Work summary generation method and device, electronic equipment and storage medium

Abstract

The present disclosure provides a work summary generation method and device, electronic equipment and storage medium, relating to the technical field of computers, especially the technical field of artificial intelligence such as large language models, natural language processing and deep learning, which can be applied to scenarios such as office software. The specific implementation scheme is: receiving a work summary generation request sent by a first client, wherein the generation request includes a target period and professional information of a user to which the first client belongs; determining a target data source according to the professional information; obtaining target work data associated with the user in the target period from the target data source; inputting the target work data and a first preset prompt information into a first large model to obtain a work summary of the user in the target period output by the first large model.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and in particular to the fields of artificial intelligence technology such as large language models, natural language processing, and deep learning. Specifically, it relates to a method, apparatus, electronic device, and storage medium for generating work summaries. Background Technology

[0002] In enterprises, users have a need to periodically summarize their work progress. Therefore, how to generate work summaries more accurately and quickly is the key to improving users' work efficiency. Summary of the Invention

[0003] This disclosure aims to at least partially address one of the technical problems in the related art.

[0004] Therefore, the purpose of this disclosure is to provide a method, apparatus, electronic device, and storage medium for generating work summaries, which not only ensures the accuracy and personalization of the generated work summaries, but also reduces the time users spend writing work summaries and improves users' work efficiency.

[0005] According to the first aspect of this disclosure, a method for generating a work summary is provided, comprising:

[0006] Receive a work summary generation request sent by a first client, wherein the generation request includes a target time period and the occupational information of the user to whom the first client belongs;

[0007] Based on the occupational information, the target data source is determined;

[0008] Obtain the target work data associated with the user within the target time period from the target data source;

[0009] The target work data and the first preset prompt information are input into the first large model to obtain the work summary of the user during the target time period output by the first large model.

[0010] According to a second aspect of this disclosure, an apparatus for generating a work summary is provided, comprising:

[0011] The first receiving module is used to receive a work summary generation request sent by the first client, wherein the generation request includes the target time period and the occupational information of the user to which the first client belongs;

[0012] The first determining module is used to determine the target data source based on the occupational information;

[0013] The first acquisition module is used to acquire the target work data associated with the user within the target time period from the target data source;

[0014] The second acquisition module is used to input the target work data and the first preset prompt information into the first large model to obtain the work summary of the user in the target time period output by the first large model.

[0015] According to a third aspect of this disclosure, an electronic device is provided, comprising:

[0016] At least one processor; and

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method for generating a work summary as described in the first aspect.

[0019] According to a fourth aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions for causing the computer to perform a method for generating a work summary as described in the first aspect.

[0020] According to a fifth aspect of this disclosure, a computer program product is provided, including computer instructions that, when executed by a processor, implement the steps of the method for generating a work summary as described in the first aspect.

[0021] The method, apparatus, electronic device, and storage medium for generating the work summary provided in this disclosure have the following beneficial effects:

[0022] In this disclosure, by utilizing a large model to process target work data obtained based on the user's professional information and target time period, a user's work summary can be generated. This not only ensures the accuracy and personalization of the generated work summary, but also reduces the time users spend writing work summaries and improves their work efficiency.

[0023] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0024] The above and / or additional aspects and advantages of this disclosure will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, which are provided for a better understanding of the present invention and are not intended to limit the scope of this disclosure, wherein:

[0025] Figure 1 This is a flowchart illustrating a method for generating a work summary according to an embodiment of this disclosure;

[0026] Figure 2 This is a flowchart illustrating a method for generating a work summary according to another embodiment of this disclosure;

[0027] Figure 3 This is a schematic diagram of the structure of a work summary generation apparatus according to an embodiment of the present disclosure;

[0028] Figure 4 A block diagram of an exemplary electronic device suitable for implementing embodiments of the present disclosure is shown. Detailed Implementation

[0029] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0030] This disclosure relates to the fields of artificial intelligence technologies such as large models and deep learning.

[0031] Artificial Intelligence (AI) is a new technological science that studies and develops theories, methods, technologies, and application systems to simulate, extend, and expand human intelligence.

