often very complicated to construct. The calculation is huge and the efficiency is low.

technical problem

In view of this, the embodiments of the present application provide a semantic role analysis method, a computer readable storage medium, a terminal device, and a device, so as to solve the current semantic role analysis method, and the entire sentence is determined in a single neural network model. As a result of the impact, neural network models are often constructed with very complex, computationally intensive and inefficient problems.

Technical solution

The first aspect of the embodiment of the present application provides a semantic role analysis method, which may include:

Performing word-cutting on the sentence text to obtain each participle constituting the text of the sentence;

Searching for a word vector of each word segment in a preset word vector database, and respectively constructing a first input matrix and a second input matrix of each word segment according to the word vector, wherein the word vector database is between the record word and the word vector a database of correspondences;

Inputting a first input matrix of each participle into a preset first neural network model to obtain a first output vector of each participle, the first neural network model being a neural network model for performing positive-sequence part-of-speech analysis;

The second input matrix of each participle is input into a preset second neural network model to obtain a second output vector of each participle, and the second neural network model is a neural network model for performing reverse-sequence part-of-speech analysis;

Determining the part of speech type of each participle according to the first output vector and the second output vector of each participle;

Searching for a part-of-speech vector corresponding to the part-of-speech type of each participle in a preset part of speech vector database, and constructing a third input matrix and a fourth input matrix of each participle according to the part-of-speech vector, the part-of-speech vector database is a recorded part of speech type a database of correspondences with part of speech vectors;

The third input matrix of each participle is input into a preset third neural network model to obtain a third output vector of each participle, and the third neural network model is a neural network model for performing positive sequence semantic role analysis;

The fourth input matrix of each participle is respectively input into a preset fourth neural network model to obtain a fourth output vector of each participle, and the fourth neural network model is a neural network model for performing reverse order semantic role analysis;

The semantic role type of each word segment is determined according to the third output vector and the fourth output vector of each participle.

A second aspect of embodiments of the present application provides a computer readable storage medium storing computer readable instructions that, when executed by a processor, implement the semantic role analysis method described above step.

A third aspect of embodiments of the present application provides a semantic role analysis terminal device including a memory, a processor, and computer readable instructions stored in the memory and executable on the processor, the processor executing The computer readable instructions implement the steps of the semantic role analysis method described above.

A fourth aspect of the embodiments of the present application provides a semantic role analysis apparatus, which may include a module for implementing the steps of the semantic role analysis method described above.

Beneficial effect

The beneficial effects of the embodiment of the present application compared with the prior art are: the embodiment of the present application splits the originally complicated neural network model into a relatively simple neural network model, and then comprehensively processes the output of each neural network model. As a result, due to the simplification of the neural network model structure, the amount of calculation is greatly reduced, and the analysis efficiency is improved.

DRAWINGS

FIG. 1 is a flowchart of an embodiment of a semantic role analysis method according to an embodiment of the present application;

2 is a schematic flow chart of a processing procedure of a first neural network model;

3 is a schematic flow chart of a processing procedure of a second neural network model;

4 is a schematic flow chart of a processing procedure of a third neural network model;

5 is a schematic flow chart of a processing procedure of a fourth neural network model;

FIG. 6 is a structural diagram of an embodiment of a semantic role analysis apparatus according to an embodiment of the present application;

FIG. 7 is a schematic block diagram of a semantic role analysis terminal device according to an embodiment of the present application.

Embodiments of the invention

Referring to FIG. 1, an embodiment of a semantic role analysis method in an embodiment of the present application may include:

Step S101, performing word segmentation on the sentence text to obtain each word segment constituting the sentence text.

The word processing refers to dividing a sentence text into a single word, that is, each of the word segments. In this embodiment, the sentence text can be segmented according to the general dictionary to ensure that the words that are separated are normal words. If the words are not in the dictionary, the words are separated. When the current rear direction can be a word, for example, "require God", it will be divided according to the size of the statistical word frequency. For example, if the word "required" is high, the word "requirement/god" will be separated. / Ask God."

Step S102: Search for a word vector of each word segment in a preset word vector database, and respectively construct a first input matrix and a second input matrix of each word segment according to the word vector.

The word vector database is a database for recording a correspondence between words and word vectors, and the word vectors may be corresponding word vectors obtained by training words according to the word2vec model.

Specifically, the first input matrix of each word segment can be separately constructed according to the following formula:

Where n is the serial number of the word segmentation in order of precedence, 1≤n≤N, N is the total number of word segmentation of the sentence text, cl is the line number of the first input matrix, 1≤cl≤CoupLen, CoupLen For a preset coupling length, wvl is the column number of the first input matrix, 1≤wvl≤wVecLen, wVecLen is the length of the word vector of any one of the participles, and the word vector of the nth participle is WordVec _n , and

WordVec _n = (WdVecEm _n,1 , WdVecEm _n,2 ,...,WdVecEm _n,vl ,...,WdVecEm _n,wVecLen ),

FwWdMatrix _n is the first input matrix of the nth participle.

The second input matrix of each participle is constructed according to the following formula:

BkWdMatrix _n is the second input matrix of the nth participle.

Step S103, the first input matrix of each word segment is separately input into a preset first neural network model, and a first output vector of each word segment is obtained.

The first neural network model is a neural network model for performing positive-sequence part-of-speech analysis, and the processing process of the first neural network model may specifically include the steps shown in FIG. 2:

In step S1031, the first composite vector of each participle is calculated separately.

Specifically, the first composite vector of each participle can be separately calculated according to the following formula:

FwWdCpVec _n = (FwWdCpEm _n,1 , FwWdCpEm _n,2 ,..., FwWdCpEm _n,wvl ,...,FwWdCpEm _n,wVecLen ) where

Ln is a natural logarithm function, tanh is a hyperbolic tangent function, and FwWdWt _wvl and FwWdWt' _wvl are preset weight coefficients.

