Advertisement delivery method and apparatus, and computing device cluster

Abstract

An advertisement delivery method, comprising: acquiring retrieval data input by an information retriever; on the basis of the retrieval data, obtaining a search result and first advertisement data, wherein the search result is obtained by means of a search engine performing a search on the basis of the retrieval data, and the first advertisement data is obtained by means of an advertisement engine performing a search on the basis of the retrieval data; at least fusing the search result and the first advertisement data, so as to obtain a processing result, wherein in the process of fusing the search result and the first advertisement data, the first advertisement data is optimized by means of the search result, and / or the search result is optimized by means of the first advertisement data; and presenting the processing result to the information retriever. By means of the method, mixed pushing of search content and advertisement can be realized while both the user experience and the advertisement revenue of a platform are ensured.

Description

Advertisement delivery method, device, and computing device cluster

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office of China on September 12, 2023, with application number 202311170892.5 and application name “A method, device and computing device cluster for advertising delivery”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of information technology (IT), and in particular to an advertisement delivery method, apparatus, and computing device cluster. Background Art

[0003] With the emergence of various emerging Internet products in recent decades, computational advertising technology based on Internet traffic has become one of the important business models for monetizing Internet traffic. It has continuously promoted the long-term development of online advertising forms such as search advertising, information flow advertising, and native advertising, and has an important share in the revenue of many leading Internet companies.

[0004] Current online advertising technology generally uses a forced insertion model, inserting ads into lists of organic content found during searches. Organic content and ads are generated independently by separate search engines, with the results merged using forced insertion only during the final display phase. This advertising method's greatest advantages are its simplicity and high reliability, and it has been widely used across the internet industry for over a decade.

[0005] However, with the development of large language model (LLM) technologies such as generative pre-trained transformer (GPT), conversational search products are gaining more and more popularity among users. In conversational search, how to achieve a mixed push of content and ads to ensure a good user experience while increasing the platform's advertising revenue is a key technical challenge.

[0006] Summary of the Invention

[0007] The present application provides an advertising delivery method, apparatus, computing device cluster, computer storage medium, and computer product, which can achieve mixed push of content and advertisements while ensuring both user experience and platform advertising revenue.

[0008] In a first aspect, the present application provides an advertising delivery method, comprising: obtaining search data input by an information searcher; obtaining search results and first advertisement data based on the search data, wherein the search results are obtained by a search engine based on the search data, and the first advertisement data are obtained by an advertisement engine based on the search data; at least fusing the search results and the first advertisement data to obtain a processed result, wherein, in the process of fusing the search results and the first advertisement data, the first advertisement data is optimized using the search results, and / or the search results are optimized using the first advertisement data; and presenting the processed result to the information searcher. Exemplarily, the search method of the search engine may be a conversational search.

[0009] In this way, when delivering ads, by integrating ad data and search results, and optimizing each other during the fusion process, the ads and search results in the resulting fusion process are strongly correlated, making the delivered ads more similar to the content users expect to search for, improving the relevance between the two. This allows ads to be delivered without users noticing, improving their search engine experience. This allows for a hybrid push of search content and ads while ensuring both user experience and platform advertising revenue.

[0010] In a possible implementation, the retrieval data is the first data most recently input by the information searcher in the first dialogue window, and the first dialogue window is used for the information searcher to communicate with the search engine.

[0011] Furthermore, at least the search results and the first advertisement data are merged to obtain a processing result, including: fusing the search results, the first advertisement data and the historical data in the first session window to obtain a processing result.

[0012] In one possible implementation, at least the search results and the first advertising data are merged to obtain a processing result, specifically including: obtaining a first task prompt based on the search results and the first advertising data, wherein the first task prompt is used to characterize the quality requirements for the processing results; adding the search results and the first advertising data to the first task prompt, and adding the processing method for processing the search results and / or the first advertising data to the first task prompt to obtain a second task prompt; decomposing the second task prompt to obtain a third task prompt, wherein the third task prompt includes N (N≥2) executable subtasks, wherein, among the N subtasks, the subtasks from the 1st subtask to the N-1th subtask are used to process the search results and / or advertising data using the first plug-in to obtain N-1 processing results, and the Nth subtask is used to merge the search results, the first advertising data and at least one of the N-1 processing results to obtain a processing result; executing the subtasks in the third task prompt to obtain the processing result. In this way, the advertisement generation task can be broken down into multiple executable subtasks, thereby improving the efficiency of advertisement generation. In the advertisement generation process, the search results and / or advertisement data can be processed to obtain data adapted to the preferences of the information searcher. The advertisement data and search results can also be optimized with each other (for example, the data processed from the advertisement data can be related to the search results, and / or the data processed from the search results can be related to the advertisement data), thereby improving the effect of subsequent fusion results.

[0013] In one possible implementation, the second task prompt further includes constraining the input and output formats of data during the processing of the search results and / or the first advertisement data. This allows for a unified format for the processed search results and / or advertisement data, facilitating subsequent integration of the two.

[0014] In one possible implementation, the processing method is obtained through the plug-in knowledge of the plug-in, and the plug-in is determined based on the search results and the first advertisement data. In this way, the plug-in can be used to generate non-text advertisements, thereby improving the generalization of advertisement generation.

[0015] In one possible implementation, the search data is the most recently entered first data by the information searcher in the first dialog window, which is used for the information searcher to communicate with the search engine. At this point, adding the search results and the first advertisement data to the first task prompt includes adding the search results, the first advertisement data, and the historical data from the first dialog window to the first task prompt. This ensures that the content presented in the processing result is somewhat relevant to the conversation context, avoiding the abrupt display of the delivered advertisement and further improving the effectiveness of the advertising and user experience.

[0016] In one possible implementation, before presenting the processing results to the information seeker, the process further includes: evaluating the processing results in at least one dimension to obtain an evaluation result; and determining that the evaluation result indicates that the processing results are qualified, wherein the at least one dimension includes at least one of fluency, elegance, and credibility. This allows users to see high-quality processing results, further improving the user experience.

[0017] In one possible implementation, if the evaluation result indicates that the processing result is unsatisfactory, at least the search results and second advertisement data are merged to update the processing result, where the second advertisement data includes the first advertisement data and the processing result. During the fusion of the search results and the second advertisement data, the second advertisement data is optimized using the search results, and / or the search results are optimized using the second advertisement data. This allows for optimization of the processing result to ensure that the quality of the obtained processing result meets requirements.

