Text content processing method and device based on large model, equipment and storage medium

Abstract

The invention provides a text content processing method and device based on a large model, equipment and a storage medium, and relates to the field of artificial intelligence such as deep learning, large models and natural language processing. The method comprises the following steps: determining first content to be processed, wherein the first content comprises text content of a target text; acquiring context information of the first content from the target text; obtaining first enterprise knowledge information matched with the first content according to enterprise knowledge information included in an enterprise knowledge base; and generating processing result information corresponding to the first content according to the first content, the context information and the first enterprise knowledge information. By applying the scheme of the invention, the time cost of the user can be reduced, the information acquisition efficiency of the user can be improved, and the accuracy of the obtained processing result information can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular to a text content processing method, apparatus, device and storage medium based on a big model in the fields of deep learning, big models and natural language processing. Background Art

[0002] When users read text, they often need to interpret or translate the text (such as translating English into Chinese). For example, if a user does not understand the meaning of a certain word, the traditional way is to manually copy the word and search for it in a search engine. The implementation process is complicated, which increases the time cost and reduces the efficiency of information acquisition. Summary of the invention

[0003] The present disclosure provides a text content processing method, apparatus, device and storage medium based on a large model.

[0004] A text content processing method based on a large model, comprising:

[0005] Determining first content to be processed, wherein the first content includes text content of a target text;

[0006] Acquire context information of the first content from the target text;

[0007] Acquire first enterprise knowledge information matching the first content according to the enterprise knowledge information included in the enterprise knowledge base;

[0008] Processing result information corresponding to the first content is generated according to the first content, the context information and the first enterprise knowledge information.

[0009] A text content processing device based on a large model, comprising: a content determination module, an information acquisition module and a result generation module;

[0010] The content determination module is used to determine a first content to be processed, wherein the first content includes text content of a target text;

[0011] The information acquisition module is used to acquire context information of the first content from the target text, and acquire first enterprise knowledge information matching the first content based on the enterprise knowledge information included in the enterprise knowledge base;

[0012] The result generation module is used to generate processing result information corresponding to the first content according to the first content, the context information and the first enterprise knowledge information.

[0013] An electronic device, comprising:

[0014] at least one processor; and

[0015] a memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform the method described above.

[0017] A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method as described above.

[0018] A computer program product comprises a computer program / instruction, wherein the computer program / instruction implements the method described above when executed by a processor.

[0019] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0021] Figure 1 It is a flowchart of an embodiment of a text content processing method based on a large model described in the present disclosure;

[0022] Figure 2 It is a schematic diagram of the overall implementation process of the text content processing method based on the large model described in the present disclosure;

[0023] Figure 3 It is a schematic diagram of the composition structure of an embodiment 300 of a text content processing device based on a large model described in the present disclosure;

[0024] Figure 4 A schematic block diagram of an electronic device 400 that can be used to implement an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0025] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0026] In addition, it should be understood that the term "and / or" in this article is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0027] Figure 1 Flow chart of an embodiment of the text content processing method based on a large model described in the present disclosure. Figure 1 As shown, the following specific implementation methods are included.

[0028] In step 101, a first content to be processed is determined, where the first content includes text content of a target text.

[0029] In step 102, context information of the first content is obtained from the target text, and first enterprise knowledge information matching the first content is obtained based on the enterprise knowledge information included in the enterprise knowledge base.

[0030] In step 103, processing result information corresponding to the first content is generated according to the first content, the context information and the first enterprise knowledge information.

[0031] By adopting the scheme described in the above method embodiment, the context information and the matching first enterprise knowledge information of the first content to be processed can be automatically obtained, and the processing result information corresponding to the first content can be automatically generated in combination with this information, thereby simplifying user operations, reducing the user's time cost and improving the user's information acquisition efficiency. Moreover, the above-mentioned multiple information can be combined to determine the final processing result information, thereby improving the accuracy of the obtained processing result information. In addition, it can also achieve effective utilization of the knowledge base content within the enterprise.

[0032] The target text may refer to any text, such as a text that the user is reading (browsing).

