Information processing method based on large language model, medium, electronic equipment and product

Abstract

The invention provides an information processing method based on a large language model, a medium, electronic equipment and a product, and relates to the technical field of computers.The method comprises the steps that a search control including a first sub-control used for selecting a to-be-analyzed target object is displayed on a first page; then in response to an analysis operation for a target object indicated by the first sub-control, an analysis result obtained by the large language model based on the event flow of the target object is displayed on the first page, and the analysis result at least comprises interaction summary information used for describing interaction actions and interaction requirements existing in the target object; the event stream is analyzed by using the large language model, so that the analysis efficiency of the target object can be improved, and the analysis result can be more accurate.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular, to an information processing method, medium, electronic device, and product based on a large language model. Background Art

[0002] In the related art, event streams can be analyzed to better carry out marketing activities. For example, by analyzing event streams, interaction patterns at different activity stages can be determined, providing a basis for subsequent marketing strategy adjustments. However, due to the extremely complex and numerous event tagging information, the analysis of objects still relies on a large amount of information query by staff, resulting in low analysis efficiency. Summary of the Invention

[0003] This Summary of the Invention section is provided to introduce concepts in a brief form, which will be described in detail in the following Detailed Implementation section. This Summary of the Invention section is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0004] In a first aspect, the present disclosure provides an information processing method based on a large language model, including: Displaying a search control on a first page, the search control including a first sub-control for selecting a target object to be analyzed; In response to an analysis operation on the target object indicated by the first sub-control, displaying, on the first page, an analysis result obtained by the large language model based on the event stream of the target object, the analysis result including at least interaction summary information for describing interaction actions and interaction requirements of the target object.

[0005] In a second aspect, the present disclosure provides a computer-readable medium having a computer program stored thereon, and when the computer program is executed by a processing device, the steps of the method in the first aspect are implemented.

[0006] In a third aspect, the present disclosure provides an electronic device, including: A storage device having a computer program stored thereon; A processing device for executing the computer program in the storage device to implement the steps of the method in the first aspect.

[0007] In a fourth aspect, the present disclosure provides a computer program product including a computer program, and when the computer program is executed by a processor, the steps of the method in the first aspect are implemented.

[0008] Based on the above technical solution, by displaying a search control including a first sub-control for selecting a target object to be analyzed on a first page, and then in response to an analysis operation on the target object indicated by the first sub-control, displaying on the first page the analysis result obtained by the large language model based on the event stream of the target object, the analysis result at least includes interaction summary information for describing the interaction actions and interaction requirements existing for the target object. By using the large language model to analyze the event stream, not only can the analysis efficiency of the target object be improved, but also the analysis result can be made more accurate.

[0009] Other features and advantages of the present disclosure will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In combination with the accompanying drawings and with reference to the following specific implementation manners, the above and other features, advantages and aspects of the various embodiments of the present disclosure will become more obvious. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic, and the original components and elements are not necessarily drawn to scale. In the drawings: Figure 1 is a flowchart of an information processing method based on a large language model shown according to an exemplary embodiment.

[0011] Figure 2 is a schematic diagram of a first page shown according to an exemplary embodiment.

[0012] Figure 3 is a schematic diagram of a fourth sub-control shown according to an exemplary embodiment.

[0013] Figure 4 is a schematic diagram of a second page shown according to an exemplary embodiment.

[0014] Figure 5 is a schematic diagram of a second page shown according to another exemplary embodiment.

[0015] Figure 6 is a schematic diagram of saving an association relationship shown according to an exemplary embodiment.

[0016] Figure 7 is a schematic structural diagram of an information processing device based on a large language model shown according to an exemplary embodiment.

[0017] Figure 8 is a schematic structural diagram of an electronic device shown according to an exemplary embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not used to limit the protection scope of the present disclosure.

[0019] It should be understood that the various steps recited in the method embodiments of the present disclosure can be executed in a different order and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this regard.

[0020] The term "including" and its variations used herein are open-ended, i.e., "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0021] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependent relationships.

[0022] It should be noted that the modifications of "one" and "plural" mentioned in the present disclosure are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly specified in the context, it should be understood as "one or more".

[0023] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0024] It can be understood that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, usage scopes, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0025] For example, when responding to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested by the user will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, an application program, a server or a storage medium that performs the operations of the technical solutions of the present disclosure according to the prompt message.

[0026] As an optional but non-limiting implementation, in response to receiving an active request from a user, the way to send a prompt message to the user can be, for example, in the form of a pop-up window, and the prompt message can be presented in text in the pop-up window. In addition, the pop-up window can also carry selection controls for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0027] It can be understood that the above notification and user authorization acquisition process is only illustrative and does not limit the implementation of the present disclosure. Other methods that comply with relevant laws and regulations can also be applied to the implementation of the present disclosure.

