Content generation method, apparatus, device, and medium

Abstract

This disclosure provides a content generation method, apparatus, device, and medium, relating to the field of artificial intelligence technology, particularly to machine learning, deep learning, and large model technologies, and capable of being used in application scenarios such as generative search, intelligent document editing, intelligent assistants, virtual assistants, and intelligent e-commerce. The method includes: acquiring a user's query request to a search system, multiple search results corresponding to the query request, and user behavior data of the multiple search results; determining a satisfaction index for each of the multiple search results based on the user behavior data; determining whether content generation is needed for the query request based on the satisfaction index; in response to the determination that content generation is needed, determining content generation parameters for a generative artificial intelligence model based on the multiple search results and the satisfaction index; and invoking the generative artificial intelligence model based on the query request and the content generation parameters to obtain the target content.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, particularly to the fields of machine learning, deep learning, and large models, and can be used in application scenarios such as generative search, intelligent document editing, intelligent assistants, virtual assistants, and intelligent e-commerce. Specifically, it relates to a content generation method, a content generation device, an electronic device, a computer-readable storage medium, and a computer program product. Background Technology

[0002] Artificial intelligence (AI) is the study of enabling computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, and planning). It encompasses both hardware and software technologies. AI hardware technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, and big data processing. AI software technologies mainly include natural language processing, computer vision, speech recognition, machine learning / deep learning, big data processing, and knowledge graph technologies.

[0003] The methods described in this section are not necessarily methods that had been previously conceived or adopted. Unless otherwise specified, no method described in this section should be assumed to be prior art simply because it is included in this section. Similarly, unless otherwise specified, the issues mentioned in this section should not be considered to be accepted in any prior art. Summary of the Invention

[0004] This disclosure provides a content generation method, a content generation apparatus, an electronic device, a computer-readable storage medium, and a computer program product.

[0005] According to one aspect of this disclosure, a content generation method is provided, comprising: acquiring a user's query request for a search system, multiple search results corresponding to the query request, and user behavior data of the multiple search results; determining a satisfaction index for each of the multiple search results based on the user behavior data; determining whether content generation is required for the query request based on the satisfaction index; in response to determining that content generation is required, determining content generation parameters for a generative artificial intelligence model based on the multiple search results and the satisfaction index; and invoking the generative artificial intelligence model based on the query request and the content generation parameters to obtain target content.

[0006] According to another aspect of this disclosure, a content generation apparatus is provided, comprising: an acquisition unit configured to acquire a user's query request for a search system, multiple search results corresponding to the query request, and user behavior data of the multiple search results; a first determination unit configured to determine a satisfaction index for each of the multiple search results based on the user behavior data; a determination unit configured to determine, based on the satisfaction index, whether content generation is required for the query request; a second determination unit configured to, in response to the determination that content generation is required, determine content generation parameters for a generative artificial intelligence model based on the multiple search results and the satisfaction index; and a generation unit configured to invoke the generative artificial intelligence model based on the query request and the content generation parameters to obtain target content.

[0007] According to another aspect of this disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the methods described above.

[0008] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are used to cause a computer to perform the above-described method.

[0009] According to another aspect of this disclosure, a computer program product is provided, including a computer program, wherein the computer program implements the above-described method when executed by a processor.

[0010] According to one or more embodiments of this disclosure, by obtaining the search results corresponding to the query request and their user behavior data, and determining whether to trigger content generation based on a satisfaction index, this effectively reduces the computational and storage resource consumption of generative artificial intelligence models in redundant production scenarios for search systems. Furthermore, by using search results and satisfaction indexes to determine content generation parameters, the production end can perform targeted generation based on feedback data from the distribution end, improving the matching degree between the target content and the query request intent, and enhancing the content supply quality and user experience of the search system.

[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0012] The accompanying drawings exemplify embodiments and form part of the specification, serving together with the textual description to explain exemplary implementations of the embodiments. The illustrated embodiments are for illustrative purposes only and do not limit the scope of the claims. Throughout the drawings, the same reference numerals refer to similar but not necessarily identical elements.

