Data processing method and device, computer equipment and computer readable storage medium

Abstract

The invention discloses a data processing method and device, computer equipment and a computer readable storage medium. The method comprises the steps of obtaining to-be-processed information and model operation information; the model operation information comprises model information and operation prompt information; based on model information in the model operation information, determining a target model corresponding to the to-be-processed information; and processing the to-be-processed information based on the target model and operation prompt information in the model operation information to obtain a target processing result. According to the application, the difficulty of using patents by non-professional users can be reduced, and the patent information use efficiency and viscosity can be improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a data processing method, apparatus, computer equipment and computer-readable storage medium. Background Art

[0002] Patents are the most effective carrier of technical information. Compared with other forms of documents, they are more novel and practical. However, patent language is relatively difficult to understand and reading, and it takes a lot of time and effort to capture key information. Therefore, patent tools are needed in the prior art to process patent documents. However, the use of various existing patent tools requires a high level of professionalism, and it is difficult for non-professional users to use various patent tools to process patent documents, resulting in relatively low efficiency in the mining and utilization of high-value patents. Summary of the invention

[0003] The embodiments of the present application provide a data processing method, apparatus, computer device, and computer-readable storage medium, which can reduce the difficulty of using patents for non-professional users and improve the efficiency and viscosity of patent information use.

[0004] The technical solution adopted by the present invention to solve the problem is as follows:

[0005] On the one hand, the present application provides a data processing method, comprising:

[0006] Obtain information to be processed and model operation information; model operation information includes model information and operation prompt information;

[0007] Based on the model information in the model operation information, determining a target model corresponding to the information to be processed;

[0008] The information to be processed is processed based on the target model and the operation prompt information in the model operation information to obtain the target processing result.

[0009] In some implementation schemes of the present application, obtaining information to be processed and model operation information includes:

[0010] Receiving a selection instruction input by a user in a data processing function interface;

[0011] Generate model operation information based on the selection instruction and load the information input interface;

[0012] Get the information to be processed entered by the user in the information input interface.

[0013] In some embodiments of the present application, the information to be processed is processed based on the target model and the operation prompt information in the model operation information to obtain the target processing result, including:

[0014] Pre-process the information to be processed to obtain feature vector information;

[0015] The characteristic vector information is processed based on the target model and the operation prompt information in the model operation information to obtain the target processing result.

[0016] In some implementation schemes of the present application, the information to be processed is pre-processed to obtain feature vector information, including:

[0017] Acquiring type information of the information to be processed; the type information includes patent document information, and / or, non-patent document information, and / or, non-document format information;

[0018] If the type information is patent document information, performing first feature extraction on the information to be processed to obtain first feature information, and performing vectorization processing on the first feature information to obtain feature vector information;

[0019] If the type information is non-patent document information, performing second feature extraction on the information to be processed to obtain second feature information, and performing vectorization processing on the second feature information to obtain feature vector information;

[0020] If the type information is non-document format information, vectorization is performed on the information to be processed to obtain feature vector information.

[0021] In some embodiments of the present application, the target processing result includes summary information, and the target model includes a retrieval model, a first information processing model, and a first structured processing model;

[0022] The feature vector information is processed based on the target model and the operation prompt information in the model operation information to obtain the target processing result, including:

[0023] Based on the information to be processed, determine whether related data retrieval is required;

[0024] When it is necessary to retrieve associated data, the feature vector information is input into the retrieval model, and the target associated data is output through the retrieval model;

[0025] Inputting the target association data and the feature vector information into a first information processing model, and outputting the text feature information through the first information processing model;

[0026] The text feature information is input into the first structured processing model, and the summary information is output through the first structured processing model.

[0027] In some embodiments of the present application, the target processing result includes classification label information, and the target model includes a second information processing model;

[0028] The feature vector information is processed based on the target model and the operation prompt information in the model operation information to obtain the target processing result, including:

[0029] Get feature screening information;

[0030] Performing feature screening on the feature vector information based on the feature screening information to obtain screened feature vector information;

[0031] The filtered feature vector information is input into the second information processing model, and the classification label information is output through the second information processing model.

[0032] In some embodiments of the present application, the target processing result includes target retrieval information, and the target model includes a retrieval model;

[0033] The feature vector information is processed based on the target model and the operation prompt information in the model operation information to obtain the target processing result, including:

[0034] Input the feature vector information into the retrieval model, and output the retrieval result information through the retrieval model;

[0035] Based on the search result information, generate a search result display interface;

[0036] Receiving a result filtering operation issued by a user based on a search result display interface;

[0037] Based on the result filtering operation, the target retrieval information is determined.

