A recommendation method and system based on user product usage behavior

Abstract

The present invention discloses a recommendation method and system based on user product usage behavior, belonging to the technical field of deep learning, including: providing software usage documents; storing the original text content in a database and simultaneously storing it in a word vector library in the form of word vectors, establishing a mapping relationship between the data in the database and the word vector library; obtaining user behavior through data logging, combining the user's IO input information with the software usage logic, converting the user behavior into text information, and querying the word vector library for consecutive multiple pieces of sorted text information; asking a large model after splicing the query results of the word vector library with prompt words; performing a function mapping between the answer result of the large model and the current system software, obtaining the function specified by the corresponding answer result, and returning the recommendation result to the page. By implementing targeted data logging technology and collecting user operation behavior data, and combining the analysis ability of the large model, the present invention can achieve accurate prediction of the user's next behavior.

Description

Technical Field

[0001] The present invention relates to the technical field of deep learning, and particularly relates to a recommendation method and system based on user product usage behavior. Background Art

[0002] Most of the existing software behavior recommendation applications rely on information on web pages or the software's own pages to recommend based on users, or infer user intentions based on the content currently copied in the clipboard by the user. There is a lack of coherent thinking in recommending the use of software, making it difficult to guess the intentions of the user's continuous behavior in the software. Moreover, the current technology lacks a feedback mechanism and has insufficient recognition ability for the recommended results, resulting in the inability of the recommendation ability to grow self - incrementally with the user's usage time.

[0003] In the prior art, the invention patent with the application number CN201510012093.4 and the name of a method and device for recommending solutions based on user operation behavior includes the following content:

[0004] 1. User behavior monitoring: This patent describes a method that triggers a second window object by monitoring the user's operation on the first window object;

[0005] 2. Abnormal window identification: When the second window object is an abnormal window, this method can determine the type of the abnormal window by extracting the text information in the window;

[0006] 3. Intelligent recommendation of solutions: According to the type of the abnormal window and the user's operation behavior on the abnormal window, the system will recommend corresponding solutions;

[0007] 4. Reduction of user operations: This method reduces the steps that users need to manually search for solutions when encountering problems through intelligent recommendation;

[0008] 5. Enhancement of user experience: Through the user - friendly intelligent recommendation function, it helps users solve problems encountered during computer use, thereby improving the user's usage experience.

[0009] However, the existing page recommendations have the following disadvantages:

[0010] Lack of flexibility: The current web pages or software are too single - functioned in implementing the next - step recommendation and speculation on the page, and cannot make recommendations for the user's next behavior based on page data, attributes, and user operation behavior. Since different pages may have large differences, the recommendation function very much needs to be manually written, and the implemented recommendation ability is too single.

[0011] Inability to achieve self - growth of recommendation capabilities: In previous recommendation pages, there was no feedback mechanism added to the recommendation function, and users could not provide feedback after selection. Therefore, the recommendation function could not achieve self - growth with the behaviors recognized by users.

[0012] Over - reliance on page - specific data: The recommendation function of previous pages needed to make guesses based on data such as the current page where the page was located or clipboard data, and did not incorporate data points to capture user behaviors. Therefore, it was highly dependent on the current page state and current page data, and thus the recommendation capabilities were restricted.

[0013] Too weak in understanding ability: Previous recommended options were directly written by programs and did not change much, lacking the ability to understand what users wanted to achieve. As a result, it was impossible to predict the possible next behavioral operations based on user behaviors.

[0014] In summary, the problems of the existing technical solutions in the page recommendation function can be summarized as follows:

[0015] Lack of personalization and dynamic adaptability: The current recommendation systems show obvious limitations in the page recommendation function. They usually rely too much on preset rules and the current page state, without fully considering the personalized needs and behavior patterns of users. This leads to the singularity and lack of flexibility of the recommendation function, and it cannot be dynamically adjusted according to users' real - time operations and preferences. In addition, due to the lack of an effective feedback mechanism, the recommendation system cannot learn from users' selections and optimize its recommendation strategy, thus unable to achieve self - growth of recommendation capabilities.