[0013] Figure 1 A schematic diagram of an exemplary system in which the various methods described herein may be implemented according to embodiments of the present disclosure is shown; Figure 2 A flowchart of a content generation method according to an embodiment of the present disclosure is shown; Figure 3 A flowchart illustrating the determination of content generation parameters for a generative artificial intelligence model according to embodiments of the present disclosure is shown; Figure 4 A structural block diagram of a content generation apparatus according to an embodiment of the present disclosure is shown; Figure 5 A structural block diagram of an exemplary electronic device that can be used to implement embodiments of the present disclosure is shown. Detailed Implementation

[0014] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0015] In this disclosure, unless otherwise stated, the use of terms such as "first," "second," etc., to describe various elements is not intended to limit the positional, temporal, or importance relationships of these elements; such terms are merely used to distinguish one element from another. In some examples, the first element and the second element may refer to the same instance of that element, while in other cases, based on the context, they may refer to different instances.

[0016] The terminology used in the description of the various examples in this disclosure is for the purpose of describing particular examples only and is not intended to be limiting. Unless the context explicitly indicates otherwise, an element may be one or more unless the number of elements is specifically limited. Furthermore, the term "and / or" as used in this disclosure covers any one of the listed items and all possible combinations thereof.

[0017] Among related technologies, existing methods for introducing artificial intelligence-generated search results in search systems have not yielded good results.

[0018] To address the aforementioned issues, this disclosure effectively reduces the computational and storage resource consumption of generative AI models in redundant production scenarios for search systems by acquiring search results and user behavior data corresponding to query requests and determining whether content generation should be triggered based on satisfaction metrics. Furthermore, by utilizing search results and satisfaction metrics to determine content generation parameters, the production end can perform targeted generation based on feedback data from the distribution end, improving the matching degree between target content and query intent, and enhancing the content supply quality and user experience of the search system.

[0019] The embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0020] Figure 1 A schematic diagram of an exemplary system 100 in which the various methods and apparatus described herein can be implemented according to embodiments of this disclosure is shown. Reference Figure 1 The system 100 includes one or more client devices 101, 102, 103, 104, 105 and 106, a server 120, and one or more communication networks 110 coupling the one or more client devices to the server 120. The client devices 101, 102, 103, 104, 105 and 106 can be configured to execute one or more applications.

[0021] In embodiments of this disclosure, server 120 may run one or more services or software applications that enable the execution of the methods of this disclosure.

[0022] In some embodiments, server 120 may also provide other services or software applications, which may include non-virtual and virtual environments. In some embodiments, these services may be provided as web-based services or cloud services, such as to users of client devices 101, 102, 103, 104, 105 and / or 106 under a Software as a Service (SaaS) model.

[0023] exist Figure 1 In the configuration shown, server 120 may include one or more components that implement the functions performed by server 120. These components may include software components, hardware components, or combinations thereof that can be executed by one or more processors. Users operating client devices 101, 102, 103, 104, 105, and / or 106 can sequentially interact with server 120 using one or more client applications to utilize the services provided by these components. It should be understood that various different system configurations are possible and may differ from system 100. Therefore, Figure 1 This is an example of a system used to implement the various methods described herein, and is not intended to be limiting.

[0024] Users can use client devices 101, 102, 103, 104, 105, and / or 106 for human-computer interaction. The client devices provide interfaces that enable users to interact with them. The client devices can also output information to the user through these interfaces. Although... Figure 1 Only six client devices are described, but those skilled in the art will understand that this disclosure can support any number of client devices.

[0025] Client devices 101, 102, 103, 104, 105, and / or 106 may include various types of computer devices, such as portable handheld devices, general-purpose computers (such as personal computers and laptops), workstation computers, wearable devices, smart screen devices, self-service terminal devices, service robots, gaming systems, thin clients, various messaging devices, sensors, or other sensing devices. These computer devices can run various types and versions of software applications and operating systems, such as Microsoft Windows, Apple iOS, UNIX-like operating systems, Linux or Linux-like operating systems (such as Google Chrome OS); or include various mobile operating systems, such as Microsoft Windows Mobile OS, iOS, Windows Phone, and Android. Portable handheld devices may include cellular phones, smartphones, tablets, personal digital assistants (PDAs), etc. Wearable devices may include head-mounted displays (such as smart glasses) and other devices. Gaming systems may include various handheld gaming devices, internet-enabled gaming devices, etc. Client devices are capable of executing various applications, such as various internet-related applications, communication applications (such as email applications), short message service (SMS) applications, and can use various communication protocols.

