Information extraction method and device, equipment and storage medium

Abstract

The invention discloses an information extraction method and device, equipment and a storage medium, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining a to-be-recognized text; determining a target information set based on the similarity between the to-be-recognized text and each piece of cargo knowledge information in a preset knowledge base; and performing information extraction on the to-be-identified text and the target information set by using a preset large model to obtain a cargo information extraction result. According to the method and the device, the target information set with relatively high similarity is obtained by retrieving each piece of cargo knowledge information in the preset knowledge base based on the to-be-recognized text, and the to-be-recognized text and the target information set are input into the large model for information extraction, so that each word does not need to be labeled; and the knowledge base provides a context related to actual data for the model, so that the model can be helped to eliminate ambiguity and improve the ability of understanding a complex text, and the accuracy and reliability of a final result are ensured.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an information extraction method, device, equipment and storage medium. Background Art

[0002] The extraction technology of key information plays an important role in many fields and tasks. It can quickly obtain key information from a large amount of text, understand the text content, etc. At present, text information extraction technology is mainly in the form of sequence annotation, that is, it is necessary to label each word in the input text to obtain sample data for training the large prediction model. However, entity categories need to be labeled with multiple information, which is time-consuming. In addition, the large prediction model has weak understanding ability, and may cause misrecognition of information for items with the same name. And when the entity information in the text is discontinuous, the model may not be able to extract unstructured information, resulting in information omissions. Summary of the invention

[0003] Based on this, it is necessary to provide an information extraction method, device, equipment and storage medium for the above-mentioned technical problems to solve at least one of the above-mentioned technical problems.

[0004] The present invention provides an information extraction method, comprising:

[0005] Get the text to be recognized;

[0006] Determine a target information set based on the similarity between the text to be recognized and each piece of cargo knowledge information in a preset knowledge base;

[0007] The preset large model is used to extract information from the text to be recognized and the target information set to obtain a cargo information extraction result.

[0008] Optionally, according to an information extraction method provided by the present invention, the step of determining a target information set based on the similarity between the text to be identified and each piece of cargo knowledge information in a preset knowledge base includes:

[0009] Determine the similarity between each piece of cargo knowledge information in the preset knowledge base of the text to be identified;

[0010] Sorting the similarities to select a preset amount of cargo knowledge information based on the similarity sorting results;

[0011] Based on the cargo knowledge information, the target information set is formed.

[0012] Optionally, according to an information extraction method provided by the present invention, before determining the target information set based on the similarity between the text to be identified and each cargo knowledge information in a preset knowledge base, the method further includes:

[0013] Obtain several text information samples;

[0014] Determine the cargo entity information and cargo attribute information in each of the text information samples, so as to use the cargo entity information and the cargo attribute information as cargo labels of the text information samples;

[0015] Based on each of the text information samples and the associated cargo labels, cargo knowledge information is generated, and the cargo knowledge information is stored in the preset knowledge base.

[0016] Optionally, according to an information extraction method provided by the present invention, the step of extracting information from the to-be-recognized text and the target information set using a preset large model to obtain a cargo information extraction result includes:

[0017] Generate target prompt information based on the text to be recognized and the target information set;

[0018] The target promp information is input into the large model to obtain the cargo information extraction result output by the large model.

[0019] Optionally, according to an information extraction method provided by the present invention, generating target prompt information based on the text to be recognized and the target information set includes:

[0020] Determine a preset prompt template, wherein the prompt template includes a preset input format and an output format, wherein the input format is a format for specifying information input into the big model, and the output format is a format corresponding to the result output by the rule big model;

[0021] Based on the text to be recognized and the target information set, the target prompt information is generated according to the prompt template.

[0022] Optionally, according to an information extraction method provided by the present invention, before extracting information from the to-be-recognized text and the target information set using a preset large model to obtain a cargo information extraction result, the method further comprises:

[0023] Get several texts to be trained;

[0024] Based on any of the texts to be trained, searching the preset knowledge base to obtain a cargo information set;

[0025] The text to be trained and its associated cargo information set are input into the large model, so as to fine-tune and train the large model according to the extraction result output by the large model and the training label of the text to be trained.

