Information processing method, apparatus and electronic device

Abstract

The present disclosure provides an information processing method, device and electronic equipment, relating to the technical fields of artificial intelligence, large model, natural language processing and the like. The method comprises: in response to a travel planning request of a target user, obtaining target preference information of the target user in a travel scenario; wherein the target preference information is obtained based on first operation data of the target user in at least one first application; generating structured travel plan information based on the target information; wherein the target information at least comprises the travel planning request and the target preference information; generating a first reply content and a first travel route based on the travel plan information; and outputting the first reply content and the first travel route to a user terminal of the target user.

Description

Technical Field

[0001] This disclosure relates to the technical fields of artificial intelligence, large models, and natural language processing, and particularly to the technical fields of intelligent travel planning and human-computer interaction, and especially to an information processing method, device, and electronic device. Background Technology

[0002] With the development of artificial intelligence technology, intelligent travel planning has been widely used. However, related intelligent travel planning solutions typically rely solely on single planning requests and popular attraction lists, failing to accurately capture users' travel preferences. Furthermore, these solutions directly output text guides and map routes, leading to inconsistencies between the text guides and other data, such as map routes. Summary of the Invention

[0003] This disclosure provides an information processing method, apparatus, and electronic device.

[0004] According to one aspect of this disclosure, an information processing method is provided, the method comprising: responding to a travel planning request from a target user, acquiring target preference information of the target user in a travel scenario; wherein the target preference information is obtained based on first operation data of the target user in at least one first application; generating structured itinerary plan information based on the target information; wherein the target information includes at least the travel planning request and the target preference information; generating a first response content and a first itinerary route based on the itinerary plan information; and outputting the first response content and the first itinerary route to the target user's user terminal.

[0005] According to another aspect of this disclosure, an information processing apparatus is provided, the apparatus comprising: a first acquisition module, configured to acquire target user's target preference information in a travel scenario in response to a target user's travel planning request; wherein the target preference information is obtained based on first operation data of the target user in at least one first application; a first generation module, configured to generate structured itinerary plan information based on the target information; wherein the target information includes at least the travel planning request and the target preference information; a second generation module, configured to generate a first response content and a first itinerary route based on the itinerary plan information; and an output module, configured to output the first response content and the first itinerary route to the target user's user terminal.

[0006] According to another aspect of this disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the information processing method proposed above in this disclosure.

[0007] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided that stores computer instructions for causing the computer to perform the information processing method described above.

[0008] According to another aspect of this disclosure, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the information processing method proposed above in this disclosure.

[0009] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0010] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0011] Figure 1 A flowchart illustrating an information processing method provided in an embodiment of this disclosure; Figure 2 This is a schematic diagram of a display window provided for an embodiment of the present disclosure; Figure 3 A flowchart illustrating another information processing method provided in an embodiment of this disclosure; Figure 4 A flowchart illustrating an information processing method provided in an embodiment of this disclosure; Figure 5 This is a schematic diagram of the structure of an information processing device provided in an embodiment of the present disclosure; Figure 6 A schematic block diagram of an example electronic device that can be used to implement embodiments of the present disclosure is shown. Detailed Implementation

[0012] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0013] Among related technologies, intelligent travel planning solutions have the following problems: First, existing travel planning solutions typically make recommendations based on static guide content, popular routes, or a single user input. They lack a technical mechanism to uniformly convert heterogeneous logs such as map point of interest (POI) queries, POI displays, navigation, route calculation, positioning, stay times, search history, and click history into calculable operational events and further build user profiles. This results in user preference profiles being noisy and unstable, making it difficult to form user profiles that are continuously updated, interpretable, and can directly serve online planning.

[0014] Second, the existing travel planning process is usually scattered across multiple stages such as searching, browsing guides, viewing maps, route planning, and navigation execution. There is a lack of a unified intermediate representation between text guide results, map route results, and subsequent interactive update results, making it difficult to maintain consistency in the spatial relationship between text content and map routes, and also making it difficult to support route rendering, replacement of candidates, and subsequent local updates.

[0015] Third, when users add new planning constraints, such as "walk less", "add nighttime activities", "replace attractions", or "switch to family / pet-friendly routes", the existing system usually updates the planning results by regenerating the entire route, which results in high computational resource consumption, poor result stability, and insufficient interactive continuity.

[0016] Fourth, existing map interactions are usually only used for display or navigation. Interactive operations such as map clicking, area zooming, route segment viewing, and mode of transportation switching are usually not further transformed into planning constraints, thus failing to form a technical closed loop of "map interaction can provide feedback, feedback can lead to replanning, and replanning can be synchronously refreshed".

[0017] Fifth, existing travel question-answering solutions based on large models typically output natural language text and lack a structured itinerary planning information middle layer for map rendering, candidate replacement, constraint management, and synchronous updates. They also lack mechanisms for handling profile confidence, insufficient evidence fields, and conflict constraints, resulting in output results that, while readable, lack computability, executability, and updability.

[0018] To address the aforementioned issues, this disclosure provides information processing methods, apparatus, and electronic devices. By utilizing user operation data from at least one application, it can comprehensively capture users' travel preferences and habits across applications, constructing highly accurate travel preference information and thus precisely characterizing users' travel preferences. By generating structured itinerary planning information, it enables itinerary data to possess computable, parsable, and locally updatable characteristics. By using structured itinerary planning information as a unified intermediate representation, it achieves precise mapping between "textual descriptions" and "spatial routes," ensuring consistency between the first response content and the first itinerary route, effectively solving the problem of information fragmentation in traditional planning. By simultaneously distributing the first response content and the first itinerary route to the user's terminal, users no longer need to repeatedly switch between multiple applications or interfaces for comparison, improving the intuitiveness of the travel planning scheme display and the user's viewing experience.

[0019] The information processing methods, apparatus, and electronic devices of this disclosure are described below with reference to the accompanying drawings. Furthermore, the acquisition, storage, and application of any type of information, such as user personal information, involved in the technical solutions of this disclosure comply with relevant laws and regulations and do not violate public order and good morals.

[0020] Figure 1 This is a flowchart illustrating an information processing method provided in an embodiment of the present disclosure.

[0021] like Figure 1 As shown, this information processing method may include the following steps: Step S101: In response to the travel planning request of the target user, obtain the target user's target preference information in the travel scenario; wherein, the target preference information is obtained based on the target user's first operation data in at least one first application.

[0022] It should be noted that the executing entity of the information processing method in this embodiment can be a hardware device with image generation capabilities and / or the necessary software to drive the hardware device. Optionally, the executing entity may include a server, a user terminal, and other smart devices. Optionally, the user terminal includes, but is not limited to, mobile phones, computers, smart voice interaction devices, etc. Optionally, the server includes, but is not limited to, a network server, an application server, or a server of a distributed system, or a server combined with blockchain, etc. This embodiment does not impose specific limitations.

[0023] The travel planning request can be a request entered by the target user in a search box or dialog box within a second application. The second application refers to a search application or map application, which can include browser applications, content-based social applications, or other applications capable of content searching. It should be noted that the second application can be a web application or a native application.