[0033] In some embodiments of the present disclosure, in response to detecting a selection behavior of a target text, the selected text content may be determined as the first content.

[0034] For example, if the user does not understand the meaning of a certain word or a certain sentence that appears in the target text, the user can use the mouse to select the word or the sentence. Accordingly, the word or the sentence selected by the user can be determined as the first content.

[0035] In the above processing method, the user only needs to perform a swiping operation to select the first content to be processed, and then the processing result information corresponding to the first content can be automatically generated, thereby simplifying the user operation and improving the processing efficiency.

[0036] Alternatively, in some embodiments of the present disclosure, in response to obtaining a processing instruction issued for any specific text content in the target text, the specific text content can be determined as the first content. The specific text content can be text content with a number of selections greater than a predetermined threshold determined based on historical selection behavior. The specific text content is marked in a predetermined manner.

[0037] In addition, in some embodiments of the present disclosure, the predetermined method may include but is not limited to at least one of the following: highlighting method, bolding method, underlining method, and in response to obtaining a processing instruction issued for any specific text content in the target text, it may include: in response to monitoring the mouse moving to the area where the specific text content is located or clicking the area where the specific text content is located with the mouse, determining that the processing instruction is obtained.

[0038] For example, the historical selection behaviors of each user on the target text within the most recent predetermined time period can be analyzed, the number of selections of the text content corresponding to each historical selection behavior can be counted, and the text content with a selection number greater than a predetermined threshold can be determined as the specific text content, and the specific value of the predetermined threshold can be determined according to actual needs. In addition, the specific value of the most recent predetermined time period can also be determined according to actual needs, such as the most recent week.

[0039] For example, specific text content can be marked in a highlighting manner. Furthermore, if the user wants to understand the meaning of a specific text content, the user can move the mouse to the area where the specific text content is located, or click the mouse to the area where the specific text content is located, thereby issuing a processing instruction. Accordingly, the specific text content can be determined as the first content, and subsequent processing can be performed in the manner described in the present disclosure.

[0040] Through the above processing, specific text content that the user has a high probability of selecting can be automatically determined, and can be marked in a predetermined manner and displayed to the user for user viewing. Accordingly, for specific text content, the user does not need to perform a selection operation, but only needs to simply operate the mouse to select it, thereby further simplifying the user's operation and improving processing efficiency.

[0041] After determining the first content, the context information of the first content can be obtained from the target text. There is no restriction on how to obtain the context information. For example, assuming that the user selects the text content in a paragraph in the target text as the first content, the text content in the two paragraphs adjacent to the paragraph can be determined as the context information of the first content. For another example, the target text can be regarded as a web page, which includes nodes one by one, and the other text content in the node where the first content is located (if there are other text contents in addition to the first content) and the text content in the previous node and the next node of the node where the first content is located can be determined as the context information of the first content.

[0042] In addition, the first enterprise knowledge information matching the first content may be determined according to the enterprise knowledge base. In some embodiments of the present disclosure, a search keyword may be determined according to the first content, and then the enterprise knowledge base may be searched using the search keyword to obtain the first enterprise knowledge information.

[0043] The search keyword may refer to one or more core words directly extracted from the first content, or may refer to a result obtained by summarizing the first content, etc. In addition, the search keyword may also be the first content itself.

[0044] An enterprise knowledge base refers to a proprietary knowledge base within an enterprise, which may include office documents, intranet information, and personal public information.

[0045] Through the above processing, enterprise knowledge information matching the first content can be efficiently and accurately retrieved from the enterprise knowledge base, and then used for subsequent processing result information generation, thereby realizing the effective utilization of the knowledge base content within the enterprise, that is, realizing the deep integration of the knowledge base content within the enterprise, which not only enhances the availability of the knowledge base content, but also brings higher information processing capabilities and competitive advantages to the enterprise.

[0046] In addition, in some embodiments of the present disclosure, at least two second enterprise knowledge information matching the first content can be obtained from the enterprise knowledge base, and the quality information of the second enterprise knowledge information can be obtained, and the second enterprise knowledge information is sorted in descending order according to the value of the quality information, and the second enterprise knowledge information in the first M positions after sorting is determined as the first enterprise knowledge information, where M is a positive integer and the value of M is less than the number of the second enterprise knowledge information. The specific value of M can be determined according to actual needs.