[0026] Optionally, according to an information extraction method provided by the present invention, the cargo information extraction result includes cargo entity information and cargo attribute information of the cargo in the text to be identified;

[0027] After extracting information from the to-be-recognized text and the target information set using the preset large model to obtain the cargo information extraction result, the method further includes:

[0028] Determine the type of transport vehicle that meets the size of the goods according to the goods entity information and the goods attribute information of the goods in the text to be identified;

[0029] The cargo order associated with the text to be recognized is pushed to the driver corresponding to the transport vehicle type for acceptance.

[0030] The present invention also provides an information extraction device, comprising:

[0031] An acquisition module is used to acquire the text to be recognized;

[0032] A determination module, used to determine a target information set based on the similarity between the text to be identified and each piece of cargo knowledge information in a preset knowledge base;

[0033] The information extraction module is used to extract information from the text to be recognized and the target information set using a preset large model to obtain a cargo information extraction result.

[0034] The present invention also provides a computer device, comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor implements the above-mentioned information extraction method when executing the computer-readable instructions.

[0035] The present invention also provides one or more readable storage media storing computer-readable instructions, and the computer-readable instructions implement the above-mentioned information extraction method when executed by a processor.

[0036] The above-mentioned information extraction method, device, equipment and storage medium include: obtaining a text to be recognized; determining a target information set based on the similarity between the text to be recognized and each cargo knowledge information in a preset knowledge base; extracting information from the text to be recognized and the target information set using a preset large model to obtain cargo information extraction results. The present invention retrieves a target information set with a high similarity from each cargo knowledge information in a preset knowledge base based on the text to be recognized, and inputs the text to be recognized and the target information set into the large model for information extraction. It is not necessary to mark each word, and the knowledge base provides the model with context related to the actual data, which can help the model eliminate ambiguity and improve the ability to understand complex texts, thereby ensuring the accuracy and reliability of the final result. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0038] Figure 1 is a flow chart of an information extraction method in one embodiment of the present invention;

[0039] Figure 2 is an information extraction flow chart provided in one embodiment of the present invention;

[0040] Figure 3 is a schematic diagram of building a preset knowledge base provided by an embodiment of the present invention;

[0041] Figure 4 It is a schematic diagram of the process of extracting information using a large model provided by an embodiment of the present invention;

[0042] Figure 5 is a schematic diagram of the structure of an information extraction device in one embodiment of the present invention;

[0043] Figure 6 is a schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] The terms used in one or more embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of the present invention. The singular forms of "a", "said" and "the" used in one or more embodiments of the present invention are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of the present invention refers to and includes any or all possible combinations of one or more associated listed items.

[0046] Key information extraction technology plays an important role in many fields and tasks. It can quickly obtain key information from a large amount of text, understand the text content, etc. The current text information extraction technology is mainly in the form of sequence annotation, that is, it is necessary to label each word in the input text to obtain sample data for training the large prediction model. However, the entity category needs to be labeled with multiple information, which is time-consuming. In addition, the large prediction model has weak understanding ability and may cause misidentification for items with the same name. For example, in the transportation scenario, "I am in bed" and "I want to transport a bed". Although there is a bed, the former is not cargo information, but an item mentioned in the text, and the latter is the cargo information that needs to be identified. In addition, when the entity information in the text is discontinuous, the model may not be able to extract unstructured information. For example, "I need to move 3, yes, yes, mattresses." The model cannot extract the information of "3".

[0047] In view of the above problems, the present invention provides the following embodiments. In one embodiment, specifically, as Figure 1 As shown, Figure 1 : is a flow chart of an information extraction method in an embodiment of the present invention. The present invention provides an information extraction method, comprising the following steps:

[0048] Step S11, obtaining the text to be recognized;

[0049] It should be noted that the application scenarios of the present invention are not limited, for example, customer service scenarios, financial information extraction scenarios, and cargo transportation scenarios, etc. The embodiments of the present invention are specifically described using the cargo transportation scenario as an example. In the transportation scenario, the text to be recognized refers to the original text data related to cargo or transportation information during the logistics or transportation process. The text to be recognized can be the text obtained by recognizing the conversational speech, or it can be the text obtained by user text input. For example, the text to be recognized is: I need to transport 2 beds.