[0024] The target preference information indicates the preferences of the target user in a travel scenario; these preferences can also be referred to as the user's travel preferences. For example, the target preference information may include the target user's travel preferences, route preferences, attraction preferences, food preferences, hotel preferences, planning and decision-making preferences, and special need preferences (such as preferences for family-friendly venues, pet-friendly venues, repeat visit preferences, niche exploration preferences, etc.). For example, each preference field may include: feature value, conclusion strength, basis, evidence list, and coverage time range.

[0025] It should be noted that the target user's target preference information can be pre-calculated during the offline phase, and then the target preference information can be directly invoked when a travel planning request from the target user is received.

[0026] The first application is a travel service application; that is, the first application refers to an application that can provide travel services for users' travel. For example, the first application may include search applications, map applications, ride-hailing applications, ticketing applications, accommodation booking applications, etc.

[0027] The first operational data refers to the interactive operation data and operation statistics of the target user within the first application. For example, the first operational data may include map POI query data, map POI display data, navigation data, route calculation data, location data, dwell trajectory data, search history data, click history data, user information feed (feed) historical interaction data, other content consumption data, as well as data such as replanning ratio, selection rate, and OD round trip frequency.

[0028] It should be noted that the first operational data can refer to the data obtained after processing the data collected in the first application. This data processing includes cleaning, parsing, unification and event-based processing, deduplication (e.g., deduplication and fusion based on time windows), and data completion. For example, data processing may include parsing address text and POI information, splitting navigation / routing records into origin and destination points, unifying time fields, mapping POI categories, completing city information, filtering duplicate records, removing abnormal data, and extracting retrieval events, display events, click events, origin-destination (OD) events, and stay events.

[0029] As an example, if the first operation data is data from a map application, then the first operation data may include: user unique identifier (user_id), event occurrence time (event_time), event type (event_type), city ID (city_id), point of interest identifier (poi_id), point of interest category (poi_category), data source type (source_type), origin point of interest (origin_poi), destination point of interest (dest_poi), dwell time (dwell_time), search keyword (query_term), click target (click_target), travel route mode (route_mode), data quality score (data_quality_score), etc.

[0030] It should be noted that after the first operation data is collected, for ease of data management, the first operation data can be converted into an operation event stream in chronological order. For example, the operation event stream may include: browsing POI events, navigation start events, attraction stay events, and search interest events.

[0031] Step S102: Generate structured itinerary plan information based on the target information; wherein the target information includes at least travel planning requests and target preference information.

[0032] The structured itinerary plan information serves as a unified intermediate representation, supporting the generation of response content and itinerary routes. For example, the structured itinerary plan information can be generated based on a large language model (also known as a large model), meaning that the target information is used as input to the large model to obtain the structured itinerary plan information output by the large model.

[0033] The target information may also include other information, such as the target user's basic preference information, historical conversation data, historical search data, and historical click data. The basic preference information is used to indicate the target user's identity.

[0034] As an example, if the travel planning request is entered by the target user in a dialog box, the target information may also include the target user's historical conversation data; if the travel planning request is entered by the target user in a search box, the target information may also include the target user's historical search data.

[0035] Structured itinerary planning information can include one or more itineraries. As an example, any itinerary in structured itinerary planning information can include the following fields: trip_id: Trip identifier; day_id: Trip day number; slot_id: The intraday time slot number; theme_tag: Theme tag; poi_seq: a sequence of tourist attractions; arrival_time and departure_time: arrival time and departure time; transport_mode: mode of transportation; route_segment_id: Route segment identifier; distance_cost and time_cost: Distance cost and time cost; constraint_tag_set: The set of constraint tags; alternative_poi_set: A set of alternative attractions; freeze_flag: The freeze flag during local replanning.

[0036] Step S103: Based on the itinerary plan information, generate the first response content and the first itinerary route.

[0037] The first response includes the itinerary information provided in the travel plan. For example, the first response can be generated based on the output format of the answers in a question-and-answer dialogue.

[0038] The first itinerary is a visualized map route, which includes visualized map routes for each itinerary. For example, when generating the first itinerary, the visualized map routes corresponding to different itineraries can be differentiated by layer, display format, or sequence number.

[0039] Step S104: Output the first reply content and the first trip route to the target user's user terminal.

[0040] The user terminal can simultaneously display the first reply content and the first trip route; for example, the first reply content and the first trip route can be displayed in the dialogue interface of a search application or map application on the user terminal.

[0041] As an example, the first reply is displayed in the first display window of the dialogue interface, and the first travel route is displayed in the second display window of the dialogue interface. The second display window shows a thumbnail of the first travel route; when the target user clicks on the second display window, they can enter the detailed display / interactive interface of the first travel route.

[0042] This disclosure does not limit the position, size, or movement of the first and second display windows. For example... Figure 2 As shown, Figure 2 This is a diagram illustrating the positions of the first and second display windows. For example, Figure 2 In (a), the position and size of the first display window are fixed, while the position and size of the second display window are not fixed. For example, the second display window is a floating window, that is, the first display window cannot be moved, while the second display window can be moved. Figure 2 In (b), the position and size of the first display window and the second display window are both fixed (i.e., the position and size of the first display window and the second display window cannot be adjusted), or the position and size of the first display window and the second display window are both not fixed (i.e., the position and size of the first display window and the second display window can be adjusted).

[0043] The information processing method provided in this disclosure can comprehensively capture users' travel preferences and habits across applications by using user operation data from at least one application, constructing highly accurate travel preference information, and thus accurately depicting users' travel preferences. By generating structured itinerary plan information, it can make the itinerary data have the characteristics of being computable, parsable, and locally updatable. By using structured itinerary plan information as a unified intermediate representation, it can achieve accurate mapping between "textual descriptions" and "spatial paths," improving the consistency between the first response content and the first itinerary route, and effectively solving the problem of information fragmentation in traditional planning. By simultaneously sending the first response content and the first itinerary route to the user terminal, the user does not need to repeatedly switch between multiple applications or interfaces for comparison, improving the intuitiveness of the travel planning scheme display and the user's viewing experience.

[0044] Figure 3 This is a flowchart illustrating another information processing method provided in an embodiment of the present disclosure.

[0045] like Figure 3 As shown, this information processing method may include the following steps: Step S301: In response to the target user's travel planning request, obtain multiple first operation data of the target user in multiple first applications.

[0046] The explanation of step S301 can be found in the relevant description in any embodiment of this disclosure, and will not be repeated here.

[0047] Step S302: Obtain the data reliability of the first operation data.

[0048] The data credibility of the first operational data is used to indicate the reliability and reference value of the first operational data in reflecting the target user's true travel preferences or intentions.

[0049] In some embodiments, a first weight is determined based on a first application associated with the first operation data; a second weight is determined based on the operation type of the first operation data; a quality coefficient of the first operation data is determined based on the data quality of the first operation data; the time interval between the data generation time of the first operation data and the current time is obtained; and data reliability is determined based on at least one of the first weight, the second weight, the quality coefficient, the time interval, and the time decay coefficient corresponding to the first operation data.

[0050] This involves pre-configuring different weights for the first application, and then determining the first weight based on the first application to which the first operation data belongs. Operations such as location, dwell, and actual navigation completion correspond to higher weights; routing operations correspond to medium weights; and operations such as retrieval, display, and clicks correspond to lower but not zero weights. For dimensions with insufficient evidence, their data credibility can be determined as "unknown" or "low confidence," and their weight will be automatically reduced in subsequent online planning.