[0047] For example, assuming that the value of M is 5, and assuming that the number of second enterprise knowledge information obtained is 20, then the quality information of these 20 second enterprise knowledge information can be obtained respectively, and these 20 second enterprise knowledge information can be sorted in descending order according to the value of the quality information, and then the second enterprise knowledge information in the top 5 after sorting can be determined as the first enterprise knowledge information.

[0048] In some embodiments of the present disclosure, the following processing can be performed for any second enterprise knowledge information: obtain the correlation information between the second enterprise knowledge information and the first content, obtain the freshness information of the second enterprise knowledge information, the closer the generation time of the second enterprise knowledge information is to the current time, the higher the value of the freshness information, obtain the source reliability information of the second enterprise knowledge information, multiply the correlation information, freshness information and source reliability information with the corresponding weights respectively to obtain three product results, and obtain the quality information of the second enterprise knowledge information by adding the three product results.

[0049] In practical applications, the quality information may be a quality score, and correspondingly, the relevance information, freshness information and source reliability information may be a relevance score, a freshness score and a source reliability score, respectively.

[0050] There is no restriction on how to obtain the relevance score, freshness score and source reliability score. For example, the second enterprise knowledge information and the first content can be input into a pre-trained scoring model to obtain an output relevance score. For another example, the time difference between the generation time of the second enterprise knowledge information and the current time can be obtained, and then the freshness score can be obtained by performing a predetermined transformation on the time difference. For another example, different scores can be set for different sources (such as office documents, intranet information, etc., which can be further subdivided if necessary), and the score corresponding to the source of the second enterprise knowledge information can be determined as the source reliability score of the second enterprise knowledge information. The larger the value of the source reliability score, the higher the reliability of the source.

[0051] In addition, corresponding weights can be set for the relevance score, freshness score, and source reliability score, respectively. The weights corresponding to different scores can be the same or different. For ease of description, the three weights are respectively called weight 1, weight 2, and weight 3. Then, the product of the relevance score and weight 1, the product of the freshness score and weight 2, and the product of the source reliability score and weight 3 can be obtained respectively. Then, the three products can be added together, and the sum of the addition is determined as the quality score of the second enterprise knowledge information.

[0052] By screening the second enterprise knowledge information, the workload of subsequent processing can be reduced. In addition, although the second enterprise knowledge information obtained is enterprise knowledge information that matches the first content, the "match" usually refers to the relevance of the content. In the above processing method, in addition to considering the relevance of the content, it further combines the freshness and source reliability to determine the quality score of the second enterprise knowledge information, that is, it combines multiple different evaluation indicators to evaluate the quality of the second enterprise knowledge information, thereby improving the accuracy of the screening results, and then ensuring the quality of the screened first enterprise knowledge information.

[0053] After respectively acquiring the first content, the context information of the first content, and the first enterprise knowledge information, the processing result information corresponding to the first content may be generated in combination with these information.

[0054] In some embodiments of the present disclosure, the first content, the context information of the first content, and the first enterprise knowledge information may be input into a large language model (LLM) to obtain output processing result information.

[0055] The big language model is a deep learning model trained based on massive text data. It can not only generate natural language text, but also deeply understand the meaning of the text, handle various natural language tasks, etc.

[0056] The solution disclosed in the present invention does not limit the structure / type of the large language model. For example, it can be a generative pre-trained model 3 (GPT-3, Generative Pre-trained Transformer 3) based on transformation or a bidirectional encoding representation model (BERT, Bidirectional Encoder Representations from Transformers) based on transformation.

[0057] With the help of the powerful reasoning ability of the large language model, the accuracy of the processing result information can be improved.

[0058] Assuming that the meaning of the first content needs to be explained, the processing result information includes specific meaning explanation information; assuming that the first content needs to be translated, the processing result information includes the translated content.