Step S1032, respectively calculating a first probability value of each part of speech type.

Specifically, the first probability value of each part of speech type may be separately calculated according to the following formula:

Where m is the number of the part of speech type, 1≤m≤M, M is the number of part of speech type, and FwWdWtVec _m is the preset weight vector corresponding to the mth part of speech type.

FwWdProb _n,m is the first probability that the nth participle is the mth part of speech type.

In step S1033, a first output vector of each participle is constructed.

Specifically, the first output vector of each participle can be constructed according to the following formula:

FwWdVec _n = (FwWdProb _n,1 , FwWdProb _n,2 ,...,FwWdProb _n,m ,...,FwWdProb _n,M )

Where FwWdVec _n is the first output vector of the nth participle.

Step S104, the second input matrix of each word segment is separately input into a preset second neural network model, and a second output vector of each word segment is obtained.

The second neural network model is a neural network model for performing reverse-sequence part-of-speech analysis, and the processing process of the second neural network model may specifically include the steps shown in FIG. 3:

In step S1041, the second composite vector of each participle is calculated separately.

Specifically, the second composite vector of each participle can be separately calculated according to the following formula:

BkWdCpVec _n = (BkWdCpEm _n,1 , BkWdCpEm _n,2 ,..., BkWdCpEm _n,wvl ,..., BkWdCpEm _n,wVecLen ) where

BkWdWt _wvl and BkWdWt' _wvl are preset weight coefficients.

Step S1042, respectively calculating a second probability value of each part of speech type.

Specifically, the second probability value of each part of speech type may be separately calculated according to the following formula:

Wherein, BkWdWtVec _m is a preset weight vector corresponding to the mth part of speech type, and BkWdProb _n,m is a second probability value that the nth participle is the mth part of speech type.

Step S1043, constructing a second output vector of each participle.

Specifically, the second output vector of each participle can be constructed according to the following formula:

BkWdVec _n = (BkWdProb _n,1 , BkWdProb _n,2 ,..., BkWdProb _n,m ,...,BkWdProb _n,M )

Where BkWdVec _n is the second output vector of the nth participle.

Step S105, determining the part of speech type of each participle according to the first output vector and the second output vector of each participle.

Specifically, the part-of-speech probability vector of each participle can be calculated according to the following formula:

WdProbVec _n = (WdProb _n,1 , WdProb _n,2 ,...,WdProb _n,m ,...,WdProb _n,M )

Wherein, WdProb _{n, m} = η ₁ * FwWdProb _{n, m} + η ₂ * BkWdProb _{n, m} , η ₁ , η ₂ are preset weight coefficients, and WdProbVec _n is the part-of-sense probability vector of the nth participle.

Determine the part of speech of each participle according to the following formula:

CharSeq _n =argmax(WdProbVec _n )

Where argmax is the largest independent variable function and CharSeq _n is the part-of-speech type number of the nth participle. It is also determined that the part of speech type corresponding to the element having the largest value in the part-of-speech probability vector of the nth participle is determined as the part of speech type of the nth participle.

Step S106: Search for a part-of-speech vector corresponding to the part-of-speech type of each participle in the preset part-of-speech vector database, and construct a third input matrix and a fourth input matrix of each participle according to the part-of-speech vector respectively.

The part of speech vector database is a database for recording the correspondence between the part of speech type and the part of speech vector. The part of speech vector is a vector form corresponding to each part of speech type, that is, the probability of occurrence of the part of speech type is represented according to context information of the part of speech type. The training of part-of-speech vectors first expresses each part of speech type into a 0-1 vector (one-hot) form, and then performs model training, using the part of speech type of n-1 words to predict the part of speech type of the nth word, neural network The intermediate process obtained after the model is predicted is used as a part of speech vector. Specifically, the one-hot vector of the part-of-speech type "noun" is set to [1, 0, 0, 0, ..., 0], and the one-hot vector of the part of speech type "adjective" is [0, 1, 0. ,0,...,0], the one-hot vector of the part of speech type "verb" is [0,0,1,0,...,0], the vector of the part of speech type "adverb"[0,0,0,1 , ..., 0], the model is trained to generate a coefficient matrix W of the hidden layer, the product of the one-hot vector of each part of speech type and the coefficient matrix is the part of speech vector of the part of speech type, and the final form will be similar to "[ A multidimensional vector such as -0.11, 0.26, -0.03, ..., 0.71]".

Specifically, the third input matrix of each word segment can be separately constructed according to the following formula:

Where n is the sequence number of the word segmentation in order of precedence, 1≤n≤N, N is the total number of word segmentation of the sentence text, cl is the line number of the third input matrix, 1≤cl≤CoupLen, CoupLen For a preset coupling length, cvl is the column number of the third input matrix, 1≤cvl≤cVecLen, cVecLen is the length of the part of speech vector of any one of the participles, and the part of speech vector of the nth participle is CharVec _n , and

CharVec _n = (CrVecEm _n,1 ,CrVecEm _n,2 ,...,CrVecEm _n,cvl ,...,CrVecEm _n,cVecLen ),

FwCrMatrix _n is the third input matrix of the nth participle.

The fourth input matrix of each participle is constructed according to the following formula:

BkCrMatrix _n is the fourth input matrix of the nth participle.

Step S107, the third input matrix of each participle is respectively input into a preset third neural network model, and a third output vector of each participle is obtained.

The third neural network model is a neural network model for performing positive sequence semantic role analysis, and the processing process of the third neural network model may specifically include the steps shown in FIG. 4:

Step S1071, respectively calculating a third composite vector of each participle.

Specifically, the third composite vector of each participle can be separately calculated according to the following formula:

FwCrCpVec _n = (FwCrCpEm _n,1 , FwCrCpEm _n,2 , . . . , FwCrCpEm _{n, vl} , . . . , FwCrCpEm _{n, cVecLen} )

Ln is a natural logarithm function, tanh is a hyperbolic tangent function, and FwCrWt _cvl and FwCrWt' _cvl are preset weight coefficients.