[0051] This involves pre-configuring weights for different operation types, and then determining the second weight based on the operation type of the first operation data. For example, The quality coefficient of the first operation data can be determined based on the field completeness, positioning accuracy, noise reduction result, or deduplication result of the first operation data.

[0052] The time decay coefficients for the first operation data of different operation types can be different; that is, the time decay coefficient for the first operation data is determined according to the operation type of the first operation data. Alternatively, the time decay coefficients for the first operation data of different operation types can be the same; that is, all the first operation data correspond to the same time decay coefficient.

[0053] Among them, the first weight and data credibility are positively correlated, the second weight and data credibility are positively correlated, the quality coefficient and data credibility are positively correlated, the time interval and data credibility are negatively correlated, and the time decay coefficient and data credibility are negatively correlated.

[0054] For example, data confidence can be determined according to the following formula:

[0055] in, Indicates data credibility. Indicates the first weight. Indicates the second weight. Indicates the quality coefficient. Indicates the time decay coefficient. Indicates a time interval.

[0056] By comprehensively judging the credibility of operational data from multiple dimensions such as application source, operation type, data quality, time interval, and time decay coefficient, the system can effectively filter out distorted, outdated, and low-quality data, accurately identify valid operational data, and improve the accuracy of user travel preference calculation.

[0057] Step S303: For at least one travel preference feature, select the second operational data associated with the target preference feature from multiple first operational data.

[0058] At least one travel preference feature can be preset.

[0059] In some embodiments, the operation type of the operation data associated with each travel preference feature can be preset, that is, the mapping relationship between the travel preference feature and the operation type of the operation data can be preset, so as to determine the second operation data associated with each travel preference feature according to the mapping relationship.

[0060] In other embodiments, second operational data associated with various travel preference features can be determined using a large model.

[0061] As an example, at least one travel preference feature may include the following characteristics: Travel preference characteristics include preferences for long-distance travel modes, local travel modes, acceptance of early morning departures, and a tendency to travel at night. The second operational data associated with these travel preference characteristics can include operational data such as transportation hub POIs, intercity travel records, navigation methods, and departure time distribution.

[0062] Route preference characteristics include route sensitivity, route stability, daily pace, and walking tolerance. The second operational data associated with these route preference characteristics may include operational data such as the proportion of replanning, comparison rate, OD round-trip frequency, single-city / cross-city coverage, daily travel distance, and duration of outings.

[0063] Attraction preference characteristics include preferences for historical and cultural sites, natural landscapes, beaches, classic attractions, photo spots, and farmers' markets. The second operational data associated with these attraction preference characteristics can include operational data such as dwell time at each attraction's Point of Interest (POI), the percentage of newly visited POIs, and visits to less popular POIs.

[0064] Hotel preference characteristics include brand preference, class preference, location preference, and special preferences. The second operational data associated with these hotel preference characteristics can include hotel name, brand, price range, and the relative location of the hotel to attractions.

[0065] Food preference characteristics include local taste preferences, travel destination preferences for trying new things, stable cross-city food category preferences, and budget preferences. The second operational data associated with these food preference characteristics can include operational data such as restaurant categories, cross-city recurrence, and name characteristics.

[0066] Planning and decision-making preference characteristics include features such as plan sensitivity, tendency to repeatedly confirm information, and tendency to make ad-hoc decisions based on local conditions. The second operational data associated with these planning and decision-making preference characteristics may include whether users query external navigation services several days in advance, whether they compare multiple Points of Interest (POIs) at once, and whether they frequently use ad-hoc navigation services after arrival.

[0067] Special needs and preferences include preferences for family-friendly venues, pet-friendly venues, repeat visits, and niche exploration.

[0068] Step S304: Based on the second operational data and the data credibility of the second operational data, determine the feature values ​​of the target user on the target preference features.

[0069] Among these methods, data analysis or data fusion can be performed on the second operational data and the data credibility of the second operational data to obtain the feature values ​​of the target user on the target preference features.

[0070] In some embodiments, a mapping function corresponding to the target preference feature is obtained; based on the mapping function, the second operation data is mapped to obtain a mapping value; and based on the mapping value corresponding to the second operation data and the data credibility of the second operation data, the feature value of the target user on the target preference feature is determined.

[0071] Different target preference features may require different mapping functions. Mapping functions can include normalization functions, key information extraction functions, encoding functions, etc. For example, a key information extraction function can extract travel route methods from the second operational data. It should be noted that the mapped value obtained based on the mapping function can be numerical, textual, or other information.

[0072] Specifically, data fusion processing can be performed based on the mapping value and data credibility corresponding to the second operational data to obtain the feature values ​​of the target user on the target preference features. This data fusion processing may include weighted summation, median value extraction, and mode selection.

[0073] For example, when the target preference feature is a numerical statistical feature, the feature value of the target user on the target preference feature can be determined by the following formula:

[0074] in, This represents the feature value of the target user on the k-th target preference feature. This represents the i-th second operation data associated with the k-th target preference feature. This represents the total number of second operational data associated with the k-th target preference feature. This represents the mapping function corresponding to the k-th target preference feature. This represents the data value in the i-th second operation data that is valuable for reference to the feature value. It should be noted that, in addition to data values ​​such as stay duration and departure time that are valuable for the feature value, a single second operation data entry typically also includes data such as the user's unique identifier and city code that are not valuable for the feature value. Therefore, This refers to the data value in the i-th second operation data that has reference value for the feature value.

[0075] For example, when the target preference feature is a non-numerical statistical feature, such as a travel mode preference feature, the travel route mode in the second operational data can be extracted using a key information extraction function. Then, the frequency of occurrence of each travel route mode is counted. Based on the travel route mode corresponding to the maximum frequency, the feature value of the travel mode preference feature is determined. For instance, if the travel route mode corresponding to the maximum frequency is high-speed rail, then the feature value of the travel mode preference feature is high-speed rail.

[0076] By mapping the original operational data using a dedicated mapping function, the data dimensions, data format, and data analysis standards can be unified. Combined with data credibility verification, the final feature values ​​can be calculated, which solves the problems of messy original travel data formats, inconsistent dimensions, and inability to accurately quantify preferences. This makes the calculation results of feature values ​​more standardized and accurate.

[0077] In some embodiments, based on the second operational data and the data credibility of the second operational data, a data analysis model is used to perform data analysis to obtain the feature values ​​of the target user on the target preference features.

[0078] The data analysis model can be a large model or a mathematical analysis model.

[0079] By integrating multi-source operational data and their credibility through data analysis models for intelligent inference, it is possible to handle complex semantics and implicit preferences, thereby improving the ability to identify preference features.

[0080] Step S305: Obtain target preference information based on the feature values ​​of the target user on at least one travel preference feature.

[0081] The target preference information includes the feature values ​​of the target user on various travel preference characteristics.

[0082] To ensure the interpretability and verifiability of the target preference information, it can also include the reliability strength of the feature values ​​for each travel preference feature, key derivations, a list of evidence, and the time range covered. For example, for the dimension "accepting late-night departures," if there is a lack of sufficient evidence, the reliability strength of the feature value is "unknown"; for the dimension "local travel is mainly by car," evidence such as "routes were calculated 353 times" can be attached.

[0083] Step S306: Generate structured itinerary plan information based on the target information.

