Geological mineral named entity recognition method and device, electronic equipment and medium

Abstract

The invention relates to a geological mineral named entity recognition method and device, electronic equipment and a medium, and the method comprises the steps: obtaining multi-modal data corresponding to a to-be-recognized geological mineral, and the multi-modal data comprises text data, a geological image and a geological map; performing feature extraction on the text data to obtain a text feature vector, and performing feature extraction on the geological image and the geological map to obtain an image feature vector; according to the text feature vector and the image feature vector, performing named entity recognition on the to-be-recognized geological mineral to obtain a preliminary recognition result; and correcting the preliminary recognition result according to a conditional random field correction constraint method to obtain a target recognition result. According to the method, the text feature vector and the image feature vector corresponding to the multi-modal data are considered, and the initial recognition result is corrected by adopting the conditional random field correction constraint method, so that the obtained target recognition result is more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of natural language processing, and in particular to a method, device, electronic device and medium for identifying geological mineral named entities. Background Art

[0002] Current NER implementation methods are primarily categorized into three categories: rule-based, traditional machine learning, and deep learning. Rule-based methods, relying on extensive domain knowledge and manual rules, are inefficient and have poor scalability. Machine learning methods require extensive manual annotation, suffer from weak generalization capabilities, and exhibit poor overall performance. Deep learning methods based on PLMs are currently the mainstream approach.

[0003] However, the rich semantics in the representation vectors are an important foundation for feature learning in downstream models (such as BiLSTM-CRF), which is crucial for NER in vertical fields. Therefore, many vertical fields have trained corresponding domain pre-training models for this problem, such as BioBERT in the biomedical field, ClinicalBERT in the clinical medicine field, SciBERT in the scientific field, FinBERT in the financial field, and GeoBERT in the geography field.

[0004] Furthermore, the strength of a pre-trained model's representational capabilities depends largely on the knowledge learned during the pre-training phase. While the pre-trained models in the aforementioned fields primarily incorporate unstructured knowledge from sentence text, the Jaket model proposed by Yu et al. and SMedBERT proposed by Zhang et al. incorporate knowledge graphs, furthering the development of pre-trained models. However, pre-trained models require comprehensive knowledge for text representation; solely incorporating textual features is insufficient for accurate recognition of geological mineral named entities. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method, device, electronic equipment and medium for geological mineral named entity recognition, aiming to solve at least one of the above technical problems.

[0006] In a first aspect, the present invention provides a technical solution to the above-mentioned technical problem as follows: a method for identifying geological mineral named entities, the method comprising:

[0007] Acquiring multimodal data corresponding to the geological mineral to be identified, wherein the multimodal data includes text data, geological images, and geological maps;

[0008] Performing feature extraction on the text data to obtain a text feature vector, and performing feature extraction on the geological image and geological map to obtain an image feature vector;

[0009] Performing named entity recognition on the geological mineral to be identified based on the text feature vector and the image feature vector to obtain a preliminary recognition result;

[0010] The preliminary recognition result is corrected according to a conditional random field correction constraint method to obtain a target recognition result, wherein the conditional random field correction constraint method is used to constrain the prediction score of the input data.

[0011] The beneficial effects of the present invention are as follows: the text feature vectors and image feature vectors corresponding to the multimodal data are taken into account in this solution, so that the preliminary recognition results corresponding to the named entities are more accurate, and the conditional random field correction constraint method is used to correct the preliminary recognition results, so that the obtained target recognition results are more accurate.

[0012] On the basis of the above technical solution, the present invention can also be improved as follows.

[0013] Furthermore, the method further comprises:

[0014] The target recognition result is corrected by the rule engine to obtain the final recognition result.

[0015] Furthermore, the feature extraction of the text data to obtain a text feature vector includes:

[0016] Convert each geological term in the text data into a word embedding vector, and add a position code to each word embedding vector, wherein the position code represents the position of the corresponding word embedding vector in the text data;

[0017] For each of the word embedding vectors, the word embedding vector is processed through multiple Transformer encoder layers to obtain surrounding term expression information corresponding to the word embedding vector;

[0018] A text feature vector is generated according to each of the word embedding vectors and each of the surrounding term expression information.