Step S1072, respectively calculating a first probability value of each semantic role type.

Specifically, the first probability value of each semantic role type may be separately calculated according to the following formula:

Where l is the sequence number of the semantic role type, 1≤l≤L, L is the number of semantic role types, and FwCrWtVec _l is the preset weight vector corresponding to the first semantic role type.

FwCrProb _n,l is the first probability that the nth participle is the first semantic role type.

In step S1073, a third output vector of each participle is constructed.

Specifically, the third output vector of each participle can be constructed according to the following formula:

FwCrVec _n = (FwCrProb _n,1 , FwCrProb _n,2 ,...,FwCrProb _n,l ,...,FwCrProb _n,L )

Where FwCrVec _n is the third output vector of the nth participle.

Step S108, the fourth input matrix of each word segment is separately input into a preset fourth neural network model, and a fourth output vector of each word segment is obtained.

The fourth neural network model is a neural network model for performing reverse-sequence semantic role analysis, and the processing process of the third neural network model may specifically include the steps shown in FIG. 5:

In step S1081, the fourth composite vector of each participle is calculated separately.

Specifically, the fourth composite vector of each participle can be separately calculated according to the following formula:

BkCrCpVec _n = (BkCrCpEm _n,1 , BkCrCpEm _n,2 , . . . , BkCrCpEm _{n, cvl} , . . . , BkCrCpEm _{n, cVecLen} ), wherein

BkCrWt _cvl and BkCrWt' _cvl are preset weight coefficients.

Step S1082, respectively calculating a second probability value of each semantic role type.

Specifically, the second probability value of each semantic role type may be separately calculated according to the following formula:

Where BkCrWtVec _l is a preset weight vector corresponding to the first semantic role type, and BkCrProb _{n, l} is the second probability that the nth participle is the first semantic role type.

Step S1083, constructing a fourth output vector of each participle.

Specifically, the fourth output vector of each participle can be constructed according to the following formula:

BkCrVec _n = (BkCrProb _n,1 , BkCrProb _n,2 ,...,BkCrProb _n,l ,...,BkCrProb _n,L )

Where BkCrVec _n is the fourth output vector of the nth participle.

Step S109, determining a semantic role type of each word segment according to the third output vector and the fourth output vector of each participle.

Specifically, the semantic role probability vector of each word segment can be separately calculated according to the following formula:

CrProbVec _n = (CrProb _n,1 ,CrProb _n,2 ,...,CrProb _n,l ,...,CrProb _n,L )

Among them, CrProb _n,l =ξ ₁ *FwCrProb _n,l +ξ ₂ *BkCrProb _n,l ,ξ ₁ , ξ ₂ are preset weight coefficients, and CrProbVec _n is the semantic role probability vector of the nth participle.

Determine the semantic role type of each word segment according to the following formula:

RoleSeq _n =argmax(CrProbVec _n )

Where argmax is the largest independent variable function and RoleSeq _n is the semantic role type number of the nth participle. It is also determined that the semantic role type corresponding to the element with the largest value among the semantic role probability vectors of the nth participle is determined as the semantic role type of the nth participle. It is also determined that the semantic role type corresponding to the element with the largest value among the semantic role probability vectors of the nth participle is determined as the semantic role type of the nth participle.

In summary, the two embodiments of the present application use two neural network models for processing in the two most critical processes, and the previously complex neural network model is divided into relatively simple neural network models, and then The output of each neural network model is processed comprehensively to obtain the result. Due to the simplification of the neural network model structure, the calculation amount is greatly reduced, and the analysis efficiency is improved.

Corresponding to a semantic role analysis method described in the foregoing embodiment, FIG. 6 is a structural diagram of an embodiment of a semantic role analysis apparatus provided by an embodiment of the present application.

In this embodiment, a semantic role analysis apparatus may include:

The word processing module 601 is configured to perform word segmentation on the sentence text to obtain each word segment constituting the text of the sentence;

The word vector searching module 602 is configured to separately search for a word vector of each word segment in a preset word vector database, where the word vector database is a database for recording a correspondence between words and word vectors;

a word vector matrix construction module 603, configured to respectively construct a first input matrix and a second input matrix of each word segment according to the word vector;

The first processing module 604 is configured to input the first input matrix of each word segment into the preset first neural network model to obtain a first output vector of each word segment, where the first neural network model performs positive sequence word Analytical neural network model;

The second processing module 605 is configured to input the second input matrix of each participle into the preset second neural network model to obtain a second output vector of each participle, and the second neural network model is to perform reverse word analysis. Neural network model;

The part of speech type determining module 606 is configured to determine a part of speech type of each participle according to the first output vector and the second output vector of each participle;

The part of speech vector search module 607 is configured to search for a part of speech vector corresponding to the part of speech type of each participle in a preset part of speech vector database, where the part of speech vector database is a database for recording the correspondence between the part of speech type and the part of speech vector;

The part of speech vector matrix construction module 608 is configured to respectively construct a third input matrix and a fourth input matrix of each word segment according to the part of speech vector;

The third processing module 609 is configured to input the third input matrix of each word segment into the preset third neural network model to obtain a third output vector of each word segment, and the third neural network model performs positive sequence semantics Neural network model for role analysis;

The fourth processing module 610 is configured to input the fourth input matrix of each participle into a preset fourth neural network model to obtain a fourth output vector of each participle, and the fourth neural network model is to perform a reverse order semantic role. Analytical neural network model;

The semantic role type determining module 611 is configured to determine a semantic role type of each word segment according to the third output vector and the fourth output vector of each word segment.

The specific embodiment of the semantic role analysis device is substantially the same as the foregoing embodiments of the semantic role analysis method. Reference may be made to the related description in the foregoing method embodiments, and details are not described herein.