[0084] In some embodiments, generating structured itinerary plan information based on target information includes: breaking down the travel planning task into multiple sub-tasks based on the target information; determining a tool call plan for any sub-task among the multiple sub-tasks; calling tools sequentially based on the tool call plans of each sub-task according to the task execution order of the multiple sub-tasks to obtain the corresponding tool call results; and generating itinerary plan information based on the tool call results.

[0085] Specifically, a large model can be used to break down the travel planning task and determine the tool invocation plan for each sub-task. That is, the target information is input into the large model, and the model outputs multiple sub-tasks and their respective tool invocation plans. The tool invocation plan can include the tools required to execute the corresponding sub-task and the order in which these tools are invoked.

[0086] The sub-tasks may include candidate attraction recall, route calculation, transportation mode evaluation, time scheduling, constraint verification, candidate replacement, hotel selection, and catering supplementation. For example, the multiple sub-tasks obtained by breaking down the travel planning task may include the following: 1. Use a tourist attraction search tool to obtain candidate attractions in the target city; 2. Use the route calculation tool and distance / time calculation tool to obtain the distance information between attractions; 3. Use the constraint verification tool to determine whether the current candidate combination violates walking distance, parent / child / pet-friendly, or time budget limits; 4. Use the candidate replacement tool to replace attractions that do not meet the constraints.

[0087] By breaking down complex travel planning tasks into multiple sub-tasks, determining the specific tools to call the plan, and executing them in sequence, it is possible to achieve modular and process-oriented processing of itinerary planning, reduce the difficulty of processing complex planning tasks, and improve the completeness, logic, and generation efficiency of itinerary plan information.

[0088] In some embodiments, when the travel planning request is input by the target user in a second application, generating structured itinerary planning information based on target information includes: obtaining the target user's basic preference information, as well as the target user's historical dialogue and third operation data in the second application; wherein the second application is a search application or a map application; obtaining real-time map service data associated with the travel planning request; obtaining target information based on at least one of the basic preference information, historical dialogue, third operation data, and real-time map service data, as well as the travel planning request and target preference information; and generating structured itinerary planning information based on the target information.

[0089] Among them, basic preference information is used to indicate the identity of the target user; the third operation data refers to the target user's historical search data, historical click data, and other historical interaction operation data in the second application; the real-time map service data associated with the travel planning request refers to real-time traffic data of areas such as travel destinations and routes, which may include road congestion data, traffic restriction data, highway closure data, etc.

[0090] As an example, the target information may include basic preference information, historical dialogues, third-party operational data and real-time map service data, travel planning requests and target preference information; or, the target information may include one or more of the following: basic preference information, historical dialogues, third-party operational data and real-time map service data, as well as travel planning requests and target preference information.

[0091] By integrating basic preference information, historical dialogues, third-party operation data, real-time map service data, travel planning requests, and target preference information, the dimensions of target information are enriched, which can enhance the understanding of user needs and scenario context, and improve the timeliness and usability of itinerary planning information.

[0092] In some embodiments, based on the travel planning request and target preference information, a matching reference route is queried from the candidate routes; and based on the target information and the reference route, trip plan information is generated.

[0093] Multiple candidate routes can be pre-built, each with structured route tags. For example, route tags may include pace tags, walking intensity tags, mode of transportation tags, family-friendly tags, pet-friendly tags, niche appeal tags, repeat visit appeal tags, nightlife tags, budget tags, and accommodation convenience tags.

[0094] First, based on the travel planning request, multiple candidate itineraries associated with the travel destination can be identified; then, the target preference information can be matched with the route tags of the candidate itineraries to obtain a matching reference itinerary.

[0095] In some embodiments, generating itinerary plan information based on target information and reference itinerary routes includes: breaking down the travel planning task into multiple sub-tasks based on the target information and reference itinerary routes; determining a tool call plan for any sub-task among the multiple sub-tasks; sequentially calling tools according to the task execution order of the multiple sub-tasks based on the tool call plan of each sub-task to obtain the corresponding tool call results; and generating itinerary plan information based on the tool call results.

[0096] The explanations of task breakdown and tool invocation can be found in the descriptions of the relevant content above, and will not be repeated here.

[0097] By first filtering reference routes and then combining them to generate itinerary information, we can not only improve the efficiency of travel planning while achieving personalized planning, but also reduce computing power consumption and shrink the search space for subsequent tool calls.

[0098] Step S307: Based on the itinerary plan information, generate the first response content and the first itinerary route.

[0099] In some embodiments, generating a first response content based on itinerary planning information includes: performing natural language generation processing based on the itinerary planning information to obtain the first response content; wherein the first response content includes an itinerary summary and supplementary descriptions of the itinerary summary.

[0100] One approach is to use a large model for natural language generation, which means taking the itinerary plan information as input to the large model and obtaining the first response content output by the large model.

[0101] The supplementary explanations may include reasons for recommending the attraction, route instructions, transportation advice, explanations of restrictions, tips to avoid pitfalls, alternative suggestions, or answers to follow-up questions.

[0102] By performing natural language processing on structured itinerary information, a first response containing an itinerary summary and supplementary explanations is obtained. This transforms rigid and difficult-to-understand structured data into easy-to-understand natural language content, allowing users to quickly grasp the core information of the itinerary and improving the intuitiveness of the itinerary display and the user's reading and comprehension experience.

[0103] In some embodiments, generating a first travel route based on travel plan information includes: calling a map service engine based on the travel plan information to obtain the first travel route generated by the map service engine.

[0104] The map service engine receives structured travel plan information and performs operations such as route calculation, route stitching, and multi-day route rendering to obtain a visualized map route.

[0105] By generating travel routes through a professional map service engine, the accuracy and reliability of the generated routes can be improved by leveraging the capabilities of the map service engine.

[0106] Step S308: Output the first reply content and the first trip route to the target user's user terminal.

[0107] The explanation of step S308 can be found in the relevant description in any embodiment of this disclosure, and will not be repeated here.

[0108] The information processing method provided in this disclosure first determines the data credibility of the operation data and matches the associated operation data of each travel preference feature, and then quantifies and calculates the feature value of each travel preference feature to generate the target user's target preference information. This achieves a refined breakdown and quantitative characterization of user travel preferences, improves the accuracy of constructing user preferences in travel scenarios, and can provide accurate data support for customized itinerary planning.

[0109] Figure 4 This is a flowchart illustrating an information processing method provided in an embodiment of the present disclosure.

[0110] like Figure 4 As shown, this information processing method may include the following steps: Step S401: In response to the travel planning request of the target user, obtain the target user's target preference information in the travel scenario.

[0111] Step S402: Generate structured itinerary plan information based on the target information.

[0112] In some embodiments, generating structured itinerary plan information based on target information includes: breaking down the travel planning task into multiple sub-tasks based on the target information; determining a tool call plan for any sub-task among the multiple sub-tasks; calling tools sequentially based on the tool call plans of each sub-task according to the task execution order of the multiple sub-tasks to obtain the corresponding tool call results; and generating itinerary plan information based on the tool call results.

[0113] The explanations of task breakdown and tool invocation can be found in the relevant descriptions in the foregoing embodiments, and will not be repeated here.

[0114] Step S403: Based on the itinerary plan information, generate the first response content and the first itinerary route.

[0115] Step S404: Output the first reply content and the first trip route to the target user's user terminal.

[0116] The explanation of steps S401-S404 can be found in the relevant descriptions in any embodiment of this disclosure, and will not be repeated here.