[0026] Network 110 can be any type of network well known to those skilled in the art, and can support data communication using any of a variety of available protocols (including but not limited to TCP / IP, SNA, IPX, etc.). By way of example only, one or more networks 110 can be a local area network (LAN), an Ethernet-based network, a token ring network, a wide area network (WAN), the Internet, a virtual network, a virtual private network (VPN), an intranet, an extranet, a blockchain network, a public switched telephone network (PSTN), an infrared network, a wireless network (e.g., Bluetooth, WIFI), and / or any combination of these and / or other networks.

[0027] Server 120 may include one or more general-purpose computers, special-purpose server computers (e.g., PC (personal computer) servers, UNIX servers, mid-range servers), blade servers, mainframe computers, server clusters, or any other suitable arrangement and / or combination. Server 120 may include one or more virtual machines running a virtual operating system, or other computing architectures involving virtualization (e.g., one or more flexible pools of logical storage devices that can be virtualized to maintain virtual storage devices for servers). In various embodiments, server 120 may run one or more services or software applications that provide the functionality described below.

[0028] The computing unit in server 120 can run one or more operating systems, including any of the aforementioned operating systems and any commercially available server operating system. Server 120 can also run any of a variety of additional server applications and / or middleware applications, including HTTP servers, FTP servers, CGI servers, JAVA servers, database servers, etc.

[0029] In some implementations, server 120 may include one or more applications to analyze and merge data feeds and / or event updates received from users of client devices 101, 102, 103, 104, 105 and / or 106. Server 120 may also include one or more applications to display data feeds and / or real-time events via one or more display devices of client devices 101, 102, 103, 104, 105 and / or 106.

[0030] In some implementations, server 120 can be a server for a distributed system or a server integrated with blockchain. Server 120 can also be a cloud server, or an intelligent cloud computing server or intelligent cloud host with artificial intelligence technology. A cloud server is a host product in the cloud computing service system, designed to address the shortcomings of traditional physical hosts and Virtual Private Server (VPS) services, such as high management difficulty and weak business scalability.

[0031] System 100 may also include one or more databases 130. In some embodiments, these databases may be used to store data and other information. For example, one or more of the databases 130 may be used to store information such as audio files and video files. Databases 130 may reside in various locations. For example, a database used by server 120 may be local to server 120, or it may be located away from server 120 and may communicate with server 120 via a network-based or dedicated connection. Databases 130 may be of different types. In some embodiments, the database used by server 120 may be, for example, a relational database. One or more of these databases may store, update, and retrieve data from and from the databases in response to commands.

[0032] In some embodiments, one or more of the databases 130 may also be used by an application to store application data. The databases used by the application may be of different types, such as key-value stores, object stores, or regular stores supported by a file system.

[0033] Figure 1 The system 100 can be configured and operated in various ways to enable the application of the various methods and apparatus described in this disclosure.

[0034] According to one aspect of this disclosure, a content generation method is provided. Figure 2 This is a flowchart illustrating a content generation method 200 according to an exemplary embodiment.

[0035] refer to Figure 2 In step S201, the user's query request to the search system, multiple search results corresponding to the query request, and user behavior data of the multiple search results are obtained.

[0036] A query request is information, such as text, voice, or images, entered by a user into the input interface of a search system to express their search intent. A search system typically includes a search box and its backend processing engine. Search results are the information entries retrieved and presented to the user in a ranked manner based on the query request, such as web links, articles, images, or videos. User behavior data refers to the feedback data generated by users during their interaction with the search system.

[0037] For example, user behavior data includes, but is not limited to, click data, impression data, page dwell time, and video playback time. This data can be extracted from the search system's logs or obtained by subscribing to real-time data streams. For obtaining search results, a snapshot of the current search results page can be captured based on the query request to obtain content data that approximates the user's visit.

[0038] In step S202, based on user behavior data, the satisfaction index of each of the multiple search results is determined.

[0039] Satisfaction index is a parameter used to quantify how well search results satisfy a user's query intent. This index reflects the quality performance of search results at the distribution end.

[0040] In one exemplary embodiment, a pre-defined evaluation model can be used to perform statistical analysis on user behavior data to calculate the satisfaction index for each search result. For example, for text-type search results, the click-through rate and dwell time can be weighted to obtain the satisfaction index; for video-type search results, the video completion rate can be used as the satisfaction index.