[0028] Meanwhile, it can be understood that the data involved in the present technical solution (including but not limited to the data itself, the acquisition or use of the data) should comply with the requirements of corresponding laws, regulations and related provisions.

[0029] Figure 1 is a flowchart of an information processing method based on a large language model shown according to an exemplary embodiment. As Figure 1 shown, the embodiments of the present disclosure provide an information processing method based on a large language model. Specifically, this method can be executed by an information processing device based on a large language model, and this device can be implemented in a software and / or hardware manner. As Figure 1 shown, this method can include the following steps.

[0030] In step 110, a search control is displayed on a first page, and the search control includes a first sub-control for selecting a target object to be analyzed.

[0031] Here, the first page can refer to a page for analyzing an object. For example, the first page can be a page providing analysis tools. In the implementation of the present disclosure, the object can refer to a user to be analyzed. A search control is displayed on the first page, and this search control at least includes a first sub-control for selecting a target object to be analyzed.

[0032] Exemplarily, the first sub-control can be an input box displayed on the first page, and the user can input an object identifier corresponding to the target object to be analyzed (such as Universally Unique Identifier, UUID, Universal Unique Identification Code) on the first sub-control to indicate the target object to be analyzed.

[0033] Figure 2 is a schematic diagram of the first page shown according to an exemplary embodiment. As Figure 2 shown, a search control 201 can be displayed on the first page 201, and this search control 201 includes a first sub-control 202.

[0034] In step 120, in response to the analysis operation of the target object indicated for the first sub-control, the analysis result obtained by the large language model based on the event stream of the target object is displayed on the first page. The analysis result at least includes interaction summary information for describing the interaction actions and interaction requirements for the existence of the target object.

[0035] Here, the analysis operation of the target object indicated for the first sub-control can be triggered by triggering the search button on the search control. As Figure 2 shown, the search control 201 includes a search button 203. When the user clicks the search button 203, the analysis operation of the target object indicated for the first sub-control 202 is triggered.

[0036] The event stream of the target object refers to a series of events that occur to the target object within a specific time period, that is, the interaction actions of the target object shown in the order of events occurring on the client side. In the implementation manner of the present disclosure, the large language model (LLM) analyzes the event stream of the target object to obtain the analysis result, and the analysis result obtained by the large language model is displayed on the first page.

[0037] It should be understood that the large language model can analyze the event stream of the target object through pre-trained logic to obtain the analysis result corresponding to the target object.

[0038] It should be noted that the large language model can be deployed locally on the electronic device or deployed in the server. That is to say, the electronic device can receive the analysis result sent by the server and display the analysis result on the first page.

[0039] Among them, the analysis result can at least include interaction summary information for describing the interaction actions and interaction requirements for the existence of the target object. The interaction action can refer to the interaction performed by the target object in the historical time, such as receiving a coupon, clicking on the XX live broadcast room, etc. The interaction requirement refers to the requirement implied by a series of interaction actions of the target object obtained through the event stream. For example, the interaction requirement can refer to finding low-price goods, etc.

[0040] By analyzing the event stream of the target object, the large language model obtains the interaction actions that the target object has in the historical time and the interaction requirements implied by the interaction actions. For example, the interaction summary information can be "the target object has searched for the live broadcast room related to XX coupon collection many times within the last day and received multiple coupons, and there may be a shopping price comparison". Among them, "searching for the live broadcast room related to XX coupon collection many times within the last day and receiving multiple coupons" represents the interaction action of the target object, and "shopping price comparison" is the interaction requirement.

[0041] Of course, the analysis results can also include object information of the target object, object characteristics of the target object, and so on. Among them, the object information can refer to the basic information of the target object, such as device information, UUID, registration time, and other information. The object characteristics can refer to the labels of the target object, and the object characteristics can be obtained by a large language model. In the first page, in addition to displaying the interaction summary information, the object information and object characteristics can also be displayed in the first page.

[0042] As Figure 2 shown, in the first page 200, the interaction summary information 204 can be displayed through the module of the interaction flow, and the object information and object characteristics can be displayed in the first page 200.

[0043] It is worth noting that by analyzing the event stream of the target object through a large language model, the entire event stream of the target object can be analyzed, greatly improving the efficiency of object analysis, and the obtained analysis results are more accurate.

[0044] Thus, by displaying a search control including a first sub-control for selecting a target object to be analyzed on the first page, and then in response to an analysis operation on the target object indicated by the first sub-control, displaying on the first page the analysis results obtained by the large language model based on the event stream of the target object, the analysis results at least including interaction summary information for describing the interaction actions and interaction requirements of the target object, by using the large language model to analyze the event stream, not only can the analysis efficiency of the target object be improved, but also the analysis results can be made more accurate.