[0059] In some embodiments of the present disclosure, the output of the large language model may also include: at least part of the third enterprise knowledge information in the input first enterprise knowledge information is used by the large language model and used to generate processing result information. Accordingly, display content information corresponding to the third enterprise knowledge information can also be generated, and the display content information and the processing result information can be output.

[0060] Assuming that five pieces of first enterprise knowledge information are input to the large language model, the large language model will determine by itself which of the first enterprise knowledge information are used to generate the processing result information. Assuming that two of the first enterprise knowledge information are used to generate the processing result information, these two pieces of first enterprise knowledge information can be determined as the third enterprise knowledge information. Accordingly, display content information corresponding to the third enterprise knowledge information can be generated. There is no restriction on what specific content is included in the display content information. For example, it can include the third enterprise knowledge information itself, or include the corresponding link address, or include summary information obtained by summarizing the third enterprise knowledge information, etc. Furthermore, the display content information can be displayed together with the processing result information.

[0061] Through the above processing, the user can obtain more information related to the first content, thereby further improving the user's information acquisition efficiency.

[0062] In some embodiments of the present disclosure, after obtaining the processing result information output by the large language model, subsequent operation suggestion information can also be generated and displayed, and / or, in response to determining that the report generation conditions are met, a statistical report can be generated and displayed based on the generated processing result information.

[0063] For example, a chat window may be displayed, and subsequent operation suggestion information may be displayed in the chat window. For example, if the processing result information includes an uncommon English word, the user may be asked whether it needs to be translated.

[0064] In addition, if it is determined that the user has finished reading the target text, then it can be considered that the report generation conditions are met, or if a report generation instruction from the user is received, it can also be considered that the report generation conditions are met. Taking the previous case as an example, assuming that the user has selected the text content 4 times in total during this reading process, a statistical report can be generated and displayed in combination with the 4 generated processing result information. There is no restriction on what specific content is included in the statistical report, such as the number of selections and an overview of the processing result information generated each time.

[0065] Through the above processing, the interaction with the user is no longer limited to a single response, but a continuous and interactive process, thereby improving the user experience and the user's operational convenience.

[0066] Combined with the above introduction, Figure 2 Schematic diagram of the overall implementation process of the text content processing method based on the large model described in the present disclosure. Figure 2As shown, for the first content in the target text, the context information of the first content can be obtained from the target text, and the first enterprise knowledge information matching the first content can be obtained according to the enterprise knowledge information included in the enterprise knowledge base, such as obtaining at least two second enterprise knowledge information matching the first content from the enterprise knowledge base, and the second enterprise knowledge information can be screened, such as obtaining the quality scores of the second enterprise knowledge information respectively, and sorting the second enterprise knowledge information in descending order according to the quality scores, and determining the second enterprise knowledge information in the first M positions after sorting as the first enterprise knowledge information, where M is a positive integer, and the value of M is less than the number of second enterprise knowledge information. Then, the first content, the context information and the first enterprise knowledge information can be input into the large language model, and the processing result information of the output first content can be obtained and displayed. Among them, before the context information is input into the large language model, the context information can also be optimized, such as deleting the content with low relevance to the first content to improve the processing efficiency of the large language model. With the help of context information, the large language model can better understand the first content, especially in the scenario involving polysemous words.

[0067] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are expressed as a series of action combinations, but those skilled in the art should be aware that the present disclosure is not limited by the described order of actions, because according to the present disclosure, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present disclosure.

[0068] The above is an introduction to the method embodiment. The following is a further explanation of the scheme disclosed in the present invention through an apparatus embodiment.

[0069] Figure 3 FIG. 3 is a schematic diagram of the structure of the text content processing device embodiment 300 based on the large model described in the present disclosure. Figure 3 As shown, it includes: a content determination module 301, an information acquisition module 302 and a result generation module 303.

[0070] The content determination module 301 is used to determine the first content to be processed, where the first content includes the text content of the target text.

[0071] The information acquisition module 302 is used to acquire context information of the first content from the target text, and acquire first enterprise knowledge information matching the first content according to the enterprise knowledge information included in the enterprise knowledge base.