[0050] Step S12, determining a target information set based on the similarity between the text to be recognized and each piece of cargo knowledge information in a preset knowledge base;

[0051] In one embodiment, each cargo knowledge information in the preset knowledge base includes a text information sample and its associated cargo label, and the cargo label is used to identify the cargo entity information of the cargo in the text information sample. In other embodiments, the cargo label may include the cargo entity information and cargo attribute information of the cargo. For example, the text information sample is: 2 beds and a 200-liter refrigerator need to be transported; in this case, the cargo entity information is: beds; the corresponding cargo attribute information is: 2 beds; the cargo entity information is: refrigerator; the corresponding cargo attribute information is: 200 liters. It should be noted that multiple text information samples are collected in advance, and then the cargo entity information and cargo attribute information of the cargo in the text information sample are manually annotated to obtain the cargo label.

[0052] Specifically, the similarity between each cargo knowledge information in the preset knowledge base of the to-be-recognized text is calculated, for example, cosine similarity, Jaccard similarity, etc. Further, based on multiple cargo knowledge information with high similarity, the target information set is formed.

[0053] Step S13, using a preset large model to extract information from the text to be recognized and the target information set to obtain a cargo information extraction result.

[0054] Specifically, the text to be recognized and the target information set are input into the large model for information extraction. In other embodiments, based on the text to be recognized and the target information set, the target prompt information is generated according to a preset prompt template, and then the target prompt information is input into the large model for information extraction to obtain a structured cargo information extraction result. It should be noted that the cargo information extraction result includes cargo entity information and cargo attribute information of the cargo in the text to be recognized. The model recognition process is guided by the target information set to help the model better understand the intention in the text to be recognized, and ensure the accuracy and reliability of the final information extraction result.

[0055] In addition, in one embodiment, in order to improve the accuracy of the large model recognition, the large model is fine-tuned and trained in advance, specifically as follows: a number of texts to be trained are obtained, wherein the training labels associated with the texts to be trained are retrieved from the preset knowledge base based on any of the texts to be trained to obtain multiple cargo knowledge information with high similarity to the texts to be trained, and then a cargo information set is formed based on the multiple cargo knowledge information with high similarity; further, the large model is used to recognize the texts to be trained and the cargo information set associated with them, and the model loss value is determined according to the extraction results output by the large model and the training labels of the texts to be trained, and then the large model is fine-tuned and trained according to the model loss value, so that the large model can accurately recognize the cargo information of the cargo transportation scenario.

[0056] The embodiment of the present invention, through the above scheme, includes: obtaining a text to be recognized; determining a target information set based on the similarity between the text to be recognized and each cargo knowledge information in a preset knowledge base; extracting information from the text to be recognized and the target information set using a preset large model to obtain a cargo information extraction result. The present invention retrieves a target information set with a high similarity from each cargo knowledge information in a preset knowledge base based on the text to be recognized, and inputs the text to be recognized and the target information set into the large model for information extraction. It is not necessary to mark each word, and the knowledge base can provide the large model with context related to the actual data, helping the model to eliminate ambiguity and improve the ability to understand complex texts, thereby ensuring the accuracy and reliability of the final result.

[0057] In one embodiment of the present invention, the determining of the target information set based on the similarity between the text to be recognized and each piece of cargo knowledge information in a preset knowledge base includes:

[0058] Determine the similarity between each piece of cargo knowledge information in the preset knowledge base of the text to be identified; sort each of the similarities to select a preset number of cargo knowledge information based on the similarity sorting result; and form the target information set based on each piece of cargo knowledge information.