[0117] Step S405: Obtain the target user's planning adjustment request; wherein, the planning adjustment request is generated based on the target user's dialogue instructions, or based on the target user's interactive operation on the first trip route.

[0118] Among them, the planning adjustment request refers to the supplementary needs of the target user through text dialogue or operation of the travel route.

[0119] The interactive operations of the target user on the first itinerary route may include deleting attractions, adding attractions, adjusting departure time, viewing details of a certain attraction, clicking on a certain route segment, zooming in on a certain area, switching modes of transportation, and confirming or abandoning a candidate route.

[0120] Step S406: In response to the planning adjustment request, determine the first trip segment and the second trip segment in the trip plan information; wherein, the first trip segment is the trip segment to be adjusted in the trip plan information, and the second trip segment is the trip segment in the trip plan information other than the first trip segment.

[0121] The first travel segment refers to the travel segment that needs to be adjusted, and is used to indicate the travel segment that can be replaced, rearranged or recalculated during replanning; the second travel segment can also be called the frozen segment, and is used to indicate the travel segment that remains unchanged during replanning.

[0122] For example, based on a planning adjustment request, a large model is used to determine the first and second trip segments in the trip planning information.

[0123] In some embodiments, in response to a planning adjustment request, determining a first trip segment in the trip plan information includes: in response to the planning adjustment request, determining an interaction event associated with the planning adjustment request; identifying a new intent of the target user based on at least one of the trip plan information, target preference information, the target user's historical dialogues, and real-time map service data associated with the planning adjustment request, as well as the interaction event; and determining a first trip segment related to the new intent in the trip plan information.

[0124] Interactive events include, but are not limited to, deleting attractions, adding attractions, switching to family-friendly or pet-friendly routes, adjusting departure time, adjusting travel intensity, viewing details of a specific attraction, clicking on a route segment, zooming in on a specific area, switching modes of transportation, and confirming / abandoning a route.

[0125] As an example, the new intent of a target user can be identified based on one or more of the following: itinerary planning information, target preference information, the target user's historical dialogues, and real-time map service data associated with the planning adjustment request, as well as interaction events; or, the new intent of a target user can be identified based on itinerary planning information, target preference information, the target user's historical dialogues, real-time map service data associated with the planning adjustment request, and interaction events.

[0126] Among these methods, the new intent of the target user can be identified through a large model. For example, trip planning information, target preference information, target user's historical dialogues, real-time map service data associated with the planning adjustment request, and interaction events can be input into the large model to obtain the new intent output by the large model.

[0127] After identifying the new intent, it can determine which dates, time periods, attraction sets, or route segments in the current itinerary plan information are related to the new intent, and identify these related parts as the first itinerary segment to be adjusted, and the irrelevant parts as the second itinerary segment.

[0128] By combining multi-dimensional information such as interaction events, user preferences, historical dialogues, and real-time map data to identify users' intentions to make new adjustments, the accuracy of user intention recognition can be improved, thereby accurately locating the trip segments that need adjustment and improving the accuracy of trip adjustments.

[0129] Step S407: While keeping the second trip segment unchanged, adjust the first trip segment in the trip plan information based at least on the planning adjustment request and target preference information to obtain the adjusted trip plan information.

[0130] Specifically, the first itinerary segment in the itinerary planning information can be adjusted using a large model to obtain the adjusted itinerary planning information output by the large model. For example, the planning adjustment request and target preference information can be used as inputs to the large model, or the planning adjustment request and target information can be used as inputs to the large model. The target information includes one or more of the following: basic preference information, historical dialogues, third-party operational data, and real-time map service data, as well as the travel planning request and target preference information.

[0131] During the adjustment process, external tools can be re-called only for the first trip segment to be adjusted, and based on the results returned by the tool, operations such as candidate replacement, path recalculation, time adjustment or order rearrangement can be performed on the first trip segment, while keeping the frozen segments unchanged.

[0132] The adjusted itinerary plan information may include: updated first itinerary segment, second itinerary segment, updated order of attractions, updated mode of transportation, updated route, and updated constraint labels.

[0133] Step S408: Based on the adjusted itinerary plan information, generate a second response content and a second itinerary route, and output the second response content and the second itinerary route to the user terminal.

[0134] The second response may include one or more of the following: a response to the request for planning adjustments; an explanation of the replaced attractions; an explanation of the modified route rhythm; an explanation of changes in transportation methods; and an explanation of the unchanged segments remaining unchanged.

[0135] Since both the second response content and the second itinerary route originate from the same adjusted itinerary plan information, it is possible to maintain consistency between the structured itinerary content after partial replanning, the second response content, the attractions, route segments, and modes of transportation displayed in the second itinerary route.

[0136] The second reply can be displayed in the first display window of the dialogue interface, and the second route can be displayed in the second display window of the dialogue interface.

[0137] The information processing method provided in this disclosure, in response to a planning adjustment request, determines a first trip segment and a second trip segment in the trip plan information, and adjusts the first trip segment in the trip plan information based at least on the planning adjustment request and target preference information while keeping the second trip segment unchanged, to obtain the adjusted trip plan information; it can use a partial segment adjustment mode to optimize only the trip segment to be adjusted, retain the original reasonable trip content, without the need for overall trip reconstruction, thereby reducing the amount of calculation for trip adjustment and the scope of re-retrieval, and improving the efficiency of trip adjustment.

[0138] This disclosure also provides an information processing system, which includes: a multi-source data acquisition module, an operation extraction and standardization module, an operation fusion and completion module, a travel preference construction module, an offline storage module, a route label management and candidate matching module, an online feature fusion module, a planning control module, a tool registration and invocation orchestration module, an itinerary generation and structured output module, a question-and-answer generation module, a map service engine, an impact range identification module, and a front-end interactive feedback module. Wherein: The multi-source data acquisition module is used to collect multi-source operation data generated by users in the first application (map application and search application). Multi-source operation data includes, but is not limited to, map POI query data, map POI display data, navigation data, route calculation data, location data, dwell trajectory data, search history data, click history data, basic preference data, and optional feed history or other content consumption data. This data is collected only with user authorization or under legally permissible conditions, and undergoes user identification de-identification, field minimization, or tiered access control processing before entering the travel preference construction process.

[0139] The operation extraction and standardization module is used to clean, parse, unify, and event-based process operation data from different sources to form standardized operation events, i.e., the first operation data. This processing includes at least address text and POI information parsing, splitting navigation / routing records into start and end points, unifying time fields, mapping POI categories, completing city information, filtering duplicate records, removing abnormal data, and extracting retrieval events, display events, click events, OD events, and dwell events.

[0140] The operation fusion and completion module is used to merge events such as positioning, navigation, route calculation, map retrieval and search history according to time and space dimensions, and to complete missing city information, POI name mapping, permanent and non-permanent location markings, etc.

[0141] The travel preference construction module is used to extract travel preference features and form structured travel preference information, i.e., target preference information, based on standardized operational events and their data credibility. Travel preference information includes at least: user basic information, travel preferences, route preferences, attraction preferences, food preferences, hotel preferences, planning and decision-making preferences, and special needs preferences. Each preference field must include at least: feature value, conclusion strength, basis, evidence list, and coverage time range.