[0072] The result generation module 303 is used to generate processing result information corresponding to the first content according to the first content, the context information and the first enterprise knowledge information.

[0073] By adopting the scheme described in the above-mentioned device embodiment, the context information and the matching first enterprise knowledge information of the first content to be processed can be automatically obtained, and the processing result information corresponding to the first content can be automatically generated in combination with this information, thereby simplifying the user operation, reducing the user's time cost and improving the user's information acquisition efficiency. Moreover, the above-mentioned multiple information can be combined to determine the final processing result information, thereby improving the accuracy of the obtained processing result information. In addition, it can also achieve effective utilization of the knowledge base content within the enterprise.

[0074] In some embodiments of the present disclosure, in response to detecting a selection behavior of a target text, the content determination module 301 may determine the selected text content as the first content.

[0075] Alternatively, in some embodiments of the present disclosure, the content determination module 301 may determine any specific text content in the target text as the first content in response to obtaining a processing instruction issued for the specific text content in the target text. The specific text content may be text content with a selection count greater than a predetermined threshold determined based on historical selection behavior, and the specific text content is marked in a predetermined manner.

[0076] In some embodiments of the present disclosure, the predetermined method may include but is not limited to at least one of the following: highlighting method, bolding method, and underlining method. The content determination module 301 can determine that the processing instruction is obtained in response to detecting that the mouse moves to the area where the specific text content is located or clicking the area where the specific text content is located by the mouse.

[0077] The information acquisition module 302 can obtain context information of the first content from the target text, and can determine the first enterprise knowledge information matching the first content according to the enterprise knowledge base. In some embodiments of the present disclosure, the information acquisition module 302 can determine a search keyword according to the first content, and then use the search keyword to search the enterprise knowledge base to obtain the first enterprise knowledge information.

[0078] Alternatively, in some embodiments of the present disclosure, the information acquisition module 302 can obtain at least two second enterprise knowledge information matching the first content from the enterprise knowledge base, and can obtain quality information of the second enterprise knowledge information, sort the second enterprise knowledge information in descending order of the values ​​of the quality information, and determine the second enterprise knowledge information in the first M positions after sorting as the first enterprise knowledge information, where M is a positive integer and the value of M is less than the number of second enterprise knowledge information.

[0079] In some embodiments of the present disclosure, for any second enterprise knowledge information, the information acquisition module 302 can perform the following processing respectively: obtain the correlation information between the second enterprise knowledge information and the first content, obtain the freshness information of the second enterprise knowledge information, the closer the generation time of the second enterprise knowledge information is to the current time, the higher the value of the freshness information, obtain the source reliability information of the second enterprise knowledge information, multiply the correlation information, freshness information and source reliability information with the corresponding weights respectively to obtain three product results, and obtain the quality information by adding the three product results.

[0080] In addition, in some embodiments of the present disclosure, the result generation module 303 may input the first content, the context information of the first content, and the first enterprise knowledge information into the large language model to obtain processing result information output by the large language model.

[0081] In some embodiments of the present disclosure, the output of the large language model may also include: at least part of the third enterprise knowledge information in the input first enterprise knowledge information is used by the large language model and is used to generate processing result information. Accordingly, the result generation module 303 may also generate display content information corresponding to the third enterprise knowledge information, and may output the display content information and the processing result information.

[0082] In addition, in some embodiments of the present disclosure, after the result generation module 303 obtains the processing result information output by the large language model, it can also generate and display subsequent operation suggestion information, and / or, in response to determining that the report generation conditions are met, it can generate and display a statistical report based on the generated processing result information.

[0083] Figure 3 The specific working process of the illustrated device embodiment can refer to the relevant description in the aforementioned method embodiment and will not be described in detail.

[0084] In short, the use of the solution described in the present disclosure can reduce the user's time cost, improve the user's information acquisition efficiency and improve the accuracy of the processing result information, and can achieve effective utilization of the knowledge base content within the enterprise. In addition, the solution described in the present disclosure can be applied to various industries that require text processing, such as law, finance, education, etc., and can be applied to enterprise-level applications, that is, it has wide applicability.