[0019] Furthermore, the feature extraction of the geological image and geological map to obtain the image feature vector includes:

[0020] A convolutional neural network is used to extract features from the geological image and geological map to generate image feature vectors.

[0021] Furthermore, the named entity recognition is performed on the geological mineral to be identified based on the text feature vector and the image feature vector to obtain a preliminary recognition result, including:

[0022] Fusing the text feature vector with the image feature vector to obtain a fused feature;

[0023] The fused features are input into a hybrid neural network that combines a bidirectional gated recurrent unit with a self-attention mechanism to obtain preliminary recognition results.

[0024] In a second aspect, in order to solve the above technical problems, the present invention further provides a geological mineral named entity recognition device, which includes:

[0025] An acquisition module is used to acquire multimodal data corresponding to the geological mineral to be identified, wherein the multimodal data includes text data, geological images and geological maps;

[0026] A feature extraction module is used to extract features from the text data to obtain a text feature vector, and to extract features from the geological image and geological map to obtain an image feature vector;

[0027] A preliminary recognition module, configured to perform named entity recognition on the geological mineral to be identified based on the text feature vector and the image feature vector to obtain a preliminary recognition result;

[0028] The correction module is used to correct the preliminary recognition result according to the conditional random field correction constraint method to obtain a target recognition result. The conditional random field correction constraint method is used to constrain the prediction score of the input data.

[0029] In the third aspect, in order to solve the above-mentioned technical problems, the present invention also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, the geological mineral named entity recognition method of the present application is implemented.

[0030] In a fourth aspect, in order to solve the above-mentioned technical problems, the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the geological mineral named entity recognition method of the present application is implemented.

[0031] Additional aspects and advantages of the present application will be given in part in the following description, which will become apparent from the following description, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments of the present invention.

[0033] Figure 1 A schematic flow chart of a method for identifying geological mineral named entities provided by one embodiment of the present invention;

[0034] Figure 2A schematic flow chart of another method for identifying geological mineral named entities provided by one embodiment of the present invention;

[0035] Figure 3 A schematic structural diagram of a geological mineral named entity recognition device provided by one embodiment of the present invention;

[0036] Figure 4 The present invention provides a schematic structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0037] The principles and features of the present invention are described below. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0038] The following describes in detail the technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems using specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The following embodiments of the present invention are described in conjunction with the accompanying drawings.

[0039] The solution provided by the embodiments of the present invention can be applied to any application scenario requiring geological mineral named entity recognition. The solution provided by the embodiments of the present invention can be executed by any electronic device, for example, a user's terminal device, including at least one of the following: a smartphone, tablet computer, laptop computer, desktop computer, smart speaker, smart watch, smart TV, or smart car device.

[0040] The embodiment of the present invention provides a possible implementation method, such as Figure 1 As shown, a flowchart of a method for identifying geological mineral named entities is provided. The solution can be executed by any electronic device, for example, a terminal device, or by a terminal device and a server. For ease of description, the method provided by the embodiment of the present invention will be described below using a terminal device as an example of the execution subject. Figure 1 As shown in the flowchart, the method may include the following steps:

[0041] S10, acquiring multimodal data corresponding to the geological mineral to be identified, wherein the multimodal data includes text data, geological images, and geological maps;

[0042] S20, performing feature extraction on the text data to obtain a text feature vector, and performing feature extraction on the geological image and geological map to obtain an image feature vector;

[0043] S30, performing named entity recognition on the geological mineral to be identified based on the text feature vector and the image feature vector to obtain a preliminary recognition result;

[0044] S40, correcting the preliminary recognition result according to a conditional random field correction constraint method to obtain a target recognition result, wherein the conditional random field correction constraint method is used to constrain the prediction score of the input data.

[0045] This scheme takes into account the text feature vectors and image feature vectors corresponding to multimodal data, making the preliminary recognition results corresponding to named entities more accurate. In addition, the conditional random field correction constraint method is used to correct the preliminary recognition results, making the obtained target recognition results more accurate.