[0016] Reliance on page data while ignoring user behaviors: Existing recommendation systems often over - rely on data of the page itself, such as the current page content or clipboard information, while ignoring the importance of capturing user behaviors through data - point technology. This approach limits the understanding ability of the recommendation system and makes it unable to accurately predict users' next behaviors. Due to the lack of in - depth analysis of user behaviors, the recommendation system cannot provide more accurate and forward - looking suggestions, thus affecting the user experience and satisfaction. Summary of the Invention

[0017] The present invention aims to solve the obvious limitations shown by the recommendation systems in the prior art in the recommendation function, and proposes a recommendation method and system based on users' product usage behaviors. By implementing targeted data - point technology, collecting users' operation behavior data, and combining the analysis ability of large models, it realizes accurate prediction of users' next behaviors.

[0018] To achieve the above - mentioned invention objectives, the technical solution of the present invention is as follows:

[0019] A recommendation method based on users' product usage behaviors, comprising:

[0020] Provide software usage documents, which include documents related to software operations and some preset sample user behavior recommendation documents;

[0021] Store the original text content of the software usage documents in a database and simultaneously store it in the word vector library in the form of word vectors, establishing a mapping relationship between the data in the database and the word vector library;

[0022] Obtain the user behavior based on the operations of the IO input device during the user's software usage, place markers at important functional nodes in the front-end and back-end of the program or database operations, and define the specific meaning of the currently triggered function according to the software usage logic;

[0023] Combine the obtained user behavior and the marker operations, combine the IO input information with the software usage logic, convert the user behavior into text information, and after organizing multiple consecutive text messages, query the word vector library to obtain the query result;

[0024] Use the query result of the word vector library as reference information, combine it with the prompt words to form a complete question for asking the large model;

[0025] Start the large model as an independent web service to answer the combined complete question;

[0026] By obtaining the answer result of the large model, perform function mapping with the current system software, obtain the function specified by the corresponding answer result, and return the recommended result to the page.

[0027] Further, the step of storing the original text content of the software usage documents in a database and simultaneously storing it in the word vector library in the form of word vectors, establishing a mapping relationship between the data in the database and the word vector library, includes: cutting the plain text content according to the number of words, and sliding back and forth during each cut to ensure data coherence; splitting the segmented text data into the word vector library through a word vector tool, and generating a unique id for the stored word vector data and returning it; after obtaining the corresponding id, store it together with the original document content data, thereby realizing the mapping storage of vectors and original data.

[0028] Further, in the step of using the query result of the word vector library as reference information, combining it with the prompt words to form a complete question for asking the large model, it is necessary to combine the current software design information, set different prompt words for different modules, pages, and interfaces, and optimize the prompt word information for each different function to make it adapt to different scenarios.

[0029] Further, obtaining the answer result of the large model, performing function mapping with the current system software, and obtaining the function specified by the corresponding answer result includes: through a question-and-answer operation, the natural language model returns its guess of the user's next behavior. Based on multiple possible results inferred by the large model and combined with the current software design operation, the corresponding operation behavior is obtained, and this behavior is returned as multiple final recommended results to the page.

[0030] Further, it also includes judging whether the user selects and clicks on the page recommendation option, recording whether the user accepts the recommendation result, and giving feedback: if the user clicks on the recommendation option, it is considered that the current recommendation is reasonable and the user agrees with the recommendation result. This behavior is regarded as a high-quality recommendation, and the user behavior and result are vectorized and stored in the database as reference information for the next recommendation; otherwise, the current recommendation process is abandoned as a useless recommendation without recording.

[0031] Further, it also includes performing the corresponding option operation after the user clicks on the option corresponding to the page recommendation result, tracking the user behavior, and giving feedback on the execution result: if it is not executed correctly, the recommendation result is regarded as an abnormal result, and feedback is given on the final result, and the stored reference information is retrieved or changed in a timely manner. This behavior is regarded as a high-quality recommendation but the recommendation result needs to be optimized; if the execution fails, it means that the user agrees with the behavior guess recommendation, but there is a logical problem with the recommendation result, and feedback is also given on it.