[0085] The scheme disclosed in the present invention can be applied to the field of artificial intelligence, especially to the fields of deep learning, large models and natural language processing. Artificial intelligence is a discipline that studies how to use computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.). It has both hardware-level and software-level technologies. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, and big data processing. Artificial intelligence software technologies mainly include computer vision technology, speech recognition technology, natural language processing technology, as well as machine learning / deep learning, big data processing technology, knowledge graph technology, and other major directions.

[0086] In addition, the texts and the like in the embodiments described in this disclosure are not intended for a specific user and do not reflect the personal information of a specific user. In the technical solution of this disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0087] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0088] Figure 4 A schematic block diagram of an electronic device 400 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0089] like Figure 4 As shown, the device 400 includes a computing unit 401, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 402 or a computer program loaded from a storage unit 408 to a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the device 400 can also be stored. The computing unit 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0090] A number of components in the device 400 are connected to the I / O interface 405, including: an input unit 406, such as a keyboard, a mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a disk, an optical disk, etc.; and a communication unit 409, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 409 allows the device 400 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0091] The computing unit 401 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI, Artificial Intelligence) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSP, Digital Signal Processing), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 401 performs the various methods and processes described above, such as the methods described in the present disclosure. For example, in some embodiments, the methods described in the present disclosure may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 408. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 400 via the ROM 402 and / or the communication unit 409. When the computer program is loaded into the RAM 403 and executed by the computing unit 401, one or more steps of the methods described in the present disclosure may be performed. Alternatively, in other embodiments, the computing unit 401 may be configured to execute the method described in the present disclosure in any other appropriate manner (for example, by means of firmware).

[0092] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard parts (ASSPs), system on chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0093] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0094] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory (EPROM, Electronically Programmable Read-Only Memory), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM, Compact Disc Read-Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0095] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a cathode ray tube (CRT) or a liquid crystal display (LCD) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0096] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0097] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0098] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0099] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A text content processing method based on a large model, comprising: Determining first content to be processed, wherein the first content includes text content of a target text; Acquire context information of the first content from the target text; Acquire first enterprise knowledge information matching the first content according to the enterprise knowledge information included in the enterprise knowledge base; Processing result information corresponding to the first content is generated according to the first content, the context information and the first enterprise knowledge information.

2. The method according to claim 1, wherein: Determining the first content to be processed includes: In response to detecting a selection behavior of the target text, the selected text content is determined as the first content.

3. The method according to claim 2, wherein: Determining the first content to be processed includes: In response to obtaining a processing instruction issued for any specific text content in the target text, the specific text content is determined as the first content. The specific text content is text content whose number of selections determined based on historical selection behavior is greater than a predetermined threshold. The specific text content is marked in a predetermined manner.

4. The method according to claim 3, wherein: The predetermined mode includes at least one of the following: highlight mode, bold mode, underline mode; The processing instruction issued in response to obtaining any specific text content in the target text includes: In response to monitoring that the mouse moves to the area where the specific text content is located or clicking the area where the specific text content is located by the mouse, it is determined that the processing instruction is acquired.

5. The method according to claim 1, wherein: The acquiring, according to the enterprise knowledge information included in the enterprise knowledge base, first enterprise knowledge information matching the first content comprises: Determine a search keyword according to the first content; The enterprise knowledge base is searched using the search keyword to obtain the first enterprise knowledge information.

6. The method according to claim 1, wherein: The acquiring, according to the enterprise knowledge information included in the enterprise knowledge base, first enterprise knowledge information matching the first content comprises: Acquire at least two second enterprise knowledge information matching the first content from the enterprise knowledge base; Acquiring quality information of the second enterprise knowledge information; Sorting the second enterprise knowledge information in descending order according to the values ​​of the quality information; The second enterprise knowledge information in the first M positions after sorting is determined as the first enterprise knowledge information, where M is a positive integer and the value of M is less than the number of the second enterprise knowledge information.