[0142] The offline storage module is used to save travel preference information in the form of JSON, key-value maps, sparse vectors or other computable data structures for online reading and retrieval.

[0143] The route tag management and candidate matching module is used to maintain a pre-built candidate route library, route template library, or POI candidate set, and to configure structured route tags for each candidate route or route segment. Route tags include, but are not limited to: pace tags, walking intensity tags, mode of transportation tags, family-friendly tags, pet-friendly tags, niche tags, repeat visit suitability tags, nightlife tags, budget tags, and accommodation convenience tags.

[0144] The online feature fusion module is used to read the user's current query (i.e., travel planning request), multi-turn dialogue history, travel preference information, basic preference information, search history, click history, and real-time map service data during the online phase, and integrate them into a unified planning context input.

[0145] The planning and control module is used to reason, break down, and control the travel planning task based on travel preference information, the current query, multi-turn dialogue history, and real-time map service data, determining which external tools to invoke and in what order. This module not only generates user-readable question-and-answer content but also performs task decomposition, tool selection, invocation control, result integration, and structured itinerary generation. The planning and control module can be implemented using a large language model, or collaboratively using a large language model with a rule engine, ranking module, candidate recall module, and constraint validation module.

[0146] The tool registration and invocation orchestration module manages the capability descriptions, input parameters, output formats, and invocation interfaces of various tourism planning tools. It executes the corresponding tool invocation when the planning control module initiates a request, and then returns the standardized results to the planning control module. These tools include, but are not limited to: attraction search tools, route calculation tools, distance calculation tools, time estimation tools, transportation mode query tools, hotel candidate query tools, restaurant candidate query tools, and constraint validation tools.

[0147] The itinerary generation and structured output module is used to generate structured itinerary plan information based on the planning results returned by the planning control module. The structured itinerary plan information serves as a unified intermediate representation and includes at least date identifiers, time period identifiers, attraction nodes, arrival and departure times, transportation modes, route segments, distance costs, time costs, constraint labels, alternative candidate sets, and freeze flags. It can further output text descriptions, recommendations, and operational suggestions corresponding to the structured itinerary plan information.

[0148] The question-and-answer generation module is used to generate corresponding natural language question-and-answer output content (i.e., the first response content) based on structured itinerary planning information. The question-and-answer output content includes at least one itinerary summary presented to the user, reasons for attraction recommendations, route descriptions, transportation suggestions, constraint explanations, pitfall avoidance tips, alternative suggestions, or replies to follow-up questions. The text fragments output by the question-and-answer generation module are mapped to dates, time periods, attraction nodes, route segments, or constraint tags in the structured itinerary planning information to enable partial updates during subsequent interactive updates.

[0149] The map service engine receives structured travel plan information and performs route calculations, route stitching, and multi-day route rendering to obtain the first travel route corresponding to the structured travel plan information. The map service engine supports multiple modes of transportation, including walking, driving, cycling, and public transportation. It supports differentiating map results by layer, style, or sequence based on date, route segment, and candidate routes, and also supports partial refreshing of the travel route.

[0150] The impact scope identification module is used to identify the affected dates, time periods, attraction sets and route segments based on the current structured itinerary plan information after receiving new constraints (i.e. planning adjustment requests), and set freeze marks for unaffected segments, thereby supporting local replanning only for the affected segments (the first itinerary segment).

[0151] The front-end interactive feedback module displays both Q&A output and map / trip routes on the front end, and collects dialogue and map interaction events. Map interaction events include clicking on attractions, viewing route segments, zooming in on areas, switching modes of transportation, confirming candidate routes, deleting attractions, and replacing attractions; dialogue events include adding requirements, adjusting pace, adding themes, switching between family / pet-friendly mode and niche / classic mode. Furthermore, the front-end interactive feedback module converts the collected interaction events into new constraints and transmits them to the planning control module and the impact range identification module to trigger subsequent local update processes.

[0152] It should be noted that the information processing methods or systems provided in this disclosure can be applied to scenarios such as tourism vertical result pages in search engines, travel itinerary planning and navigation in map products, travel customization in smart personal assistants, personalized guide generation in online travel services, and destination navigation services that combine map display and question-and-answer interaction.

[0153] The following explanation uses specific application scenarios as examples: Users enter travel-related queries into a search engine, such as "Beijing 3-day travel guide," "Xiamen travel guide," or "Shanghai food guide." After recognizing the query as a travel planning request, the system first retrieves offline travel preference information (i.e., target preference information) based on the user's identifier and displays lightweight itinerary planning results on the search results page. These lightweight itinerary planning results include at least: an initial route preview, a daily theme summary, a brief Q&A explanation, and interactive information guiding the user to a details page for further planning. The brief Q&A explanation summarizes the reasons for the recommendation, its pace, or the characteristics of the target audience, helping users quickly understand the recommendations. It should be noted that the initial route can be a travel itinerary planned based on the user's basic preference information, travel preference information, conversation history, search history, and click history, or it can be a reference itinerary obtained by matching travel preference information with pre-built candidate itineraries.

[0154] When a user clicks to enter the details page, or continues to supplement their needs in the dialog interface, the system enters the online detailed planning stage. At this time, the system retrieves offline travel preference information based on the user's identifier, and combines it with basic preference information, search history, click history, and real-time map service data to construct the context input for the planning control module. This allows the planning control module to break down the travel planning task into sub-tasks such as attraction candidate retrieval, route calculation, transportation mode evaluation, visited attraction filtering, constraint verification, and itinerary arrangement, and call the corresponding tools to obtain the results required for planning. Subsequently, the system generates structured itinerary plan information based on the tool's return results, and generates corresponding Q&A output content (first response content) based on the itinerary plan information. The map service engine then generates the corresponding route set, attraction locations, and map display results based on the itinerary plan information, resulting in the first itinerary route.

[0155] For example, for users who prefer family-friendly venues, museums, moderate travel pace, and advance planning, when they input "Beijing travel guide," the system can prioritize searching for candidate attractions that are family-friendly, have moderate walking intensity, and are suitable for family travel pace, based on the user's travel preference information and real-time map service data. It then uses route calculation and constraint verification tools to complete multi-day route planning and outputs a structured itinerary plan that includes family-friendly attractions, a suitable pace for family travel, and transportation suggestions. Simultaneously, the system can also generate question-and-answer output corresponding to the structured itinerary plan information, such as "Why prioritize museums and family-friendly venues?", "Why is less walking planned for the second day?", and "Which attractions are suitable for bringing children?", to help users understand the planning basis and ask further questions.

[0156] As the user continues to interact, for example, when the user enters in the dialogue "I want to relax the next afternoon," "Add a night market," or "Don't plan too much walking," or clicks on a scenic spot on the map, views a route segment, zooms in on a certain area, or switches to public transportation mode, the system can recognize this input as a new constraint and trigger a local replanning process. After identifying the affected segments, the planning control module only re-initiates tool calls for the affected time periods, sets of scenic spots, or route segments to obtain alternative scenic spots, update routes, distances, travel times, and modes of transportation, and generates updated structured itinerary plan information while keeping the unaffected segments stable.

[0157] On this basis, the system further generates updated question-and-answer output content based on the updated structured travel itinerary information, and refreshes the map display result synchronously. The system only refreshes the question-and-answer content related to the affected segments, such as the arrangement reason for the newly added night market, the description of the pace change after reducing walking, the explanation of the public transport replacement result, and keeps the description of unchanged segments unchanged, thereby realizing linked updating among the structured travel itinerary result, the question-and-answer explanation result and the map display result.