[0032] The present invention also proposes a recommendation system based on user product usage behavior, including:

[0033] Software usage documentation, which contains all information related to the software;

[0034] Word vector storage module, which stores the original text content of the software usage documentation in the word vector library in the form of word vectors;

[0035] Original data storage module, which stores the original text content in the database and establishes a mapping relationship with the word vector data;

[0036] User behavior acquisition module, which acquires user behavior according to the operations of the IO input device during the user's software usage process;

[0037] System buried point recording module, which buries points at important functional nodes of the program front and back ends or database operations, and defines the specific meaning of the currently triggered function according to the software usage logic;

[0038] Record query module. The record query module combines the IO input information with the software usage logic through the captured user behaviors and buried point operations, transforms the user's behaviors into text information, and after organizing multiple consecutive text information, queries the information in the vector library and obtains the results;

[0039] Prompt word and query result splicing module. The prompt word and query result splicing module uses the vector library query result as reference information and splices it with the prompt word into a complete question for asking the large model;

[0040] Large model answer module. The large model answer module is started as an independent web service to provide large model question and answer functions;

[0041] Result generation module. The result generation module obtains the answer result of the large model, maps it with the current system software to obtain the function specified by the corresponding result, and returns the recommended result to the page.

[0042] Furthermore, the software usage document includes a software operation manual, the software interface architecture and hierarchical relationship, software operation precautions, and also includes some preset sample user behavior recommendation documents.

[0043] Furthermore, it also includes a recommended result feedback module. The recommended result feedback module records whether the user accepts the recommended result by judging whether the user selects and clicks the page recommended option and gives feedback.

[0044] Furthermore, it also includes an execution result feedback module. The execution result feedback module performs the corresponding option operation after the user clicks the option corresponding to the page recommended result, and gives feedback on the execution result by tracking the user's behavior.

[0045] In summary, the present invention has the following advantages:

[0046] 1. The recommendation method of the present invention based on the user's product usage behavior incorporates a large model and constructs a large model recommendation feedback function based on a word vector library. It is more flexible in the recommendation method. By implementing targeted buried point technology to collect the user's operation behavior data and combining the analysis ability of the large model, it can achieve accurate prediction of the user's next behavior;

[0047] 2. The present invention records the user's current multiple consecutive behaviors through the user's software operation behaviors. After combining information such as the large model and software document reference content, it jointly recommends and guesses the user's next behavior, and transforms the behavior into click options, quickly simplifying the multi-level operations that the user wants to perform into single-step operations, realizing fast internal software jumps and high-quality speculation of user behaviors;

[0048] 3. In order to further improve the performance of the recommendation system, the present invention optimizes the tracking strategy, defines the specific meaning of the current trigger function according to the development operation logic of the software function, and collects detailed information on user behavior at each tracking point by designing rich parameters to ensure that the captured user behavior data is highly relevant and clear, which will help the big model to more accurately understand the user's operation intention and provide more accurate recommendations;

[0049] 4. The present invention establishes an effective user feedback mechanism, analyzes the user's clicks and execution of recommendation results, continuously optimizes the recommended content, and implements a two-way optimization strategy, which helps to distinguish between performance problems and data accuracy problems of the recommendation system, thereby improving the recommendation results and data quality in a targeted manner, thereby improving the accuracy of recommendations and user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is an implementation flow chart of the present invention. DETAILED DESCRIPTION

[0051] In order to explain the present invention more clearly, the present invention is further described below in conjunction with preferred embodiments and drawings. It should be understood by those skilled in the art that the content described below is illustrative rather than restrictive, and should not be used to limit the scope of protection of the present invention.

[0052] Example 1

[0053] The present invention provides a recommendation method based on user product usage behavior, such as Figure 1 As shown, the following steps are included:

[0054] Provide software usage documentation: Software usage documentation contains all information related to the software. To cooperate with the current program operation and operation, the injected documents include software operation manual, software interface architecture and hierarchical relationship, software usage and operation precautions, and other documents related to software operation and use. It also contains some preset user behavior recommendation document samples. Software usage documentation is jointly written and maintained by product, development, and test personnel, and is updated synchronously with the current software development and design.