[0041] In step S203, based on the satisfaction index, it is determined whether content generation is required for the query request.

[0042] This step enables the identification of supply and demand gaps in search scenarios, that is, determining whether existing search results have fully met user needs.

[0043] In some embodiments, the determination logic may include: if the overall satisfaction index of multiple search results is generally lower than a preset threshold, and the pageview index of the query request is at a high level, then it is determined that there is a supply-demand gap, and targeted content production is needed for the query request. Conversely, if the feedback data of existing results is excellent, indicating that the demand has been well met, then it is determined that no content generation is needed, thereby saving computing resources.

[0044] In step S204, in response to the determination that content generation is required, content generation parameters for the generative artificial intelligence model are determined based on multiple search results and satisfaction indicators.

[0045] This step enables a closed-loop transfer of data from the distribution end to the production end control parameters, mapping the user's post-hoc behavior on the search end into specific control signals that guide the generation of AI-based content.

[0046] In some embodiments, content that performs well (high satisfaction) in existing search results can be analyzed to uncover the common features of these high-satisfaction content and transform these features into parameters that the generative model can recognize.

[0047] In one exemplary embodiment, if analysis reveals that structured and concise text content has a higher user dwell time for a certain type of query request, then the degree of structure and conciseness are set as content generation parameters. If it is found that videos with specific cover visual features have a higher completion rate, then the corresponding visual features are parameterized to guide subsequent video generation tasks.

[0048] In step S205, a generative artificial intelligence model is invoked based on the query request and content generation parameters to obtain the target content.

[0049] In some embodiments, generative artificial intelligence models include large language models, multimodal generative models, and may also include other AI-based content generation models.

[0050] In one exemplary embodiment, content generation parameters are incorporated as input variables into the production script or prompt word template, and combined with the original intent represented by the query request, the corresponding generative artificial intelligence model is invoked. In this way, the model can generate target content that better meets user needs, such as high-quality articles or short videos, under specific parameter constraints.

[0051] According to embodiments of this disclosure, by acquiring search results and user behavior data corresponding to a query request, and determining whether to trigger content generation based on a satisfaction index, this effectively reduces the computational and storage resource consumption of generative artificial intelligence models in redundant production scenarios for search systems. Furthermore, by using search results and satisfaction indexes to determine content generation parameters, the production end can perform targeted generation based on feedback data from the distribution end, improving the matching degree between target content and query request intent, and enhancing the content supply quality and user experience of the search system.

[0052] Specifically, this content production mechanism based on user satisfaction with search results ensures that every piece of AI-generated content corresponds to a genuine search need. By transforming post-hoc user behavior data into generation parameters for the AI ​​model, the generated AIGC content possesses superior characteristics validated by big data search, improving overall user satisfaction and thus enhancing the user experience. Simultaneously, by pre-intercepting low-value production tasks, redundant overhead on large-scale model computing power and subsequent storage and indexing resources is reduced.

[0053] According to some embodiments, such as Figure 3 As shown, step S204, in response to the determination that content generation is required, and based on multiple search results and satisfaction index, determining the content generation parameters for the generative artificial intelligence model may include: step S301, filtering at least one reference search result from multiple search results whose satisfaction index is higher than a first preset threshold; step S302, performing content feature identification on at least one reference search result to obtain the content features of each of the at least one reference search result; and step S303, determining the content generation parameters based on the content features of each of the at least one reference search result.

[0054] In some embodiments, the first preset threshold can be a benchmark value used to measure whether the satisfaction level of the search results meets the quality standard, and can be preset by statistical analysis of user behavior feedback data in historical search scenarios or by other means. Reference search results refer to search results with excellent user behavior feedback that can serve as benchmark samples for subsequent content generation. Content features refer to attribute information reflecting the content organization, expression structure, or visual presentation style of the reference search results.

[0055] In one embodiment of step S302, a feature extraction model based on preset engineering rules can be used to parse the reference search results and extract the content features of the reference search results in terms of layout structure, information density, or distribution of key information.

[0056] In one embodiment of step S303, the extracted content features can be converted into standardized configuration parameters, and the configuration parameters can be integrated into the production script of the generative artificial intelligence model as input control signals to constrain model generation.