FIG. 7 is a schematic block diagram of a semantic role analysis terminal device provided by an embodiment of the present application. For convenience of description, only parts related to the embodiment of the present application are shown.

In this embodiment, the semantic role analysis terminal device 7 may be a computing device such as a mobile phone, a tablet computer, a desktop computer, a notebook, a palmtop computer, and a cloud server. The semantic role analysis terminal device 7 may include a processor 70, a memory 71, and computer readable instructions 72 stored in the memory 71 and executable on the processor 70, such as performing the semantic role analysis method described above. Computer readable instructions. The processor 70 executes the steps in the embodiments of the various semantic role analysis methods described above when the computer readable instructions 72 are executed.

The functional units in the various embodiments of the present application may be stored in a computer readable storage medium if implemented in the form of a software functional unit and sold or used as a separate product. Based on such understanding, the technical solution of the present application, in essence or the contribution to the prior art, or all or part of the technical solution may be embodied in the form of a software product stored in a storage medium. A number of computer readable instructions are included to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present application.

Claims (20)

1. A semantic role analysis method, comprising:

   Performing word-cutting on the sentence text to obtain each participle constituting the text of the sentence;

   Searching for a word vector of each word segment in a preset word vector database, and respectively constructing a first input matrix and a second input matrix of each word segment according to the word vector, wherein the word vector database is between the record word and the word vector a database of correspondences;

   Inputting a first input matrix of each participle into a preset first neural network model to obtain a first output vector of each participle, the first neural network model being a neural network model for performing positive-sequence part-of-speech analysis;

   The second input matrix of each participle is input into a preset second neural network model to obtain a second output vector of each participle, and the second neural network model is a neural network model for performing reverse-sequence part-of-speech analysis;

   Determining the part of speech type of each participle according to the first output vector and the second output vector of each participle;

   Searching for a part-of-speech vector corresponding to the part-of-speech type of each participle in a preset part of speech vector database, and constructing a third input matrix and a fourth input matrix of each participle according to the part-of-speech vector, the part-of-speech vector database is a recorded part of speech type a database of correspondences with part of speech vectors;

   The third input matrix of each participle is input into a preset third neural network model to obtain a third output vector of each participle, and the third neural network model is a neural network model for performing positive sequence semantic role analysis;

   The fourth input matrix of each participle is respectively input into a preset fourth neural network model to obtain a fourth output vector of each participle, and the fourth neural network model is a neural network model for performing reverse order semantic role analysis;

   The semantic role type of each word segment is determined according to the third output vector and the fourth output vector of each participle.
2. The semantic role analysis method according to claim 1, wherein the constructing the first input matrix and the second input matrix of each word segment according to the word vector respectively comprises:

   The first input matrix of each participle is constructed according to the following formula:

   

   Where n is the serial number of the word segmentation in order of precedence, 1≤n≤N, N is the total number of word segmentation of the sentence text, cl is the line number of the first input matrix, 1≤cl≤CoupLen, CoupLen For a preset coupling length, wvl is the column number of the first input matrix, 1≤wvl≤wVecLen, wVecLen is the length of the word vector of any one of the participles, and the word vector of the nth participle is WordVec _n , and

   WordVec _n = (WdVecEm _n,1 , WdVecEm _n,2 ,...,WdVecEm _n,vl ,...,WdVecEm _n,wVecLen ),

   

   FwWdMatrix _n is the first input matrix of the nth participle;

   The second input matrix of each participle is constructed according to the following formula:

   

   

   BkWdMatrix _n is the second input matrix of the nth participle.
3. The semantic role analysis method according to claim 2, wherein the processing of the first neural network model comprises:

   Calculate the first composite vector of each participle according to the following formula:

   FwWdCpVec _n = (FwWdCpEm _n,1 , FwWdCpEm _n,2 ,..., FwWdCpEm _n,wvl ,...,FwWdCpEm _n,wVecLen ) where

   

   Ln is a natural logarithm function, tanh is a hyperbolic tangent function, and FwWdWt _wvl and FwWdWt' _wvl are preset weight coefficients;

   Calculate the first probability value of each part of speech type according to the following formula:

   

   Where m is the number of the part of speech type, 1≤m≤M, M is the number of part of speech type, and FwWdWtVec _m is the preset weight vector corresponding to the mth part of speech type.

   

   FwWdProb _n,m is the first probability that the nth participle is the mth part of speech type;

   Construct the first output vector of each participle according to the following formula:

   FwWdVec _n = (FwWdProb _n,1 , FwWdProb _n,2 ,...,FwWdProb _n,m ,...,FwWdProb _n,M )

   Where FwWdVec _n is the first output vector of the nth participle;

   The processing process of the second neural network model includes:

   Calculate the second composite vector of each participle according to the following formula:

   BkWdCpVec _n = (BkWdCpEm _n,1 , BkWdCpEm _n,2 ,..., BkWdCpEm _n,wvl ,..., BkWdCpEm _n,wVecLen ) where

   

   BkWdWt _wvl and BkWdWt' _wvl are preset weight coefficients;

   Calculate the second probability value of each part of speech type according to the following formula:

   

   Wherein, BkWdWtVec _m is a preset weight vector corresponding to the mth part of speech type, and BkWdProb _n,m is a second probability value that the nth participle is the mth part of speech type;

   Construct a second output vector for each participle according to the following formula:

   BkWdVec _n = (BkWdProb _n,1 , BkWdProb _n,2 ,..., BkWdProb _n,m ,...,BkWdProb _n,M )

   Where BkWdVec _n is the second output vector of the nth participle.
4. The semantic role analysis method according to claim 3, wherein the determining the part of speech type of each word segment according to the first output vector and the second output vector of each word segment comprises:

   Calculate the part-of-speech probability vector of each participle according to the following formula:

   WdProbVec _n = (WdProb _n,1 , WdProb _n,2 ,...,WdProb _n,m ,...,WdProb _n,M )