[0038] In a second aspect, the present application provides a data processing device, including:

[0039] An information acquisition unit, used to acquire information to be processed and model operation information; the model operation information includes model information and operation prompt information;

[0040] A model determination unit, used to determine a target model corresponding to the information to be processed based on the model information in the model operation information;

[0041] The data processing unit is used to process the information to be processed based on the target model and the operation prompt information in the model operation information to obtain the target processing result.

[0042] In a third aspect, the present application further provides a computer device, the computer device comprising:

[0043] one or more processors;

[0044] Memory; and

[0045] One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement any one of the data processing methods in the first aspect.

[0046] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, and the computer program is loaded by a processor to execute the steps in the data processing method of any one of the first aspects.

[0047] The beneficial effects of the present invention are as follows: the target model corresponding to the information to be processed is determined based on the model information, the information to be processed is processed based on the target model and the operation prompt information, and the intermediate processing process of the information to be processed is black-boxed through the target model. When the user needs to process the patent document, only a simple input operation is required to obtain the processing result, which reduces the difficulty of using patents for non-professional users and improves the efficiency and viscosity of using patent information. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0049] Figure 1 is a flow chart of a data processing method provided by an embodiment of the present invention;

[0050] Figure 2 is a flow chart of a specific embodiment of a data processing method provided by an embodiment of the present invention;

[0051] Figure 3 is a flow chart of an application embodiment of the data processing method provided by an embodiment of the present invention;

[0052] Figure 4 is a principle block diagram of a data processing device provided by an embodiment of the present invention;

[0053] Figure 5 It is a schematic diagram of the structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0055] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, which are only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second", "third", "fourth", etc. are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first", "second", "third", "fourth", etc. may explicitly or implicitly include one or more features. In the description of the present application, "multiple" means two or more, and "several" means one or more, unless otherwise clearly and specifically defined.

[0056] In this application, the word "exemplary" is used to mean "used as an example, illustration, or description." Any embodiment described in this application as "exemplary" is not necessarily to be construed as being preferred or advantageous over other embodiments. The following description is given to enable any technician in the field to implement and use the present application. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present application can be implemented without using these specific details. In other instances, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present application with unnecessary details. Therefore, the present application is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present application.

[0057] It should be noted that since the method of the embodiment of the present application is executed in a computer device, the processing objects of each computer device exist in the form of data or information. For example, time is actually time information. It can be understood that if size, quantity, position, etc. are mentioned in subsequent embodiments, they are all corresponding data for processing by the computer device. The details will not be repeated here.

[0058] As the most effective carrier of technological information, patents cover more than 90% of the world's latest technological intelligence, which is 5 to 6 years earlier than the information provided by general technical publications. In addition, 70% to 80% of inventions are only disclosed through patent documents and are not found in other scientific and technological documents. Therefore, compared with other forms of documents, patents have more obvious characteristics of novelty and practicality.

[0059] The inventor has found through research that, compared with other document forms, patent language is relatively difficult to understand and read, and it is time-consuming and laborious to capture key information. Therefore, patent tools are needed in the prior art to process patent documents. However, the use of various existing patent tools requires a high level of professionalism, and it is difficult for non-professional users to use various patent tools to process patent documents, resulting in relatively low efficiency in the mining and utilization of high-value patents.

[0060] Based on this, in an embodiment of the present application, information to be processed and model operation information are obtained; the model operation information includes model information and operation prompt information; based on the model information in the model operation information, the target model corresponding to the information to be processed is determined; based on the target model and the operation prompt information in the model operation information, the information to be processed is processed to obtain a target processing result, and the intermediate processing process of the information to be processed is black-boxed through the target model. When the user needs to process the patent document, he only needs to perform a simple input operation to obtain the processing result, which reduces the difficulty of using patents for non-professional users and improves the efficiency and viscosity of using patent information.

[0061] The content of this application is further illustrated below through the description of embodiments in conjunction with the accompanying drawings.

[0062] This embodiment provides a data processing method, such as Figure 1 As shown in , the method includes:

[0063] Step S10, obtaining information to be processed and model operation information; the model operation information includes model information and operation prompt information.

[0064] The information to be processed is information input by the user through the information input interface. The information input interface may display multiple preset information. The information to be processed may be one or more of the multiple preset information, or may be information manually input by the user on the information input interface. The information to be processed may include document information and / or non-document information, for example, the information to be processed may include patent documents, technical solution documents in word or PDF format, literature, keywords, etc.

[0065] The model operation information includes model information and operation prompt information. The model information is information about the model to be used to process the information to be processed. Based on the model information, the model to be used to process the information to be processed can be determined. For example, the model information is the model number of the model to be used. The operation prompt information is prompt information for model operation. The operation prompt information includes but is not limited to the model operation sequence, model parameter information when the model is running, etc. For example, the retrieval model, information processing model and text structured processing model are required to process the information to be processed. The operation prompt information includes the prompt information of "run the retrieval model first, then run the information processing model, and finally run the text structured processing model".

