Role representation multi-round learning method and system for text-level event argument extraction

Abstract

The invention provides a role representation multi-round learning method and system for text-level event argument extraction, and the method proposes the construction of an event mode-instance graph, and builds role association of an abstract event mode level, span association of a specific event instance level, and role-span association between a mode and an instance. A multi-round learning network of node-edge interaction is provided, node and edge representation in a graph is iteratively updated through an optimized graph attention network, a chapter-level event argument extraction model and system of the multi-round learning network of fusion event mode-instance graph construction and node-edge interaction are developed, and effective role representation is trained and learned. According to the method, potential semantics of roles are integrally captured from the graph through the event mode-instance graph, a role memory fusion module is designed to obtain a new round of role representation in a multi-round learning process, and semantic association of elements in the graph is fully considered to optimize an argument boundary predicted in the new round.

Description

Technical Field

[0001] The present invention relates to the technical field of information extraction, and in particular to a multi-round learning method and system for role representation for extracting event arguments at the paragraph level. Background Art

[0002] Chapter-level event argument extraction is a key task in information extraction, which involves extracting event-related arguments from a chapter and accurately identifying their roles. Existing role-based span selection strategies mainly capture role representation and chapter representation through pre-trained language models and prompt tuning techniques, thereby constructing a span selector to predict spans for each role. These methods have limitations: (1) they treat arguments as independent units, ignoring the interrelationships between arguments in event instances; (2) they rely on pre-trained language models to obtain role representation and chapter representation, or only capture the semantics of event types (roles) at the event pattern level, or only capture the semantics of roles at the overall chapter level.

[0003] Therefore, it is necessary to develop a role learning model that can comprehensively capture the role semantics implied by various direct or indirect connections between roles in event patterns, between arguments in event instances, and between pattern-instance matches. Summary of the invention

[0004] In view of the above situation, the main purpose of the present invention is to propose a multi-round learning method and system for role representation for chapter-level event argument extraction to solve the above technical problems.

[0005] The present invention proposes a multi-round learning method for role representation for extracting event arguments at the chapter level, the method comprising the following steps: Step 1: Obtain the initial role representation from the text and use it to predict the argument span; Step 2: Based on the event pattern information, event instance information and argument span in the text, establish the role association at the abstract event pattern level, the span association at the specific event instance level and the role-span association between the pattern and the instance to obtain the event pattern-instance graph; Step 3: Use the edges in the event pattern-instance graph as virtual nodes, use the graph attention network to iteratively update the nodes and edges in the event pattern-instance graph, and output the updated role representation; Step 4: Repeat steps 1 to 3 in an iterative manner for multiple rounds of learning. In each round of learning, the initial role representation, the historical role representation, and the updated role representation of the current round are fused as the prediction input of the argument span of the next round to gradually correct the span boundary. After learning is completed, the final role representation is output.

[0006] The present invention also proposes a role representation multi-round learning system for extracting chapter-level event arguments, wherein the system applies the role representation multi-round learning method for extracting chapter-level event arguments as described above, and the system comprises: Graph building blocks for: Obtaining initial role representation from the text and using the initial role representation to predict argument span; According to the event pattern information, event instance information and argument span in the text, role associations at the abstract event pattern level, span associations at the specific event instance level and role-span associations between patterns and instances are established to obtain an event pattern-instance graph. Iterative interactive update module for: The edges in the event pattern-instance graph are used as virtual nodes, and the graph attention network is used to iteratively update the nodes and edges in the event pattern-instance graph, and the updated role representation is output; Role memory fusion module, used for: In each round of learning, the initial role representation, historical role representation, and updated role representation of the current round are fused as the prediction input of the argument span of the next round; Multi-round span prediction optimization module, used for: Multiple rounds of learning are performed in an iterative manner to gradually correct the span boundary. After learning is completed, the final role representation is output.

[0007] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention proposes the construction of an event pattern-instance graph, establishes role associations at the abstract event pattern level, span associations at the specific event instance level, and role-span associations between patterns and instances, and realizes the integrated capture of the potential semantics of roles from the graph.

[0008] 2. This paper proposes a multi-round learning network with node-edge interaction. Through an optimized graph attention network, the node and edge representations in the graph are iteratively updated, and a role memory fusion module is designed to obtain a new round of role representations and optimize the argument boundaries predicted in the new round. 3. The present invention develops a chapter-level event argument extraction scheme that integrates the construction of event pattern-instance graph and multi-round learning network of node-edge interaction.