   Wherein, WdProb _{n, m} = η ₁ * FwWdProb _{n, m} + η ₂ * BkWdProb _{n, m} , η ₁ , η ₂ are preset weight coefficients, and WdProbVec _n is a part-of-speech probability vector of the nth participle;

   Determine the part of speech of each participle according to the following formula:

   CharSeq _n =arg max(WdProbVec _n )

   Where arg max is the largest independent variable function and CharSeq _n is the part-of-speech type number of the nth participle.
5. The semantic role analysis method according to claim 1, wherein the constructing the third input matrix and the fourth input matrix of each word segment according to the part of speech vector respectively comprises:

   The third input matrix of each participle is constructed according to the following formula:

   

   Where n is the sequence number of the word segmentation in order of precedence, 1≤n≤N, N is the total number of word segmentation of the sentence text, cl is the line number of the third input matrix, 1≤cl≤CoupLen, CoupLen For a preset coupling length, cvl is the column number of the third input matrix, 1≤cvl≤cVecLen, cVecLen is the length of the part of speech vector of any one of the participles, and the part of speech vector of the nth participle is CharVec _n , and

   CharVec _n = (CrVecEm _n,1 ,CrVecEm _n,2 ,...,CrVecEm _n,cvl ,...,CrVecEm _n,cVecLen ),

   

   FwCrMatrix _n is the third input matrix of the nth participle;

   The fourth input matrix of each participle is constructed according to the following formula:

   

   

   BkCrMatrix _n is the fourth input matrix of the nth participle.
6. The semantic role analysis method according to claim 5, wherein the processing of the third neural network model comprises:

   Calculate the third composite vector of each participle according to the following formula:

   FwCrCpVec _n = (FwCrCpEm _n,1 , FwCrCpEm _n,2 , . . . , FwCrCpEm _{n, vl} , . . . , FwCrCpEm _{n, cVecLen} )

   

   Ln is a natural logarithm function, tanh is a hyperbolic tangent function, and FwCrWt _cvl and FwCrWt' _cvl are preset weight coefficients;

   Calculate the first probability value of each semantic role type according to the following formula:

   

   Where l is the sequence number of the semantic role type, 1≤l≤L, L is the number of semantic role types, and FwCrWtVec _l is the preset weight vector corresponding to the first semantic role type.

   

   FwCrProb _n,l is the first probability that the nth participle is the first semantic role type;

   Construct a third output vector for each participle according to the following formula:

   FwCrVec _n = (FwCrProb _n,1 , FwCrProb _n,2 ,...,FwCrProb _n,l ,...,FwCrProb _n,L )

   Where FwCrVec _n is the third output vector of the nth participle;

   The processing process of the fourth neural network model includes:

   Calculate the fourth composite vector of each participle according to the following formula:

   BkCrCpVec _n = (BkCrCpEm _n,1 , BkCrCpEm _n,2 , . . . , BkCrCpEm _{n, cvl} , . . . , BkCrCpEm _{n, cVecLen} ) wherein

   

   BkCrWt _cvl and BkCrWt' _cvl are preset weight coefficients;

   Calculate the second probability value of each semantic role type according to the following formula:

   

   Where BkCrWtVec _l is a preset weight vector corresponding to the first semantic role type, and BkCrProb _{n, l} is the second probability that the nth participle is the first semantic role type;

   The fourth output vector of each participle is constructed according to the following formula:

   BkCrVec _n = (BkCrProb _n,1 , BkCrProb _n,2 ,...,BkCrProb _n,l ,...,BkCrProb _n,L )

   Where BkCrVec _n is the fourth output vector of the nth participle.
7. The semantic role analysis method according to claim 6, wherein the determining the semantic role types of each word segment according to the third output vector and the fourth output vector of each word segment includes:

   Calculate the semantic role probability vector of each participle according to the following formula:

   CrProbVec _n = (CrProb _n,1 ,CrProb _n,2 ,...,CrProb _n,l ,...,CrProb _n,L )

   Wherein, CrProb _n,l =ξ ₁ *FwCrProb _n,l +ξ ₂ *BkCrProb _n,l ,ξ ₁ , ξ ₂ are preset weight coefficients, and CrProbVec _n is the semantic role probability vector of the nth participle;

   Determine the semantic role type of each word segment according to the following formula:

   RoleSeq _n =arg max(CrProbVec _n )

   Where arg max is the largest independent variable function and RoleSeq _n is the semantic role type number of the nth participle.
8. A computer readable storage medium storing computer readable instructions, wherein the computer readable instructions, when executed by a processor, implement the following steps:

   Performing word-cutting on the sentence text to obtain each participle constituting the text of the sentence;

   Searching for a word vector of each word segment in a preset word vector database, and respectively constructing a first input matrix and a second input matrix of each word segment according to the word vector, wherein the word vector database is between the record word and the word vector a database of correspondences;

   Inputting a first input matrix of each participle into a preset first neural network model to obtain a first output vector of each participle, the first neural network model being a neural network model for performing positive-sequence part-of-speech analysis;

   The second input matrix of each participle is input into a preset second neural network model to obtain a second output vector of each participle, and the second neural network model is a neural network model for performing reverse-sequence part-of-speech analysis;

   Determining the part of speech type of each participle according to the first output vector and the second output vector of each participle;

   Searching for a part-of-speech vector corresponding to the part-of-speech type of each participle in a preset part of speech vector database, and constructing a third input matrix and a fourth input matrix of each participle according to the part-of-speech vector, the part-of-speech vector database is a recorded part of speech type a database of correspondences with part of speech vectors;

   The third input matrix of each participle is input into a preset third neural network model to obtain a third output vector of each participle, and the third neural network model is a neural network model for performing positive sequence semantic role analysis;

   The fourth input matrix of each participle is respectively input into a preset fourth neural network model to obtain a fourth output vector of each participle, and the fourth neural network model is a neural network model for performing reverse order semantic role analysis;