[0032] Large Language Models (LLMs) are deep learning models trained on large amounts of text data that can generate natural language text or understand the meaning of language text. LLMs can handle various natural language tasks, such as text classification, question answering, and dialogue, and are an important pathway to artificial intelligence.

[0033] Natural Language Processing (NLP) is an important field within computer science and artificial intelligence. It studies the theories and methods that enable effective communication between humans and computers using natural language.

[0034] Deep learning learns the inherent patterns and hierarchical representations of sample data. The information gained during this learning process greatly aids in interpreting data such as text, images, and sound. The ultimate goal of deep learning is to enable machines to possess analytical and learning capabilities similar to humans, allowing them to recognize data such as text, images, and sound.

[0035] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0036] The following description, with reference to the accompanying drawings, outlines a method, apparatus, electronic device, and storage medium for generating work summaries according to embodiments of this disclosure.

[0037] It should be noted that the execution entity of the work summary generation method in this embodiment is a work summary generation device, which can be implemented by software and / or hardware. This device can be configured in an electronic device, which may include, but is not limited to, a terminal or a server. This embodiment uses the example of a work summary generation device being configured into a work summary generation system for illustration.

[0038] Figure 1 This is a flowchart illustrating a method for generating a work summary according to an embodiment of this disclosure.

[0039] like Figure 1 As shown, the method for generating this work summary includes:

[0040] S101: Receive the work summary generation request sent by the first client.

[0041] The generated request includes the target time period and the occupational information of the user to whom the first client belongs.

[0042] It should be noted that the target period can be one week or one quarter, etc., and the occupational information of the user can include occupational name, occupational nature, occupational content, etc., which are not limited in this disclosure.

[0043] S102: Determine the target data source based on occupational information.

[0044] Because users in different professions have different job content, the data sources they refer to when generating work summaries may differ. Therefore, this disclosure allows for the determination of different types of data to generate work summaries based on users' different professional information. For example, candidate data sources include meeting data, document data, business trip data, order data, etc. For marketing personnel, the target data might be business trip data, order data, etc., while for clerical personnel, the target data might be document data, meeting data, etc.

[0045] S103: Obtain the target work data associated with the user within the target time period from the target data source.

[0046] In this embodiment of the disclosure, the creation time and / or modification time of the data can be queried in the attribute information of each target data source, so that the target time period for generating a work summary based on the user's goal can be obtained from the target data source, and the target work data with the corresponding creation time and / or modification time within the target time period can be obtained.

[0047] In this embodiment of the disclosure, when the number of associated work data within the target time period is greater than the number threshold, the feature value of the work data in each dimension can be determined. Then, based on the feature value, the score corresponding to each work data is calculated by methods such as weighted summation. Then, based on the score, a portion of the work data is selected as the target work data. This may be selecting the N work data with the highest scores, or it may be selecting the work data with corresponding scores greater than the score threshold, etc. This disclosure does not limit this.

[0048] S104: Input the target work data and the first preset prompt information into the first large model to obtain the user's work summary for the target time period output by the first large model.

[0049] The first major model refers to the Large Language Model (LLM), which can be a model like Wenxin Yiyan. The first preset prompt instructs the major model to output a work summary based on the target work data and in the target output format. For example, the first preset prompt could be: "You are a work summary assistant. Based on the user's written documents and attended meetings, summarize the user's work content at the topic level, strictly imitating the output format shown in the example. Do not create your own format. Example: ..., Input: {Target Work Data}, Output: ...".

[0050] It should be noted that the examples in the first preset prompt information are used to standardize the format of the output information of the large model, and can be determined according to actual needs. This disclosure does not limit this.

[0051] It should be noted that when the length of a single target work data is long, inputting multiple target work data together into the first large model to generate a summary may result in an exceptionally long input information. Since the input length of the large model is limited, and redundant information may also affect the accuracy of the generated summary, it is advisable to generate summary data corresponding to each target work data before generating the work summary.