[0066] In a specific implementation, step S10 includes:

[0067] Step S11, receiving a selection instruction input by a user in a data processing function interface;

[0068] Step S12: Generate model operation information based on the selection instruction and load the information input interface;

[0069] Step S13: Obtain the information to be processed input by the user in the information input interface.

[0070] The data processing function is a function for processing the information to be processed, and the data processing function includes but is not limited to patent information summary display, related patent search, patent classification label and technical solution optimization, etc. Selecting different data processing functions to process the information to be processed can obtain different processing results. For example, selecting the patent information summary display function to process the information to be processed can obtain the patent information summary; selecting the related patent search function to process the information to be processed can obtain patent search information; selecting the patent classification label function to process the information to be processed can obtain patent label information; selecting the technical solution optimization function to process the information to be processed can obtain optimization suggestion information and / or technical comparison results.

[0071] The data processing function interface can be a selection interface for multiple data processing functions. The selection instruction is a selection operation for a specific data processing function input by the user in the data processing function interface. The selection instruction includes but is not limited to touch instructions, mouse instructions, remote control instructions, voice instructions, etc. For example, when the user wants to select the "Patent Information Summary Display" function, he directly clicks the "Patent Information Summary Display" icon on the touch screen, or the user clicks the "Patent Information Summary Display" icon on the touch screen with a mouse, or the user issues a voice command of "Patent Information Summary Display".

[0072] After the computer device receives the selection instruction input by the user in the data processing function interface, it will generate model operation information corresponding to the data processing function based on the data processing function selected by the user. For example, when the user selects the patent information summary display function, the computer device will generate model operation information corresponding to the patent information summary display function; when the user selects the associated patent search function, the computer device will generate model operation information corresponding to the associated patent search function; when the user selects the patent classification label function, the computer device will generate model operation information corresponding to the patent classification label function; when the user selects the technical solution optimization function, the computer device will generate model operation information corresponding to the technical solution optimization function.

[0073] The information input interface is an interface for users to input information to be processed. Multiple preset information can be displayed on the information input interface. Users can directly select one piece of information from the multiple preset information as the information to be processed. Users can also manually input or upload information to be processed on the information input interface. For example, users manually input multiple keywords, manually upload briefing documents in word format, comparison files in PDF format, etc.

[0074] Step S20: Based on the model information in the model operation information, determine the target model corresponding to the information to be processed.

[0075] The target model is the model that needs to be used to process the information to be processed, such as Figure 2 As shown, in this embodiment, a model library is pre-built, and the model library includes multiple candidate models. After obtaining the model information, the target model corresponding to the information to be processed is determined from the model library based on the model information. Among them, the multiple candidate models include but are not limited to feature extraction models, information processing models, text structured processing models, and retrieval models.

[0076] In a specific implementation, the feature extraction model is used to extract features of the information to be processed. For example, the feature extraction model can use BERT-BiLSTM-CRF, Word3vec, etc.; the information processing model is used to process the extracted feature vector information. For example, the information processing model can use a fine-tuned large model, such as ChatGPT, ChatGLM and other large language models (Large Language Model, LLM); the text structured processing model is used to structure the information output by the information processing model to obtain summary information, and / or optimization suggestions, and / or technical comparison results. For example, the text structured processing model can use conditional random field (conditional random field, CRF), support vector machine (Support Vector Machine, SVM), etc.; the retrieval model is a functional plug-in for searching jump link data. The retrieval model can retrieve patent documents, literature, etc. related to the information to be processed. For example, the retrieval model can perform patent search based on the keywords entered by the user or the uploaded briefing text jump link patent data.

[0077] The target model varies with the data processing function selected by the user. For example, when the data processing function selected by the user is patent information summary display and / or technical solution optimization, the target model includes a retrieval model, an information processing model and a text structured processing model; when the data processing function selected by the user is associated patent retrieval, the target model includes a retrieval model; when the data processing function selected by the user is patent classification labeling, the target model includes an information processing model and a text structured processing model.

[0078] In order to improve the accuracy of data processing results, the models in the model library can be optimized and upgraded regularly or irregularly by iteratively updating the sample set. For example, for the feature extraction model in the model library, the sample set can be optimized and upgraded every week by iteratively updating the sample set.

[0079] Step S30: Process the information to be processed based on the target model and the operation prompt information in the model operation information to obtain a target processing result.

[0080] The target processing result is the processing result obtained by processing the information to be processed based on the target model and the operation prompt information. The target processing result varies with the data processing function selected by the user. For example, when the data processing function selected by the user is patent information summary display, the target processing result is the patent information summary; when the data processing function selected by the user is associated patent retrieval, the processing result is patent retrieval information; when the data processing function selected by the user is patent classification labeling, the processing result is patent classification label information; when the data processing function selected by the user is technical solution optimization, the processing result is technical solution optimization suggestions and / or technical comparison results.