   The semantic role type of each word segment is determined according to the third output vector and the fourth output vector of each participle.
9. The computer readable storage medium according to claim 8, wherein the constructing the first input matrix and the second input matrix of each word segment according to the word vector respectively comprises:

   The first input matrix of each participle is constructed according to the following formula:

   

   Where n is the serial number of the word segmentation in order of precedence, 1≤n≤N, N is the total number of word segmentation of the sentence text, cl is the line number of the first input matrix, 1≤cl≤CoupLen, CoupLen For a preset coupling length, wvl is the column number of the first input matrix, 1≤wvl≤wVecLen, wVecLen is the length of the word vector of any one of the participles, and the word vector of the nth participle is WordVec _n , and

   WordVec _n = (WdVecEm _n,1 , WdVecEm _n,2 ,...,WdVecEm _n,vl ,...,WdVecEm _n,wVecLen ),

   

   FwWdMatrix _n is the first input matrix of the nth participle;

   The second input matrix of each participle is constructed according to the following formula:

   

   

   BkWdMatrix _n is the second input matrix of the nth participle.
10. The computer readable storage medium according to claim 9, wherein the processing of the first neural network model comprises:

    Calculate the first composite vector of each participle according to the following formula:

    FwWdCpVec _n = (FwWdCpEm _n,1 , FwWdCpEm _n,2 ,..., FwWdCpEm _n,wvl ,...,FwWdCpEm _n,wVecLen ) where

    

    Ln is a natural logarithm function, tanh is a hyperbolic tangent function, and FwWdWt _wvl and FwWdWt' _wvl are preset weight coefficients;

    Calculate the first probability value of each part of speech type according to the following formula:

    

    Where m is the number of the part of speech type, 1≤m≤M, M is the number of part of speech type, and FwWdWtVec _m is the preset weight vector corresponding to the mth part of speech type.

    

    FwWdProb _n,m is the first probability that the nth participle is the mth part of speech type;

    Construct the first output vector of each participle according to the following formula:

    FwWdVec _n = (FwWdProb _n,1 , FwWdProb _n,2 ,...,FwWdProb _n,m ,...,FwWdProb _n,M )

    Where FwWdVec _n is the first output vector of the nth participle;

    The processing process of the second neural network model includes:

    Calculate the second composite vector of each participle according to the following formula:

    BkWdCpVec _n = (BkWdCpEm _n,1 , BkWdCpEm _n,2 ,..., BkWdCpEm _n,wvl ,..., BkWdCpEm _n,wVecLen ) where

    

    BkWdWt _wvl and BkWdWt' _wvl are preset weight coefficients;

    Calculate the second probability value of each part of speech type according to the following formula:

    

    Wherein, BkWdWtVec _m is a preset weight vector corresponding to the mth part of speech type, and BkWdProb _n,m is a second probability value that the nth participle is the mth part of speech type;

    Construct a second output vector for each participle according to the following formula:

    BkWdVec _n = (BkWdProb _n,1 , BkWdProb _n,2 ,..., BkWdProb _n,m ,...,BkWdProb _n,M )

    Where BkWdVec _n is the second output vector of the nth participle.
11. A semantic role analysis terminal device comprising a memory, a processor, and computer readable instructions stored in the memory and executable on the processor, wherein the processor executes the computer readable instructions The following steps are implemented:

    Performing word-cutting on the sentence text to obtain each participle constituting the text of the sentence;

    Searching for a word vector of each word segment in a preset word vector database, and respectively constructing a first input matrix and a second input matrix of each word segment according to the word vector, wherein the word vector database is between the record word and the word vector a database of correspondences;

    Inputting a first input matrix of each participle into a preset first neural network model to obtain a first output vector of each participle, the first neural network model being a neural network model for performing positive-sequence part-of-speech analysis;

    The second input matrix of each participle is input into a preset second neural network model to obtain a second output vector of each participle, and the second neural network model is a neural network model for performing reverse-sequence part-of-speech analysis;

    Determining the part of speech type of each participle according to the first output vector and the second output vector of each participle;

    Searching for a part-of-speech vector corresponding to the part-of-speech type of each participle in a preset part of speech vector database, and constructing a third input matrix and a fourth input matrix of each participle according to the part-of-speech vector, the part-of-speech vector database is a recorded part of speech type a database of correspondences with part of speech vectors;

    The third input matrix of each participle is input into a preset third neural network model to obtain a third output vector of each participle, and the third neural network model is a neural network model for performing positive sequence semantic role analysis;

    The fourth input matrix of each participle is respectively input into a preset fourth neural network model to obtain a fourth output vector of each participle, and the fourth neural network model is a neural network model for performing reverse order semantic role analysis;

    The semantic role type of each word segment is determined according to the third output vector and the fourth output vector of each participle.
12. The semantic role analysis terminal device according to claim 11, wherein the first input matrix and the second input matrix for respectively constructing each word segment according to the word vector include:

    The first input matrix of each participle is constructed according to the following formula:

    

    Where n is the serial number of the word segmentation in order of precedence, 1≤n≤N, N is the total number of word segmentation of the sentence text, cl is the line number of the first input matrix, 1≤cl≤CoupLen, CoupLen For a preset coupling length, wvl is the column number of the first input matrix, 1≤wvl≤wVecLen, wVecLen is the length of the word vector of any one of the participles, and the word vector of the nth participle is WordVec _n , and

    WordVec _n = (WdVecEm _n,1 , WdVecEm _n,2 ,...,WdVecEm _n,vl ,...,WdVecEm _n,wVecLen ),

    

    FwWdMatrix _n is the first input matrix of the nth participle;

    The second input matrix of each participle is constructed according to the following formula:

    

    

    BkWdMatrix _n is the second input matrix of the nth participle.
13. The semantic role analysis terminal device according to claim 12, wherein the processing process of the first neural network model comprises:

    Calculate the first composite vector of each participle according to the following formula:

    FwWdCpVec _n = (FwWdCpEm _n,1 , FwWdCpEm _n,2 ,..., FwWdCpEm _n,wvl ,...,FwWdCpEm _n,wVecLen ) where

    

    Ln is a natural logarithm function, tanh is a hyperbolic tangent function, and FwWdWt _wvl and FwWdWt' _wvl are preset weight coefficients;

    Calculate the first probability value of each part of speech type according to the following formula:

    

    Where m is the number of the part of speech type, 1≤m≤M, M is the number of part of speech type, and FwWdWtVec _m is the preset weight vector corresponding to the mth part of speech type.