[0045] It is worth noting that by obtaining the analysis results through a large language model, advertisers can intuitively understand the interaction patterns of the target object at different stages, providing a reference basis for advertisers in subsequent marketing strategy adjustments. For example, through the analysis results obtained by the large language model, the interaction actions of the target object can be summarized, and the implicit interaction requirements in the interaction actions can be found, so as to optimize the subsequent marketing strategies to facilitate conversion. For example, for objects that often browse but do not generate purchase actions, exclusive coupons can be sent.

[0046] In some implementable embodiments, the search control further includes a second sub-control for selecting an analysis scenario. As Figure 2 shown, a second sub-control 205 is displayed on the search control 201, and the user can select an analysis scenario through the second sub-control 205. For example, by clicking on the second sub-control 205, one or more analysis scenarios available for selection can be displayed in the second sub-control 205, and then the user clicks on the corresponding analysis scenario to complete the selection of the analysis scenario.

[0047] It should be noted that the analysis scenario refers to a specific scenario. For example, the analysis scenario can refer to a low-price search scenario, an empty search scenario, etc. In different analysis scenarios, the large language model can use the analysis logic corresponding to the analysis scenario to analyze the event stream of the target object.

[0048] Correspondingly, in step 120, in response to the selection operation for the second sub-control, the target analysis scenario can be determined. In response to the analysis operation for the target object indicated by the first sub-control, the analysis result is displayed on the first page. The analysis result is obtained by the large language model analyzing the event stream of the target object based on the analysis logic corresponding to the target analysis scenario.

[0049] Here, the electronic device can determine the analysis scenario selected by the user on the second sub-control as the target analysis scenario in response to the selection operation for the second sub-control. It should be noted that one or more candidate analysis scenarios displayed on the second sub-control can be pre-configured. For example, for accounts in different business lines, different types of analysis scenarios can be configured. Of course, one or more candidate analysis scenarios displayed on the second sub-control can also be saved by the user himself. For example, after the user uses the analysis ability of the large language model to complete the analysis of a target object in a specific analysis scenario, the user can save the analysis logic of the large language model so that in the subsequent process, for the same specific analysis scenario, a consistent analysis logic can be used to perform object analysis.

[0050] It should be understood that after the user selects the target analysis scenario, the target analysis scenario selected by the user can be displayed in the second sub-control.

[0051] Different analysis scenarios can correspond to different analysis logics. For example, the "low-price search" analysis scenario corresponds to an analysis logic of the large language model, and the "empty search" analysis scenario corresponds to an analysis logic of the large language model. The analysis logic corresponding to the analysis scenario can refer to the event range corresponding to the large language model in this analysis scenario and the analysis method used to analyze the event stream.

[0052] After determining the target analysis scenario, the large language model uses the analysis logic corresponding to the target analysis scenario to analyze the event stream of the target object, obtains the analysis result of the target object, and displays the analysis result obtained by the large language model on the first page.

[0053] Thus, through the above implementation method, the user can select different analysis scenarios and perform accurate analysis on the event stream in different analysis scenarios to meet the needs of different scenarios, greatly improving the analysis efficiency.

[0054] In some implementable embodiments, the search control may further include a third sub-control for receiving a natural language description. Among them, the third sub-control may be a text input box for inputting natural language.

[0055] As Figure 2 shown, a third sub-control 206 is displayed on the search control 201, and the user can input a natural language description through the third sub-control 206, so that the large language model analyzes the event stream of the target object according to the input natural language description.

[0056] It should be noted that the natural language description may refer to the natural text input by the user. For example, the user can input a natural language description of "why there is no conversion" in the third sub-control. By inputting the natural language description on the third sub-control, the large language model can perform targeted analysis.

[0057] Correspondingly, in step 120, in response to an input operation on the third sub-control, the target language description can be determined, and then in response to an analysis operation on the target object indicated by the first sub-control, the analysis result is displayed on the first page, and the analysis result is obtained by the large language model based on the target language description and analyzing the event stream of the target object.

[0058] Here, the electronic device can determine the text input in the third sub-control as the target language description in response to an input operation on the third sub-control. It should be understood that the target language description can actually be understood as the user defining the analysis scope and expected analysis logic of the large language model.

[0059] After determining the target language description, the large language model analyzes the event stream of the target object according to the target language description input by the third sub-control, obtains the analysis result of the target object, and displays the analysis result obtained by the large language model on the first page.

[0060] Thus, through the above embodiments, the user can filter the event stream through natural language to improve the analysis efficiency of the target object.

[0061] In some implementable embodiments, the search control may further include a fourth sub-control for configuring filtering parameters. As Figure 2 shown, a fourth sub-control 207 is displayed on the search control 201, and the user can configure parameters through the fourth sub-control 207.

[0062] It should be noted that the filtering parameters may refer to the parameters configured for the target object, and the filtering parameters are used to define the analysis scope of the large language model. For example, the filtering parameters may refer to event types, business buying points, event filtering parameters, object filtering parameters, and so on.