[0018] In one possible implementation, the processing result includes text and images, or text and video, or text, images, and video, so as to display advertisements with richer content and enhance user experience.

[0019] In a second aspect, the present application provides an advertising delivery device, comprising: an acquisition module and a processing module. The acquisition module is used to acquire the search data input by the information searcher. The processing module is used to obtain search results and first advertisement data based on the search data, wherein the search results are obtained by the search engine based on the search data, and the first advertisement data is obtained by the advertisement engine based on the search data. The processing module is also used to at least fuse the search results and the first advertisement data to obtain a processing result, and after obtaining the processing result, present the processing result to the information searcher; wherein, in the process of fusing the search results and the first advertisement data, the first advertisement data is optimized through the search results, and / or the search results are optimized through the first advertisement data.

[0020] In a possible implementation, the retrieval data is the first data most recently input by the information searcher in the first dialogue window, and the first dialogue window is used for the information searcher to communicate with the search engine.

[0021] In a possible implementation, when the processing module fuses at least the search results and the first advertisement data to obtain the processing result, it is specifically configured to fuse the search results, the first advertisement data, and the historical data in the first session window to obtain the processing result.

[0022] In one possible implementation, when the processing module fuses at least the search results and the first advertising data to obtain the processing results, it is specifically used to: obtain a first task prompt based on the search results and the first advertising data, wherein the first task prompt is used to characterize the quality requirements for the processing results; add the search results and the first advertising data to the first task prompt, and add the processing method for processing the search results and / or the first advertising data to the first task prompt to obtain a second task prompt; decompose the second task prompt to obtain a third task prompt, and the third task prompt includes N (N≥2) executable subtasks, wherein, among the N subtasks, the subtasks from the 1st subtask to the N-1th subtask are used to process the search results and / or advertising data using the first plug-in to obtain N-1 processing results, and the Nth subtask is used to fuse the search results, the first advertising data and at least one of the N-1 processing results to obtain the processing result; execute the subtasks in the third task prompt to obtain the processing result.

[0023] In a possible implementation, the processing module is further configured to: constrain, in the second task prompt, the input and output formats of data in the process of processing the search results and / or the first advertisement data.

[0024] In a possible implementation, the processing method is obtained through plug-in knowledge of the plug-in, and the plug-in is determined based on the search result and the first advertisement data.

[0025] In one possible implementation, the search data is the most recently entered first data by the information searcher in a first dialog window, where the first dialog window is used for a dialog between the information searcher and the search engine. When adding the search results and the first advertisement data to the first task prompt, the processing module is specifically configured to: add the search results, the first advertisement data, and the historical data in the first dialog window to the first task prompt.

[0026] In one possible implementation, before presenting the processing result to the information searcher, the processing module is further used to: evaluate the processing result in at least one dimension to obtain an evaluation result, and determine whether the evaluation result indicates that the processing result is qualified, wherein the at least one dimension includes at least one of fluency, sophistication and credibility.

[0027] In one possible implementation, the processing module is further configured to, if the evaluation result indicates that the processing result is unsatisfactory, merge at least the search results with second advertisement data to update the processing result, where the second advertisement data includes the first advertisement data and the processing result. In the process of merging the search results and the second advertisement data, the second advertisement data is optimized using the search results, and / or the search results are optimized using the second advertisement data.

[0028] In a possible implementation, the processing result includes text and pictures, or text and video, or text, pictures and video.

[0029] In a third aspect, the present application provides a computing device cluster comprising at least one computing device, each computing device comprising a processor and a memory; the processor of at least one computing device is used to execute instructions stored in the memory of at least one computing device, so that the computing device cluster performs the method described in the first aspect or any possible implementation of the first aspect.

[0030] In a fourth aspect, the present application provides a computer-readable storage medium comprising computer program instructions. When the computer program instructions are executed by a computing device, the computing device performs the method described in the first aspect or any possible implementation of the first aspect; alternatively, when the computer program instructions are executed by a computing device cluster, the computing device cluster performs the method described in the first aspect or any possible implementation of the first aspect. Exemplarily, the computing device cluster may include one or more computing devices.

[0031] In a fifth aspect, the present application provides a computer program product comprising instructions, which, when executed by a computing device, causes the computing device to perform the method described in the first aspect or any possible implementation of the first aspect; or, when executed by a computing device cluster, causes the computing device cluster to perform the method described in the first aspect or any possible implementation of the first aspect. Exemplarily, the computing device cluster may include one or more computing devices.

[0032] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The following is a brief introduction to the drawings required for describing the embodiments or prior art.

[0034] FIG1 is a schematic diagram of the technical architecture of an advertising delivery system provided in an embodiment of the present application;

[0035] FIG2 is a schematic diagram of the structure of an advertising delivery system provided in an embodiment of the present application;

[0036] FIG3 is a schematic diagram of a process for generating an advertisement generation task based on search results and advertisement data provided by an embodiment of the present application;

[0037] FIG4 is a flow chart of an advertisement delivery method according to an embodiment of the present application;

[0038] FIG5 is a schematic diagram of a conversation window when an information searcher interacts with a search engine according to an embodiment of the present application;

[0039] FIG6 is a schematic structural diagram of an advertising delivery device provided in an embodiment of the present application;

[0040] FIG7 is a schematic diagram of the structure of a computing device provided in an embodiment of the present application;

[0041] FIG8 is a schematic diagram of the structure of a computing device cluster provided in an embodiment of the present application;

[0042] FIG9 is a schematic diagram of the structure of another computing device cluster provided in an embodiment of the present application. DETAILED DESCRIPTION

[0043] The term "and / or" as used herein describes an association between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. The symbol " / " as used herein indicates that the related objects are in an "or" relationship, for example, A / B means either A or B.

[0044] The terms "first" and "second" in this specification and claims are used to distinguish different objects rather than to describe a specific order of objects. For example, "first response message" and "second response message" are used to distinguish different response messages rather than to describe a specific order of response messages.

[0045] In the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of this application should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0046] In the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more, for example, multiple processing units means two or more processing units, etc.; multiple elements means two or more elements, etc.

[0047] It should be understood that the logistics robots mentioned in the embodiments of this application can also be replaced by other devices capable of autonomous navigation, such as vehicles, etc., without limitation here. For ease of description, the following devices capable of autonomous navigation, such as logistics robots or vehicles, are referred to as "terminal devices."