[0055] Word vector storage: The original text content of the software usage document is stored in the word vector library in the form of word vectors for subsequent query of similar user behavior content. In this embodiment, the implementation of converting the original text content into a word vector library mainly relies on word embedding techniques in natural language processing (NLP), such as Word2Vec, GloVe, and FastText. These techniques learn the vector representations of words by training a large amount of text data, making words that are semantically similar close to each other in the vector space. Taking GloVe as an example, it learns word vectors by statistically analyzing the co-occurrence information of words in the corpus. GloVe uses a global matrix to capture the co-occurrence relationship between words, considering the co-occurrence frequency of words in the entire corpus.

[0056] Original data storage: The original text content is stored in the database and a mapping relationship is established with the data in the word vector library. Since the natural language model only supports text content input and does not support vector input, the original data storage enables more friendly acquisition of the original text content during subsequent similarity queries. During implementation, by reading the document or text content, the text content is split according to the document content or specified data format: the plain text content is cut by the number of characters, and slides are made before and after each cut to ensure data coherence; the segmented text data is stored in the word vector library through a word vector tool, and a unique id is generated and returned for the stored word vector data; after obtaining the corresponding id, it is jointly stored with the original document content data, thereby realizing the mapping and storage of vectors and original data.

[0057] User behavior acquisition: It includes acquiring the operations of the IO input device during the user's use of the software, and obtaining user behavior by capturing the input status of the physical IO device and the software application IO input status. For example, IO input device operations such as mouse sliding, mouse hovering, mouse clicking, keyboard input, and shortcut key operations during the user's use of the software. The implementation of software IO input requires software cooperation. Taking a browser as an example, the event listener built into the browser can capture input content, mouse click behavior, mouse sliding trajectory, mouse-selected data, etc. These information will also be recorded as user usage behavior. In addition, different products will run on different devices and software, and all records that can capture user behavior will be recorded by the program.

[0058] System Logging: Add behavior capture programs at important functional nodes in the front-end, back-end, or database operations of the program to obtain the true operation intentions of users. Since the physical operations of users do not have much practical significance and it is impossible to perceive users' operation intentions, the present invention defines the specific meaning of the currently triggered function based on the development operation logic of software functions at important nodes such as the front-end, back-end, or database operations. During implementation, at each logging point, rich parameters are designed to collect detailed information about users' behaviors. For example, which button the user clicks, the page position at the time of clicking, the time the user's mouse hovers, etc. These parameters can record users' behaviors more comprehensively, enabling the large model to better understand users' current behaviors. For example, when the user clicks on the upper right corner of the web page, the back-end will capture the corresponding request and, in combination with the front-end operation, identify this behavior as "viewing personal information", thereby determining what specific operation the user is performing in the current software. The targeted logging set in this step focuses on valuable operation behaviors of users. These operation points are all important behaviors when users interact with the product, such as clicking buttons, submitting forms, copying text, mouse hovering, etc., ensuring that the captured user operation behaviors are highly relevant and are all clear rather than unknown behaviors, so that the recommendation results of the large model can be accurately optimized around the current software for recommendation.

[0059] Record Query: Based on the obtained user behaviors and logging operations, transform the user's behaviors into text information. After organizing multiple consecutive text messages, query them in the vector library to obtain the vector library query results for inferring the user's subsequent behaviors. In this step, through the logging operation, the IO input information is combined with the software usage logic to transform the behaviors into text, thereby obtaining a natural language description of the current user operation behaviors. Organize the user's recent multiple consecutive operation behaviors into consecutive text information, such as: "It is currently 9 pm, the user is on the home page of the software, the user copied an IP information, the user clicked on the data panel, and the user selected a data SQL query", and input this text into the vector library to query the relevant information of the documents in the vector and the preset behavior records, so as to obtain the behavior most similar to the user's current behavior and the specific function information corresponding to the current behavior.

[0060] Prompt and Query Result Concatenation: Use the query results as reference information and concatenate them with prompts to form a complete question. In this step, based on the current user behavior information, obtain the appropriate prompts through the prompt management tool, and concatenate the prompts, the user behavior information, and the relevant reference information queried from the vector library together to form a complete question for asking the large model. By combining the design information of the current software, different modules, pages, and interfaces are set with different prompts. Optimizing the prompt information for each different function can make it adapt to different scenarios.