[0063] Correspondingly, in step 120, in response to a configuration operation on the fourth sub-control, the target screening parameter can be determined, and then in response to an analysis operation on the target object indicated by the first sub-control, the analysis result is displayed on the first page. The analysis result is obtained by the large language model analyzing the event stream of the target object based on the target screening parameter.

[0064] Here, the electronic device can determine the parameter configured in the fourth sub-control as the target screening parameter in response to a configuration operation on the fourth sub-control. It should be understood that the target screening parameter can actually be understood as the scope of the event stream that the user defines for the large language model to analyze, in order to analyze specific events.

[0065] After determining the target screening parameter, the large language model analyzes the event stream of the target object according to the target screening parameter configured in the fourth sub-control, obtains the analysis result of the target object, and displays the analysis result obtained by the large language model on the first page.

[0066] Figure 3 is a schematic diagram of the fourth sub-control shown according to an exemplary embodiment. As Figure 3 shown, when the user clicks on the fourth sub-control 207, at least one configuration option 2071 can be displayed. For example, the configuration option 2071 can include an event screening option for defining events, an object screening option for defining objects, and an option for defining forward / backward queries. The user can configure the corresponding target screening parameter through the displayed configuration option 2071.

[0067] Of course, in other embodiments, the target screening parameter configured by the user can be saved as a template, and during subsequent analysis processes, the user can select any saved template to directly configure the target screening parameter.

[0068] Thus, through the above-described embodiments, the user can screen the event stream by configuring the target screening parameter to improve the analysis efficiency of the target object and perform personalized object analysis.

[0069] Continuing with the above-described embodiments, the analysis result can be displayed on the first page. The analysis result is obtained by the large language model analyzing the event stream of the target object based on the analysis logic corresponding to the target analysis scenario and / or the target language description and / or the target screening parameter.

[0070] That is to say, the user can select the corresponding target analysis scenario through the second sub-control, input the corresponding target language description through the third sub-control, and configure the corresponding target screening parameter through the fourth sub-control, so that the large language model analyzes the event stream of the target object through the analysis logic and / or the target language description and / or the target screening parameter of the target analysis scenario.

[0071] It should be noted that whether it is the target analysis scenario, natural language description, or target screening parameters, they essentially define the analysis scope, analysis objectives, and analysis methods used by the large language model to obtain analysis results that are more in line with the target analysis scenario and target language description.

[0072] In some implementable embodiments, at least one scenario label and / or analysis control related to the interaction summary information may be displayed on the first page. As Figure 2 shown, the scenario label 208 and the analysis control 209 are displayed on the first page 200.

[0073] Here, the scenario label is used to respond to the trigger operation for the scenario label, display the second page for conversing with the large language model, and ask questions to the large language model in the second page based on the questions associated with the scenario label.

[0074] When the user clicks any one of the at least one displayed scenario labels, the second page for conversing with the large language model can be displayed, and questions associated with the scenario label are used to ask questions to the large language model in the second page, so that the large language model analyzes the target object based on the questions.

[0075] Among them, each scenario label corresponds to a question. When the user selects the corresponding scenario label, it means using the question corresponding to the scenario label to ask questions to the large language model. It should be understood that each scenario label actually corresponds to an analysis scenario. When the user selects the corresponding scenario label, it is equivalent to instructing the large language model to analyze the event stream of the target object through the analysis logic corresponding to the scenario label. It should be noted that the scenario labels can be recommended by the large language model.

[0076] The analysis control is used to respond to the trigger operation for the analysis control and display the second page for conversing with the large language model.

[0077] Among them, the trigger operation for the analysis control may refer to the click operation on the analysis control. When the user clicks the analysis control, the second page for conversing with the large language model can be displayed, so that the user can optimize the analysis of the target object by conversing with the large language model in the second page.

[0078] Thus, by displaying the scenario labels and / or analysis controls, the user can perform a deeper analysis of the event stream of the target object.

[0079] In some implementable embodiments, in response to a question sent on a second page, interaction features matching the analysis scenario determined by the large language model for the question are displayed on the second page, and then in response to determining the interaction features, target events matching the interaction features found by the large language model in the event stream are displayed on the second page.

[0080] Here, when the user sends a question to the large language model on the second page, interaction features matching the analysis scenario determined by the large language model for the question are displayed on the second page. The large language model determines the analysis scenario required by the user through the question sent by the user on the second page, and then the large language model determines the analysis intention in this analysis scenario and determines the corresponding interaction features through the analysis intention. For example, the analysis intention is "find low price", and the interaction features of this analysis intention are "price search" and "price reverse sorting".

[0081] Determining the interaction features may mean that the user does not modify the displayed interaction features within a preset time interval, or the user no longer asks questions to the large language model.

[0082] After determining the displayed interaction features, the large language model finds target events matching the interaction features in the event stream based on the interaction features and displays the found target events on the second page.