[0048] For example, FIG1 shows a schematic diagram of the technical architecture of an advertising delivery system provided by an embodiment of the present application. As shown in FIG1 , the technical architecture of the advertising delivery system may include: a search engine and an advertising engine, and the search engine and the advertising engine are independent. When a user uses a conversational search engine, the search engine can search the web page library based on the data entered by the user to obtain search results, and the advertising engine can search the advertising library based on the data entered by the user to obtain advertising data. Finally, the conversational search engine can forcibly insert the advertising data into the search results (for example, inserting it into the end of the search results, etc.) and present it to the user. Since the correlation between the advertising data and the search results is weak and the form is abrupt, it is easy to interrupt the coherence of the conversation content and reduce the user experience of using the conversational search engine. At the same time, this weakly correlated display form will become white noise in the results after long-term use, and it is easy to be ignored by users, resulting in low conversion rate and reduced advertising revenue of the platform.

[0049] In view of this, an embodiment of the present application provides an advertising delivery method that can provide an advertisement with a more user-friendly display format when delivering advertisements. Specifically, the method can fuse the search results searched by the search engine with the advertisement data searched by the advertising engine, and during the fusion process, the two can be mutually optimized. This allows the advertisements and search results in the fused processing results to be strongly correlated, thereby making the delivered advertisements more similar to the content that the user expects to search for, improving the relevance between the two, and thus completing the delivery of advertisements without the user's perception, thereby improving the user's experience of using the search engine.

[0050] For example, Figure 2 shows a schematic diagram of the architecture of an advertisement delivery system provided by an embodiment of the present application. As shown in Figure 2, the advertisement delivery system 200 may include: a client 210 and a server 220. The client 210 and the server 220 may communicate with each other via a network, but is not limited to a network.

[0051] The client 210 is primarily responsible for receiving search data entered by the information seeker and transmitting the search data to the server 220. It also receives processing results from the server 220 and presents the processing results to the information seeker. The processing results are a fusion of search results from the search engine on the server 220 and advertising data from the advertising engine. During the fusion process, the server 220 can use the search results to optimize the advertising data and / or optimize the search results using the advertising data. The processing results may include text, images, videos, text and images, text and videos, or text, images, and videos. The client 210 may be a desktop application, a mobile application, a web application, or a web-based application. In this embodiment, the information seeker can enter search data on the client 210. The search data may be, but is not limited to, words, phrases, sentences, or questions. After the information seeker completes the input and triggers the information search task on the client 210, the client 210 may send the search data entered by the information seeker to the server 220. For example, after completing the input, the information searcher can click a button on the client 210 for executing the information search task. For example, when the client 210 is a Web application, the information searcher can click the "enter" button on the keyboard to trigger the information search task.

[0052] The server 220 is primarily responsible for receiving the search data from the client 210, and, based on the search data, obtaining processing results related to the search data and including search results and advertising data, and sending the processing results to the client 210. The server 220 is typically deployed on a server and has certain computing and storage resources. Exemplarily, the server on which the server 220 is deployed may be, but is not limited to, a cloud server, a physical server, or a virtual private server. The server 220 includes a search engine 221, an advertising engine 222, an LLM-based task decomposition device 224, an LLM-based generation device 224, a plug-in management device 225, and a quality judgment device 226.

[0053] Search engine 221 is primarily responsible for searching the webpage library based on the search data entered by the information seeker on client 210 to obtain search results related to the search data. In some embodiments, search engine 221 may employ, but is not limited to, a conversational search approach, i.e., search engine 2221 is a conversational search engine. For example, the conversation window between the information seeker and search engine 221 may be in the form of conversation window 51 as shown in FIG5 .

[0054] The advertising engine 222 is mainly responsible for searching the advertising library based on the search data entered by the information searcher on the client 210 to obtain advertising data related to the search data. Exemplarily, the advertising data may include: advertising materials or advertising creatives obtained by combining advertising materials. Advertising materials refer to the content displayed in the advertisement. The basic forms of advertising materials include text, pictures, rich media, videos, etc. Exemplarily, advertising materials may include: one or more of the following contents: advertising copy, advertising video, product description, product core selling point, advertising landing page, etc. An advertisement can be, but is not limited to, composed of multiple advertising materials, and the same advertising material can also be used by multiple different advertisements. An advertising creative is a basic advertising delivery unit.

[0055] The LLM-based task decomposition device 224 is mainly responsible for receiving search results from the search engine 221 and receiving advertising data from the advertising engine 222, and processing the search results and advertising data through LLM to generate an advertisement generation task. The advertisement generation task is mainly used to fuse the search results and the advertising data, and in the fusion process, optimize the advertising data using the search results, and / or optimize the search results using the advertising data to obtain a processing result with a more friendly display form. The advertisement generation task may include: a text content generation task, a non-text content generation task, and a multimodal fusion generation task. The text content generation task is used to fuse the text data contained in the search results and the text data contained in the advertising data into one, so as to obtain a processing result containing only text data. The non-text content generation task is used to fuse the non-text data contained in the search results and the non-text data contained in the advertising data into one, so as to obtain a processing result containing only non-text data. For example, non-text data may include data such as pictures, videos, or audio. The multimodal fusion generation task is used to fuse the text data and non-text data contained in the search results, as well as the text data and non-text data contained in the advertisement data, into one, so as to obtain a processing result that contains both text data and non-text data. The LLM-based task decomposition device 224 can determine what kind of advertisement generation task to generate based on the search results and the advertisement data. For example, if both the search results and the advertisement data only contain text data, it can be determined that the generated advertisement generation task is a text content generation task. If both the search results and the advertisement data only contain non-text data, it can be determined that the generated advertisement generation task is a non-text content generation task. If the search results contain text data and the advertisement data contain non-text data, or if the search results contain non-text data and the advertisement data contain text data, it can be determined that the generated advertisement generation task is a multimodal fusion generation task.