[0057] Therefore, by selecting reference search results with high satisfaction indicators and extracting content features as generation parameters, the content generation process can reversely reference the features of existing high-quality content, improving the matching degree between the target content and the query intent. Generating content based on content features validated by distribution data avoids AIGC models generating content that does not match user preferences, reduces the ineffective computational power consumption of generative artificial intelligence models, and improves the efficiency of high-quality content production and processing.

[0058] According to some embodiments, at least one reference search result may include a first search result of text type. Step S302, performing content feature recognition on at least one reference search result to obtain the content features of each of the at least one reference search result, may include: extracting a text summary of the first search result; and inputting the query request and the text summary into a large language model to perform content feature recognition to obtain the first content features of the first search result.

[0059] The first search result can refer to webpages or articles presented in plain text or a combination of text and images. A text summary can be an extraction of the core information from the first search result. A large language model can be an artificial intelligence model with natural language understanding and generation capabilities. First content features can refer to the structural, logical, or expressive characteristics of the text content.

[0060] In some embodiments, a text summary can be obtained by extracting key paragraphs or titles from the first search result, or by processing the first search result using an extractive summarization algorithm. The query request can be used as background intent guidance, and the text summary can be input into a large language model. Through the semantic analysis capabilities of the large language model, primary content features such as degree of structure, information conciseness, or emphasis on key points can be output.

[0061] In one exemplary embodiment, the large language model analyzes the first search result of the text type, and the output first content feature may be, for example, "structured, concise and highlighting key points". By transforming this first content feature into content generation parameters for a generative artificial intelligence model, the model can output high-quality content that meets this feature.

[0062] Therefore, by leveraging a large language model to perform deep semantic analysis on text summaries, we can capture the expressive advantages of highly satisfying text content and improve the accuracy of feature extraction. Inputting text summaries instead of the full text into the large language model reduces the amount of data processed by the model, saves computational overhead, and improves processing efficiency.

[0063] According to some embodiments, at least one reference search result may include a second search result of video type. Step S302, performing content feature recognition on at least one reference search result to obtain the content features of each of the at least one reference search result, may include: using a multimodal model to perform content understanding on the second search result to obtain a content understanding result; and inputting the query request and the content understanding result into a large language model to perform content feature recognition to obtain the second content features of the second search result.

[0064] In some embodiments, the second search result may refer to resources presented in video format. A multimodal model is a machine learning model capable of processing multiple modalities of data, including visual, auditory, and textual data, simultaneously. Content understanding results may be structured textual descriptions of information such as video footage, audio, and subtitles. Second content features may refer to the characteristics of the video content in terms of visual presentation, pacing, or the use of rich media elements.

[0065] In some embodiments, a multimodal model can be used to extract the cover image of the second search result and keyframe images within a preset time period, thereby generating a description containing the information content of the cover image and the information density of the initial stage image as the content understanding result. The query request and the above-mentioned structured content understanding result can be input into a large language model to analyze and obtain second content features such as the attractiveness of the initial stage content and the rich media presentation format.

[0066] In one exemplary embodiment, the large language model analyzes the content understanding results of the second search result, and the output second content feature could be, for example, "the cover image is information-rich, and the first 3 seconds of rich media content are eye-catching." By transforming this second content feature into content generation parameters for a generative artificial intelligence model, the model can output high-quality content that meets this feature.

[0067] Therefore, by transforming unstructured video data into analyzable content understanding results through a multimodal model, and then using a large language model for feature extraction, the deep-seated advantages of multimedia content are uncovered, improving the accuracy of feature extraction. This process enables the features of video-based reference search results to be quantified, providing reliable parameters for the targeted generation of subsequent AIGC content, thus improving the quality of generated content and user satisfaction.

[0068] According to some embodiments, the satisfaction metrics of multiple search results may include at least one of click-through rate, dwell time, and video completion rate.

[0069] In some embodiments, click-through rate (CTR) can be the ratio of the number of times a user clicks on a specific search result to the number of times that search result is displayed. Dwell time can be the length of time a user remains on the details page corresponding to a specific search result. Video completion rate can be the ratio of the actual playback time of video-based multimedia content to the total duration of the video. These metrics can be obtained in real-time or offline through user behavior logs recorded by the search system's near-line engineering.