[0059] Specifically, the similarity between each cargo knowledge information in the preset knowledge base of the text to be identified is calculated, and then each similarity is sorted to obtain a similarity sorting result. Based on the similarity sorting result, a preset number of cargo knowledge information with higher similarity is selected. Further, based on the preset number of cargo knowledge information, the target information set is formed. Figure 2 , Figure 2 This is an information extraction flow chart provided in an embodiment of the present invention. The text to be recognized is "a refrigerator with 2 beds and 200 liters". The text to be recognized is first searched in a cargo information knowledge base (the preset knowledge base in this embodiment) to retrieve cargo knowledge information with a high similarity to the text. For example, 10 cargo knowledge information are selected to finally form a target information set {cargo knowledge information 1, cargo knowledge information 2, ..., cargo knowledge information 10}. By introducing the cargo knowledge information in the knowledge base, the accuracy of the subsequent large model recognition results can be effectively improved.

[0060] In one embodiment of the present invention, before determining the target information set based on the similarity between the text to be recognized and each piece of cargo knowledge information in the preset knowledge base, the method further includes:

[0061] Acquire a plurality of text information samples; determine the cargo entity information and cargo attribute information in each of the text information samples, so as to use the cargo entity information and the cargo attribute information as cargo labels of the text information samples; generate cargo knowledge information based on each of the text information samples and its associated cargo label, and store the cargo knowledge information in the preset knowledge base.

[0062] It should be noted that cargo entity information refers to the specific name of the cargo, and cargo attribute information includes the quantity, volume, size and other attribute information of the cargo. Figure 3 , Figure 3 This is a schematic diagram of building a preset knowledge base provided by an embodiment of the present invention. A number of text information samples are collected in advance, and then the cargo entity information and cargo attribute information in each of the text information samples are determined. Further, the cargo entity information and the cargo attribute information are used as cargo labels for the text information samples. Optionally, the text may not contain the cargo quantity, volume, cargo size and other attribute information. In this case, there is no need to mark the cargo attribute information. For example, Figure 3 In the question "Do I need to select a cargo if I want to transport a bed?", the cargo entity information in the cargo label is: bed; and the text information sample does not have relevant cargo attribute information. Further, based on each of the text information samples and its associated cargo label, cargo knowledge information is generated, and the cargo knowledge information is stored in the preset knowledge base.

[0063] The embodiment of the present invention collects a number of text information samples and marks the cargo entity information and cargo attribute information in the text information samples, thereby constructing a knowledge base. In the process of information recognition by the large model, the knowledge base can provide the large model with context related to the actual data, help the model eliminate ambiguity and improve the ability to understand complex texts, thereby ensuring the accuracy and reliability of the final result.

[0064] In one embodiment of the present invention, the preset large model is used to extract information from the text to be recognized and the target information set to obtain the cargo information extraction result, including:

[0065] Based on the text to be recognized and the target information set, target prompt information is generated; the target prompt information is input into the large model to obtain the cargo information extraction result output by the large model.

[0066] Specifically, a preset prompt template is determined, wherein the prompt template includes a preset input format and output format, wherein the input format is a format for specifying information input into the big model, and the output format is a format corresponding to the result output by the rule big model; further, based on the text to be recognized and the target information set, the target prompt information is generated according to the prompt template. Then, the target prompt information is input into the big model for information extraction, and the cargo information extraction result output by the big model is obtained. For example, referring to Figure 4 , Figure 4 This is a flow chart of information extraction by a large model provided by an embodiment of the present invention. The text to be recognized is: there are 2 beds and a 200-liter refrigerator. The prompt template is as follows: the input is the text to be recognized and the target information set, and the output format is name, quantity, length, width, etc. When there are multiple goods, multiple results are output. The goods information extraction results output by the large model are: 1. Goods: beds, quantity: 2 beds. 2. Goods: refrigerator, volume: 200 liters.

[0067] The embodiment of the present invention helps the model eliminate ambiguity and improve the ability to understand complex texts by inputting the text to be recognized and the target information set into a large model for information extraction, thereby ensuring the accuracy and reliability of the final result.

[0068] In one embodiment of the present invention, after extracting information from the to-be-recognized text and the target information set using a preset large model to obtain a cargo information extraction result, the method further includes:

[0069] According to the cargo entity information and cargo attribute information of the cargo in the text to be identified, the type of transport vehicle that meets the size of the cargo is determined; and the cargo order associated with the text to be identified is pushed to the driver corresponding to the type of transport vehicle for acceptance.