[0052] Optionally, if the amount of target work data exceeds a data volume threshold, the target work data and a second preset prompt message are input into the second large model to obtain summary data corresponding to the target work data output by the second large model. Then, the summary data and the first preset prompt message are input into the first large model to obtain a summary of the user's work during the target time period output by the first large model.

[0053] The data volume threshold can be determined based on the maximum input length of the large model, etc., and this disclosure does not impose any limitations on it. The second large model can be the same as the first large model, or it can be a different large language model.

[0054] The second preset prompt is used to instruct the second major model to summarize the single target working data. For example, the second preset prompt could be: "You are a document summary generation assistant. Please generate a summary of the provided document based on its content. The summary should be around 200 words."

[0055] In this embodiment, a summary is generated from the target work data, which has a large volume. Based on the generated summary and prompts, the data is then input into a large model to obtain a work summary. This improves the quality and accuracy of the work summary, reduces the processing of redundant information during the summary generation process, and increases the efficiency of work summary generation.

[0056] In this embodiment, a work summary generation request is first received from a first client. Then, based on the professional information of the user to whom the first client belongs, a target data source is determined. Target work data associated with the user within a target time period is obtained from the target data source. The target work data and a first preset prompt are then input into a first large model to obtain the user's work summary for the target time period, output by the first large model. Thus, by utilizing a large model to process the target work data obtained based on the user's professional information and the target time period, the user's work summary can be generated. This not only ensures the accuracy and personalization of the generated work summary but also reduces the time spent by the user in writing the work summary, improving the user's work efficiency.

[0057] Figure 2 This is a flowchart illustrating a method for generating a work summary according to another embodiment of this disclosure.

[0058] like Figure 2 As shown, the method for generating this work summary includes:

[0059] S201: Receive the work summary generation request sent by the first client.

[0060] S202: Determine the target data source based on occupational information.

[0061] The descriptions of S201 and S202 above can be found in the above embodiments, and will not be repeated here.

[0062] S203: Retrieve candidate job data associated with the user within the target time period from the target data source.

[0063] In this embodiment of the disclosure, there may be a situation where the amount of work data in the target data source and within the target time period is particularly large. If all the work data is summarized, the information will be redundant and will affect the generation rate of the work summary. Therefore, the work data can be scored to identify the more important target work data.

[0064] S204: When the number of candidate working data exceeds the number threshold, determine the feature value of each candidate working data in each dimension.

[0065] The quantity threshold can be a value determined based on actual needs, and this disclosure does not impose any limitations on it. The eigenvalues ​​can be used to assess the importance of each piece of working data.

[0066] It should be noted that the feature dimensions corresponding to different types of candidate working data may be the same or different.

[0067] In this embodiment of the disclosure, the dimensions of document-type work data may include at least one of the following: number of citations, number of views, number of likes, number of comments, number of favorites, and the time difference between the latest update time and the creation time. Therefore, the importance of document-type work data can be evaluated from multiple personalized perspectives, providing conditions for improving the accuracy of target work data selection.

[0068] In this embodiment of the disclosure, the dimensions of meeting-related work data may include at least one of the following: meeting duration, number of documents associated with the meeting, number of documents provided by users during the meeting, number of participants, job information of participants, correlation between participants and users, and information of the meeting initiator. Therefore, the importance of meeting-related work data can be evaluated from multiple personalized perspectives, providing conditions for improving the accuracy of target work data filtering.

[0069] It should be noted that the relevance between participants and users refers to the average intimacy score between the user and all participants. This intimacy score can be obtained from existing models and will not be elaborated upon here. Based on the meeting initiator information, corresponding feature values ​​can be determined. For example, if the initiator is the user themselves, the feature value can be 1; or, if the initiator is the user's superior, the feature value can be 1; or, if the initiator is someone else, the feature value can be the intimacy score.

[0070] S205: Determine the score value for each candidate job data based on the feature values ​​of each candidate job data in each dimension.

[0071] In this embodiment of the disclosure, the score value corresponding to the candidate working data can be calculated by weighted summation of the feature values ​​of the candidate working data in each dimension.