[0070] In some embodiments, feature extraction can be performed on query requests, including basic feature extraction and intent feature extraction. Basic features may include categories, entity words, and timeliness. Intent features may include searching for knowledge, services, resources, or tools. For different categories, click-through rate, dwell time, and video completion rate can be assigned different calculation weights, and a quantified comprehensive score is obtained through weighted summation as the final satisfaction index. In some embodiments, the calculation weights corresponding to different categories can be pre-set or determined in real time; this is not limited here.

[0071] Therefore, by introducing multi-dimensional objective posterior behavioral data as an evaluation criterion, we can more comprehensively and realistically reflect users' preferences for search results, thus improving the accuracy of identifying gaps in users' search needs. By reusing existing statistical data generated by the search system for quantitative evaluation, no additional manual evaluation and annotation are required, saving computational costs and improving processing efficiency.

[0072] According to some embodiments, the content generation method may further include: adding the target content to the content resource library of the search system.

[0073] In some embodiments, the target content can be an article or video entity generated by a generative artificial intelligence model under content generation parameter constraints. The content resource library can be a data set used by the search system backend to persistently store various content resources and support the search engine in performing operations such as retrieval and ranking.

[0074] In some embodiments, the generated target content can first undergo structured feature extraction and tag construction, and then be imported into an indexing system to build an inverted index, so that the target content can participate in the search and distribution process of subsequent similar query requests.

[0075] Therefore, the targeted content customized based on the gaps in real user search needs is fed back to the search distribution end, effectively supplementing the shortcomings of the original search results and improving the overall matching degree between the content supply and user needs of the search system. Through the closed-loop mechanism of on-demand production and targeted storage, the idle content caused by redundant production by the AIGC model is avoided, and the storage redundancy of invalid content and the consumption of index resources are reduced.

[0076] According to some embodiments, step S201, obtaining a user's query request for the search system, multiple search results corresponding to the query request, and user behavior data of the multiple search results may include: obtaining the query request through a subscription interface; and crawling web pages based on the query request to obtain multiple search results.

[0077] In some embodiments, the subscription interface is a data channel provided by the search system's near-line engineering, used to sample and output user-initiated search requests according to set rules. Web crawling refers to obtaining snapshots of the static or dynamic code presented in the front end of a webpage.

[0078] In some embodiments, representative query requests can be extracted from the nearline data stream of the search system according to preset category filtering rules or preset time intervals.

[0079] In some embodiments, for the sampled query requests, the search access scenario of a real user can be simulated to obtain the full-page sorted content returned by the current search engine and the webpage data snapshots corresponding to each result, as multiple search results.

[0080] Therefore, by using subscription interfaces and simulated real-world web crawling, the acquired feature data is highly consistent with the actual search scenarios faced by users. By employing a mechanism that combines near-line subscription with sampling crawling, the enormous system pressure caused by analyzing the full real-time traffic of the search engine is avoided, saving computational overhead and improving processing efficiency.

[0081] According to some embodiments, step S203, determining whether content generation is required for the query request based on the satisfaction index, may include: obtaining the pageview index of the query request; and determining that content generation is required in response to the pageview index being higher than a second preset threshold and the satisfaction index of each of the multiple search results meeting preset conditions.

[0082] In some embodiments, the pageview metric can refer to the number of pageviews for the query request over a specific time period, which can be calculated from historical logs. The second preset threshold can serve as a baseline for determining whether the query request has sufficient traffic. The preset conditions are used to evaluate the overall performance of multiple search results and whether they adequately meet user needs.

[0083] In one exemplary embodiment, the preset condition may be that the combined score calculated based on the satisfaction indicators of multiple search results is less than a preset score. The combined score may be calculated, for example, by summing, weighting, or otherwise calculating the satisfaction indicators of the multiple search results. The preset score can be set based on actual needs.

[0084] In one exemplary embodiment, the preset condition may be that the satisfaction index of each of the multiple search results is lower than a third preset threshold.

[0085] In some embodiments, the total number of search impressions of the query request in the hourly, daily, weekly, or monthly dimensions can be counted to assess its potential lead value.

[0086] In some embodiments, when the pageview metric exceeds a second preset threshold representing high potential traffic, and the click-through rate and dwell time of the current full-page search results are both lower than values ​​representing a passing level, it indicates that there are a large number of unmet user needs, and the system can determine to trigger the content generation process.