[0046] The present invention will be further described below with reference to the following specific embodiments. Figure 2 The method for identifying geological mineral named entities provided in this embodiment may include the following steps:

[0047] S10, acquiring multimodal data corresponding to the geological mineral to be identified, wherein the multimodal data includes text data, geological images, and geological maps;

[0048] The geological minerals to be identified can be obtained from earth science-related technical reports, journal articles, books, and other literature data that contain geological mineral named entities. The geological minerals to be identified may be included in text data, geological images, and geological maps.

[0049] S20, performing feature extraction on the text data to obtain a text feature vector, and performing feature extraction on the geological image and geological map to obtain an image feature vector;

[0050] One possible implementation of the feature extraction of the text data in S20 to obtain a text feature vector is as follows:

[0051] The text data is subjected to feature extraction by an improved pre-trained model to obtain a text feature vector. Optionally, the improved pre-trained model may be BERT, RoBERTa, or ERNIE.

[0052] The improved pre-training model includes multiple Transformer encoders, each of which can be called a Transformer encoder layer. The specific implementation of S20 can be:

[0053] S201, converting each geological term in the text data into a word embedding vector, and adding a position code to each word embedding vector, wherein the position code represents the position of the corresponding word embedding vector in the text data;

[0054] S202, for each of the word embedding vectors, the word embedding vector is processed through multiple Transformer encoder layers to obtain surrounding term expression information corresponding to the word embedding vector; the context information of the geological term corresponding to the word embedding vector can be captured through the Transformer encoder layer.

[0055] S203: Generate a text feature vector based on each word embedding vector and each surrounding term expression information.

[0056] In the above S203, the text feature vector may be a high-dimensional representation.

[0057] One possible implementation of the feature extraction of the geological image and geological map in S20 to obtain the image feature vector is as follows:

[0058] A convolutional neural network is used to extract features from the geological image and geological map to generate image feature vectors.

[0059] S30, performing named entity recognition on the geological mineral to be identified based on the text feature vector and the image feature vector to obtain a preliminary recognition result;

[0060] Optionally, the preliminary recognition result may be a classification result, specifically a plurality of prediction scores, each prediction score being represented as a probability value of the corresponding named entity, and a larger probability value indicates a greater possibility of the corresponding named entity.

[0061] One possible implementation of the above S30 is:

[0062] S301, fusing the text feature vector with the image feature vector to obtain a fused feature;

[0063] S302: Input the fused features into a hybrid neural network combining a bidirectional gated recurrent unit (BiGRU) and a self-attention mechanism to obtain a preliminary recognition result.

[0064] The training method of the above hybrid neural network is:

[0065] In the BiGRU model, the input sequence (fused features) is first encoded by the embedding layer and then passed to the bidirectional GRU layer for processing. Each GRU layer has multiple memory cells and gating units that can selectively read or ignore different parts of the input sequence and remember previous state information. Among them, the reset gate can control which information is retained in the memory cell, the update gate can control the input of new information, and the output gate can control the output of information from the memory cell, adaptively learning useful features in the input sequence. Ultimately, the BiGRU model combines the output of the bidirectional GRU layer with the self-attention mechanism to generate a representation with the same length as the input sequence, which is the preliminary recognition result.

[0066] Optionally, the reset gate calculation formula is as follows:

[0067] r t =σ(W r ·[h t-1 , x t ]+b r )

[0068] Among them, r t is the activation vector of the reset gate, W r is the weight matrix, b r represents the bias term, σ represents the sigmoid activation function, h t-1 represents the hidden state of the previous time step, x t Represents the input at the current time step.

[0069] Optionally, the update gate calculation formula is as follows:

[0070] z t =σ(W z ·[h t-1 , x t ]+b z )

[0071] Among them, z t is the activation vector of the update gate, W z is the weight matrix, b z represents the bias term, σ represents the sigmoid activation function, h t-1 represents the hidden state of the previous time step, x t Represents the input at the current time step.

[0072] Optionally, the output gate calculation formula is as follows:

[0073]

[0074] Among them, h t represents the hidden state at the current time step, Represents the candidate hidden state, that is, the initial recognition result.