[0009] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description or learned through embodiments of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A flowchart of a multi-round learning method for role representation for extracting event arguments at the chapter level proposed by the present invention; Figure 2 This is a framework diagram of the multi-round learning method for role representation for extracting event arguments at the chapter level proposed by the present invention; Figure 3 An example diagram constructed for the event pattern-instance diagram proposed by the present invention; Figure 4 This is a structural diagram of the multi-round learning system for role representation proposed by the present invention for extracting event arguments at the paragraph level. DETAILED DESCRIPTION

[0011] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0012] These and other aspects of the embodiments of the present invention will be apparent with reference to the following description and accompanying drawings. In these descriptions and accompanying drawings, some specific implementations of the embodiments of the present invention are specifically disclosed to represent some ways of implementing the principles of the embodiments of the present invention, but it should be understood that the scope of the embodiments of the present invention is not limited thereto.

[0013] See also Figure 1 and Figure 2 , in the figure represents the product operation, Represents the attention coefficient calculation operation.

[0014] This embodiment provides a multi-round learning method for role representation for extracting event arguments at the chapter level, the method comprising the following steps: Step 1: Obtain the initial role representation from the text and use it to predict the argument span; As a further preferred embodiment of the present invention, a method for obtaining an initial role representation from a passage and predicting an argument span using the initial role representation comprises the following steps: Given a prompt, input the prompt and the text into the pre-trained language model BART to obtain the role representation and text representation of the prompt; Initial span selection layer: First, follow the prompts P The expression is selected k Type of role r k The representation of is used as the initial role representation, and then the preliminary prediction span is obtained based on the initial role representation and the chapter representation. The corresponding process has the following relationship: ; ; ; ; ; in, and Represent two different learnable parameters, and Represents roles r k The role representation of the start and end positions of and Represents roles r k The probability distribution of the start and end positions of the corresponding prediction span in the entire context, represents the initial role representation, represents the role representation of round u, Indicates chapter The chapter indicates; u Indicates the number of rounds the model is currently learning. u ∈U, U is the total number of rounds of model learning; when u =1, That is ;when u >1 hour, For the calculation of , see the last formula in the final character representation acquisition part.

[0015] As a further preferred embodiment of the present invention, a method for obtaining a role representation of a prompt by inputting the prompt and the passage into a pre-trained language model BART comprises the following steps: Define two different special tags <t>< / t> and ; Locate the trigger words in the article, use two different special tags to wrap the trigger words, and generate the preprocessed article ;in, , Indicates the first n words, Indicates trigger words, <t>< / t> and Respectively represent two different special marks; The preprocessed passage is input into the pre-trained language model BART encoder to obtain the passage representation of the passage. The corresponding process has the following relationship: ; in, represents the pre-trained language model BART encoder, The chapter that represents the chapter indicates that Represents the chapter after preprocessing; Given a prompt corresponding to a certain event type, the chapter representation and the prompt are input into the pre-trained language model BART decoder to obtain the entire prompt P The role representation of the corresponding process has the following relationship: ; in, represents the pre-trained language model BART decoder, Indicates a given prompt; Indicates the entire prompt P The representation includes the representation of all roles under this event type.

[0016] Step 2: Based on the event pattern information, event instance information and argument span in the text, establish the role association at the abstract event pattern level, the span association at the specific event instance level and the role-span association between the pattern and the instance to obtain the event pattern-instance graph; See also Figure 3 As a further preferred embodiment of the present invention, a method for obtaining an event pattern-instance graph by establishing role associations at the abstract event pattern level, span associations at the specific event instance level, and role-span associations between patterns and instances based on event pattern information, event instance information, and argument spans in a passage includes the following steps: Constructing event pattern subgraphs: Generate corresponding nodes according to the event types involved in the chapter, and generate corresponding nodes for each role under the event type; Establish an edge of attribute type between the event type node and the role node to obtain the event pattern subgraph; Constructing event instance subgraph: Generate trigger word nodes according to the event trigger words given by the events in the chapter, and generate corresponding span nodes according to the events involved, and fill the predicted span into the corresponding span nodes; Establish edges between trigger word nodes and span nodes to obtain event instance subgraphs, where the edge type is played by the role describing the corresponding span; Subgraph matching association construction: Build an instance type edge between the event type node and the trigger word node, and build a value type edge between the role node and its corresponding span node to obtain an event pattern-instance graph; Among them, the event instance refers to the predicted event triggered by a given event trigger word, which is composed of the trigger word and the span predicted for each role.