[0056] In this embodiment, after determining what type of advertisement generation task to generate, the LLM-based task decomposition device 224 can generate an advertisement generation task based on the search results and advertisement data. Specifically, as shown in FIG3 , the following steps can be included:

[0057] S301. Based on the search results and advertisement data, a first task prompt is screened from the template library, and the search results and advertisement data are added to the first task prompt. Specifically, the advertisement generation task to be generated can be determined first through the search results and advertisement data. Then, the advertisement generation task is used to screen from the preset template library to select a first task prompt. The first task prompt is associated with the determined advertisement generation task. The first task prompt can be used to characterize the quality requirements for the processing results obtained by the fusion. Exemplarily, when the advertisement generation task is a text content generation task, the task prompt can be as follows:

[0058] When the ad generation task is a non-text content generation task, the task prompt can be as follows:

[0059] When the ad generation task is a multimodal fusion generation task, the task prompt can be as follows:

[0060] After the first task prompt is screened, the search results and advertising data can be added to the first task prompt to complete the initialization of the first task prompt. It should be understood that when the first task prompt is initialized, in order to improve the matching degree between the subsequently generated processing results and the information searcher, the personalized characteristics of the information searcher (such as: historical search results related to the information searcher, etc.) can also be added to the first task prompt. At this time, the item "personalized characteristics" can be added to the first task prompt. When the search engine 221 is a conversational search engine, the conversation context of the information searcher's multiple rounds of conversations in the conversational search engine (that is, the historical data generated by the interaction between the information searcher and the search engine, such as the historical data in the conversation window used by the two to interact, etc.) can also be added to the first task prompt. At this time, the item "conversation context" can be added to the first task prompt.

[0061] S302: Based on the search results and advertisement data, at least one first plugin is selected from multiple plugins, and the plugin knowledge of the first plugin is added to the first task prompt to obtain a second task prompt. Since the LLM can only process language-related tasks and cannot process non-language tasks such as images and videos, external tools can be used to enable the LLM to complete non-language tasks. This external tool is a plugin included in the plugin management device 226. The plugin can be used to process the search results and / or advertisement data to obtain data that is tailored to the information seeker's preferences. Each plugin has plugin knowledge. The plugin knowledge of the plugin can be used to indicate how the plugin processes the search results and / or advertisement data. Adding plugin knowledge can be understood as adding a processing method for the search results and / or advertisement data. In some embodiments, the plugin knowledge of the plugin can also be used to indicate the data input and output format of the plugin. In this case, adding plugin knowledge to the first task prompt can constrain the data input and output format during the processing of the search results and / or the first advertisement data. In this case, adding plugin knowledge can be understood as adding constraints on the data input and output format during the processing of the search results and / or the first advertisement data.

[0062] In this embodiment, the LLM-based task decomposition device 224 can scan the plug-in library in the plug-in management device 226 based on the advertisement generation task it determines to be generated, and insert the data contained in the required plug-in into the first task prompt to insert the plug-in knowledge into the first task prompt, thereby obtaining a second task prompt. For example, when the advertisement generation task is a text content generation task, the plug-in screened out can be a plug-in that can generate text data; when the advertisement generation task is a non-text content generation task, the plug-in screened out can be a plug-in that can generate non-text data; when the advertisement generation task is a multimodal fusion generation task, the plug-in screened out can be a plug-in that can generate text data and a plug-in that generates non-text data. After filtering out the required plug-in, the plug-in knowledge of the plug-in can be added to the first task prompt to obtain a second task prompt. Exemplarily, when the advertisement generation task is a multimodal fusion generation task, the second task prompt can be as follows:

[0063] S303. Decompose the second task prompt to obtain a third task prompt, wherein the third task prompt includes N (N≥2) executable subtasks. Among the N subtasks, the subtasks from the 1st subtask to the N-1th subtask are used to process the search results and / or advertising data using the first plug-in to obtain N-1 processing results; the Nth subtask is used to fuse the search results, the first advertising data and at least one of the N-1 processing results to obtain the required processing results. Once the third task prompt is obtained, the advertisement generation task is obtained. In addition, different subtasks can have different execution orders; of course, different subtasks can also have the same execution order, in which case different subtasks can be executed in parallel, but the Nth subtask needs to be executed last because this subtask requires data fusion processing. Exemplarily, when the second task prompt is the task prompt shown in S302, the third task prompt can be as follows:

[0064] {Output@1} and {Output@2} are template replacement expressions. When the third task prompt is used, they are replaced with the outputs of subtasks 1 and 2 in the third task prompt. In the third task prompt in the above example, the numbers in the left column can be understood as the order in which the subtasks are executed. Furthermore, the URL in subtask 1 in the third task prompt can be obtained based on the webpage parsing plug-in and advertisement data retrieval in the second task prompt; the URL in subtask 2 can be obtained based on the image semantic information extraction plug-in and advertisement data retrieval in the second task prompt. Subtask 1 can be used to execute a task that can be performed by the webpage parsing plug-in. Subtask 2 can be used to execute a task that can be performed by the image semantic information extraction plug-in. Subtask 3 can be used to fuse the results of subtasks 1 and 2, as well as the search results and advertisement data, to obtain a processing result. After generating the advertisement generation task, the LLM-based task decomposition device 224 can transmit the advertisement generation task to the LLM-based generation device 225, which then generates the processing result.

[0065] The LLM-based generation device 225 is mainly responsible for executing the advertisement generation task generated by the LLM-based task decomposition device 224 through the LLM. In particular, when executing the task, the LLM-based generation device 225 can execute each subtask through the LLM and according to the execution order of the subtasks included in the third task prompt, and can call the plug-in in the plug-in management device 226 during the execution process to generate the processing result. In particular, when executing the last subtask in the third task prompt, the LLM-based generation device 225 can optimize the advertisement data through the LLM and use the search results, and / or optimize the search results using the advertisement data, so as to merge the two and obtain a processing result with a more friendly display form.

[0066] The plug-in management device 226 may include at least one plug-in, wherein the plug-in may include tasks that the plug-in can complete, as well as input and output formats.

[0067] The quality judgment device 227 is primarily responsible for judging the quality of the processing result generated by the LLM-based generation device 225 to determine whether the quality of the processing result meets the quality standards. If the quality of the processing result meets the quality standards, the quality judgment device 227 can transmit the processing result to the client 210 so that the processing result can be presented to the information searcher through the client 210.