[0072] Optionally, the weights corresponding to the feature values ​​of each dimension in any candidate work data and the scoring pattern of any candidate work data can be determined based on the type of any candidate work data. Then, based on the scoring pattern, the score value of any candidate work data is determined according to the feature values ​​and corresponding weights of each dimension.

[0073] The scoring model refers to the score calculation formula corresponding to the dimensions of different types of work data.

[0074] In this embodiment, for different types of work data, the weights corresponding to feature values ​​under the same dimension may be the same or different. For example, the weights for document-type data under the document citation count dimension may differ from those for meeting-type data. Furthermore, since the feature value dimensions for different types of candidate work data can be different, a corresponding scoring mode can be determined based on the type of candidate work data. Then, by substituting the feature values ​​and corresponding weights of any candidate work data in each dimension into the scoring mode, a score value for any candidate work data can be obtained. This allows for targeted scoring of each candidate work data, improving the accuracy of the score values ​​and providing conditions for improving the reliability of the work summary.

[0075] Optionally, if the weight value corresponding to any dimension is less than a weight threshold, the feature value of each candidate working data in any dimension can be determined. Then, if the feature value of any candidate working data in any dimension is greater than a feature value threshold, the feature value of that dimension is compressed to obtain the compressed feature value of that dimension. Afterwards, based on the scoring mode, and using the compressed feature value and corresponding weight of each candidate working data in each dimension, the score value of that candidate working data is determined.

[0076] In this embodiment, if a feature value is greater than a feature value threshold in any dimension where the corresponding weight value is less than the weight threshold, the influence of the weight may be amplified, leading to inaccurate scores. Therefore, feature values ​​in that dimension can be compressed, for example, by multiplying the feature value by a compression coefficient, to obtain compressed feature values, which are then used to determine the score for any candidate work data. This reduces the influence of unimportant feature values ​​in the score calculation process, further improving the accuracy and reliability of the scores, and providing conditions for improving the reliability of the work summary.

[0077] Optionally, a compression coefficient can be determined based on the weight value corresponding to any dimension and / or the correlation between any dimension and each type of data. Then, based on the compression coefficient, the feature values ​​of any dimension are compressed to obtain the compressed feature values ​​of any dimension.

[0078] In this embodiment of the disclosure, the compression coefficient can be determined based on the weight value corresponding to the dimension; the smaller the weight value, the larger the compression coefficient. Alternatively, if the weights corresponding to two dimensions are the same, and one dimension is only related to one type of data, it indicates that the dimension is relatively important to this type of data, and therefore the corresponding compression coefficient may be smaller than the compression coefficient corresponding to the other dimension.

[0079] In this embodiment of the disclosure, the compression coefficient and the feature value can be substituted into formula (1) to calculate the compressed feature value, as shown in formula (1) below:

[0080]

[0081] Where alpha is the compression coefficient and x is the eigenvalue of any dimension.

[0082] In this embodiment, the compression coefficient is determined by the weight value corresponding to any dimension and / or the correlation between any dimension and each type of data, thereby obtaining the compressed feature value. This not only ensures the reliability of feature value compression and further improves the accuracy of the score value, but also improves the efficiency of the compression process.

[0083] For example, the rating of document-type work data can be calculated using the rating model shown in formula (2) below:

[0084] score = (Number of citations * weight of citation feature + sigmoid(number of views, alpha = 0.05) * weight of view feature + number of likes * weight of like feature + number of comments * weight of comment feature + number of favorites * weight of favorites feature + sigmoid(time difference, alpha = 0.09) * weight of time difference feature) (2)

[0085] The weights for each dimension are shown in Table 1 below.

[0086] Referenced feature weights 5 Browsing feature weights 3 Like feature weights 4 Collection feature weight 4 Time difference feature weights 3

[0087] Table 1

[0088] S206: Select target job data from the candidate job data based on the score value of each candidate job data.

[0089] In this embodiment of the disclosure, the top N data points with the highest scores among the candidate working data can be determined as the target working data, where N can be any value, such as 5. Alternatively, candidate working data points with scores greater than a scoring threshold can also be determined as the target working data. This disclosure does not limit this approach.