[0070] Specifically, in the cargo order transportation scenario, the cargo size, weight and other information can be determined based on the cargo entity information and cargo attribute information of the cargo in the to-be-recognized text, so as to determine the type of transportation vehicle that meets the cargo size, such as a small van, a medium-sized van, etc., to ensure that the vehicle has sufficient capacity to complete the loading. Then, the cargo order associated with the to-be-recognized text is pushed to the driver corresponding to the type of transportation vehicle for acceptance, so as to realize automated and intelligent cargo transportation scheduling based on the cargo information extraction results.

[0071] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0072] In one embodiment, an information extraction device is provided, which corresponds one-to-one to the information extraction method in the above embodiment. Figure 5 As shown, Figure 5 : is a schematic diagram of a structure of an information extraction device in one embodiment of the present invention, the information extraction device comprises:

[0073] An acquisition module 21 is used to acquire the text to be recognized;

[0074] A determination module 22, configured to determine a target information set based on the similarity between the text to be identified and each piece of cargo knowledge information in a preset knowledge base;

[0075] The information extraction module 23 is used to extract information from the text to be recognized and the target information set using a preset large model to obtain a cargo information extraction result.

[0076] The determination module 22 is further used for:

[0077] Determine the similarity between each piece of cargo knowledge information in the preset knowledge base of the text to be identified;

[0078] Sorting the similarities to select a preset amount of cargo knowledge information based on the similarity sorting results;

[0079] Based on the cargo knowledge information, the target information set is formed.

[0080] The information extraction device also includes:

[0081] A text information sample acquisition module, used to acquire a number of text information samples;

[0082] A cargo information determination module, used to determine cargo entity information and cargo attribute information in each of the text information samples, so as to use the cargo entity information and the cargo attribute information as cargo labels of the text information samples;

[0083] A generating module is used to generate cargo knowledge information based on each of the text information samples and the cargo labels associated therewith, and store the cargo knowledge information in the preset knowledge base.

[0084] The information extraction module 23 is also used for:

[0085] Generate target prompt information based on the text to be recognized and the target information set;

[0086] The target promp information is input into the large model to obtain the cargo information extraction result output by the large model.

[0087] The information extraction module 23 is also used for:

[0088] Determine a preset prompt template, wherein the prompt template includes a preset input format and an output format, wherein the input format is a format for specifying information input into the big model, and the output format is a format corresponding to the result output by the rule big model;

[0089] Based on the text to be recognized and the target information set, the target prompt information is generated according to the prompt template.

[0090] The information extraction device also includes:

[0091] A module for acquiring texts to be trained, used for acquiring a number of texts to be trained;

[0092] A retrieval module, used for searching the preset knowledge base based on any of the texts to be trained to obtain a set of cargo information;

[0093] The training module is used to input the text to be trained and its associated cargo information set into the large model, so as to fine-tune and train the large model according to the extraction results output by the large model and the training labels of the text to be trained.

[0094] The information extraction device also includes:

[0095] The cargo information extraction result includes cargo entity information and cargo attribute information of the cargo in the text to be identified;

[0096] The information extraction device also includes:

[0097] A vehicle determination module, used to determine the type of transport vehicle that meets the size of the goods according to the goods entity information and the goods attribute information of the goods in the text to be identified;

[0098] A push module is used to push the cargo order associated with the text to be recognized to the driver corresponding to the type of transport vehicle for acceptance.

[0099] For the specific definition of the information extraction device, please refer to the definition of the information extraction method above, which will not be repeated here. Each module in the above-mentioned information extraction device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0100] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 6 As shown, Figure 6: is a schematic diagram of a computer device in one embodiment of the present invention. The computer device includes a processor, a memory, a network interface and a database connected by a device bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium and an internal memory. The readable storage medium stores an operating device, a computer-readable instruction and a database. The internal memory provides an environment for the operation of the operating device and the computer-readable instructions in the readable storage medium. The database of the computer device is used to store data involved in the information extraction method. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer-readable instruction is executed by the processor, an information extraction method is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.