[0068] When the quality of the processing result does not meet the quality standards, the quality judgment device 227 can transmit the processing result to the LLM-based task decomposition device 223 to iteratively regenerate the advertisement generation task. When the LLM-based task decomposition device 223 iteratively generates the advertisement generation task, the input to the LLM-based task decomposition device 223 is the search results, advertisement data, and the processing result that does not meet the quality standards. The processing result that does not meet the quality standards can be understood as another type of advertisement data. During the iteration process, the LLM-based task decomposition device 223 can merge the search results, advertisement data, and the previously generated processing result that did not meet the quality standards to update the processing result that did not meet the quality standards. Exemplarily, the quality judgment device 227 can evaluate the processing result based on multiple dimensions such as fluency, sophistication, and credibility to obtain an evaluation result. If the evaluation result indicates that the processing result is qualified, it indicates that the generated processing result is of high quality, i.e., meets the quality standards. Otherwise, it indicates that the quality is low, i.e., does not meet the quality standards. Exemplarily, the quality judgment device 227 can be, but is not limited to, a neural network model.

[0069] In some embodiments, in addition to integrating search results and advertisement data through the aforementioned LLM-based task decomposition device 224, LLM-based generation device 224, and plug-in management device 226, after obtaining the search results and advertisement data, some words or phrases in the search results can be replaced with some keywords contained in the advertisement data, or some keywords in the advertisement data can be replaced with some words or phrases in the search results, wherein the words before replacement should have the same or similar semantics as the words after replacement. In this way, through this keyword replacement method, the advertisement and search results can be integrated. In addition, after a sentence (or paragraph) in the search results, the sentence (or paragraph) can be explained by example, and the content contained in the advertisement data can be used in the explanation. In this way, the search results are optimized through the advertisement data, making the search results easier to understand. In this way, the advertisement is integrated into the search results through example. Of course, when explaining by example, the keywords in the search results can also be used to modify / optimize the keywords in the advertisement data so that the advertisement described in the example is not too abrupt, thereby improving the user experience.

[0070] The above is an introduction to the advertising delivery system 200 provided in the embodiment of the present application. Next, based on the above content, an advertising delivery method provided in the embodiment of the present application is introduced. Part or all of the content of the advertising delivery method can be referred to the description in the above content.

[0071] For example, FIG4 shows a flow chart of an advertisement delivery method provided by an embodiment of the present application. It is understood that the method can be executed by any device, equipment, platform, or device cluster with computing and processing capabilities. For ease of description, the following is an introduction using the aforementioned server 220 execution as an example. Among them, the execution processes of other devices, equipment, platforms, and device clusters can all refer to the execution process of the server 220. As shown in FIG4 , the advertisement delivery method may include the following steps:

[0072] S401: Acquire search data input by an information searcher.

[0073] In this embodiment, the server 220 can obtain the search data entered by the information seeker from the aforementioned client 210. In some embodiments, a first conversation window can be displayed on the client 210. This first conversation window can be used for the information seeker to communicate with the search engine. The client 210 can be a client associated with the search engine. The search data obtained can be the most recent data entered by the information seeker in the first conversation window. Furthermore, the search engine can be a conversational search engine. The interaction between the information seeker and the search engine can be a question-and-answer format, where the information seeker asks a question in the conversation window, and the search engine returns the answer to the question in the conversation window. For example, as shown in FIG5 , a conversation window 51 can be shown in which the question posed by the information seeker and the answer returned by the search engine can be displayed. This conversation window 51 can also include an input control 52, through which the information seeker can enter the data they wish to retrieve.

[0074] S402. Obtain search results and first advertisement data based on the search data, wherein the search results are obtained by the search engine based on the search data, and the first advertisement data are obtained by the advertisement engine based on the search data.

[0075] In this embodiment, the search engine in the server 220 can search the webpage library based on the search data to obtain search results related to the search data. The advertising engine in the server 220 can search the advertisement library based on the search data to obtain advertisement data related to the search data.

[0076] In some embodiments, before executing S402, it may be determined whether the information searcher's search behavior has commercial intent, such as commercial intent related to advertising, such as purchasing goods, downloading applications, etc. If the information searcher's search behavior has commercial intent, S420 may be executed. If there is no commercial intent, S402 and subsequent steps may not be executed, and the search engine may be directly used to search the web page library based on the retrieval data to obtain search results, and the search results may be presented to the user. In other words, when the information searcher's search behavior does not have commercial intent, advertising may not be placed; when the information searcher's search behavior has commercial intent, advertising may be placed, that is, S402 to S404 may be executed.

[0077] S403. At least merge the search results and the first advertisement data to obtain a processing result, wherein, in the process of merging the search results and the first advertisement data, the first advertisement data is optimized through the search results, and / or the search results are optimized through the first advertisement data.

[0078] In this embodiment, the server 220 can at least fuse the search results and the advertising data to obtain a processing result with a more user-friendly display format. For example, the processing result can include text and images, or text and video, or text, images, and video. The process of obtaining the processing result can be seen in the description of FIG. 3 and the operation of the LLM-based generation device 225, and will not be repeated here.

[0079] In some embodiments, in order to enable information searchers to see high-quality processing results, the processing results can also be evaluated in at least one dimension (such as fluency, exquisiteness or credibility, etc.). When the evaluation result indicates that the processing result is qualified, it means that the quality of the processing result meets the requirements, and S404 can be executed at this time. When the evaluation result indicates that the processing result is unqualified, it means that the quality of the processing result does not meet the requirements. At this time, at least the search results and the second advertising data can be integrated into one to update the processing results. The second advertising data includes the first advertising data and the processing results. In the process of fusing the search results and the second advertising data, the second advertising data can be optimized through the search results, and / or the search results can be optimized through the second advertising data. The specific fusion process can be referred to the process of fusing the search results and the first advertising data, which will not be repeated here.

[0080] In some embodiments, when the search engine is a conversational search engine, the search results, the first advertisement data, and historical data from the conversation window between the information seeker and the search engine may be integrated to generate a processed result. This allows the content presented in the processed result to be somewhat relevant to the conversation context, avoiding unobtrusive display of the delivered advertisement and further improving the effectiveness of advertising and user experience.

[0081] S404: Present the processing results to the information searcher.

[0082] In this embodiment, the server 220 can present the processing result to the information searcher. For example, the server 220 can transmit the processing result to the client 210 so as to display the processing result to the user through the client 210.

[0083] In this way, since the processing results presented to information searchers are obtained by the fusion of search results and advertising data, they contain both search results and advertising data, and the two can optimize each other during the fusion process. This makes the advertisements and search results in the fused processing results strongly correlated, making the delivered advertisements more similar to the content that users expect to search for, thereby improving the correlation between the two, and then completing the delivery of advertisements without users noticing, thereby improving the user experience of using search engines.