[0090] S207: Select target job data from the candidate job data based on the score value of each candidate job data.

[0091] The description of S207 above can be found in the above embodiments, and will not be repeated here.

[0092] In this embodiment, candidate work data associated with the user within the target time period are first obtained from the target data source. Then, if the number of candidate work data exceeds a threshold, the feature value of each candidate work data in each dimension is determined. Based on the feature values ​​of each candidate work data in each dimension, a score value is determined for each candidate work data. Finally, based on the score values ​​of each candidate work data, the target work data is selected from the candidate work data. Therefore, by scoring the candidate work data to determine the target work data, the reliability of the data used to generate the work summary is ensured, further improving the efficiency and accuracy of work summary generation.

[0093] Figure 3 This is a schematic diagram of the structure of a work summary generation device according to an embodiment of the present disclosure.

[0094] like Figure 3 As shown, the apparatus 300 for generating this work summary includes:

[0095] The first receiving module 301 is used to receive a work summary generation request sent by the first client, wherein the generation request includes the target time period and the occupational information of the user to whom the first client belongs;

[0096] The first determining module 302 is used to determine the target data source based on occupational information;

[0097] The first acquisition module 303 is used to acquire the target work data associated with the user within the target time period from the target data source;

[0098] The second acquisition module 304 is used to input the target work data and the first preset prompt information into the first large model to obtain the user's work summary during the target time period output by the first large model.

[0099] In some embodiments, the first acquisition module 303 is specifically used for:

[0100] Retrieve candidate job data associated with users within the target time period from the target data source;

[0101] If the number of candidate work data exceeds a threshold, determine the feature value of each candidate work data in each dimension;

[0102] Based on the feature values ​​of each candidate job data in each dimension, determine the score value of each candidate job data;

[0103] Based on the score of each candidate job data, the target job data is selected from the candidate job data.

[0104] In some embodiments, the first acquisition module 303 is specifically used for:

[0105] Based on the type of any candidate work data, determine the weight corresponding to each dimension feature value in any candidate work data and the scoring mode of any candidate work data;

[0106] Based on the scoring model, the score of any candidate work data is determined according to the feature value and corresponding weight of each dimension.

[0107] In some embodiments, the first acquisition module 303 is specifically used for:

[0108] If the weight value corresponding to any dimension is less than the weight threshold, determine the feature value of each candidate working data in any dimension;

[0109] If the feature value of any candidate working data in any dimension is greater than the feature value threshold, the feature value of any dimension is compressed to obtain the compressed feature value of any dimension.

[0110] Based on the scoring model, the score of any candidate work data is determined by the compressed feature values ​​and corresponding weights of each dimension.

[0111] In some embodiments, the first acquisition module 303 is specifically used for:

[0112] The compression factor is determined based on the weight value corresponding to any dimension and / or the correlation between any dimension and each type of data.

[0113] Based on the compression coefficient, the feature value of any dimension is compressed to obtain the compressed feature value of any dimension.

[0114] In some embodiments, the dimensions of document-type work data include at least one of the following: number of citations, number of views, number of likes, number of comments, number of favorites, and the time difference between the latest update time and the creation time.

[0115] In some embodiments, the dimensions of meeting-related work data include at least one of the following: meeting duration, number of documents associated with the meeting, number of documents provided by users during the meeting, number of participants, job information of participants, degree of association between participants and users, and information of the meeting initiator.

[0116] In some embodiments, the second acquisition module 304 is specifically used for:

[0117] If the amount of target working data exceeds the data volume threshold, the target working data and the second preset prompt information are input into the second large model to obtain the summary data corresponding to the target working data output by the second large model.

[0118] The summary data and the first preset prompt information are input into the first large model to obtain the user's work summary for the target time period output by the first large model.

[0119] It should be noted that the foregoing explanation of the method for generating work summaries also applies to the work summary generation device of this embodiment, and will not be repeated here.