    

    FwWdProb _n,m is the first probability that the nth participle is the mth part of speech type;

    Construct the first output vector of each participle according to the following formula:

    FwWdVec _n = (FwWdProb _n,1 , FwWdProb _n,2 ,...,FwWdProb _n,m ,...,FwWdProb _n,M )

    Where FwWdVec _n is the first output vector of the nth participle;

    The processing process of the second neural network model includes:

    Calculate the second composite vector of each participle according to the following formula:

    BkWdCpVec _n = (BkWdCpEm _n,1 , BkWdCpEm _n,2 ,..., BkWdCpEm _n,wvl ,..., BkWdCpEm _n,wVecLen ) where

    

    BkWdWt _wvl and BkWdWt' _wvl are preset weight coefficients;

    Calculate the second probability value of each part of speech type according to the following formula:

    

    Wherein, BkWdWtVec _m is a preset weight vector corresponding to the mth part of speech type, and BkWdProb _n,m is a second probability value that the nth participle is the mth part of speech type;

    Construct a second output vector for each participle according to the following formula:

    BkWdVec _n = (BkWdProb _n,1 , BkWdProb _n,2 ,..., BkWdProb _n,m ,...,BkWdProb _n,M )

    Where BkWdVec _n is the second output vector of the nth participle.
14. The semantic role analysis terminal device according to claim 13, wherein the determining the part of speech type of each word segment according to the first output vector and the second output vector of each word segment comprises:

    Calculate the part-of-speech probability vector of each participle according to the following formula:

    WdProbVec _n = (WdProb _n,1 , WdProb _n,2 ,...,WdProb _n,m ,...,WdProb _n,M )

    Wherein, WdProb _{n, m} = η ₁ * FwWdProb _{n, m} + η ₂ * BkWdProb _{n, m} , η ₁ , η ₂ are preset weight coefficients, and WdProbVec _n is the part-of-speech probability vector of the nth participle;

    Determine the part of speech of each participle according to the following formula:

    CharSeq _n =arg max(WdProbVec _n )

    Where arg max is the largest independent variable function and CharSeq _n is the part-of-speech type number of the nth participle.
15. The semantic role analysis terminal device according to claim 11, wherein the third input matrix and the fourth input matrix for respectively constructing each word segment according to the part of speech vector include:

    The third input matrix of each participle is constructed according to the following formula:

    

    Where n is the sequence number of the word segmentation in order of precedence, 1≤n≤N, N is the total number of word segmentation of the sentence text, cl is the line number of the third input matrix, 1≤cl≤CoupLen, CoupLen For a preset coupling length, cvl is the column number of the third input matrix, 1≤cvl≤cVecLen, cVecLen is the length of the part of speech vector of any one of the participles, and the part of speech vector of the nth participle is CharVec _n , and

    CharVec _n = (CrVecEm _n,1 ,CrVecEm _n,2 ,...,CrVecEm _n,cvl ,...,CrVecEm _n,cVecLen ),

    

    FwCrMatrix _n is the third input matrix of the nth participle;

    The fourth input matrix of each participle is constructed according to the following formula:

    

    

    BkCrMatrix _n is the fourth input matrix of the nth participle.
16. The semantic role analysis terminal device according to claim 15, wherein the processing process of the third neural network model comprises:

    Calculate the third composite vector of each participle according to the following formula:

    FwCrCpVec _n = (FwCrCpEm _n,1 , FwCrCpEm _n,2 , . . . , FwCrCpEm _{n, vl} , . . . , FwCrCpEm _{n, cVecLen} )

    

    Ln is a natural logarithm function, tanh is a hyperbolic tangent function, and FwCrWt _cvl and FwCrWt' _cvl are preset weight coefficients;

    Calculate the first probability value of each semantic role type according to the following formula:

    

    Where l is the sequence number of the semantic role type, 1≤l≤L, L is the number of semantic role types, and FwCrWtVec _l is the preset weight vector corresponding to the first semantic role type.

    

    FwCrProb _n,l is the first probability that the nth participle is the first semantic role type;

    Construct a third output vector for each participle according to the following formula:

    FwCrVec _n = (FwCrProb _n,1 , FwCrProb _n,2 ,...,FwCrProb _n,l ,...,FwCrProb _n,L )

    Where FwCrVec _n is the third output vector of the nth participle;

    The processing process of the fourth neural network model includes:

    Calculate the fourth composite vector of each participle according to the following formula:

    BkCrCpVec _n = (BkCrCpEm _n,1 , BkCrCpEm _n,2 , . . . , BkCrCpEm _{n, cvl} , . . . , BkCrCpEm _{n, cVecLen} ) wherein

    

    BkCrWt _cvl and BkCrWt' _cvl are preset weight coefficients;

    Calculate the second probability value of each semantic role type according to the following formula:

    

    Where BkCrWtVec _l is a preset weight vector corresponding to the first semantic role type, and BkCrProb _{n, l} is the second probability that the nth participle is the first semantic role type;

    The fourth output vector of each participle is constructed according to the following formula:

    BkCrVec _n = (BkCrProb _n,1 , BkCrProb _n,2 ,...,BkCrProb _n,l ,...,BkCrProb _n,L )