[0061] Answer from the large model: The large model is started as an independent web service to answer the complete spliced question. In this step, an independent process is set up to build the large model service, which is isolated from the current software application. An independent large model Q&A service function is provided to facilitate the independent management and deployment of the large model Q&A ability. For example, ensuring that actions such as switching parameter information and replacing high-quality models will not affect the normal operation of the main program. The model Q&A with the prompt words will output the specified result structure for subsequent function transformation.

[0062] Result generation: By obtaining the answer result of the large model, performing function mapping with the current system software, obtaining the function specified by the corresponding result, and returning the recommended result to the page. In this step, through the Q&A operation, the natural language model will return its guess of the user's next behavior. For example, based on the user's current click operation and reference document information, it is speculated that the user may want to enter a certain page to view the corresponding content. Based on multiple possible results speculated by the large model, combined with the current software design operation, the corresponding operation behavior is obtained and returned to the page as the final multiple recommended results. For example, after the user selects an IP address, a recommended list can appear on the page, speculating that the user may want to perform SQL queries, data searches, data parsing, IP jumps, etc. The user can quickly jump to the specified page to complete the specified operation according to the reasoning result, without the need for multi-level clicks and other operations, and all operations will be completed in one step.

[0063] The present invention incorporates a large model and constructs a large model recommendation feedback function based on a word vector library, which is more flexible in the recommendation method. By implementing targeted data tracking technology, collecting the user's operation behavior data, and combining the analysis ability of the large model, it is possible to achieve accurate prediction of the user's next behavior.

[0064] Embodiment 2

[0065] Based on Embodiment 1, this embodiment proposes a recommendation method based on the user's product usage behavior. The difference from Embodiment 1 is as follows:

[0066] Furthermore, it also includes recommendation result feedback: By judging whether the user selects and clicks on the page recommendation option, recording whether the user accepts the recommendation result, and feeding back the result record to optimize the database information. Specifically, when the user clicks on the recommendation option, it is considered that the current recommendation is reasonable and the user agrees with the recommendation result. This behavior is regarded as a high-quality recommendation, and the user's behavior and result will be vectorized and stored in the database as reference information for the next recommendation. Conversely, the current recommendation process will be abandoned as a useless recommendation without recording.

[0067] Further, it also includes execution result feedback: after the user clicks an option and corresponding option operations are performed, such as page jumping, content searching, generating text content, etc., by tracking the user's behavior, the execution result is fed back to ensure that the user's selection is correctly executed. Specifically, if it is not correctly executed, the recommended result is regarded as an abnormal result. For example, behaviors such as recommending to jump to a non-existent page require feedback on the final result, and the stored reference information needs to be retrieved or changed in a timely manner. This behavior is regarded as a high-quality recommendation, but the recommended result needs to be optimized; if the execution fails, it means that the user agrees with the behavior guess recommendation, but there are logical problems with the recommended result, and feedback is also required.

[0068] In this embodiment, through the user's click and actual execution result feedback, the feedback content is accurately optimized to ensure that the data is more accurate in the reference result. Through feedback, it can be confirmed whether it is the lack of recommendation ability or data error in the recommended final result. If the user selects the recommended result, it means that the recommended result meets the expectations. For example, in the background record, program buried points will record the user's next operation after selecting the recommended option. If the user clicks to jump and then immediately cancels the jump behavior, or selects to accept the content continuation of the large model and then immediately deletes the continued content, similar operations, then the program determines that the recommendation ability of the large model is insufficient, that is, the recommended direction meets the expected result, but lacks accuracy. If there is a problem with the click execution, it means that there are certain data errors in the result that meets the expectations, and the result still needs to be optimized. For example, if the user selects the recommendation of the large model, but the subsequent execution result of the recommendation fails or reports an error, it is regarded as an error.

[0069] The present invention establishes an effective user feedback mechanism, analyzes the user's click and execution of the recommended result, continuously optimizes the recommended content, implements a two-way optimization strategy, helps to distinguish the performance problems and data accuracy problems of the recommendation system, and thus improves the recommended result and data quality in a targeted manner, and can improve the accuracy of recommendation and user satisfaction.

[0070] Embodiment 3

[0071] The present invention provides a recommendation system based on user product usage behavior, including a software usage document, a word vector storage module, an original data storage module, a user usage behavior acquisition module, a system buried point record module, a record query module, a prompt word and query result splicing module, a large model answer module, and a result generation module.