[0083] It should be noted that the large language model can determine the buried point definition corresponding to the interaction feature based on the corresponding interaction feature, and then find the target event matching the buried point definition in the event stream through the buried point definition. It should be noted that each event in the event stream can be associated with a corresponding buried point definition, and the corresponding target event can be determined through the determined buried point definition.

[0084] For example, the buried point definition corresponding to the interaction feature of "price search" is "search_price", and the buried point definition corresponding to the interaction feature of "price reverse sorting" is "Filter". Then, target events are found in the event stream through the buried point definition of "search_price", and target events are found in the event stream through the buried point definition of "Filter".

[0085] Figure 4 is a schematic diagram of a second page shown according to an exemplary embodiment. As Figure 4 shown, when the user clicks Figure 2 the scenario label 208 (such as the scenario label of "find low price") in the first page 200 shown, the Figure 4 shown second page 400 is displayed, and the question 401 corresponding to the scenario label 208 is displayed on the second page 400.

[0086] The large language model responds with problem 401, and on the second page 200, the interaction features 402 that match the analysis scenario corresponding to problem 401 determined by the large language model are displayed. When the user determines the displayed interaction features 402, on the second page 401, the target events 403 that match the interaction features 402 found by the large language model in the event stream are displayed.

[0087] Figure 5 is a schematic diagram of the second page shown according to another exemplary embodiment. As Figure 5 shown, when the user clicks on the analysis control 209 in the first page 200 shown Figure 2 shown, the second page 400 shown Figure 5 is displayed. The user sends problem 401 to the large language model on the second page 400 to call the large language model to analyze the event stream through the problem 401 sent by the user.

[0088] It should be noted that when the problem 401 sent by the user is inaccurate, a modification control 408 can be displayed to prompt the user to modify the sent problem 401 to indicate that the large language model rethinks to obtain an accurate analysis result.

[0089] Thus, through the above implementation, the user can have a conversation with the large language model to instruct the large language model to perform a detailed analysis of the event stream of the target object to obtain an accurate analysis result.

[0090] In some implementable embodiments, in response to the operation of saving the analysis logic of the large language model, an association relationship can be established between the analysis scenario corresponding to the problem and the analysis logic of the target event corresponding to the analysis scenario obtained by the large language model for the problem, and the association relationship is saved.

[0091] Here, the operation of saving the analysis logic of the large language model can be triggered by a save control displayed on the second page. For example, the user can trigger the operation of saving the analysis logic of the large language model by clicking on the save control displayed on the second page.

[0092] As Figure 4 shown, when the large language model finishes outputting the target event, a save control 404 can be displayed on the second page 400, and the user triggers the operation of saving the analysis logic of the large language model by clicking on the save control 404.

[0093] Figure 6 is a schematic diagram of saving the association relationship according to an exemplary embodiment. As Figure 6 shown, when the user saves the control 404, it is displayed on the second page 400 Figure 6For the pop-up window 601 shown, the user can fill in the analysis logic name (corresponding to the analysis scenario) and the analysis logic description through this pop-up window 601. When the user clicks the confirmation button in the pop-up window 601, the association relationship between the analysis scenario corresponding to the problem and the analysis logic of the target event corresponding to the analysis scenario obtained by the large language model is saved.

[0094] Saving the analysis logic of the large language model means saving the thinking process of the large language model for the target event corresponding to the problem sent by the user.

[0095] In response to the operation of saving the analysis logic of the large language model, the electronic device establishes the association relationship between the analysis scenario corresponding to the problem and the analysis logic of the target event corresponding to the analysis scenario obtained by the large language model, and saves this association relationship.

[0096] It should be noted that by saving the above association relationship, when the user encounters an analysis scenario identical to the above association relationship, the user can directly select the analysis scenario to directly instruct the large language model to call the corresponding analysis logic to analyze the event flow of the target object.

[0097] It should be understood that the association relationship saved through the above implementation manner can be presented in Figure 2 the second sub-control 205 shown, so that the user can quickly select the analysis logic of the large language model.

[0098] Thus, by maintaining the association relationship between the analysis scenario corresponding to the problem and the analysis logic of the target event corresponding to the analysis scenario obtained by the large language model, the user can quickly call the analysis logic corresponding to the analysis scenario, greatly improving the analysis efficiency.

[0099] In some implementable embodiments, the second page further includes a calibration control for adjusting interaction features and / or a positioning control for positioning the target event. As Figure 4 shown, a calibration control 405 and a positioning control 404 can be displayed on the second page 400.

[0100] Correspondingly, the interaction features can be adjusted in response to the trigger operation for the calibration control; and / or, the target event can be located in the event flow in response to the trigger operation for the positioning control Here, when the user believes that the interaction features determined by the large language model are inaccurate, the user can click the calibration control to manually calibrate the interaction features determined by the large language model.

[0101] It should be understood that adjusting the interaction features can be that the user directly fills in the corresponding interaction features, or that the user sends a new question to instruct the large language model to adjust the determined interaction features.