    Where BkCrVec _n is the fourth output vector of the nth participle.
17. The semantic role analysis terminal device according to claim 16, wherein the determining the semantic role type of each word segment according to the third output vector and the fourth output vector of each word segment includes:

    Calculate the semantic role probability vector of each participle according to the following formula:

    CrProbVec _n = (CrProb _n,1 ,CrProb _n,2 ,...,CrProb _n,l ,...,CrProb _n,L )

    Wherein, CrProb _n,l =ξ ₁ *FwCrProb _n,l +ξ ₂ *BkCrProb _n,l ,ξ ₁ , ξ ₂ are preset weight coefficients, and CrProbVec _n is the semantic role probability vector of the nth participle;

    Determine the semantic role type of each word segment according to the following formula:

    RoleSeq _n =arg max(CrProbVec _n )

    Where arg max is the largest independent variable function and RoleSeq _n is the semantic role type number of the nth participle.
18. A semantic role analysis device, comprising:

    a word processing module for performing word segmentation on the sentence text to obtain respective word segments constituting the text of the sentence;

    a word vector search module, configured to separately search for a word vector of each participle in a preset word vector database, where the word vector database is a database for recording a correspondence between a word and a word vector;

    a word vector matrix building module, configured to respectively construct a first input matrix and a second input matrix of each word segment according to the word vector;

    a first processing module, configured to input a first input matrix of each word segment into a preset first neural network model, to obtain a first output vector of each word segment, where the first neural network model performs positive sequence word analysis Neural network model;

    a second processing module, configured to input a second input matrix of each participle into a preset second neural network model to obtain a second output vector of each participle, where the second neural network model is for performing reverse word analysis Neural network model;

    a part of speech type determining module, configured to determine a part of speech type of each participle according to a first output vector and a second output vector of each participle;

    a part of speech vector search module, configured to respectively search for a part of speech vector corresponding to a part of speech type of each participle in a preset part of speech vector database, wherein the part of speech vector database is a database for recording a correspondence between a part of speech type and a part of speech vector;

    a part of speech vector matrix building module, configured to respectively construct a third input matrix and a fourth input matrix of each word segment according to the part of speech vector;

    a third processing module, configured to input a third input matrix of each word segment into a preset third neural network model to obtain a third output vector of each word segment, where the third neural network model performs a positive sequence semantic role Analytical neural network model;

    a fourth processing module, configured to input a fourth input matrix of each participle into a preset fourth neural network model, to obtain a fourth output vector of each participle, and the fourth neural network model is to perform inverse semantic role analysis Neural network model;

    The semantic role type determining module is configured to determine a semantic role type of each word segment according to the third output vector and the fourth output vector of each participle.
19. The semantic role analysis apparatus according to claim 18, wherein the word vector matrix construction module comprises:

    a first input matrix construction unit for constructing a first input matrix of each word segment according to the following formula:

    

    Where n is the serial number of the word segmentation in order of precedence, 1≤n≤N, N is the total number of word segmentation of the sentence text, cl is the line number of the first input matrix, 1≤cl≤CoupLen, CoupLen For a preset coupling length, wvl is the column number of the first input matrix, 1≤wvl≤wVecLen, wVecLen is the length of the word vector of any one of the participles, and the word vector of the nth participle is WordVec _n , and

    WordVec _n = (WdVecEm _n,1 , WdVecEm _n,2 ,...,WdVecEm _n,vl ,...,WdVecEm _n,wVecLen ),

    

    FwWdMatrix _n is the first input matrix of the nth participle;

    a second input matrix building unit is configured to respectively construct a second input matrix of each word segment according to the following formula:

    

    

    BkWdMatrix _n is the second input matrix of the nth participle.
20. The semantic role analysis apparatus according to claim 19, wherein the first processing module comprises:

    The first composite vector calculation unit is configured to separately calculate the first composite vector of each word segment according to the following formula:

    FwWdCpVec _n = (FwWdCpEm _n,1 , FwWdCpEm _n,2 ,..., FwWdCpEm _n,wvl ,...,FwWdCpEm _n,wVecLen ) where

    

    Ln is a natural logarithm function, tanh is a hyperbolic tangent function, and FwWdWt _wvl and FwWdWt' _wvl are preset weight coefficients;

    The part of speech first probability value calculation unit is configured to respectively calculate a first probability value of each part of speech type according to the following formula:

    

    Where m is the number of the part of speech type, 1≤m≤M, M is the number of part of speech type, and FwWdWtVec _m is the preset weight vector corresponding to the mth part of speech type.

    

    FwWdProb _n,m is the first probability that the nth participle is the mth part of speech type;

    a first output vector building unit for constructing a first output vector of each word segment according to the following formula:

    FwWdVec _n = (FwWdProb _n,1 , FwWdProb _n,2 ,...,FwWdProb _n,m ,...,FwWdProb _n,M )

    Where FwWdVec _n is the first output vector of the nth participle;

    The second processing module includes:

    a second composite vector calculation unit, configured to separately calculate a second composite vector of each word segment according to the following formula:

    BkWdCpVec _n = (BkWdCpEm _n,1 , BkWdCpEm _n,2 ,..., BkWdCpEm _n,wvl ,..., BkWdCpEm _n,wVecLen ) where

    

    BkWdWt _wvl and BkWdWt' _wvl are preset weight coefficients;

    The part of speech second probability value calculation unit is configured to separately calculate a second probability value of each part of speech type according to the following formula:

    

    Wherein, BkWdWtVec _m is a preset weight vector corresponding to the mth part of speech type, and BkWdProb _n,m is a second probability value that the nth participle is the mth part of speech type;

    a second output vector building unit for constructing a second output vector of each word segment according to the following formula:

    BkWdVec _n = (BkWdProb _n,1 , BkWdProb _n,2 ,..., BkWdProb _n,m ,...,BkWdProb _n,M )

    Where BkWdVec _n is the second output vector of the nth participle.

Images