[0017] Step 3: Use the edges in the event pattern-instance graph as virtual nodes, use the graph attention network to iteratively update the nodes and edges in the event pattern-instance graph, and output the updated role representation; As a further preferred embodiment of the present invention, the edges in the event pattern-instance graph are used as virtual nodes, and the nodes and edges in the event pattern-instance graph are iteratively updated using a graph attention network. The method for outputting the updated role representation specifically includes the following steps: In terms of node embedding representation, except for the prediction span to be filled, for other nodes in the event pattern-instance graph, the BERT model is first used to obtain the representation of all words contained in each node in the event pattern-instance graph, and then the average pooling method is used to obtain the representation of the entire node. For the edges in the graph, since the types of edges in the event pattern-instance graph are represented by corresponding words, the BERT model is also used to initialize the vectors to obtain the representation of the corresponding edges. k Type of role r k In the u The representation of the span node of the round prediction is generated by the preliminary span selection layer. According to the obtained prediction span, the representations of several words contained in the prediction span are fused to obtain the representation of the predicted span node. The specific calculation is as follows: ; ; ; ; ; in, represents the BERT model, represents average fusion, Indicates the trigger word n words, Indicates the event type n words, Representing roles k No. n words, Indicates k The representation of role edges, Indicates trigger words, Indicates the event type. Indicates k The role represents The edges represent the attribute type, instance type, and value type respectively. The edges representing attribute type, instance type, and value type respectively. and Represents rolesr k The corresponding start and end positions of the prediction span, express and In the range j The expression of a word, Indicates k Type of role r k In the u Representation of span nodes for round prediction; Multi-round learning network for node-edge interaction: Since the event pattern-instance graph has some particularities, the goal is to enrich the semantics of role representation, and there are both role nodes and role edges in the graph. Therefore, when using the graph attention network (GAT) to update the node representation, its connecting edge is also regarded as a virtual node and included in the information transmission range. The specific process is to splice the edge representation to the corresponding neighbor node representation, and then use the softmax function to normalize the masked attention coefficient to obtain the final attention distribution, and then update the node representation according to the attention distribution; i and j The attention coefficient between nodes, the normalization process and the i The representations of the nodes are as follows: ; ; ; in, Represents a splicing operation, represents the attention vector, represents the activation function, represents the transpose operation, Representation Node i The expression, Representation Node j , and , the representation of a node is determined by the node type itself, Indicates i Nodes and j The representation of the edges between nodes, and , the representation of the edge is determined by the edge type between nodes, represents a learnable weight matrix used to transform node features. Representation Node i and nodes j The original attention distribution between represents the exponential function, Representation Node i The neighbor set of Representation Node i and nodes j The final attention distribution between Indicates m The attention head is l Node in layer i and nodes j The attention coefficient between Indicates m The attention head is l The weight matrix in the layer, Indicates l- 1st layer midpoint j The expression, Indicates l Node in layer i The updated representation is, represents the activation function, Represents a regularization operation. In order to prevent overfitting, dropout needs to be applied to the attention coefficient. represents the number of attention heads, Indicates i The set of neighbor nodes of a node, l ∈ L , L Indicates the total number of layers of GAT.

[0018] After the node representation is updated, the representation of the role edge is updated by fusing the following: role node, trigger word node, updated span node and edge representation. This method fully considers the semantic association of each element in the graph. u Round l Tier k Representation of role edges As shown below: ; in, represents the edge weight matrix, , , Respectively represent u Round l Tier k The span node representation of the role u Round l The representation of the trigger word node of the first u Round l Tier k The representation of a role node. Indicates u Round l- 1st floor kThrough average fusion, all relevant semantic features are integrated into the new edge representation, making the semantics of the edge representation richer.

[0019] In 1~ L During the update process of the layer, the node representation and edge representation will affect each other, and the latest role node representation and role edge representation will be continuously generated. U Round L Layer, the role node representation is merged with the role edge representation to obtain the role representation that is finally output by the multi-round learning network, as shown below: ; Among them, g represents the event pattern-instance graph structure. U The meaning of round learning, Indicates U Round L Tier k The representation of role edges, Indicates U Round L Tier k The representation of a role node. Represents the character representation of the final output of the multi-round learning network.