[0158] Wherein, the present disclosure can also be applied to subdivided travel scenarios such as revisits, pet-friendly travel, parent-child travel, urban walking, niche exploration, etc. For revisit users, based on their historical visit records and current travel preference information, the system can reduce the recommendation proportion of visited attractions through the visited attraction filtering tool, increase candidate attractions that have not been experienced or are more in line with their preferences, and explain "why reduce the arrangement of classic attractions and increase the recommendation of niche routes" through question-and-answer output content; for pet-friendly or parent-child users, the system can introduce corresponding special demand constraints in the process of tool calling and constraint checking, give priority to selecting suitable attractions, restaurants and hotels, and generate corresponding constraint explanations and frequently asked questions on precautions; for users who prefer less walking or a relaxed pace, the system can reduce walking intensity and decrease high-density check-in arrangements during the path calculation and itinerary planning stage, and present the corresponding adjustment results of the travel route and reason explanation in map display and question-and-answer output.

[0159] In summary, by integrating offline travel preference information, online planning control, structured travel itinerary information generation, question-and-answer output generation and map rendering display into the same processing link, the system can integrate the traditional decentralized process of "searching information first, switching maps then, and sorting out routes manually" into a unified link, reducing cross-platform switching and repeated operations, thereby improving the overall efficiency from information retrieval to completion of itinerary decision-making. In addition, when it is necessary to adjust the itinerary planning, local update can be performed only on the affected segments after identifying the affected local range.

[0160] Figure 5 is a structural schematic diagram of an information processing apparatus provided by an embodiment of the present disclosure.

[0161] As shown in Figure 5 , the information processing apparatus 500 of the embodiment of the present disclosure includes a first obtaining module 501, a first generating module 502, a second generating module 503 and an output module 504.

[0162] the first obtaining module 501 is configured to, in response to a travel planning request from a target user, obtain target preference information of the target user in a travel scenario; wherein the target preference information is obtained based on first operation data of the target user in at least one first application; The first generation module 502 is used to generate structured itinerary plan information based on target information; wherein, the target information includes at least travel planning requests and target preference information; The second generation module 503 is used to generate the first response content and the first itinerary route based on the itinerary plan information; The output module 504 is used to output the first reply content and the first travel route to the target user's user terminal.

[0163] In one embodiment of this disclosure, the first acquisition module 501 is specifically configured to: acquire multiple first operation data of a target user in at least one first application; acquire the data credibility of the first operation data; for a target preference feature among at least one travel preference feature, select second operation data associated with the target preference feature from the multiple first operation data; determine the feature value of the target user on the target preference feature based on the second operation data and the data credibility of the second operation data; and obtain target preference information based on the feature value of the target user on at least one travel preference feature.

[0164] In one embodiment of this disclosure, the first acquisition module 501 is specifically used for: determining a first weight based on a first application associated with the first operation data; determining a second weight based on the operation type of the first operation data; determining a quality coefficient of the first operation data based on the data quality of the first operation data; acquiring the time interval between the data generation time of the first operation data and the current time; and determining data credibility based on at least one of the first weight, the second weight, the quality coefficient, the time interval, and the time decay coefficient corresponding to the first operation data.

[0165] In one embodiment of this disclosure, the first acquisition module 501 is specifically used for: acquiring a mapping function corresponding to the target preference feature; performing mapping processing on the second operation data based on the mapping function to obtain a mapping value; and determining the feature value of the target user on the target preference feature based on the mapping value corresponding to the second operation data and the data credibility of the second operation data.

[0166] In one embodiment of this disclosure, the first acquisition module 501 is specifically used to: perform data analysis using a data analysis model based on the second operation data and the data credibility of the second operation data, and obtain the feature values ​​of the target user on the target preference features.

[0167] In one embodiment of this disclosure, the apparatus further includes: a second acquisition module, a determination module, an adjustment module, and a third generation module; wherein the second acquisition module is used to acquire a planning adjustment request from a target user; wherein the planning adjustment request is generated based on the target user's dialogue instructions, or based on the target user's interactive operation on the first travel route; The determination module is used to determine the first trip segment and the second trip segment in the trip plan information in response to the planning adjustment request; wherein the first trip segment is the trip segment to be adjusted in the trip plan information, and the second trip segment is the trip segment in the trip plan information other than the first trip segment. The adjustment module is used to adjust the first trip segment in the trip plan information based at least on the planning adjustment request and target preference information, while keeping the second trip segment unchanged, to obtain the adjusted trip plan information; The third generation module is used to generate a second response and a second itinerary based on the adjusted itinerary plan information, and to output the second response and the second itinerary to the user terminal.

[0168] In one embodiment of this disclosure, the determining module is configured to: in response to a planning adjustment request, determine an interaction event associated with the planning adjustment request; identify a new intent of the target user based on at least one of trip plan information, target preference information, the target user's historical dialogues and real-time map service data associated with the planning adjustment request, and the interaction event; and determine a first trip segment related to the new intent in the trip plan information.

[0169] In one embodiment of this disclosure, the travel planning request is input by the target user in a second application; the first generation module 502 is configured to: obtain the target user's basic preference information, as well as the target user's historical dialogue and third operation data in the second application; wherein the second application is a search application or a map application; obtain real-time map service data associated with the travel planning request; obtain target information based on at least one of the basic preference information, historical dialogue, third operation data, and real-time map service data, as well as the travel planning request and target preference information; and generate structured itinerary plan information based on the target information.

[0170] In one embodiment of this disclosure, the first generation module 502 is configured to: decompose the travel planning task based on the target information to obtain multiple sub-tasks; determine the tool call plan for any one of the multiple sub-tasks; perform tool calls sequentially based on the tool call plan of each sub-task according to the task execution order of the multiple sub-tasks to obtain the corresponding tool call results; and generate itinerary plan information based on the tool call results.

[0171] In one embodiment of this disclosure, the first generation module 502 is configured to: query a matching reference itinerary route from candidate itinerary routes based on the travel planning request and target preference information; and generate itinerary plan information based on the target information and the reference itinerary route.

[0172] In one embodiment of this disclosure, the second generation module 503 is used to: perform natural language generation processing based on the itinerary plan information to obtain a first response content; wherein the first response content includes an itinerary summary and auxiliary descriptions of the itinerary summary.

[0173] In one embodiment of this disclosure, the second generation module 503 is used to: call the map service engine based on the trip plan information to obtain the first trip route generated by the map service engine.

[0174] The information processing apparatus provided in this disclosure can comprehensively capture users' travel preferences and habits across applications by using user operation data from at least one application, constructing highly accurate travel preference information, and thus accurately depicting users' travel preferences. By generating structured itinerary planning information, it enables itinerary data to have the characteristics of being computable, parsable, and locally updatable. By using structured itinerary planning information as a unified intermediate representation, it can achieve accurate mapping between "textual descriptions" and "spatial paths," improving the consistency between the first response content and the first itinerary route, and effectively solving the problem of information fragmentation in traditional planning. By simultaneously sending the first response content and the first itinerary route to the user terminal, users do not need to repeatedly switch between multiple applications or interfaces for comparison, improving the intuitiveness of the travel planning scheme display and the user's viewing experience. According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0175] Figure 6 A schematic block diagram of an example electronic device 600 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0176] like Figure 6As shown, device 600 includes a computing unit 601, which can perform various appropriate actions and processes based on computer programs / instructions stored in read-only memory (ROM) 602 or loaded from storage unit 608 into random access memory (RAM) 603. RAM 603 may also store various programs and data required for the operation of device 600. The computing unit 601, ROM 602, and RAM 603 are interconnected via bus 604. Input / output (I / O) interface 605 is also connected to bus 604.