[0081] After determining the target model corresponding to the information to be processed, this embodiment processes the information to be processed based on the target model and the operation prompt information to obtain the processing result of the information to be processed. The intermediate processing process of the information to be processed is black-boxed through the target model. The user only needs to perform simple input operations to process the patent document, which reduces the difficulty of using patents for non-professional users and improves the efficiency and viscosity of using patent information.

[0082] In a specific implementation, step S30 includes:

[0083] Step S31, pre-processing the information to be processed to obtain feature vector information;

[0084] Step S32: Process the feature vector information based on the target model and the operation prompt information in the model operation information to obtain a target processing result.

[0085] The feature vector information is a vectorized representation of the information to be processed obtained by pre-processing the information to be processed, such as Figure 3 As shown, when the present embodiment processes the information to be processed based on the target model and the operation prompt information in the model operation information, the information to be processed is first pre-processed to obtain feature vector information, and then the feature vector information is processed based on the target model and the operation prompt information in the model operation information to obtain the target processing result.

[0086] In a specific implementation, step S31 includes:

[0087] Step S311, obtaining type information of the information to be processed; the type information includes patent document information, and / or, non-patent document information, and / or, non-document format information;

[0088] Step S312: if the type information is patent document information, perform first feature extraction on the information to be processed to obtain first feature information, and perform vectorization processing on the first feature information to obtain feature vector information;

[0089] Step S313: if the type information is non-patent document information, extract the second feature of the information to be processed to obtain second feature information, and vectorize the second feature information to obtain feature vector information;

[0090] Step S314: If the type information is non-document format information, vectorize the information to be processed to obtain feature vector information.

[0091] The type information is information related to the type of information to be processed, and the type information includes patent document information, and / or non-patent document information, and / or non-document format information. For example, when the information to be processed is a patent in PDF format, the type information of the information to be processed is patent document information; when the information to be processed is a technical solution in word format, the type information of the information to be processed is technical document information; when the information to be processed is a keyword, the type information of the information to be processed is non-document format information.

[0092] Continue to refer to Figure 3 As shown, when the present embodiment performs pre-processing on the information to be processed, the type information of the information to be processed is obtained. If the type information of the information to be processed is patent document information, a first feature extraction is performed on the information to be processed to obtain first feature information, and then the first feature information is quantized to obtain feature vector information; if the type information of the information to be processed is non-patent document information, a second feature extraction is performed on the information to be processed to obtain second feature information, and the second feature information is vectorized to obtain feature vector information; if the type information of the information to be processed is non-document format information, the information to be processed is vectorized to obtain feature vector information.

[0093] For example, the information to be processed input by the user includes a briefing document in word format and a patent comparison file of the briefing document. When pre-processing the information to be processed, this embodiment performs a first feature extraction on the patent comparison file to obtain first feature information, and at the same time performs a second feature extraction on the briefing document to obtain second feature information, and then quantizes the first feature information and the second feature information respectively to obtain first feature vector information and second feature vector information.

[0094] In a specific implementation, the feature extraction model in the model library includes a first feature extraction model. The first feature extraction model can adopt BERT-BiLSTM-CRF, Word3vec, etc. The step of performing a first feature extraction on the information to be processed to obtain the first feature information specifically includes: inputting the information to be processed into the first feature extraction model, and performing a first feature extraction on the information to be processed through the first feature extraction model to obtain the first feature information.

[0095] Furthermore, the feature extraction model includes a second feature extraction model. The second feature extraction model can adopt BERT-BiLSTM-CRF, Word3vec, etc. The step of performing second feature extraction on the information to be processed to obtain second feature information specifically includes: inputting the information to be processed into the second feature extraction model, and performing second feature extraction on the information to be processed through the second feature extraction model to obtain second feature information.

[0096] In a specific implementation, the target processing result includes summary information, the target model includes a retrieval model, a first information processing model and a first structured processing model, and step S32 includes:

[0097] Step S321: Determine whether related data retrieval is required based on the information to be processed;

[0098] Step S322: when it is necessary to search for associated data, the feature vector information is input into the search model, and the target associated data is output through the search model;

[0099] Step S323, inputting the target association data and the feature vector information into the first information processing model, and outputting the text feature information through the first information processing model;

[0100] Step S324: input the text feature information into the first structured processing model, and output summary information through the first structured processing model.

[0101] When determining whether to perform a related data search based on the information to be processed, it can be determined whether the information to be processed contains related data. If the information to be processed contains related data, it is determined that it is not necessary to perform a related data search. If the information to be processed does not contain related data, it is determined that it is necessary to perform a related data search. For example, if the information to be processed contains a comparative document, it is determined that it is not necessary to perform a patent search. If the information to be processed does not contain a comparative document, it is determined that a patent search is necessary.