[0084] It should be understood that FIG4 is described with the server as the execution entity. In some embodiments, the advertising delivery method shown in FIG4 can also be executed by the client. In other embodiments, the advertising delivery method shown in FIG4 can also be executed jointly by the client and the server. In this case, the client can execute S401 and transmit the retrieved data to the server. The server can execute S402 and S403, and after obtaining the processing results, transmit the processing results to the client. Then, the client can execute S404.

[0085] It can be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0086] The above is an introduction to the advertisement delivery method provided by the embodiment of the present application. Next, based on the method in the above embodiment, an advertisement delivery device provided by the embodiment of the present application is introduced.

[0087] Exemplarily, FIG6 shows a schematic structural diagram of an advertising delivery device provided by an embodiment of the present application. As shown in FIG6 , the advertising delivery device 600 may include: an acquisition module 601 and a processing module 602. The acquisition module 601 is used to acquire the retrieval data input by the information searcher. The processing module 602 is used to obtain search results and first advertisement data based on the retrieval data, wherein the search results are obtained by the search engine based on the retrieval data, and the first advertisement data are obtained by the advertisement engine based on the retrieval data. The processing module 602 is also used to at least fuse the search results and the first advertisement data to obtain a processing result, and after obtaining the processing result, present the processing result to the information searcher; wherein, in the process of fusing the search results and the first advertisement data, the first advertisement data is optimized through the search results, and / or the search results are optimized through the first advertisement data.

[0088] In a possible implementation, the retrieval data is the first data most recently input by the information searcher in the first dialogue window, and the first dialogue window is used for the information searcher to communicate with the search engine.

[0089] In a possible implementation, when the processing module 602 fuses at least the search results and the first advertisement data to obtain the processing result, it is specifically configured to fuse the search results, the first advertisement data, and the historical data in the first session window to obtain the processing result.

[0090] In some embodiments, when the processing module 602 integrates at least the search results and the first advertising data into one to obtain the processing results, it is specifically used to: obtain a first task prompt based on the search results and the first advertising data, wherein the first task prompt is used to characterize the quality requirements for the processing results; add the search results and the first advertising data to the first task prompt, and add the processing method for processing the search results and / or the first advertising data to the first task prompt to obtain a second task prompt; decompose the second task prompt to obtain a third task prompt, and the third task prompt includes N (N≥2) executable subtasks, wherein, among the N subtasks, the subtasks in the 1st subtask to the N-1th subtask are used to process the search results and / or advertising data using the first plug-in to obtain N-1 processing results, and the Nth subtask is used to integrate the search results, the first advertising data and at least one of the N-1 processing results to obtain the processing result; execute the subtasks in the third task prompt to obtain the processing result.

[0091] In some embodiments, the processing module 602 is further configured to: constrain, in the second task prompt, the input and output formats of data in the process of processing the search results and / or the first advertisement data.

[0092] In some embodiments, the processing method is obtained through plug-in knowledge of the plug-in, and the plug-in is determined based on the search result and the first advertisement data.

[0093] In some embodiments, the search data is the most recently entered first data in the first dialog window used by the information searcher to communicate with the search engine. When adding the search results and the first advertisement data to the first task prompt, processing module 602 is specifically configured to: add the search results, the first advertisement data, and the historical data in the first dialog window to the first task prompt.

[0094] In some embodiments, before presenting the processing results to the information searcher, the processing module 602 is also used to: evaluate the processing results in at least one dimension to obtain an evaluation result, and determine whether the evaluation result indicates that the processing results are qualified, wherein the at least one dimension includes: at least one of: fluency, sophistication and credibility.

[0095] In some embodiments, processing module 602 is further configured to, if the evaluation result indicates that the processing result is unsatisfactory, merge at least the search results with the second advertisement data to update the processing result, where the second advertisement data includes the first advertisement data and the processing result. In the process of merging the search results and the second advertisement data, the second advertisement data is optimized using the search results, and / or the search results are optimized using the second advertisement data.

[0096] In some embodiments, the processing result includes text and pictures, or text and video, or text, pictures and video.

[0097] In some embodiments, both the acquisition module 601 and the processing module 602 shown in FIG6 can be implemented via software or hardware. For example, the implementation of the acquisition module 601 will be described below using the acquisition module 601 as an example. Similarly, the implementation of the processing module 602 can refer to the implementation of the acquisition module 601.

[0098] As an example of a software functional unit, the acquisition module 601 may include code running on a computing instance. The computing instance may include at least one of a physical host (computing device), a virtual machine, and a container. Furthermore, the computing instance may be one or more. For example, the acquisition module 601 may include code running on multiple hosts / virtual machines / containers. It should be noted that the multiple hosts / virtual machines / containers used to run the code may be distributed in the same region or in different regions. Furthermore, the multiple hosts / virtual machines / containers used to run the code may be distributed in the same availability zone (AZ) or in different AZs, each AZ including one data center or multiple geographically close data centers. Typically, a region may include multiple AZs.

[0099] Similarly, multiple hosts / virtual machines / containers running the code can be distributed within the same virtual private cloud (VPC) or across multiple VPCs. Typically, a VPC is set up within a region. Cross-region communication between two VPCs within the same region, or between VPCs in different regions, requires a communication gateway within each VPC to interconnect the VPCs.

[0100] As an example of a hardware functional unit, the acquisition module 601 may include at least one computing device, such as a server. Alternatively, the acquisition module 601 may be implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD). The PLD may be a complex programmable logical device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0101] The multiple computing devices included in acquisition module 601 can be distributed in the same region or in different regions. The multiple computing devices included in acquisition module 601 can be distributed in the same AZ or in different AZs. Similarly, the multiple computing devices included in acquisition module 601 can be distributed in the same VPC or in multiple VPCs. The multiple computing devices can be any combination of servers, ASICs, PLDs, CPLDs, FPGAs, GALs, and other computing devices.

[0102] It should be noted that, in other embodiments, the acquisition module 601 can be used to execute any step in the advertising delivery method described in the above embodiments, and the processing module 602 can be used to execute any step in the advertising delivery method described in the above embodiments. The steps that the acquisition module 601 and the processing module 602 are responsible for implementing can be specified as needed. By respectively implementing different steps in the advertising delivery method described in the above embodiments by the acquisition module 601 and the processing module 602, all the functions of the advertising delivery device 600 shown in Figure 6 can be realized.