[0087] Therefore, by combining the pageview metrics, which represent traffic volume, and the satisfaction metrics, which represent distribution performance, a comprehensive evaluation is conducted to ensure that the content production system only intervenes in scenarios with clear production value. This avoids the blind production of low-value long-tail keywords, reduces resource consumption, and improves processing efficiency.

[0088] In some embodiments, step S203, determining whether content generation is required for the query request based on the satisfaction index, may include: in response to the number of multiple search results being greater than or equal to a preset number, and the satisfaction index of multiple search results being higher than a third preset threshold, determining that content generation is not required for the query request.

[0089] The third preset threshold is a high standard limit for measuring whether the current search results are of sufficient quality and whether user needs have been fully met.

[0090] In some embodiments, when it is identified that the entire page displays multiple rich search results and the user's satisfaction level exceeds a third preset threshold, the system identifies the current supply and demand as being in a saturated and high-quality state, and directly terminates the production triggering process for the query request.

[0091] In some embodiments, step S203, determining whether content generation is needed for the query request based on the satisfaction index, may include: in response to the fact that the satisfaction index of each of the multiple search results is within a preset range and the content satisfaction dimension of the multiple search results is singular, determining that differentiated content needs to be supplemented to generate content for the query request.

[0092] In some embodiments, the preset range refers to an intermediate level of user feedback that is neither extremely poor nor extremely good, and its specific value can be set in various ways. "Single content satisfaction dimension" means that existing search results only contain one or a few homogeneous types in terms of intent coverage, content structure, or media presentation. "Differentiated content" refers to incremental content that differs from existing results in its expressive characteristics.

[0093] In some embodiments, if the satisfaction index is generally poor, the system performs feature clustering on the search content of the current page. If it finds that all high-frequency results only cover the single dimension of finding basic definitions and lack in-depth service or resource dimensions, it determines that there is room for production based on the browsing volume of the query request and triggers an instruction to generate supplementary differentiated content.

[0094] According to another aspect of this disclosure, a content generation apparatus is provided. For example... Figure 4 As shown, the device 400 includes: an acquisition unit 410 configured to acquire a user's query request for a search system, multiple search results corresponding to the query request, and user behavior data of the multiple search results; a first determination unit 420 configured to determine a satisfaction index for each of the multiple search results based on the user behavior data; a judgment unit 430 configured to determine whether content generation is required for the query request based on the satisfaction index; a second determination unit 440 configured to determine content generation parameters for a generative artificial intelligence model based on the multiple search results and the satisfaction index in response to the determination that content generation is required; and a generation unit 450 configured to invoke the generative artificial intelligence model based on the query request and the content generation parameters to obtain target content.

[0095] It is understood that the operation and effects of units 410-450 in device 400 can be referred to the description of steps S201-S205 in method 200 above, and will not be repeated here.

[0096] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0097] According to embodiments of this disclosure, an electronic device, a readable storage medium, and a computer program product are also provided.

[0098] refer to Figure 5 The present invention describes a structural block diagram of an electronic device 500 that can serve as a server or client of the present disclosure, which is an example of a hardware device that can be applied to various aspects of the present disclosure. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0099] like Figure 5 As shown, the electronic device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. The RAM 503 may also store various programs and data required for the operation of the electronic device 500. The computing unit 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0100] Multiple components in electronic device 500 are connected to I / O interface 505, including: input unit 506, output unit 507, storage unit 508, and communication unit 509. Input unit 506 can be any type of device capable of inputting information to electronic device 500. Input unit 506 can receive input digital or character information and generate key signal inputs related to user settings and / or function control of electronic device, and may include, but is not limited to, a mouse, keyboard, touchscreen, trackpad, trackball, joystick, microphone, and / or remote control. Output unit 507 can be any type of device capable of presenting information, and may include, but is not limited to, a monitor, speaker, video / audio output terminal, vibrator, and / or printer. Storage unit 508 may include, but is not limited to, disk and optical disk. Communication unit 509 allows electronic device 500 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers, and / or chipsets, such as Bluetooth devices, 802.11 devices, WiFi devices, WiMax devices, cellular communication devices, and / or the like.