7. The method according to claim 6, wherein: The obtaining of the quality information of the second enterprise knowledge information includes: For any of the second enterprise knowledge information, the following processing is performed respectively: Acquire correlation information between the second enterprise knowledge information and the first content; Acquire freshness information of the second enterprise knowledge information, the closer the generation time of the second enterprise knowledge information is to the current time, the higher the value of the freshness information; Obtaining source reliability information of the second enterprise knowledge information; Multiplying the relevance information, the freshness information, and the source reliability information by corresponding weights respectively to obtain three product results; The quality information is obtained by adding the three multiplication results.

8. The method according to any one of claims 1 to 7, wherein: The generating of processing result information corresponding to the first content according to the first content, the context information and the first enterprise knowledge information includes: The first content, the context information and the first enterprise knowledge information are input into a large language model to obtain the output processing result information.

9. The method according to claim 8, wherein: The output of the large language model also includes: at least part of the third enterprise knowledge information in the input first enterprise knowledge information that is used by the large language model and is used to generate the processing result information; The method further comprises: Generating display content information corresponding to the third enterprise knowledge information; The display content information and the processing result information are output.

10. The method according to claim 8, further comprising: After obtaining the output processing result information, subsequent operation suggestion information is generated and displayed.

11. A text content processing device based on a large model, comprising: content determination module, information acquisition module and result generation module; The content determination module is used to determine a first content to be processed, wherein the first content includes text content of a target text; The information acquisition module is used to acquire context information of the first content from the target text, and acquire first enterprise knowledge information matching the first content based on the enterprise knowledge information included in the enterprise knowledge base; The result generation module is used to generate processing result information corresponding to the first content according to the first content, the context information and the first enterprise knowledge information.

12. The device according to claim 11, wherein: In response to detecting a selection action on the target text, the content determination module determines the selected text content as the first content.

13. The device according to claim 12, wherein: In response to obtaining a processing instruction issued for any specific text content in the target text, the content determination module determines the specific text content as the first content. The specific text content is text content with a number of selections greater than a predetermined threshold determined based on historical selection behavior, and the specific text content is marked in a predetermined manner.

14. The device according to claim 13, wherein: The predetermined mode includes at least one of the following: highlight mode, bold mode, underline mode; The content determination module determines that the processing instruction is acquired in response to monitoring that the mouse moves to the area where the specific text content is located or clicking the area where the specific text content is located by the mouse.

15. The device according to claim 11, wherein The information acquisition module determines a search keyword according to the first content, and uses the search keyword to search the enterprise knowledge base to obtain the first enterprise knowledge information.

16. The device according to claim 11, wherein The information acquisition module acquires at least two second enterprise knowledge information matching the first content from the enterprise knowledge base, acquires quality information of the second enterprise knowledge information, sorts the second enterprise knowledge information in descending order according to the values ​​of the quality information, and determines the second enterprise knowledge information in the first M positions after sorting as the first enterprise knowledge information, where M is a positive integer and the value of M is less than the number of the second enterprise knowledge information.

17. The device according to claim 16, wherein: The information acquisition module performs the following processing for any second enterprise knowledge information: obtains the correlation information between the second enterprise knowledge information and the first content, obtains the freshness information of the second enterprise knowledge information, the closer the generation time of the second enterprise knowledge information is to the current time, the higher the value of the freshness information, obtains the source reliability information of the second enterprise knowledge information, multiplies the correlation information, the freshness information and the source reliability information with the corresponding weights respectively to obtain three product results, and obtains the quality information by adding the three product results.

18. The device according to any one of claims 11 to 17, wherein: The result generation module inputs the first content, the context information and the first enterprise knowledge information into a large language model to obtain the output processing result information.

19. The device according to claim 18, wherein The output of the large language model also includes: at least part of the third enterprise knowledge information in the input first enterprise knowledge information that is used by the large language model and is used to generate the processing result information; The result generation module is further used to generate display content information corresponding to the third enterprise knowledge information, and output the display content information and the processing result information.

20. The device according to claim 18, wherein The result generation module is further used to generate and display subsequent operation suggestion information after obtaining the output processing result information.

21. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 10.

22. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to make a computer execute the method according to any one of claims 1 to 10.

23. A computer program product, comprising a computer program / instruction, wherein when the computer program / instruction is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

Images