[0075] S40, correcting the preliminary recognition result according to a conditional random field correction constraint method to obtain a target recognition result, wherein the conditional random field correction constraint method is used to constrain the prediction score of the input data.

[0076] Optionally, the aforementioned conditional random field correction constraint method can be a constraint, for example, requiring that the predicted score corresponding to the preliminary recognition result be no less than a set score. Based on this constraint, the predicted scores corresponding to the preliminary recognition result can be screened to improve the accuracy of the preliminary recognition result. This can also ensure the legitimacy of the output sequence and reduce the probability of outputting an illegal sequence.

[0077] The above-mentioned conditional random field correction constraint method can also be applied to a hybrid neural network. Then, one implementation of the above-mentioned 302 is:

[0078] Inputting the fused features into a preliminary prediction network in a hybrid neural network combining a bidirectional gated recurrent unit (BiGRU) and a self-attention mechanism to obtain a preliminary recognition result;

[0079] The preliminary recognition result is corrected according to the conditional random field correction constraint method in the hybrid neural network to obtain a target recognition result.

[0080] Optionally, the conditional random field correction constraint method (CRF correction constraint method) considers the label information of adjacent data (the real named entities of adjacent geological terms) through CRF and constrains the predicted scores (initial recognition results) output by the hybrid neural network to ensure that legal sequences are output as much as possible and reduce the probability of illegal sequence output. By finding the most likely path on the predicted sequence (the sequence corresponding to the predicted score) output by the preliminary prediction network in the hybrid neural network, this method can determine the label of the output given entity (the given entity refers to the entity of the known named entity, and the label is used to characterize the real named entity corresponding to the entity), that is, more accurately learn the named entity features of the known entity, thereby achieving entity identification. Therefore, the goal of hybrid neural network training is to maximize the probability p(y|X). The larger the probability, the higher the accuracy of the model, and the better the training, the more accurate the recognition results output by the hybrid neural network.

[0081] The above maximization probability p(y|X) can be achieved by maximizing the log-likelihood, as shown below:

[0082]

[0083] Among them, y iis the true label sequence of the i-th known entity (i.e., the label of the known entity), X i is the input sequence corresponding to the i-th known entity, W and b are the weight matrix and bias term respectively, and n is the number of known entities in the training sample.

[0084] After S40, the method further includes:

[0085] S50, correcting the target recognition result through a rule engine to obtain a final recognition result.

[0086] Among them, the rule engine can call the knowledge base and rule base obtained by experts in the geological field to perform secondary verification and correction on the entity labels or target recognition results output by CRF to ensure the accuracy and reliability of the final recognition results.

[0087] The method provided by this scheme can be applied to multiple fields such as geological exploration, mineral resource assessment, geological disaster warning, etc., and can effectively improve the accuracy and efficiency of geological mineral named entity recognition.

[0088] Compared with the prior art, the solution of the present invention has the following beneficial effects:

[0089] The present invention achieves efficient and accurate identification and extraction of key information through automatic information extraction of geoscience-related literature. The use of BERT and deep neural network combined with conditional random field processing technology significantly improves the accuracy in identifying mineral types, geological structures, rocks and geological time. Specifically, by preprocessing literature data and using the GMNER deep neural network model, the system can accurately extract the required information from Chinese geological literature, greatly improving research efficiency and data processing reliability. This method has important application value in geoscience research and significantly improves the utilization efficiency of literature data.

[0090] Based on Figure 1 Based on the same principle as the method shown in , the embodiment of the present invention also provides a geological mineral named entity recognition device 20, such as Figure 3 As shown in , the geological mineral named entity recognition device 20 may include an acquisition module 210, a feature extraction module 220, a preliminary recognition module 230 and a correction module 240, wherein:

[0091] An acquisition module 210 is configured to acquire multimodal data corresponding to the geological mineral to be identified, wherein the multimodal data includes text data, geological images, and geological maps;

[0092] A feature extraction module 220 is used to perform feature extraction on the text data to obtain a text feature vector, and to perform feature extraction on the geological image and geological map to obtain an image feature vector;

[0093] A preliminary recognition module 230 is configured to perform named entity recognition on the geological mineral to be identified based on the text feature vector and the image feature vector to obtain a preliminary recognition result;

[0094] The correction module 240 is used to correct the preliminary recognition result according to the conditional random field correction constraint method to obtain a target recognition result. The conditional random field correction constraint method is used to constrain the prediction score of the input data.