[0102] When the target event is displayed on the second page, if the user needs to directly view the found target event, the user can click on the positioning control to locate the target event in the event stream. It should be noted that locating the target event in the event stream can be to directly highlight the corresponding target event in the event stream. For example, when the user needs to view the events before and after the target event, the user can click on the positioning control to directly locate the target event in the event stream.

[0103] It should be noted that the event stream of the target object can be displayed on the second page. When the user clicks on the positioning control, the target event can be directly highlighted and located in the displayed event stream.

[0104] Thus, through the above calibration control and / or positioning control, the user can manually calibrate the interaction features to ensure the accuracy of the analysis results, and the user can also directly locate the target event in the event stream through the positioning control.

[0105] In some implementable embodiments, the search process of the large language model for finding the target event can also be displayed on the second page, and when each sub-process in the search process is completed, the sub-process is highlighted.

[0106] Here, the search process of the large language model for finding the target event can refer to the steps of the large language model processing the user input and identifying the target event. For example, the search process of the large language model for finding the target event can include four sub-processes: "detailed description of the target analysis scenario", "search for interaction features that meet the intention", "analysis of the buried point definition corresponding to the interaction", and "result output".

[0107] When each sub-process in the search process is completed, the sub-process is highlighted. Among them, highlighting the completed sub-process can be to use highlighting, bold, color, or animation effects to highlight the completed sub-process.

[0108] It is worth noting that the sub-processes included in the search process can include sub-processes that need to be completed by the user and sub-processes that need to be completed by the large language model. For example, the sub-process of "detailed description of the target analysis scenario" is a sub-process that needs to be completed by the user, and the sub-processes of "search for interaction features that meet the intention", "analysis of the buried point definition corresponding to the interaction", and "result output" are sub-processes that need to be completed by the large language model.

[0109] That is to say, after the user details the target analysis scenario, the sub-process of "detailed description of the target analysis scenario" will be highlighted to indicate that the sub-process of "detailed description of the target analysis scenario" has been completed.

[0110] Thus, by displaying the search process of the large language model to find the target event, users can intuitively understand the steps that have been carried out, enabling them to instruct the large language model to search for the target event according to the prompts of the search process, greatly improving the analysis efficiency.

[0111] In some realizable embodiments, the first page further includes a filtering control for filtering events. The filtering control is used for users to filter specific events from the event stream through the filtering control. The filtering control can support options such as event attributes, clients to which the events belong, event highlighting, attribute highlighting, etc. for users to filter specific events.

[0112] Such as Figure 2 As shown, the filtering control 210 can be displayed on the first page 200, and users can filter events through the parameters provided by the filtering control 210 so that they can locate the events they need to view.

[0113] Correspondingly, the target event found by the large language model can also be filled back to the filtering control to display the target event through the filtering control.

[0114] Here, after the large language model determines the target event, the target event found by the large language model can be filled back to the filtering control to display the target event found by the large language model in the filtering control.

[0115] It is worth noting that by filling back the target event found by the large language model to the filtering control, the target event can be directly displayed on the first page, facilitating users to directly view the target event and greatly improving the analysis efficiency.

[0116] It should be understood that the target event found by the large language model can locate the key events that have not been converted, enabling advertisers to intuitively understand the whole process of the target object from generating a purchase plan to finally not being converted based on the events before and after the target event, thus facilitating advertisers to adjust subsequent marketing strategies.

[0117] Figure 7 is a schematic structural diagram of an information processing device based on a large language model shown according to an exemplary embodiment. Such as Figure 7 As shown, the embodiment of the present disclosure provides an information processing device 700 based on a large language model. The information processing device 700 based on a large language model includes: A first display module 701, configured to display a search control on a first page, where the search control includes a first sub-control for selecting a target object to be analyzed; The second display module 702 is configured to, in response to an analysis operation on a target object indicated by the first sub-control, display, on the first page, an analysis result obtained by the large language model based on the event stream of the target object, where the analysis result at least includes interaction summary information for describing interaction actions and interaction requirements for the existence of the target object.

[0118] Optionally, the search control further includes a second sub-control for selecting an analysis scenario; the second display module 702 includes: A first determination unit configured to determine a target analysis scenario in response to a selection operation on the second sub-control; A first result display unit configured to, in response to an analysis operation on a target object indicated by the first sub-control, display, on the first page, the analysis result, where the analysis result is obtained by the large language model analyzing the event stream of the target object based on the analysis logic corresponding to the target analysis scenario.

[0119] Optionally, the search control further includes a third sub-control for receiving a natural language description; the second display module 702 includes: A second determination unit configured to determine a target language description in response to an input operation on the third sub-control; A second result display unit configured to, in response to an analysis operation on a target object indicated by the first sub-control, display, on the first page, the analysis result, where the analysis result is obtained by the large language model analyzing the event stream of the target object based on the target language description.