[0120] In this embodiment, a work summary generation request is first received from a first client. Then, based on the professional information of the user to whom the first client belongs, a target data source is determined. Target work data associated with the user within a target time period is obtained from the target data source. The target work data and a first preset prompt are then input into a first large model to obtain the user's work summary for the target time period, output by the first large model. Thus, by utilizing a large model to process the target work data obtained based on the user's professional information and the target time period, the user's work summary can be generated. This not only ensures the accuracy and personalization of the generated work summary but also reduces the time spent by the user in writing the work summary, improving the user's work efficiency.

[0121] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0122] Figure 4 A schematic block diagram of an example electronic device 400 that can be used to implement embodiments 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, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0123] like Figure 4 As shown, device 400 includes a computing unit 401, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 402 or a computer program loaded from storage unit 408 into random access memory (RAM) 403. RAM 403 may also store various programs and data required for the operation of device 400. The computing unit 401, ROM 402, and RAM 403 are interconnected via bus 404. Input / output (I / O) interface 405 is also connected to bus 404.

[0124] Multiple components in device 400 are connected to I / O interface 405, including: input unit 406, such as keyboard, mouse, etc.; output unit 407, such as various types of monitors, speakers, etc.; storage unit 408, such as disk, optical disk, etc.; and communication unit 409, such as network card, modem, wireless transceiver, etc. Communication unit 409 allows device 400 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0125] The computing unit 401 can be a variety of general-purpose and / or special-purpose 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 special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 401 performs the various methods and processes described above, such as the method for generating a work summary. For example, in some embodiments, the method for generating a work summary may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 408. In some embodiments, part or all of the computer program may be loaded and / or installed on device 400 via ROM 402 and / or communication unit 409. When the computer program is loaded into RAM 403 and executed by the computing unit 401, one or more steps of the method for generating a work summary described above may be performed. Alternatively, in other embodiments, the computing unit 401 may be configured to perform the method for generating a work summary by any other suitable means (e.g., by means of firmware).

[0126] 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 products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0127] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0128] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0129] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, 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 sound input, voice input, or tactile input).

[0130] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), the Internet, and blockchain networks.

[0131] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service ecosystem, addressing the shortcomings of traditional physical hosts and VPS (Virtual Private Server, or simply "VPS") services, such as high management difficulty and weak business scalability. Servers can also be servers for distributed systems or servers incorporating blockchain technology.

[0132] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0133] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this disclosure, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified. In the description of this disclosure, the words "if" and "suppose" as used may be interpreted as "when," "in response to a determination," or "in the circumstances."

[0134] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for generating a work summary, comprising: Receive a work summary generation request sent by a first client, wherein the generation request includes a target time period and the occupational information of the user to whom the first client belongs; Based on the occupational information, the target data source is determined; Obtain the target work data associated with the user within the target time period from the target data source; The target work data and the first preset prompt information are input into the first large model to obtain the work summary of the user during the target time period output by the first large model. The step of obtaining the target work data associated with the user within the target time period from the target data source includes: Obtain candidate job data associated with the user within the target time period from the target data source; If the number of candidate working data is greater than a threshold, determine the feature value of each candidate working data in each dimension; Based on the feature values ​​of each candidate job data in each dimension, a score value is determined for each candidate job data. Based on the score value of each candidate job data, the target job data is selected from the candidate job data; The step of determining the score value of each candidate job data based on the feature values ​​of each candidate job data in each dimension includes: Based on the type of any candidate work data, determine the weight corresponding to each dimension feature value in the candidate work data and the scoring mode of the candidate work data; Based on the scoring model, the score value of any candidate work data is determined according to the feature value and corresponding weight of each dimension. The step of determining the score value of any candidate work data based on the scoring model, using the feature values ​​and corresponding weights of each dimension, includes: If the weight value corresponding to any dimension is less than the weight threshold, determine the feature value of each candidate working data in that dimension; If the feature value of any candidate working data in any dimension is greater than the feature value threshold, the feature value of any dimension is compressed to obtain the compressed feature value of any dimension. Based on the scoring model, the score value of any candidate working data is determined according to the compressed feature value and corresponding weight of each dimension.