[0020] Step 4: Repeat steps 1 to 3 in an iterative manner for multiple rounds of learning. In each round of learning, the initial role representation, the historical role representation, and the updated role representation of the current round are fused as the prediction input of the argument span of the next round to gradually correct the span boundary. After learning is completed, the final role representation is output.

[0021] As a further preferred embodiment of the present invention, fusing the initial role representation, the historical role representation and the role representation after the current round of update specifically includes the following steps: For the role r k , assuming that the representation in the memory unit is (Initially empty), the current ( u Historical role representation , multi-round learning network u Role representation of round output , Initial role representation In addition, to ensure the comparability of different role representations (historical role representation, representation output by multi-round learning network, and initial role representation) and avoid dimensionality deviation, each representation is first normalized. The obtained role representation is as follows: ; ; ; in, Indicates u The historical role of the wheel indicates that Represents the multi-round learning network u The role representation of the round output, represents the normalization operation, Represents the normalized u The historical role of the wheel indicates that Represents the normalized multi-round learning network u The role representation of the round output, represents the normalized initial role representation; Next, the normalized representations are averaged and fused to generate the final character representation. u The calculation of the character representation after round fusion is as follows: ; in, Represents the historical role learning matrix.

[0022] As a further preferred embodiment of the present invention, in executing the above steps 1 to 4, the argument extraction model is used, and the training method of the argument extraction model includes the following steps: go through U Round learning, using the final character representation containing rich semantic information ; Adopt the span selection strategy in the preliminary span selection layer and output the probability distribution of the start and end positions of the predicted span in the entire context; Generate a set of candidate spans through greedy search according to the probability distribution of the start and end positions of the predicted span in the entire context; Based on minimizing the matching cost, the Hungarian algorithm is used to assign a unique true span label to each predicted candidate span; The cross entropy loss function is constructed based on the probability distribution of the start and end positions of the predicted span in the entire context and the cross entropy of the corresponding true span label. The corresponding process has the following relationship: ; in, D represents the number of chapters in the training set, represents the optimal allocation calculated by the Hungarian algorithm, and Respectively represent k The gold starting position and gold ending position of each character; The argument extraction model is trained by updating the weights and learning the parameters to minimize the loss.

[0023] Please refer to Figure 4 This embodiment also provides a role representation multi-round learning system for extracting chapter-level event arguments, wherein the system applies the role representation multi-round learning method for extracting chapter-level event arguments as described above, and the system includes: Graph building blocks for: Obtaining initial role representation from the text and using the initial role representation to predict argument span; According to the event pattern information, event instance information and argument span in the text, role associations at the abstract event pattern level, span associations at the specific event instance level and role-span associations between patterns and instances are established to obtain an event pattern-instance graph. Iterative interactive update module for: The edges in the event pattern-instance graph are used as virtual nodes, and the graph attention network is used to iteratively update the nodes and edges in the event pattern-instance graph, and the updated role representation is output; Role memory fusion module, used for: In each round of learning, the initial role representation, historical role representation, and updated role representation of the current round are fused as the prediction input of the argument span of the next round; Multi-round span prediction optimization module, used for: Multiple rounds of learning are performed in an iterative manner to gradually correct the span boundary. After learning is completed, the final role representation is output.

[0024] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0025] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0026] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0027] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A multi-round learning method for role representation for extracting event arguments at the chapter level, characterized in that: The method comprises the following steps: Step 1: Obtain the initial role representation from the text and use it to predict the argument span; Step 2: Based on the event pattern information, event instance information and argument span in the text, establish the role association at the abstract event pattern level, the span association at the specific event instance level and the role-span association between the pattern and the instance to obtain the event pattern-instance graph; Step 3: Use the edges in the event pattern-instance graph as virtual nodes, use the graph attention network to iteratively update the nodes and edges in the event pattern-instance graph, and output the updated role representation; Step 4: Repeat steps 1 to 3 in an iterative manner for multiple rounds of learning. In each round of learning, the initial role representation, the historical role representation, and the updated role representation of the current round are fused as the prediction input of the argument span of the next round to gradually correct the span boundary. After learning is completed, the final role representation is output.