[0102] The target associated data is the data associated with the information to be processed obtained by the retrieval model through data retrieval based on the feature vector information. For example, the target associated data is the patent document with a high similarity to the briefing document solution obtained by the retrieval model through patent retrieval based on the feature vector information. When the retrieval model performs associated data retrieval, it can jump to the link database and calculate the similarity between the data in the database and the feature vector information, and determine the data with a similarity greater than a preset threshold as the target associated data of the information to be processed. For example, Figure 3 As shown, when the retrieval model performs patent retrieval, it can jump to the patent database and calculate the similarity between the patents in the patent database and the feature vector information, and determine the patents with similarity greater than a preset threshold as target patents.

[0103] In a specific implementation, continue to refer to Figure 3 As shown, the information processing model in the database includes a first information processing model. The first information processing model can adopt a fine-tuned large model, such as ChatGPT, ChatGLM and other large language models (LLM). The text structured processing model includes a first structured processing model. The first structured processing model can adopt a conditional random field (CRF), a support vector machine (SVM), etc.

[0104] When the data processing function selected by the user is the summary display of patent information, the target model includes a retrieval model, a first information processing model and a first structured processing model, and the target processing result includes summary information. Then, the steps of processing the feature vector information based on the target model and the operation prompt information in the model operation information specifically include: determining whether it is necessary to perform associated data retrieval based on the information to be processed; when it is necessary to perform associated data retrieval, inputting the feature vector information into the retrieval model, and outputting the target associated data through the retrieval model; inputting the target associated data and the feature vector information into the first information processing model, and outputting the text feature information through the first information processing model; inputting the text feature information into the first structured processing model, and outputting the summary information through the first structured processing model. For example, when a user inputs a patent, the summary information of the patent can be output after data processing. When reading the patent, the user does not need to read the entire patent but only needs to read the summary information, which can reduce the difficulty of reading the patent for the user and enable the user to grasp the key information of the patent more quickly.

[0105] On the contrary, when it is determined that associated data retrieval is not required, the feature vector information is directly input into the first information processing model, and the text feature information is output through the first information processing model; the text feature information is input into the first structured processing model, and the summary information is output through the first structured processing model.

[0106] In a specific implementation, the target processing result includes classification label information, the target model includes a second information processing model, and step S32 includes:

[0107] Step S325, obtaining feature screening information;

[0108] Step S326: Perform feature screening on the feature vector information based on the feature screening information to obtain screened feature vector information;

[0109] Step S327: input the filtered feature vector information into the second information processing model, and output the classification label information through the second information processing model.

[0110] The feature screening information is the screening information input by the user through the information input interface, and the feature screening information is used to perform feature screening on the feature vector information. During the data processing process, when the user needs to screen the feature vector information, the feature screening information can be input through the information input interface. Specifically, the information input interface can preset multiple feature options, and the user can input the feature screening information by checking multiple feature options. The user can also manually input the feature screening information directly in the information input interface, and this application does not limit this.

[0111] In a specific implementation, continue to refer to Figure 3 As shown, the information processing model in the database includes a second information processing model, and the second information processing model can adopt a fine-tuned large model, such as ChatGPT, ChatGLM and other large language models (Large Language Model, LLM). When the data processing function selected by the user is a patent classification label, the target model includes a second information processing model, and the target processing result includes classification label information. Then, the step of processing the feature vector information based on the target model and the operation prompt information in the model operation information specifically includes: obtaining feature screening information, performing feature screening on the feature vector information based on the feature screening information, obtaining the screened feature vector information, inputting the screened feature vector information into the second information processing model, and outputting the classification label information through the second information processing model. For example, the screened feature vector information is input into the second information processing model, and the patent label information is output through the second information processing model.

[0112] In a specific implementation, the target processing result includes target retrieval information, the target model includes a retrieval model, and step S32 includes:

[0113] Step S328, input the feature vector information into the retrieval model, and output the retrieval result information through the retrieval model;

[0114] Step S329: Generate a search result display interface based on the search result information;

[0115] Step S330: receiving a result screening operation issued by the user based on the search result display interface;

[0116] Step S331: Determine target retrieval information based on the result screening operation.

[0117] In a specific implementation, the target processing result includes target search information, the target model includes a search model, and when the feature vector information is processed based on the target model and the operation prompt information in the model operation information, the feature vector information is first input into the search model, and the search model can jump to the link database, and calculate the similarity between the data in the database and the feature vector information, and determine the data with a similarity greater than a preset threshold as the search result information. For example, the search model jumps to the patent database, and calculates the similarity between the patent data in the patent database and the feature vector information, and determines the patent with a similarity greater than a preset threshold as the target patent.