2. The method as described in claim 1, wherein, The step of compressing the feature value of any dimension to obtain the compressed feature value of any dimension includes: The compression coefficient is determined based on the weight value corresponding to any dimension and / or the correlation between any dimension and each type of data. Based on the compression coefficient, the feature value of any dimension is compressed to obtain the compressed feature value of any dimension.

3. The method as described in claim 1, wherein, Document-related work data dimensions include at least one of the following: number of citations, number of views, number of likes, number of comments, number of favorites, and the time difference between the latest update time and the creation time.

4. The method of claim 1, wherein, The dimensions of meeting-related work data include at least one of the following: meeting duration, number of documents associated with the meeting, number of documents provided by the user in the meeting, number of participants, job information of the participants, degree of association between the participants and the user, and information of the meeting initiator.

5. The method as described in any one of claims 1-4, wherein, The step of inputting the target work data and the first preset prompt information into the first large model to obtain the user's work summary during the target time period output by the first large model includes: If the amount of target working data is greater than the data amount threshold, the target working data and the second preset prompt information are input into the second large model to obtain the summary data corresponding to the target working data output by the second large model. The summary data and the first preset prompt information are input into the first large model to obtain the user's work summary during the target time period output by the first large model.

6. A device for generating a work summary, comprising: The first receiving module is used to receive a work summary generation request sent by the first client, wherein the generation request includes the target time period and the occupational information of the user to which the first client belongs; The first determining module is used to determine the target data source based on the occupational information; The first acquisition module is used to acquire the target work data associated with the user within the target time period from the target data source; The second acquisition module is used to input the target work data and the first preset prompt information into the first large model to obtain the work summary of the user in the target time period output by the first large model. The first acquisition module is specifically used for: Obtain candidate job data associated with the user within the target time period from the target data source; If the number of candidate working data is greater than a threshold, determine the feature value of each candidate working data in each dimension; Based on the feature values ​​of each candidate job data in each dimension, a score value is determined for each candidate job data. Based on the score value of each candidate job data, the target job data is selected from the candidate job data; The first acquisition module is specifically used for: Based on the type of any candidate work data, determine the weight corresponding to each dimension feature value in the candidate work data and the scoring mode of the candidate work data; Based on the scoring model, the score value of any candidate work data is determined according to the feature value and corresponding weight of each dimension. The first acquisition module is specifically used for: If the weight value corresponding to any dimension is less than the weight threshold, determine the feature value of each candidate working data in that dimension; If the feature value of any candidate working data in any dimension is greater than the feature value threshold, the feature value of any dimension is compressed to obtain the compressed feature value of any dimension. Based on the scoring model, the score value of any candidate working data is determined according to the compressed feature value and corresponding weight of each dimension.

7. The apparatus of claim 6, wherein, The first acquisition module is specifically used for: The compression coefficient is determined based on the weight value corresponding to any dimension and / or the correlation between any dimension and each type of data. Based on the compression coefficient, the feature value of any dimension is compressed to obtain the compressed feature value of any dimension.

8. The apparatus of claim 6, wherein, Document-related work data dimensions include at least one of the following: number of citations, number of views, number of likes, number of comments, number of favorites, and the time difference between the latest update time and the creation time.

9. The apparatus of claim 6, wherein, The dimensions of meeting-related work data include at least one of the following: meeting duration, number of documents associated with the meeting, number of documents provided by the user in the meeting, number of participants, job information of the participants, degree of association between the participants and the user, and information of the meeting initiator.

10. The apparatus according to any one of claims 6-9, wherein, The second acquisition module is specifically used for: If the amount of target working data is greater than the data amount threshold, the target working data and the second preset prompt information are input into the second large model to obtain the summary data corresponding to the target working data output by the second large model. The summary data and the first preset prompt information are input into the first large model to obtain the user's work summary during the target time period output by the first large model.

11. 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 executable by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the method for generating a work summary according to any one of claims 1-5.

12. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, in, The computer instructions are used to cause the computer to execute the method for generating a work summary according to any one of claims 1-5.

13. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the steps of the method for generating a work summary according to any one of claims 1-5.

Images