[0095] Optionally, the device further comprises:

[0096] The final recognition result determination module is used to correct the target recognition result through a rule engine to obtain a final recognition result.

[0097] Optionally, when the feature extraction module 220 extracts features from the text data to obtain a text feature vector, it is specifically configured to:

[0098] Convert each geological term in the text data into a word embedding vector, and add a position code to each word embedding vector, wherein the position code represents the position of the corresponding word embedding vector in the text data;

[0099] For each of the word embedding vectors, the word embedding vector is processed through multiple Transformer encoder layers to obtain surrounding term expression information corresponding to the word embedding vector;

[0100] A text feature vector is generated according to each of the word embedding vectors and each of the surrounding term expression information.

[0101] Optionally, when the feature extraction module 220 extracts features from the geological image and geological map to obtain an image feature vector, it is specifically configured to:

[0102] A convolutional neural network is used to extract features from the geological image and geological map to generate image feature vectors.

[0103] Optionally, when the preliminary recognition module 230 performs named entity recognition on the geological mineral to be identified based on the text feature vector and the image feature vector to obtain a preliminary recognition result, it is specifically configured to:

[0104] Fusing the text feature vector with the image feature vector to obtain a fused feature;

[0105] The fused features are input into a hybrid neural network that combines a bidirectional gated recurrent unit with a self-attention mechanism to obtain preliminary recognition results.

[0106] The geological mineral named entity recognition device of the embodiment of the present invention can execute the geological mineral named entity recognition method provided by the embodiment of the present invention, and its implementation principle is similar. The actions performed by each module and unit in the geological mineral named entity recognition device in each embodiment of the present invention correspond to the steps in the geological mineral named entity recognition method in each embodiment of the present invention. For the detailed functional description of each module of the geological mineral named entity recognition device, please refer to the description in the corresponding geological mineral named entity recognition method shown in the previous text, and will not be repeated here.

[0107] Among them, the above-mentioned geological mineral named entity recognition device can be a computer program (including program code) running in a computer device, for example, the geological mineral named entity recognition device is an application software; the device can be used to execute the corresponding steps in the method provided in the embodiment of the present invention.

[0108] In some embodiments, the geological mineral named entity recognition device provided by the embodiment of the present invention can be implemented by a combination of software and hardware. As an example, the geological mineral named entity recognition device provided by the embodiment of the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the geological mineral named entity recognition method provided by the embodiment of the present invention. For example, the processor in the form of a hardware decoding processor can adopt one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs) or other electronic components.

[0109] In other embodiments, the geological mineral named entity recognition device provided by the embodiment of the present invention can be implemented in a software manner. Figure 3 A geological mineral named entity recognition device stored in a memory is shown, which can be software in the form of a program and a plug-in, and includes a series of modules, including an acquisition module 210, a feature extraction module 220, a preliminary recognition module 230 and a correction module 240, for implementing the geological mineral named entity recognition method provided by an embodiment of the present invention.

[0110] The modules involved in the embodiments of the present invention may be implemented in software or hardware, wherein the name of a module does not necessarily limit the module itself.

[0111] Based on the same principle as the method shown in the embodiments of the present invention, an electronic device is also provided in the embodiments of the present invention, which may include but is not limited to: a processor and a memory; the memory is used to store computer programs; the processor is used to execute the method shown in any embodiment of the present invention by calling the computer program.

[0112] In an alternative embodiment, an electronic device is provided, such as Figure 4 As shown, Figure 4 The electronic device 4000 shown includes: a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, for example, via a bus 4002. Optionally, the electronic device 4000 may further include a transceiver 4004, which can be used for data exchange between the electronic device and other electronic devices, such as data transmission and / or data reception. It should be noted that in actual applications, the number of transceivers 4004 is not limited to one, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0113] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the present disclosure. Processor 4001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0114] Bus 4002 may include a path for transmitting information between the above components. Bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. Bus 4002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0115] The memory 4003 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to these.