[0101] In one embodiment, a computer device is provided. The computer device may be a terminal device, and its internal structure diagram may be as follows: Figure 6 As shown. The computer device includes a processor, a memory, and a network interface connected through a device bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium. The readable storage medium stores computer-readable instructions. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer-readable instructions are executed by the processor, an information extraction method is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.

[0102] In one embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, and the processor implements the steps of the above-mentioned information extraction method when executing the computer-readable instructions.

[0103] In one embodiment, a readable storage medium is provided, the readable storage medium stores computer-readable instructions, and the computer-readable instructions are executed by the processor to implement the above-mentioned information extraction method steps. It can be understood by a person of ordinary skill in the art that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When the computer-readable instructions are executed, they may include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0104] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0105] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. An information extraction method, characterized in that: include: Get the text to be recognized; Determining a target information set based on the similarity between the text to be recognized and each piece of cargo knowledge information in a preset knowledge base; The preset large model is used to extract information from the text to be recognized and the target information set to obtain a cargo information extraction result.

2. The information extraction method according to claim 1, characterized in that: The step of determining a target information set based on the similarity between the text to be recognized and each piece of cargo knowledge information in a preset knowledge base includes: Determine the similarity between each piece of cargo knowledge information in the preset knowledge base of the text to be identified; Sorting the similarities to select a preset amount of cargo knowledge information based on the similarity sorting results; Based on the cargo knowledge information, the target information set is formed.

3. The information extraction method according to claim 1, characterized in that: Before determining the target information set based on the similarity between the text to be recognized and each cargo knowledge information in the preset knowledge base, the method further includes: Obtain several text information samples; Determine the cargo entity information and cargo attribute information in each of the text information samples, so as to use the cargo entity information and the cargo attribute information as cargo labels of the text information samples; Based on each of the text information samples and the associated cargo labels, cargo knowledge information is generated, and the cargo knowledge information is stored in the preset knowledge base.

4. The information extraction method according to claim 1, characterized in that: The method of extracting information from the to-be-recognized text and the target information set using a preset large model to obtain a cargo information extraction result includes: Generate target prompt information based on the text to be recognized and the target information set; The target promp information is input into the large model to obtain the cargo information extraction result output by the large model.

5. The information extraction method according to claim 4, characterized in that: The generating target prompt information based on the text to be recognized and the target information set includes: Determine a preset prompt template, wherein the prompt template includes a preset input format and an output format, wherein the input format is a format for specifying information input into the big model, and the output format is a format corresponding to the result output by the rule big model; Based on the text to be recognized and the target information set, the target prompt information is generated according to the prompt template.

6. The information extraction method according to claim 1, characterized in that: Before extracting information from the to-be-recognized text and the target information set using the preset large model to obtain the cargo information extraction result, the method further includes: Get several texts to be trained; Based on any of the texts to be trained, searching the preset knowledge base to obtain a cargo information set; The text to be trained and its associated cargo information set are input into the large model, so as to fine-tune and train the large model according to the extraction result output by the large model and the training label of the text to be trained.

7. The information extraction method according to claim 1, characterized in that: The cargo information extraction result includes cargo entity information and cargo attribute information of the cargo in the text to be identified; After extracting information from the to-be-recognized text and the target information set using the preset large model to obtain the cargo information extraction result, the method further includes: Determine the type of transport vehicle that meets the size of the goods according to the goods entity information and the goods attribute information of the goods in the text to be identified; The cargo order associated with the text to be recognized is pushed to the driver corresponding to the transport vehicle type for acceptance.

8. An information extraction device, characterized in that: include: An acquisition module is used to acquire the text to be recognized; A determination module, used to determine a target information set based on the similarity between the text to be identified and each piece of cargo knowledge information in a preset knowledge base; The information extraction module is used to extract information from the text to be recognized and the target information set using a preset large model to obtain a cargo information extraction result.

9. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executed on the processor, characterized in that: When the processor executes the computer-readable instructions, the information extraction method according to any one of claims 1 to 7 is implemented.

10. A readable storage medium having computer readable instructions stored thereon, characterized in that: When the computer-readable instructions are executed by a processor, the information extraction method according to any one of claims 1 to 7 is implemented.

Images