[0101] The computing unit 501 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 501 performs the various methods, processes, and / or processes described above. For example, in some embodiments, these methods, processes, and / or processes may be implemented as computer software programs tangibly contained in a machine-readable medium, such as storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, one or more steps of the methods, processes, and / or processes described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform these methods, processes, and / or processes by any other suitable means (e.g., by means of firmware).

[0102] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0103] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0104] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0105] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0106] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), the Internet, and blockchain networks.

[0107] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0108] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0109] While embodiments or examples of this disclosure have been described with reference to the accompanying drawings, it should be understood that the methods, systems, and devices described above are merely exemplary embodiments or examples, and the scope of the invention is not limited by these embodiments or examples, but only by the granted claims and their equivalents. Various elements in the embodiments or examples may be omitted or replaced by their equivalents. Furthermore, the steps may be performed in a different order than that described in this disclosure. Further, various elements in the embodiments or examples may be combined in various ways. Importantly, as the technology evolves, many elements described herein can be replaced by equivalents that appear after this disclosure.

Claims

1. A content generation method, comprising: Acquire user query requests to the search system, multiple search results corresponding to the query requests, and user behavior data of the multiple search results; Based on the user behavior data, the satisfaction index of each of the multiple search results is determined; Based on the satisfaction index, determine whether content generation is required for the query request; In response to the determination that content generation is required, the content generation parameters for the generative artificial intelligence model are determined based on the multiple search results and the satisfaction index. as well as The generative artificial intelligence model is invoked based on the query request and the content generation parameters to obtain the target content.

2. The method according to claim 1, wherein, In response to the determination that content generation is required, based on the multiple search results and the satisfaction index, the content generation parameters for the generative artificial intelligence model are determined as follows: Filter at least one reference search result whose satisfaction index is higher than a first preset threshold from the plurality of search results; Content feature identification is performed on the at least one reference search result to obtain the content features of each of the at least one reference search result; and The content generation parameters are determined based on the content features of each of the at least one reference search result.

3. The method according to claim 2, wherein, The at least one reference search result includes a first search result of text type. Content feature identification is performed on the at least one reference search result to obtain the content features of each of the at least one reference search result, including: Extract the text summary of the first search result; and The query request and the text summary are input into a large language model for content feature recognition to obtain the first content feature of the first search result.

4. The method according to claim 2, wherein, The at least one reference search result includes a second search result of video type. Content feature identification is performed on the at least one reference search result to obtain the content features of each of the at least one reference search result, including: The second search result is analyzed using a multimodal model to obtain the content understanding result; and The query request and the content understanding result are input into a large language model for content feature recognition to obtain the second content feature of the second search result.

5. The method according to any one of claims 1-4, wherein, The satisfaction metrics for the multiple search results include at least one of click-through rate, dwell time, and video completion rate.

6. The method according to any one of claims 1-5, further comprising: The target content is added to the content resource library of the search system.

7. The method according to any one of claims 1-6, wherein, Obtaining a user's query request to the search system, multiple search results corresponding to the query request, and user behavior data of the multiple search results includes: The query request is obtained through the subscription interface; and Based on the query request, web pages are crawled to obtain the multiple search results.

8. The method according to any one of claims 1-7, wherein, Determining whether content generation is needed for the query request based on the satisfaction index includes: Obtain the pageview metrics for the query request; and In response to the pageview count being higher than a second preset threshold, and the satisfaction index of each of the multiple search results meeting preset conditions, it is determined that content generation is required.

9. A content generation apparatus, comprising: The acquisition unit is configured to acquire a user's query request for the search system, multiple search results corresponding to the query request, and user behavior data of the multiple search results. The first determining unit is configured to determine the satisfaction index of each of the multiple search results based on the user behavior data. The determination unit is configured to determine, based on the satisfaction index, whether content generation is required for the query request. The second determining unit is configured to determine content generation parameters for the generative artificial intelligence model based on the multiple search results and the satisfaction index in response to the determination that content generation is required. as well as The generation unit is configured to invoke the generative artificial intelligence model based on the query request and the content generation parameters to obtain the target content.

10. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1 to 8.

11. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, Computer instructions are used to cause a computer to perform the method according to any one of claims 1 to 8.

12. A computer program product comprising a computer program, wherein, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 8.

Images