[0120] Optionally, the search control further includes a fourth sub-control for configuring filtering parameters; the second display module 702 includes: A third determination unit configured to determine target filtering parameters in response to a configuration operation on the fourth sub-control; A third result display unit configured to, in response to an analysis operation on a target object indicated by the first sub-control, display, on the first page, the analysis result, where the analysis result is obtained by the large language model analyzing the event stream of the target object based on the target filtering parameters.

[0121] Optionally, the apparatus 700 further includes: A second display module, configured to display at least one scenario tag and / or analysis control related to the interaction summary information on the first page; wherein, the scenario tag is used to respond to a trigger operation on the scenario tag, display a second page for conversing with the large language model, and ask the large language model questions based on the questions associated with the scenario tag on the second page; the analysis control is used to respond to a trigger operation on the analysis control, and display a second page for conversing with the large language model.

[0122] Optionally, the apparatus 700 further includes: A third display module, configured to display, on the second page, interaction features matching the analysis scenario corresponding to the question determined by the large language model in response to a question sent on the second page; A fourth display module, configured to display, on the second page, a target event found by the large language model in the event stream and matching the interaction feature in response to determining the interaction feature.

[0123] Optionally, the apparatus 700 further includes: A saving module, configured to establish an association relationship between the analysis scenario corresponding to the question and the analysis logic of the target event corresponding to the analysis scenario obtained by the large language model for the question, and save the association relationship in response to an operation of saving the analysis logic of the large language model.

[0124] Optionally, the second page further includes a calibration control for adjusting the interaction feature and / or a positioning control for positioning the target event; the apparatus 700 further includes: An adjustment display module, configured to adjust the interaction feature in response to a trigger operation on the calibration control; and / or, position the target event in the event stream in response to a trigger operation on the positioning control.

[0125] Optionally, the apparatus 700 further includes: A fifth display module, configured to display the search process of the large language model for finding the target event on the second page, and highlight each sub-process in the search process when each sub-process is completed.

[0126] Optionally, the first page further includes a filtering control for filtering events; the apparatus 700 further includes: A backfilling module, configured to backfill the target event found by the large language model to the filtering control, so as to display the target event through the filtering control.

[0127] Regarding the information processing apparatus 700 based on a large language model in the above embodiments, the method logics executed by each functional module have been described in detail in the section regarding the method, and will not be elaborated here.

[0128] Next, refer to Figure 8 , which shows a schematic structural diagram of an electronic device (such as a terminal device) 800 suitable for implementing the embodiments of the present disclosure. The terminal device in the embodiments of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 8 The electronic device shown is merely an example and should not impose any limitations on the functions and usage scope of the embodiments of the present disclosure.

[0129] As Figure 8 shown, the electronic device 800 may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 801, which may perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage device 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the electronic device 800 are also stored. The processing device 801, the ROM 802, and the RAM 803 are connected to each other through a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0130] Generally, the following devices may be connected to the I / O interface 805: an input device 806 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 807 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 808 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 809. The communication device 809 may allow the electronic device 800 to communicate with other devices wirelessly or wirelesly to exchange data. Although Figure 8 shows the electronic device 800 having various devices, it should be understood that it is not required to implement or have all the shown devices. More or fewer devices may be alternatively implemented or had.

[0131] In particular, according to embodiments of the present disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product that includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes program code for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device 809, or installed from a storage device 808, or installed from a ROM 802. When the computer program is executed by a processing device 801, the above-described functions defined in the methods of the embodiments of the present disclosure are performed.

[0132] It should be noted that the above-mentioned computer-readable medium in the present disclosure can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or in combination with an instruction execution system, apparatus, or device. In the present disclosure, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, in which computer-readable program code is carried. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, and the computer-readable signal medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted by any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0133] In some embodiments, the client and the server can communicate using any currently known or future-developed network protocol such as HTTP (HyperText Transfer Protocol), and can be interconnected with digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include local area networks ("LANs"), wide area networks ("WANs"), the Internet (e.g., the Internet), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0134] The above computer-readable medium can be included in the above electronic device; or it can exist separately without being assembled into the electronic device.

[0135] The above computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to: display a search control on a first page, the search control including a first sub-control for selecting a target object to be analyzed; in response to an analysis operation on the target object indicated by the first sub-control, display, on the first page, an analysis result obtained by the large language model based on the event stream of the target object, the analysis result at least including interaction summary information for describing the interaction actions and interaction requirements existing for the target object.

[0136] Computer program code for performing the operations of the present disclosure can be written in one or more programming languages or combinations thereof. The programming languages include, but are not limited to, object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., by using an Internet service provider to connect through the Internet).

[0137] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a portion of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks may occur in a different order than noted in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or by a combination of dedicated hardware and computer instructions.

[0138] The modules described in the embodiments of the present disclosure can be implemented in software or in hardware. In some cases, the name of the module does not constitute a limitation on the module itself.