2. The multi-round learning method for character representation for extracting event arguments at the chapter level according to claim 1, characterized in that: In step 1, the method of obtaining the initial role representation from the passage and using the initial role representation to predict the argument span includes the following steps: Given a prompt, input the prompt and the text into the pre-trained language model BART to obtain the role representation and text representation of the prompt; According to the prompt P The expression is selected k Type of role r k The representation of is used as the initial role representation, and then the preliminary prediction span is obtained based on the initial role representation and the chapter representation. The corresponding process has the following relationship: ； ； ； ； ； in, and Represent two different learnable parameters, and Represents roles r k The role representation of the start and end positions of and Represents roles r k The probability distribution of the start and end positions of the corresponding prediction span in the entire context, represents the initial role representation, represents the role representation of round u, Indicates chapter The chapter indicates; u Indicates the number of rounds the model is currently learning. u ∈U, U is the total number of rounds of model learning; when u =1, That is .

3. The multi-round learning method for character representation for extracting event arguments at the chapter level according to claim 2, characterized in that: Given a prompt, and inputting the prompt and the passage into a pre-trained language model BART to obtain a role representation of the prompt includes the following steps: Define two different special tags <t>< / t> and ; Locate the trigger words in the article, use two different special tags to wrap the trigger words, and generate the preprocessed article ;in, , Indicates the first n words, Indicates trigger words, <t>< / t> and Respectively represent two different special marks; The preprocessed passage is input into the pre-trained language model BART encoder to obtain the passage representation of the passage. The corresponding process has the following relationship: ； in, represents the pre-trained language model BART encoder, The chapter that represents the chapter indicates that Represents the chapter after preprocessing; Given a prompt corresponding to a certain event type, the chapter representation and the prompt are input into the pre-trained language model BART decoder to obtain the entire prompt P The role representation of the corresponding process has the following relationship: ； in, represents the pre-trained language model BART decoder, Indicates a given prompt; Indicates the entire prompt P The representation includes the representation of all roles under this event type.

4. The multi-round learning method for character representation for extracting event arguments at the chapter level according to claim 1, characterized in that: In step 2, according to the event pattern information, event instance information and argument span in the text, role associations at the abstract event pattern level, span associations at the specific event instance level and role-span associations between patterns and instances are established, and the method for obtaining an event pattern-instance graph includes the following steps: Generate corresponding nodes according to the event types involved in the chapter, and generate corresponding nodes for each role under the event type; Establish an edge of attribute type between the event type node and the role node to obtain the event pattern subgraph; Generate trigger word nodes according to the event trigger words given by the events in the chapter, and at the same time, generate corresponding span nodes according to the events involved, and fill the predicted span into the corresponding span nodes; Establish edges between trigger word nodes and span nodes to obtain event instance subgraphs; the edge type is played by the role describing the corresponding span; An instance type edge is constructed between the event type node and the trigger word node, and a value type edge is constructed between the role node and its corresponding span node. An event pattern-instance graph is obtained based on the instance type edge and the value type edge.

5. The multi-round learning method for character representation for extracting event arguments at the chapter level according to claim 2, characterized in that: In step 3, the edges in the event pattern-instance graph are used as virtual nodes, and the nodes and edges in the event pattern-instance graph are iteratively updated using a graph attention network. The method for outputting the updated role representation specifically includes the following steps: The BERT model is used to obtain the representation of all words contained in each node in the event pattern-instance graph, and then the average pooling method is used to obtain the representation of the entire node. The corresponding process has the following relationship: ； ； ； in, represents the BERT model, represents average fusion, Indicates the trigger word n words, Indicates the event type n words, Representing roles k No. n words, Indicates k The representation of role edges, Indicates trigger words, Indicates the event type. Indicates k Types of role representation; The edges in the event pattern-instance graph are initialized with BERT to obtain the representation of the corresponding edges. The corresponding process has the following relationship: ； in, The edges represent the attribute type, instance type, and value type respectively. The edge representations represent the attribute type, instance type, and value type respectively; According to the obtained prediction span, several word representations contained in the prediction span are fused to obtain the representation of the predicted span node. The corresponding process has the following relationship: ； in, and Represents roles r k The corresponding start and end positions of the prediction span, express and In the range j The expression of a word, Indicates k Type of role r k In the u Representation of span nodes for round prediction; The edge representation is concatenated to the corresponding neighbor node representation, and then the softmax function is used to normalize the masked attention coefficient to obtain the attention distribution. The corresponding process has the following relationship: ； in, Represents a splicing operation, represents the attention vector, represents the activation function, represents the transpose operation, Representation Node i Representation of; Representation Node j , and ,The representation of a node is determined by the node type itself; Indicates i Nodes and j The representation of the edges between nodes, and ,The representation of edges is determined by the edge type between nodes; Represents a learnable weight matrix used to transform node features; Representation Node i and nodes j The original attention distribution between Regularization operation is applied to the attention distribution to obtain the final attention distribution. The corresponding process has the following relationship: ； in, represents the exponential function, Representation Node i The neighbor set of represents the regularization operation, Representation Node i and nodes j The final attention distribution between According to the final attention distribution, the representation of the node is updated to obtain the updated representation of the node. The corresponding process has the following relationship: ； in, Indicates m The attention head is l Node in layer i and nodes j The attention coefficient between Indicates m The attention head is l The weight matrix in the layer, Indicates l −1st layer mid-node j The expression, Indicates l Node in layer i The updated representation is, represents the activation function, represents the number of attention heads, Indicates i The set of neighbor nodes of a node, l ∈ L , L Indicates the total number of layers of GAT; After the node representation is updated, the updated k The span node representation of the role, the updated trigger word node representation, the updated k The representation of the role node is averaged and fused with the representation of the unupdated role edge to obtain the updated role edge representation. The corresponding process has the following relationship: ； in, represents the edge weight matrix, , , Respectively represent u Round l Tier k The span node representation of the role u Round l The representation of the trigger word node of the first u Round l Tier k The representation of a role node. Indicates u Round l-1 Tier k The representation of role edges; After the update, the role node representation is merged with the role edge representation to obtain the role representation that is finally output by the multi-round learning network, as shown below: ； Among them, g represents the event pattern-instance graph structure. U The meaning of round learning, Indicates U Round L Tier k The representation of role edges, Indicates U Round L Tier k The representation of a role node. Represents the character representation of the final output of the multi-round learning network.