[0103] This application also provides a computing device 700. As shown in Figure 7, computing device 700 includes a bus 702, a processor 704, a memory 706, and a communication interface 708. Processor 704, memory 706, and communication interface 708 communicate with each other via bus 702. Computing device 700 can be a server or a terminal device. It should be understood that this application does not limit the number of processors and memories in computing device 700.

[0104] Bus 702 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, among others. Buses may be classified as address buses, data buses, control buses, and the like. For ease of illustration, FIG7 illustrates a single bus line, but this does not imply a single bus or type of bus. Bus 704 may include a path for transmitting information between various components of computing device 700 (e.g., memory 706, processor 704, and communication interface 708).

[0105] The processor 704 may include any one or more processors such as a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0106] The memory 706 may include volatile memory, such as random access memory (RAM). The processor 704 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0107] Memory 706 stores executable program code, which processor 704 executes to implement the functions of acquisition module 601 and processing module 602 shown in FIG6 , thereby implementing the advertising delivery method described in the above embodiment. In other words, memory 706 stores instructions for executing the advertising delivery method described in the above embodiment.

[0108] Alternatively, the memory 706 stores executable code, and the processor 704 executes the executable code to respectively implement the functions of the advertising delivery device 600 shown in Figure 5, thereby implementing the advertising delivery method described in the above embodiment. In other words, the memory 706 stores instructions for executing the advertising delivery method described in the above embodiment.

[0109] The communication interface 708 uses a transceiver module such as, but not limited to, a network interface card or a transceiver to implement communication between the computing device 700 and other devices or a communication network.

[0110] Embodiments of the present application also provide a computing device cluster. The computing device cluster includes at least one computing device. The computing device can be a server, such as a central server, an edge server, or a local server in a local data center. In some embodiments, the computing device can also be a terminal device such as a desktop computer, a laptop computer, or a smartphone.

[0111] As shown in Figure 8, the computing device cluster includes at least one computing device 700. The memory 706 in one or more computing devices 700 in the computing device cluster may store the same instructions for executing the advertisement delivery method described in the above embodiment.

[0112] In some possible implementations, the memory 706 of one or more computing devices 700 in the computing device cluster may also store partial instructions for executing the advertising delivery method described in the above embodiment. In other words, the combination of one or more computing devices 700 can jointly execute instructions for executing the advertising delivery method described in the above embodiment.

[0113] It should be noted that the memory 706 in different computing devices 700 in the computing device cluster can store different instructions, each for executing a portion of the functions of the advertising delivery apparatus 600 shown in FIG6 . In other words, the instructions stored in the memory 706 in different computing devices 700 can implement the functions of one or more of the acquisition module 601 and the processing module 602.

[0114] In some possible implementations, one or more computing devices in a computing device cluster may be connected via a network. The network may be a wide area network (WAN) or a local area network (LAN), among others. FIG. 9 illustrates a possible implementation. As shown in FIG. 9 , two computing devices 700A and 700B are connected via a network. Specifically, the connection to the network is achieved via a communication interface in each computing device. In this type of possible implementation, the memory 706 in the computing device 700A stores instructions for executing the functions of the acquisition module 601. Simultaneously, the memory 706 in the computing device 700B stores instructions for executing the functions of the processing module 602.

[0115] It should be understood that the functionality of the computing device 700A shown in FIG9 may also be implemented by multiple computing devices 700. Similarly, the functionality of the computing device 700B may also be implemented by multiple computing devices 700.

[0116] The present application also provides another computing device cluster. The connection relationship between the computing devices in this computing device cluster can be similar to the connection method of the computing device cluster described in Figures 8 and 9. However, the memory 706 in one or more computing devices 700 in this computing device cluster can store the same instructions for executing the method described in the above embodiment.

[0117] In some possible implementations, the memory 706 of one or more computing devices 700 in the computing device cluster may also store some instructions for executing the aforementioned advertising delivery method. In other words, the combination of one or more computing devices 700 can jointly execute the instructions for executing the aforementioned advertising delivery method.

[0118] Based on the method in the above embodiment, an embodiment of the present application provides a computer-readable storage medium, including computer program instructions. When the computer program instructions are executed by a computing device, the computing device executes the method in the above embodiment; or, when the computer program instructions are executed by a computing device cluster, the computing device cluster executes the method in the above embodiment. Exemplarily, the computer-readable storage medium can be any available medium that can be stored by a computing device or a data storage device such as a data center that contains one or more available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid-state drive), etc.

[0119] Based on the method in the above embodiment, an embodiment of the present application provides a computer program product containing instructions, which, when executed by a computing device, enables the computing device to execute the method in the above embodiment, or, when executed by a computing device cluster, enables the computing device cluster to execute the method in the above embodiment.

[0120] It is understood that the processor in the embodiments of the present application may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0121] The method steps in the embodiments of the present application can be implemented by hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, which can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, mobile hard disks, CD-ROMs or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC.

[0122] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted via the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)).

[0123] It will be understood that the various numerical numbers involved in the embodiments of the present application are merely distinctions for the convenience of description and are not intended to limit the scope of the embodiments of the present application.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the protection scope of the technical solutions of the embodiments of the present application.

Claims

1. An advertisement delivery method, characterized in that: The method comprises: Acquire search data input by information searchers; Based on the search data, a search result and first advertisement data are obtained, wherein the search result is obtained by a search engine based on the search data, and the first advertisement data is obtained by an advertisement engine based on the search data; at least fusing the search results with the first advertisement data to obtain a processing result, wherein, in the process of fusing the search results with the first advertisement data, optimizing the first advertisement data through the search results, and / or optimizing the search results through the first advertisement data; The processing result is presented to the information searcher.

2. The method according to claim 1, characterized in that The search data is the first data most recently input by the information searcher in the first dialog window, and the first dialog window is used for the information searcher to communicate with the search engine.

3. The method according to claim 2, characterized in that The at least merging the search result and the first advertisement data to obtain the processing result includes: The search result, the first advertisement data and the historical data in the first session window are merged to obtain the processing result.