[0139] The functions described above herein can be performed, at least in part, by one or more hardware logic components. By way of example, and without limitation, the types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system on a chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0140] In the context of the present disclosure, a machine-readable medium may be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, 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 a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0141] The above description is only a preferred embodiment of the present disclosure and an illustration of the applied technical principles. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosure concept. For example, the technical solutions formed by mutually replacing the above features with the technical features (but not limited to) having similar functions disclosed in the present disclosure.

[0142] In addition, although the operations are depicted in a particular order, this should not be construed as requiring that the operations be performed in the particular order shown or in sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the present disclosure. Certain features described in the context of separate embodiments may also be implemented combinatorially in a single embodiment. Conversely, the various features described in the context of a single embodiment may also be implemented separately or in any suitable sub-combination in multiple embodiments.

[0143] Although the subject matter has been described in language specific to structural features and / or methodological logical acts, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. On the contrary, the specific features and acts described above are merely example forms for implementing the claims. Regarding the apparatus in the above embodiments, the specific manner in which each module performs operations has been described in detail in the embodiments related to the method, and will not be elaborated here.

Claims

1. An information processing method based on a large language model, characterized in that, Including: Display a search control on the first page, where the search control includes a first sub-control for selecting a target object to be analyzed; In response to an analysis operation on the target object indicated by the first sub-control, display on the first page the analysis result obtained by the large language model based on the event stream of the target object, where the analysis result at least includes interaction summary information for describing the interaction actions and interaction requirements existing for the target object.

2. The method according to claim 1, wherein The search control further includes a second sub-control for selecting an analysis scenario; The step of, in response to an analysis operation on the target object indicated by the first sub-control, displaying on the first page the analysis result obtained by the large language model based on the event stream of the target object includes: In response to a selection operation on the second sub-control, determine a target analysis scenario; In response to an analysis operation on the target object indicated by the first sub-control, display on the first page the analysis result, where the analysis result is obtained by the large language model analyzing the event stream of the target object based on the analysis logic corresponding to the target analysis scenario.

3. The method according to claim 1, wherein The search control further includes a third sub-control for receiving a natural language description; The step of, in response to an analysis operation on the target object indicated by the first sub-control, displaying on the first page the analysis result obtained by the large language model based on the event stream of the target object includes: In response to an input operation on the third sub-control, determine a target language description; In response to an analysis operation on the target object indicated by the first sub-control, display on the first page the analysis result, where the analysis result is obtained by the large language model analyzing the event stream of the target object based on the target language description.

4. The method according to claim 1, wherein The search control further includes a fourth sub-control for configuring filtering parameters; The step of, in response to an analysis operation on the target object indicated by the first sub-control, displaying on the first page the analysis result obtained by the large language model based on the event stream of the target object includes: In response to a configuration operation on the fourth sub-control, determine target filtering parameters; In response to an analysis operation on the target object indicated by the first sub-control, display on the first page the analysis result, where the analysis result is obtained by the large language model analyzing the event stream of the target object based on the target filtering parameters.

5. The method according to any one of claims 1 to 4, characterized in that The method further includes: Display at least one scenario label and / or analysis control related to the interaction summary information on the first page; where the scenario label is used to, in response to a trigger operation on the scenario label, display a second page for communicating with the large language model, and ask questions to the large language model on the second page based on the questions associated with the scenario label; the analysis control is used to, in response to a trigger operation on the analysis control, display a second page for communicating with the large language model.

6. The method according to claim 5, wherein The method further includes: In response to a question sent on the second page, display on the second page the interaction features determined by the large language model that match the analysis scenario corresponding to the question. In response to determining the interaction feature, display, on the second page, the target event found by the large language model in the event stream that matches the interaction feature.

7. The method according to claim 6, wherein The method further includes: In response to an operation of saving the analysis logic of the large language model, establish an association relationship between the analysis scenario corresponding to the problem and the analysis logic of the target event corresponding to the analysis scenario obtained by the large language model for the problem, and save the association relationship.

8. The method according to claim 5, wherein The second page further includes a calibration control for adjusting the interaction feature and / or a positioning control for positioning the target event; the method further includes: In response to a trigger operation on the calibration control, adjust the interaction feature; and / or, in response to a trigger operation on the positioning control, locate the target event in the event stream.

9. The method according to claim 6, wherein The method further includes: Display, on the second page, the search process of the large language model for the target event, and highlight each sub-process in the search process when the sub-process is completed.

10. The method according to claim 6, wherein The first page further includes a filtering control for filtering events; the method further includes: Backfill the target event found by the large language model to the filtering control to display the target event through the filtering control.

11. A computer-readable medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processing device, it implements the steps of the method according to any one of claims 1-10.

12. An electronic device, characterized in that, It includes: A storage device on which a computer program is stored; A processing device for executing the computer program in the storage device to implement the steps of the method according to any one of claims 1-10.

13. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-10.

Images