[0072] Among them, the software usage documentation contains all information related to the software. To cooperate with the current program operation and operation, the injected documents include all documents related to the software operation and use, such as the software operation manual, the software interface architecture and hierarchical relationship, and the software usage and operation precautions. It also contains some preset user behavior recommendation document samples. The software usage documentation is jointly written and maintained by product, development, and test personnel, and is updated synchronously with the current software development and design.

[0073] The word vector storage module stores the original text content in the form of word vectors into the word vector library to facilitate subsequent user behavior similar content queries. In this embodiment, the implementation of text content to word vector library mainly relies on word embedding technology in natural language processing (NLP), such as Word2Vec, GloVe and FastText. These technologies learn the vector representation of words by training a large amount of text data, so that semantically similar words are close to each other in the vector space. Taking GloVe as an example, it learns word vectors by counting the co-occurrence information of words in the corpus. GloVe uses a global matrix to capture the co-occurrence relationship between words, taking into account the co-occurrence frequency of words in the entire corpus. This functional module can flexibly switch to use different word vector technologies according to the actual scenario, converting text content into word vectors for easy query.

[0074] The original data storage module stores the original text content into the database and establishes a one-to-one correspondence with the word vector data. The original data storage module enables the original text content to be obtained more friendly when performing similarity queries later. Specifically, the original data storage module reads the document or text content and splits the file content according to the document content or the specified data format: the plain text content is cut according to the number of words, and each cut will slide back and forth to ensure the continuity of the data; the text data split into segments will be stored in the word vector library through the word vector tool, and a unique id will be generated for the stored word vector data and returned. After obtaining the corresponding id, it will be stored together with the original document content data, thereby realizing the mapping and preservation of the vector and the original data.

[0075] The user behavior acquisition module acquires the IO input device operations such as mouse sliding, mouse hovering, mouse clicking, keyboard input, shortcut key operation, etc. during the user's use of the software, and acquires the user behavior by capturing the physical IO device input status and the software application IO input status. The implementation of software IO input requires the cooperation of software. Taking the browser as an example, the browser's built-in event listener can capture input content, mouse click behavior, mouse sliding trajectory, mouse selection data and other information, which will also be recorded as user usage behavior. In addition, different products will run on different devices and software, and all records that can capture relevant user behavior will be recorded by the program.

[0076] The system buried point recording module obtains the real operation intention of users by adding behavior capture programs at important function nodes such as the front and back ends of the program or database operations. Specifically, at important nodes such as the front and back ends or database operations, the system buried point recording module defines the specific meaning of the currently triggered function according to the development operation logic of the software function. For example, when a user clicks on the upper right corner of a web page, the back end will capture the corresponding request, and in combination with the front-end operation, this behavior will be recognized as "viewing personal information", so as to judge what specific operation the user is performing in the current software.

[0077] The record query module transforms the user's behavior into text information through the captured user behavior and buried point operations. After sorting out multiple consecutive text information, it queries the vector library with this information to obtain the vector library query result, so as to infer the user's next behavior. Specifically, the record query module combines the IO input information with the software usage logic through the buried point operation, transforms the behavior into text, and obtains the natural language description of the current user operation behavior. Specifically, the user's recent multiple consecutive operation behaviors are sorted into consecutive text information, such as: "It is 9 pm now, the user is on the home page of the software, the user copied an IP information, the user clicked on the data panel, and the user selected data SQL query". This paragraph of text is poured into the word vector library to query the relevant information of the document in the vector and the preset behavior record, so as to obtain the behavior most similar to the user's current behavior and the specific function information corresponding to the current behavior.

[0078] The prompt word and query result splicing module uses the query result as reference information and combines it with the prompt word to splice into a complete question. Specifically, the prompt word and query result splicing module obtains the adapted prompt word through the current user behavior information and combines it with the prompt word management tool. It splices the prompt word, the user behavior information, and the relevant reference information queried from the vector library into a complete question for asking the large model. During implementation, by combining the design information of the current software, different modules, pages, and interfaces are set with different prompt words to optimize the prompt word information for each different function, which can be adapted to different scenarios.