[0116] The memory 4003 is used to store application code (computer program) for executing the solution of the present invention, and is controlled by the processor 4001. The processor 4001 is used to execute the application code stored in the memory 4003 to implement the content shown in the above method embodiment.

[0117] Among them, the electronic device can also be a terminal device, Figure 4 The electronic device shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0118] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding contents of the aforementioned method embodiment.

[0119] According to another aspect of the present invention, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various implementations described above.

[0120] Computer program code for performing the operations of the present invention may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on the remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0121] It should be understood that the flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or can be implemented using a combination of dedicated hardware and computer instructions.

[0122] The computer-readable storage medium provided by the embodiments of the present invention may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or device.

[0123] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device executes the method shown in the above embodiment.

[0124] The above description is merely a preferred embodiment of the present invention and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of disclosure involved in the present invention is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but also includes other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned disclosed concepts. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in the present invention.

Claims

1. A method for identifying geological mineral named entities, characterized in that: The following steps are involved: Acquiring multimodal data corresponding to the geological mineral to be identified, wherein the multimodal data includes text data, geological images, and geological maps; Performing feature extraction on the text data to obtain a text feature vector, and performing feature extraction on the geological image and geological map to obtain an image feature vector; Performing named entity recognition on the geological mineral to be identified based on the text feature vector and the image feature vector to obtain a preliminary recognition result; The preliminary recognition result is corrected according to a conditional random field correction constraint method to obtain a target recognition result, wherein the conditional random field correction constraint method is used to constrain the prediction score of the input data.

2. The method according to claim 1, characterized in that The method further comprises: The target recognition result is corrected by the rule engine to obtain the final recognition result.

3. The method according to claim 1, characterized in that The step of extracting features from the text data to obtain a text feature vector includes: Convert each geological term in the text data into a word embedding vector, and add a position code to each word embedding vector, wherein the position code represents the position of the corresponding word embedding vector in the text data; For each of the word embedding vectors, the word embedding vector is processed through multiple Transformer encoder layers to obtain surrounding term expression information corresponding to the word embedding vector; A text feature vector is generated according to each of the word embedding vectors and each of the surrounding term expression information.

4. The method according to claim 1, wherein The feature extraction of the geological image and the geological map to obtain the image feature vector includes: A convolutional neural network is used to extract features from the geological image and geological map to generate image feature vectors.

5. The method according to claim 1, wherein The step of performing named entity recognition on the geological mineral to be identified based on the text feature vector and the image feature vector to obtain a preliminary recognition result includes: Fusing the text feature vector with the image feature vector to obtain a fused feature; The fused features are input into a hybrid neural network that combines a bidirectional gated recurrent unit with a self-attention mechanism to obtain preliminary recognition results.

6. A geological mineral named entity recognition device, characterized in that: include: An acquisition module is used to acquire multimodal data corresponding to the geological mineral to be identified, wherein the multimodal data includes text data, geological images and geological maps; A feature extraction module is used to extract features from the text data to obtain a text feature vector, and to extract features from the geological image and geological map to obtain an image feature vector; A preliminary recognition module, configured to perform named entity recognition on the geological mineral to be identified based on the text feature vector and the image feature vector to obtain a preliminary recognition result; The correction module is used to correct the preliminary recognition result according to the conditional random field correction constraint method to obtain a target recognition result. The conditional random field correction constraint method is used to constrain the prediction score of the input data.

7. The device according to claim 6, characterized in that The device further comprises: The final recognition result determination module is used to correct the target recognition result through a rule engine to obtain a final recognition result.

8. The device according to claim 6, characterized in that When the feature extraction module extracts features from the text data to obtain a text feature vector, it is specifically used to: Convert each geological term in the text data into a word embedding vector, and add a position code to each word embedding vector, wherein the position code represents the position of the corresponding word embedding vector in the text data; For each of the word embedding vectors, the word embedding vector is processed through multiple Transformer encoder layers to obtain surrounding term expression information corresponding to the word embedding vector; A text feature vector is generated according to each of the word embedding vectors and each of the surrounding term expression information.

9. An electronic device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 5 when executing the computer program.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images