4. The method according to claim 1 or 2, characterized in that: The at least integrating the search result and the first advertisement data into one to obtain a processing result specifically includes: Based on the search result and the first advertisement data, a first task prompt is obtained, wherein the first task prompt is used to represent the quality requirement for the processing result; adding the search result and the first advertisement data to the first task prompt, and adding a processing method for processing the search result and / or the first advertisement data to the first task prompt to obtain a second task prompt; Decomposing the second task prompt to obtain a third task prompt, wherein the third task prompt includes N (N≥2) executable subtasks, wherein, among the N subtasks, the subtasks from the 1st subtask to the N-1th subtask are used to process the search results and / or the first advertisement data to obtain N-1 processed results, and the Nth subtask is used to merge the search results, the first advertisement data and at least one of the N-1 processed results to obtain the processed result; The subtasks in the third task prompt are executed to obtain the processing result.

5. The method according to claim 4, characterized in that Also includes: The second task prompt constrains the input and output formats of data in the process of processing the search results and / or the first advertisement data.

6. The method according to claim 4 or 5, characterized in that: The processing method is obtained through plug-in knowledge of the plug-in, and the plug-in is determined based on the search result and the first advertisement data.

7. The method according to any one of claims 4 to 6, characterized in that: The search data is the first data most recently input by the information searcher in the first dialog window, and the first dialog window is used for the information searcher to communicate with the search engine; The step of adding the search result and the first advertisement data to the first task prompt includes: The search result, the first advertisement data and the historical data in the first session window are added to the first task prompt.

8. The method according to any one of claims 1 to 7, characterized in that: Before presenting the processing result to the information searcher, the method further includes: The processing result is evaluated in at least one dimension to obtain an evaluation result, and it is determined that the evaluation result indicates that the processing result is qualified, wherein the at least one dimension includes at least one of fluency, refinement, and credibility.

9. The method according to claim 8, characterized in that The method further comprises: In the case where the evaluation result indicates that the processing result is unqualified, at least the search result and second advertisement data are merged to update the processing result, wherein the second advertisement data includes the first advertisement data and the processing result; In the process of integrating the search results and the second advertisement data, the second advertisement data is optimized through the search results, and / or the search results are optimized through the second advertisement data.

10. The method according to any one of claims 1 to 9, characterized in that: The processing result includes text and pictures, or text and video, or text, pictures and video.

11. An advertisement delivery device, characterized in that: include: An acquisition module, used for acquiring search data input by an information searcher; A processing module, configured to obtain search results and first advertisement data based on the search data, wherein the search results are obtained by searching the search engine based on the search data, and the first advertisement data are obtained by searching the advertisement engine based on the search data; The processing module is further configured to at least merge the search results and the first advertisement data to obtain a processing result, wherein, in the process of merging the search results and the first advertisement data, the first advertisement data is optimized by the search results, and / or the search results are optimized by the first advertisement data; The processing module is further used to present the processing result to the information searcher after obtaining the processing result.

12. The device according to claim 11, characterized in that The search data is the first data most recently input by the information searcher in the first dialog window, and the first dialog window is used for the information searcher to communicate with the search engine.

13. The device according to claim 12, characterized in that When the processing module merges at least the search result and the first advertisement data to obtain the processing result, it is specifically used to: The search result, the first advertisement data and the historical data in the first session window are merged to obtain the processing result.

14. The device according to claim 11 or 12, characterized in that When the processing module merges at least the search result and the first advertisement data to obtain a processing result, the processing module is specifically used to: Based on the search result and the first advertisement data, a first task prompt is obtained, wherein the first task prompt is used to represent the quality requirement for the processing result; adding the search result and the first advertisement data to the first task prompt, and adding a processing method for processing the search result and / or the first advertisement data to the first task prompt to obtain a second task prompt; Decomposing the second task prompt to obtain a third task prompt, wherein the third task prompt includes N (N≥2) executable subtasks, wherein, among the N subtasks, the subtasks from the 1st subtask to the N-1th subtask are used to process the search results and / or the advertisement data using the first plug-in to obtain N-1 processed results, and the Nth subtask is used to merge the search results, the first advertisement data and at least one of the N-1 processed results to obtain the processed result; The subtasks in the third task prompt are executed to obtain the processing result.

15. The device according to claim 14, characterized in that The processing module is further used for: The second task prompt constrains the input and output formats of data in the process of processing the search results and / or the first advertisement data.

16. The device according to claim 14 or 15, characterized in that The processing method is obtained through plug-in knowledge of the plug-in, and the plug-in is determined based on the search result and the first advertisement data.

17. The device according to any one of claims 14 to 16, characterized in that: The search data is the first data most recently input by the information searcher in the first dialog window, and the first dialog window is used for the information searcher to communicate with the search engine; When the processing module adds the search result and the first advertisement data to the first task prompt, the processing module is specifically used to: The search result, the first advertisement data and the historical data in the first session window are added to the first task prompt.

18. The device according to any one of claims 11 to 17, characterized in that: Before presenting the processing result to the information searcher, the processing module is further used to: The processing result is evaluated in at least one dimension to obtain an evaluation result, and it is determined that the evaluation result indicates that the processing result is qualified, wherein the at least one dimension includes: at least one of fluency, refinement and credibility.

19. The device according to claim 18, characterized in that The processing module is further used for: In the case where the evaluation result indicates that the processing result is unqualified, at least the search result and second advertisement data are merged to update the processing result, wherein the second advertisement data includes the first advertisement data and the processing result; In the process of integrating the search results and the second advertisement data, the second advertisement data is optimized through the search results, and / or the search results are optimized through the second advertisement data.

20. The device according to any one of claims 11 to 19, characterized in that: The processing result includes text and pictures, or text and video, or text, pictures and video.

21. A computing device cluster, characterized in that: comprising at least one computing device, each computing device comprising a processor and a memory; The processor of the at least one computing device is used to execute instructions stored in the memory of the at least one computing device, so that the computing device cluster executes the method according to any one of claims 1-10.

22. A computer-readable storage medium, characterized in that: The method comprises computer program instructions, and when the instructions are executed by a computing device cluster, the computing device cluster executes the method according to any one of claims 1 to 10, wherein the computing device cluster comprises at least one computing device.

23. A computer program product comprising instructions, characterized in that When the instructions are executed by a computing device cluster, the computing device cluster executes the method according to any one of claims 1 to 10, wherein the computing device cluster includes at least one computing device.

Images