[0118] The search result display interface is an interface generated based on the search result information. The search result display interface displays the search result information. The result filtering operation is a result filtering instruction issued by the user based on the search result display interface. The result filtering instruction includes but is not limited to touch instructions, mouse instructions, remote control instructions, voice instructions, etc. For example, when the user wants to filter out the patents of applicant A from the target patents, the user directly clicks the icon of "Applicant A" on the touch screen, or the user clicks the icon of "Applicant A" on the touch screen with a mouse, or the user issues a voice instruction of "Applicant A".

[0119] After receiving the filtering operation input by the user on the search result display interface, the computer device can determine the target search information based on the filtering operation and display the target search information. For example, multiple patents are displayed on the search result display interface, and the user performs a patent filtering operation for applicant A on the search result display interface. After receiving the user's filtering operation, the computer device can filter out the patent of applicant A from the multiple patents and display the patent of applicant A.

[0120] In a specific implementation, continue to refer to Figure 3As shown, the information processing model in the database includes a third information processing model, and the third information processing model can adopt a fine-tuned large model, such as ChatGPT, ChatGLM and other large language models (Large Language Model, LLM), and the text structured processing model includes a second structured processing model. The second structured processing model can adopt conditional random field (conditional random field, CRF), support vector machine (Support Vector Machine, SVM), etc.

[0121] When the data processing function selected by the user is technical solution optimization, the target model includes a retrieval model, a third information processing model and a second structured processing model, and the target processing result includes optimization suggestions and / or technical comparison results. Then, processing the feature vector information based on the target model and the operation prompt information in the model operation information may specifically include: determining whether it is necessary to perform associated data retrieval based on the information to be processed; when associated data retrieval is required, inputting the feature vector information into the retrieval model, and outputting the target associated data through the retrieval model; inputting the target associated data and the feature vector information into the third information processing model, and outputting the optimization suggestion information through the third information processing model; inputting the optimization suggestion information into the second structured processing model, and outputting the optimization suggestion and / or technical comparison results through the second structured processing model. For example, the information to be processed includes a briefing document and a comparison file corresponding to the briefing document. After data processing, the technical comparison results of the briefing scheme and the comparison file and the optimization suggestions of the briefing scheme are output, so as to facilitate users to improve their own schemes.

[0122] On the contrary, when it is determined that there is no need to perform associated data retrieval, the feature vector information is directly input into the third information processing model, and the optimization suggestion information is output through the third information processing model; the optimization suggestion information is input into the second structured processing model, and the optimization suggestions and / or technical comparison results are output through the second structured processing model.

[0123] Among them, based on the information to be processed, it is determined whether it is necessary to perform associated data retrieval; when associated data retrieval is required, the feature vector information is input into the retrieval model, and the steps of outputting the target associated data through the retrieval model are the same as the aforementioned steps S321 and S322, and the details can refer to the discussion of the aforementioned steps S321 and S322.

[0124] Specifically, during the data processing process, the user can also re-enter the information in the information input interface and continue to refer to Figure 3As shown, when the data processing function selected by the user is technical solution optimization, the feature vector information is input into the retrieval model. After the target associated data is output through the retrieval model, it will be further determined whether the user has re-entered the information. When it is determined that the user has re-entered the information, the user input information, target associated data and feature vector information are input into the third information processing model, and optimization suggestion information is output through the third information processing model. The optimization suggestion information is input into the second structured processing model, and optimization suggestions and / or technical comparison results are output through the second structured processing model. Conversely, when it is determined that the user has not re-entered the information, the target associated data and feature vector information are directly input into the third information processing model, and optimization suggestion information is output through the third information processing model. The optimization suggestion information is input into the second structured processing model, and optimization suggestions and / or technical comparison results are output through the second structured processing model.

[0125] In order to better implement the data processing method in the embodiment of the present application, based on the data processing method, a data processing device is also provided in the embodiment of the present application, such as Figure 4 As shown, the data processing device includes:

[0126] The information acquisition unit 610 is used to acquire information to be processed and model operation information; the model operation information includes model information and operation prompt information;

[0127] A model determination unit 620, configured to determine a target model corresponding to the information to be processed based on the model information in the model operation information;

[0128] The data processing unit 630 is used to process the information to be processed based on the target model and the operation prompt information in the model operation information to obtain a target processing result.

[0129] In the embodiment of the present application, the target model corresponding to the information to be processed is determined based on the model information, the information to be processed is processed based on the target model and the operation prompt information, and the intermediate processing process of the information to be processed is black-boxed through the target model. When the user needs to process the patent document, he only needs to perform a simple input operation to obtain the processing result, which reduces the difficulty of using patents for non-professional users and improves the efficiency and viscosity of using patent information.