6. The multi-round learning method for character representation for extracting event arguments at the chapter level according to claim 2, characterized in that: In step 4, fusing the initial role representation, the historical role representation, and the role representation updated in the current round specifically includes the following steps: The first u The historical role representation of the round, the multi-round learning network u The role representation of the round output and the initial role representation are normalized respectively, and the corresponding process has the following relationship: ； ； ； in, Indicates u The historical role of the wheel indicates that Represents the multi-round learning network u The role representation of the round output, represents the normalization operation, Represents the normalized u The historical role of the wheel indicates that Represents the normalized multi-round learning network u The role representation of the round output, represents the normalized initial role representation; The normalized representations are averaged and fused to generate the fused character representation. The corresponding process has the following relationship: ； in, Represents the historical role learning matrix.

7. The multi-round learning method for character representation for extracting event arguments at the chapter level according to claim 2, characterized in that: In executing the above steps 1 to 4, the argument extraction model is used for implementation. The training method of the argument extraction model includes the following steps: go through U Round learning, using the final character representation containing rich semantic information ; Adopt the span selection strategy in the preliminary span selection layer and output the probability distribution of the start and end positions of the predicted span in the entire context; Generate a set of candidate spans through greedy search according to the probability distribution of the start and end positions of the predicted span in the entire context; Based on minimizing the matching cost, the Hungarian algorithm is used to assign a unique true span label to each predicted candidate span; The cross entropy loss function is constructed based on the start and end positions of the predicted span, the probability distribution in the entire context, and the cross entropy of the corresponding true span label. The corresponding process has the following relationship: ； in, D represents the number of chapters in the training set, represents the optimal allocation calculated by the Hungarian algorithm, and Respectively represent k The gold starting position and gold ending position of each character; The argument extraction model is trained by updating the weights and learning the parameters to minimize the loss.

8. A multi-round learning system for role representation for extracting event arguments at the chapter level, characterized in that: The system applies the role representation multi-round learning method for extracting chapter-level event arguments according to any one of claims 1 to 7, and the system comprises: Graph building blocks for: Obtaining initial role representation from the text and using the initial role representation to predict argument span; According to the event pattern information, event instance information and argument span in the text, role associations at the abstract event pattern level, span associations at the specific event instance level and role-span associations between patterns and instances are established to obtain an event pattern-instance graph. Iterative interactive update module for: The edges in the event pattern-instance graph are used as virtual nodes, and the graph attention network is used to iteratively update the nodes and edges in the event pattern-instance graph, and the updated role representation is output; Role memory fusion module, used for: In each round of learning, the initial role representation, historical role representation, and updated role representation of the current round are fused as the prediction input of the argument span of the next round; Multi-round span prediction optimization module, used for: Multiple rounds of learning are performed in an iterative manner to gradually correct the span boundary. After learning is completed, the final role representation is output.

Images