[0079] The large model answering module is started as an independent web service to provide the large model question and answer function. By setting up an independent process, a large model service is built, which is isolated from the current software application. It provides an independent large model question and answer service function, which is convenient for the independent management and deployment of the large model question and answer ability. For example, behaviors such as ensuring the switching of parameter information and replacing high-quality models will not affect the normal operation of the main program. The model question and answer with the prompt word will output a specified result structure for subsequent function transformation.

[0080] The result generation module maps the answer results of the large model to the current system software to obtain the functions specified by the corresponding results. Specifically, through the Q&A operation, the natural language model returns its guess of the user's next behavior. For example, based on the user's current click operation and the reference document information, it is speculated that the user may want to enter a certain page to view the corresponding content. Based on multiple possible results speculated by the large model and combined with the current software design operations, the corresponding operation behaviors are obtained, and these behaviors are returned as the final multiple recommended results to the page. For example, after the user selects an IP address, a recommended list can appear on the page, speculating that the user may want to perform SQL queries, data searches, data parsing, IP jumps, etc. The user can quickly jump to the specified page to complete the specified operation according to the inference results, without the need to perform multi-level clicks and other operations, and all operations will be completed in one step.

[0081] Embodiment 4

[0082] Based on Embodiment 3, the present invention provides a recommendation system based on user product usage behavior. The difference from Embodiment 3 is as follows:

[0083] Furthermore, it further includes a recommended result feedback module. The behavior feedback management module records whether the user accepts the recommended result by judging whether the user selects and clicks the recommended option, and feeds back the result record to optimize the database information. Specifically, when the user clicks the recommended option, it is considered that the current recommendation is reasonable and the user agrees with the recommended result. This behavior is regarded as a high-quality recommendation, and the user behavior and result are vectorized and stored in the database for reference information for the next recommendation; otherwise, the current recommendation process is abandoned as a useless recommendation.

[0084] Even further, it further includes an execution result feedback module. The page jump module performs corresponding option operations after the user clicks the option, such as page jumps, content searches, generating text content, etc. By tracking the user's behavior, the execution results are fed back to ensure that the user's selection is correctly executed. Specifically, if it is not correctly executed, the recommended result is regarded as an abnormal result. For example, behaviors such as recommending to jump to a non-existent page require feedback on the final result again, and the stored reference information is retrieved or changed in a timely manner. This behavior is regarded as a high-quality recommendation but the recommended result needs to be optimized; if the execution fails, it means that the user agrees with the behavior guess recommendation, but there is a logical problem with the recommended result, and it is also fed back.

[0085] In addition, as a further optimization, the system can also implement functions such as quick page jump, selection after page jump, automatic content input on the page, quick page turning, automatic click, automatic text filling, page content search, etc. within the software, and can also implement operations such as assisted writing of input text content, quick web page jump, explanation of selected content, translation of selected content, AI automatic text correction, etc. outside the software.

[0086] The above are only the preferred embodiments of the present invention, and do not impose any form of limitation on the present invention. Any simple modification or equivalent change made to the above embodiments based on the technical essence of the present invention shall fall within the protection scope of the present invention.

Claims

1. A recommendation method based on user product usage behavior, characterized in that, Including: Providing software usage documents, which include documents related to software operations and some preset sample user behavior recommendation documents; Storing the original text content of the software usage documents in a database and simultaneously storing them in a word vector library in the form of word vectors, establishing a mapping relationship between the data in the database and the word vector library; Obtaining user behavior based on the operations of the IO input device during the user's software usage, setting breakpoints at important functional nodes in the front-end and back-end of the program or database operations, and defining the specific meaning of the currently triggered function according to the software usage logic; Combining the obtained user behavior and breakpoint operations, integrating the IO input information with the software usage logic, transforming the user behavior into text information, sorting out multiple consecutive text messages and querying the word vector library to obtain a query result; Using the query result of the word vector library as reference information, combining it with prompt words to form a complete question for asking the large model; Starting the large model as an independent web service to answer the combined complete question; By obtaining the answer result of the large model, performing function mapping with the current system software, obtaining the function specified by the corresponding answer result, and returning the recommendation result to the page.