[0130] In some embodiments of the present application, the information acquisition unit 610 is specifically used to:

[0131] Receiving a selection instruction input by a user in a data processing function interface;

[0132] Generate model operation information based on the selection instruction and load the information input interface;

[0133] Get the information to be processed entered by the user in the information input interface.

[0134] In some embodiments of the present application, the data processing unit 630 is specifically used to:

[0135] Pre-process the information to be processed to obtain feature vector information;

[0136] The characteristic vector information is processed based on the target model and the operation prompt information in the model operation information to obtain the target processing result.

[0137] In some embodiments of the present application, the data processing unit 630 is further configured to:

[0138] Acquiring type information of the information to be processed; the type information includes patent document information, and / or, non-patent document information, and / or, non-document format information;

[0139] If the type information is patent document information, performing first feature extraction on the information to be processed to obtain first feature information, and performing vectorization processing on the first feature information to obtain feature vector information;

[0140] If the type information is non-patent document information, performing second feature extraction on the information to be processed to obtain second feature information, and performing vectorization processing on the second feature information to obtain feature vector information;

[0141] If the type information is non-document format information, vectorization is performed on the information to be processed to obtain feature vector information.

[0142] In some embodiments of the present application, the target processing result includes summary information, the target model includes a retrieval model, a first information processing model and a first structured processing model, and the data processing unit 630 is further specifically used to:

[0143] Based on the information to be processed, determine whether related data retrieval is required;

[0144] When it is necessary to retrieve associated data, the feature vector information is input into the retrieval model, and the target associated data is output through the retrieval model;

[0145] Inputting the target association data and the feature vector information into a first information processing model, and outputting the text feature information through the first information processing model;

[0146] The text feature information is input into the first structured processing model, and the summary information is output through the first structured processing model.

[0147] In some embodiments of the present application, the target processing result includes classification label information, the target model includes a second information processing model, and the data processing unit 630 is further configured to:

[0148] Get feature screening information;

[0149] Performing feature screening on the feature vector information based on the feature screening information to obtain screened feature vector information;

[0150] The filtered feature vector information is input into the second information processing model, and the classification label information is output through the second information processing model.

[0151] In some embodiments of the present application, the target processing result includes target retrieval information, the target model includes a retrieval model, and the data processing unit 630 is further specifically used to:

[0152] Input the feature vector information into the retrieval model, and output the retrieval result information through the retrieval model;

[0153] Based on the search result information, generate a search result display interface;

[0154] Receiving a result filtering operation issued by a user based on a search result display interface;

[0155] Based on the result filtering operation, the target retrieval information is determined.

[0156] The embodiment of the present application further provides a computer device, which integrates any one of the data processing devices provided in the embodiment of the present application, and the computer device includes:

[0157] one or more processors;

[0158] Memory; and

[0159] One or more applications, wherein the one or more applications are stored in a memory and are configured to be executed by a processor to execute the steps of the data processing method in any of the above data processing method embodiments.

[0160] The present application also provides a computer device that integrates any data processing device provided in the present application. Figure 5 As shown, it shows a schematic diagram of the structure of the computer device involved in the embodiment of the present application, specifically:

[0161] The computer device may include one or more processing core processors 801, one or more computer-readable storage media memories 802, a power supply 803, an input unit 804 and other components. Those skilled in the art will appreciate that Figure 5 The computer device structure shown in the figure does not constitute a limitation on the computer device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. Among them:

[0162] The processor 801 is the control center of the computer device. It uses various interfaces and lines to connect various parts of the entire computer device. By running or executing software programs and / or modules stored in the memory 802 and calling data stored in the memory 802, it executes various functions of the computer device and processes data, thereby monitoring the computer device as a whole. Optionally, the processor 801 may include one or more processing cores; preferably, the processor 801 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 801.

[0163] The memory 802 can be used to store software programs and modules. The processor 801 executes various functional applications and data processing by running the software programs and modules stored in the memory 802. The memory 802 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 802 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 802 may also include a memory controller to provide the processor 801 with access to the memory 802.

[0164] The computer device also includes a power supply 803 for supplying power to each component. Preferably, the power supply 803 can be logically connected to the processor 801 through a power management system, so that the power management system can manage charging, discharging, power consumption and other functions. The power supply 803 can also include one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, power status indicators and other arbitrary components.

[0165] The computer device may further include an input unit 804, which may be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0166] Although not shown, the computer device may further include a display unit, etc., which will not be described in detail herein. Specifically in this embodiment, the processor 801 in the computer device will load the executable files corresponding to the processes of one or more application programs into the memory 802 according to the following instructions, and the processor 801 will run the application programs stored in the memory 802, thereby realizing various functions, as follows:

[0167] Obtain information to be processed and model operation information; model operation information includes model information and operation prompt information;

[0168] Based on the model information in the model operation information, determining a target model corresponding to the information to be processed;

[0169] The information to be processed is processed based on the target model and the operation prompt information in the model operation information to obtain the target processing result.