[0177] Multiple components in device 600 are connected to I / O interface 605, including: input unit 606 such as keyboard, mouse, etc.; output unit 607 such as various types of monitors, speakers, etc.; storage unit 608 such as disk, optical disk, etc.; and communication unit 609 such as network card, modem, wireless transceiver, etc. Communication unit 609 allows device 600 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0178] The computing unit 601 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs the various methods and processes described above, such as information processing methods. For example, in some embodiments, the information processing method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program / instructions may be loaded and / or installed on device 600 via ROM 602 and / or communication unit 609. When the computer program / instructions are loaded into RAM 603 and executed by the computing unit 601, one or more steps of the information processing method described above may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to perform information processing methods by any other suitable means (e.g., by means of firmware).

[0179] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include: implementations in one or more computer programs / instructions that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transferring data and instructions to the storage system, the at least one input device, and the at least one output device.

[0180] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0181] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0182] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0183] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), the Internet, and blockchain networks.

[0184] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. The client-server relationship is created by computer programs / instructions running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0185] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in the disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this document does not impose any restrictions.

[0186] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. An information processing method, the method comprising: In response to a travel planning request from a target user, the system obtains the target user's target preference information in a travel scenario; wherein the target preference information is obtained based on the target user's first operation data in at least one first application. Structured itinerary planning information is generated based on the target information; wherein, the target information includes at least the travel planning request and the target preference information; Based on the aforementioned itinerary information, a first response and a first itinerary route are generated; The first reply content and the first travel route are output to the target user's user terminal.

2. The method according to claim 1, wherein, The step of obtaining the target user's target preference information in a travel scenario includes: Obtain multiple first operation data entries of the target user in multiple first applications; Obtain the data reliability of the first operation data; For at least one target preference feature among travel preference features, select second operation data associated with the target preference feature from the plurality of first operation data; Based on the second operational data and the data reliability of the second operational data, the feature value of the target user on the target preference feature is determined; The target preference information is obtained based on the feature values ​​of the target user on at least one travel preference feature.

3. The method according to claim 2, wherein, The data reliability of obtaining the first operation data includes: A first weight is determined based on the first application associated with the first operation data; Determine the second weight based on the operation type of the first operation data; Based on the data quality of the first operation data, determine the quality coefficient of the first operation data; The time interval between the data generation time of the first operation data and the current time is obtained; The data reliability is determined based on at least one of the first weight, the second weight, the quality coefficient, the time interval, and the time decay coefficient corresponding to the first operation data.

4. The method according to claim 2 or 3, wherein, The step of determining the feature value of the target user on the target preference feature based on the second operational data and the data reliability of the second operational data includes: Obtain the mapping function corresponding to the target preference feature; Based on the mapping function, the second operation data is mapped to obtain a mapped value; Based on the mapping value corresponding to the second operation data and the data credibility of the second operation data, the feature value of the target user on the target preference feature is determined.

5. The method according to claim 2 or 3, wherein, The step of determining the feature value of the target user on the target preference feature based on the second operational data and the data reliability of the second operational data includes: Based on the second operational data and the data reliability of the second operational data, a data analysis model is used to perform data analysis to obtain the feature values ​​of the target user on the target preference features.

6. The method according to claim 1, wherein, After outputting the first reply content and the first travel route to the target user's user terminal, the method further includes: Obtain the planning adjustment request from the target user; wherein the planning adjustment request is generated based on the target user's dialogue instructions, or based on the target user's interactive operation on the first travel route; In response to the planning adjustment request, a first trip segment and a second trip segment in the trip plan information are determined; wherein, the first trip segment is the trip segment to be adjusted in the trip plan information, and the second trip segment is the trip segment in the trip plan information other than the first trip segment; While keeping the second trip segment unchanged, the first trip segment in the trip plan information is adjusted based at least on the planning adjustment request and the target preference information to obtain the adjusted trip plan information; Based on the adjusted itinerary plan information, a second response and a second itinerary route are generated, and the second response and the second itinerary route are output to the user terminal.

7. The method according to claim 6, wherein, The step of determining the first itinerary segment in the itinerary plan information in response to the planning adjustment request includes: In response to the planning adjustment request, determine the interaction events associated with the planning adjustment request; Based on at least one of the trip plan information, the target preference information, the target user's historical dialogues, and real-time map service data associated with the planning adjustment request, as well as the interaction event, identify the target user's new intent; In the trip plan information, a first trip segment related to the new intention is identified.

8. The method according to claim 1, wherein, The travel planning request is entered by the target user in the second application; the generation of structured itinerary information based on the target information includes: Acquire the target user's basic preference information, as well as the target user's historical conversations and third operation data in the second application; wherein, the second application is a search application or a map application; Obtain real-time map service data associated with the travel planning request; The target information is obtained based on at least one of the basic preference information, the historical dialogue, the third operation data, and the real-time map service data, as well as the travel planning request and the target preference information; Structured itinerary planning information is generated based on target information.

9. The method according to claim 1 or 8, wherein, The generation of structured itinerary plan information based on target information includes: Based on the target information, the travel planning task is broken down into multiple sub-tasks; Determine the tool invocation plan for any one of the plurality of subtasks; Based on the task execution order of the multiple subtasks, tools are invoked sequentially according to the tool invocation plan of each subtask to obtain the corresponding tool invocation results; Based on the results of the tool call, the itinerary plan information is generated.

10. The method according to claim 8, wherein, The generation of structured itinerary plan information based on target information includes: Based on the travel planning request and the target preference information, a matching reference itinerary is queried from the candidate itinerary routes; Based on the target information and the reference itinerary route, the itinerary plan information is generated.

11. The method according to claim 1, wherein, The first response content generated based on the itinerary information includes: Based on the itinerary information, natural language processing is performed to obtain the first response content; wherein, the first response content includes an itinerary summary and supplementary explanations of the itinerary summary.

12. The method according to claim 1, wherein, Based on the aforementioned itinerary plan information, a first itinerary route is generated, including: Based on the trip plan information, the map service engine is invoked to obtain the first trip route generated by the map service engine.

13. An information processing apparatus, the apparatus comprising: The first acquisition module is configured to, in response to a travel planning request from a target user, acquire the target user's target preference information in a travel scenario; wherein, the target preference information is obtained based on the target user's first operation data in at least one first application; The first generation module is used to generate structured itinerary plan information based on target information; wherein, the target information includes at least the travel planning request and the target preference information; The second generation module is used to generate the first response content and the first itinerary route based on the itinerary plan information; The output module is used to output the first reply content and the first travel route to the target user's user terminal.

14. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method as described in any one of claims 1-12.

15. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method as described in any one of claims 1-12.

16. A computer program product comprising a computer program / instructions, wherein, When the computer program / instructions are executed by the processor, they implement the method as described in any one of claims 1-12.

Images