2. The recommendation method based on user product usage behavior according to claim 1, wherein The step of storing the original text content of the software usage documents in a database and simultaneously storing them in a word vector library in the form of word vectors, establishing a mapping relationship between the data in the database and the word vector library includes: cutting the plain text content according to the number of words, sliding before and after each cut to ensure data coherence; splitting the text data into segments and storing them in the word vector library through a word vector tool, and generating a unique id for the stored word vector data and returning it; after obtaining the corresponding id, storing it together with the original document content data, thereby realizing the mapping storage of vectors and original data.

3. A recommendation method based on user product usage behavior according to claim 1, characterized in that In the step of using the query result of the word vector library as reference information, combining it with prompt words to form a complete question for asking the large model, it is necessary to combine the current software design information, set different prompt words for different modules, pages, and interfaces, and optimize the prompt word information for each different function to make it suitable for different scenarios.

4. The recommendation method based on user product usage behavior according to claim 1, wherein, The step of performing function mapping with the current system software by obtaining the answer result of the large model and obtaining the function specified by the corresponding answer result includes: through the question-and-answer operation, the natural language model returns its guess of the user's next behavior, based on multiple possible results inferred by the large model, combining the current software design operations, obtaining the corresponding operation behavior, and returning this behavior as multiple final recommendation results to the page.

5. The recommendation method based on user product usage behavior according to claim 1, characterized in that, It also includes recording whether the user accepts the recommendation result by judging whether the user selects and clicks on the page recommendation option and providing feedback: if the user clicks on the recommendation option, it is considered that the current recommendation is reasonable and the user agrees with the recommendation result, and this behavior is regarded as a high-quality recommendation. Vectorize the user behavior and result as high-quality results and store them in the database for reference information for the next recommendation; On the contrary, abandon recording the current recommendation process as a useless recommendation.

6. The recommendation method based on user product usage behavior according to claim 5, wherein, It also includes performing the corresponding option operation after the user clicks on the option corresponding to the page recommendation result, tracking the user's behavior, and providing feedback on the execution result: if the execution is not correct, the recommendation result is regarded as an abnormal result, and feedback is provided on the final result, and the stored reference information is retrieved or changed in a timely manner. This behavior is regarded as a high-quality recommendation, but the recommendation result needs to be optimized; if the execution fails, it means that the user agrees with the behavior-guessing recommendation, but there is a logical problem with the recommendation result, and feedback is also provided for it.

7. A recommendation system based on user product usage behavior, characterized in that, It includes: A software user manual, which contains all information related to the software; A word vector storage module, which stores the original text content of the software user manual in the form of word vectors in a word vector library; An original data storage module, which stores the original text content in a database and establishes a mapping relationship with the word vector data; A user behavior acquisition module, which acquires the user's behavior according to the operations of the IO input device during the user's use of the software; A system buried point recording module, which buries points at important functional nodes of the program front-end and back-end or database operations, and defines the specific meaning of the currently triggered function according to the software usage logic; A record query module, which combines the IO input information with the software usage logic by capturing the user's behavior and buried point operations, converts the user's behavior into text information, and after sorting out multiple consecutive text messages, queries the information in the vector library and obtains the result; A prompt word and query result splicing module, which uses the vector library query result as reference information and combines it with the prompt word to splice into a complete question for asking the large model; A large model answer module, which is started as an independent web service to provide large model question and answer functions; A result generation module, which maps the answer result of the large model to the current system software, obtains the function specified by the corresponding result, and returns the recommendation result to the page.

8. A recommendation system based on user product usage behavior as claimed in claim 7, characterized in that, The software user manual includes a software operation manual, the software interface architecture and hierarchical relationship, and precautions for software use operations, and also includes some preset sample user behavior recommendation documents.

9. A recommendation system based on user product usage behavior according to claim 7, characterized in that, It also includes a recommendation result feedback module, which records whether the user accepts the recommendation result by judging whether the user selects and clicks on the page recommendation option, and provides feedback.

10. A recommendation system based on user product usage behavior according to claim 7, characterized in that, It also includes an execution result feedback module, which performs the corresponding option operation after the user clicks on the option corresponding to the page recommendation result, and provides feedback on the execution result by tracking the user's behavior.

Images