[0170] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0171] To this end, an embodiment of the present application provides a computer-readable storage medium, which may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc. A computer program is stored thereon, and the computer program is loaded by a processor to execute the steps in any data processing method provided in the embodiment of the present application. For example, the computer program can be loaded by a processor to execute the following steps:

[0172] Obtain information to be processed and model operation information; model operation information includes model information and operation prompt information;

[0173] Based on the model information in the model operation information, determining a target model corresponding to the information to be processed;

[0174] The information to be processed is processed based on the target model and the operation prompt information in the model operation information to obtain the target processing result.

[0175] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the detailed description of other embodiments above, and will not be repeated here.

[0176] In specific implementation, the above units or structures can be implemented as independent entities, or can be arbitrarily combined to be implemented as the same or several entities. The specific implementation of the above units or structures can refer to the previous method embodiments, which will not be repeated here.

[0177] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0178] The above is a detailed introduction to a data processing method, device, computer equipment and computer-readable storage medium provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the ideas of the present application, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A data processing method, characterized in that: include: Obtain information to be processed and model operation information; The model operation information includes model information and operation prompt information; Based on the model information in the model operation information, determining a target model corresponding to the information to be processed; The information to be processed is processed based on the target model and the operation prompt information in the model operation information to obtain a target processing result.

2. The method according to claim 1, characterized in that The obtaining of information to be processed and model operation information includes: Receiving a selection instruction input by a user in a data processing function interface; Generate model operation information based on the selection instruction and load the information input interface; Obtain the information to be processed input by the user in the information input interface.

3. The method according to claim 1, characterized in that The processing of the information to be processed based on the target model and the operation prompt information in the model operation information to obtain a target processing result includes: Pre-processing the information to be processed to obtain feature vector information; The feature vector information is processed based on the target model and the operation prompt information in the model operation information to obtain a target processing result.

4. The method according to claim 3, characterized in that The pre-processing of the information to be processed to obtain feature vector information includes: Acquiring type information of the information to be processed; the type information includes patent document information, and / or non-patent document information, and / or non-document format information; If the type information is patent document information, performing first feature extraction on the information to be processed to obtain first feature information, and performing vectorization processing on the first feature information to obtain feature vector information; If the type information is non-patent document information, performing second feature extraction on the information to be processed to obtain second feature information, and performing vectorization processing on the second feature information to obtain feature vector information; If the type information is non-document format information, vectorization is performed on the information to be processed to obtain feature vector information.

5. The method according to claim 3, characterized in that: The target processing result includes summary information, and the target model includes a retrieval model, a first information processing model, and a first structured processing model; The processing of the feature vector information based on the target model and the operation prompt information in the model operation information to obtain a target processing result includes: Based on the information to be processed, determining whether associated data retrieval is required; When it is necessary to retrieve associated data, the feature vector information is input into the retrieval model, and the target associated data is output through the retrieval model; Inputting the target association data and the feature vector information into the first information processing model, and outputting text feature information through the first information processing model; The text feature information is input into the first structured processing model, and summary information is output through the first structured processing model.

6. The method according to claim 3, characterized in that The target processing result includes classification label information, and the target model includes a second information processing model; The processing of the feature vector information based on the target model and the operation prompt information in the model operation information to obtain a target processing result includes: Get feature screening information; Performing feature screening on the feature vector information based on the feature screening information to obtain screened feature vector information; The filtered feature vector information is input into the second information processing model, and the classification label information is output through the second information processing model.

7. The method according to claim 3, characterized in that The target processing result includes target retrieval information, and the target model includes a retrieval model; The processing of the feature vector information based on the target model and the operation prompt information in the model operation information to obtain a target processing result includes: Inputting the feature vector information into the retrieval model, and outputting retrieval result information through the retrieval model; Based on the search result information, generate a search result display interface; Receiving a result screening operation issued by a user based on the search result display interface; Based on the result screening operation, target retrieval information is determined.

8. A data processing device, characterized in that: include: An information acquisition unit, used to acquire information to be processed and model operation information; The model operation information includes model information and operation prompt information; A model determination unit, configured to determine a target model corresponding to the information to be processed based on the model information in the model operation information; A data processing unit is used to process the information to be processed based on the target model and the operation prompt information in the model operation information to obtain a target processing result.

9. A computer device, characterized in that: The computer device comprises: one or more processors; Memory; and One or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the processor to implement the data processing method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and the computer program is loaded by a processor to execute the steps in the data processing method according to any one of claims 1 to 7.

Images