Joint matching method and system for research products and resources, computing device and medium

Abstract

The application provides a joint matching method and system of research products and resources, a computing device and a medium. The method comprises: obtaining research times and research needs of a user on a target research topic; constructing a research portrait of the user on the target research topic based on the research times and the research needs; querying a research knowledge graph based on the research portrait to obtain candidate research products and candidate resource configurations; receiving feedback information of the user on the candidate research products and the candidate resource configurations, adjusting parameters of same-layer edges and parameters of cross-layer edges based on the feedback information and the research portrait; updating the candidate research products and the candidate resource configurations based on the adjusted parameters of the same-layer edges and the parameters of the cross-layer edges to obtain target research products and target resource configurations, which can avoid the influence of inaccurate demand expression caused by insufficient user experience on the recommendation effect, avoid resource waste or insufficiency caused by ignoring individual differences of users in resource configuration, and improve the research experience and research effect of users.

Description

Technical Field

[0001] The embodiments in this specification relate to the technical field of data processing, and in particular to methods, systems, computing devices, and media for jointly matching study tour products and resources. Background Technology

[0002] With the diversification of user demands for teaching models and content, and the development of educational philosophies, study tours, with their integration of knowledge, experience, and fun, can deeply satisfy users' diverse teaching needs. In the field of study tours, accurately matching users with suitable study tour products and resources is a prerequisite for ensuring a good study tour experience.

[0003] However, existing methods for determining suitable study tour products and resources for users have the following problems: On the one hand, current methods mostly involve building a resource library where users manually filter products based on preset tags, or users input their needs (such as number of days, budget, and preferences), and the system matches corresponding products based on those needs. However, users who are new to study tours or have little experience cannot provide their deeper needs, resulting in insufficient matching between the matched products and the user, affecting the user's study tour effect and experience. On the other hand, existing solutions for determining study tour resources mostly rely on data such as the study tour content, location, and cost involved in the study tour product, combined with the experience of resource allocation personnel or pre-made resource allocation tables. However, this resource allocation method ignores individual differences among users, failing to provide differentiated resource allocation, leading to wasted or insufficient resource allocation.

[0004] In conclusion, existing technologies are insufficient to accurately match users with study tour products and resources, resulting in a lack of assurance regarding users' study tour experience and effectiveness. Summary of the Invention

[0005] In view of this, embodiments of this specification provide a method for jointly matching study tour products and resources. One or more embodiments of this specification also relate to a system for jointly matching study tour products and resources, a computing device, and a computer-readable storage medium, to address the technical deficiencies existing in the prior art.

[0006] According to a first aspect of the embodiments of this specification, a method for jointly matching study tour products and resources is provided, comprising: Obtain the number of study tours and the study tour needs of users regarding the target study tour topic; Based on the number of study tours and the needs for study tours, construct a study tour profile for the user regarding the target study tour theme; Based on the study tour profile, the study tour knowledge graph is queried to obtain candidate study tour products and candidate resource configurations. The study tour knowledge graph includes a study tour product layer and a resource configuration layer. The nodes in the study tour product layer are study tour products, and the nodes in the resource configuration layer are resource configurations. There are same-layer edges between each study tour product and cross-layer edges between any study tour product and each resource configuration. Receive user feedback on candidate study tour products and resource configurations, and adjust the parameters of same-layer edges and cross-layer edges based on the feedback and study tour profiles. Based on the adjusted parameters of the same-layer edges and cross-layer edges, the candidate study tour products and candidate resource configurations are updated to obtain the target study tour products and target resource configurations.

[0007] According to a second aspect of the embodiments of this specification, a joint matching system for study tour products and resources is provided, comprising: The acquisition module is configured to acquire the number of times a user has participated in a study tour and their study tour needs related to the target study tour topic; The module is configured to build a user profile for the target study tour theme based on the number of study tours and study tour needs. The query module is configured to query the study tour knowledge graph based on the study tour profile to obtain candidate study tour products and candidate resource configurations. The study tour knowledge graph includes a study tour product layer and a resource configuration layer. The nodes in the study tour product layer are study tour products, and the nodes in the resource configuration layer are resource configurations. There are same-layer edges between each study tour product and cross-layer edges between any study tour product and each resource configuration. The update module is configured to receive user feedback on candidate study tour products and candidate resource configurations, and adjust the parameters of the same-layer edges and cross-layer edges based on the feedback and study tour profiles. Based on the adjusted parameters of the same-layer edges and cross-layer edges, the candidate study tour products and candidate resource configurations are updated to obtain the target study tour products and target resource configurations.

[0008] According to a third aspect of the embodiments of this specification, a computing device is provided, comprising: Memory and processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions, which, when executed by the processor, implement the steps of the above-mentioned method for jointly matching research and study products and resources.

[0009] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided that stores a computer program / instructions that, when executed by a processor, implement the steps of the above-described method for jointly matching research and study products and resources.

[0010] In one embodiment of this specification, the following steps are taken: First, the number of times a user has participated in a study tour on a target study tour theme and their study tour needs are obtained. Based on these numbers and needs, a study tour profile of the user on the target study tour theme is constructed. Then, based on this profile, a study tour knowledge graph is queried to obtain candidate study tour products and candidate resource configurations. The study tour knowledge graph includes a study tour product layer and a resource configuration layer. Nodes in the study tour product layer represent study tour products, and nodes in the resource configuration layer represent resource configurations. There are same-layer edges between study tour products and cross-layer edges between any study tour product and any resource configuration. Feedback from the user on the candidate study tour products and candidate resource configurations is received. Based on the feedback and the study tour profile, the parameters of the same-layer edges and the parameters of the cross-layer edges are adjusted. Finally, based on the adjusted parameters of the same-layer edges and the cross-layer edges, the candidate study tour products and candidate resource configurations are updated to obtain the target study tour product and target resource configuration. This system leverages the number of study tours to represent users' existing experience with the target study tour theme and their differentiated needs to represent their specific requirements. By constructing a study tour profile that includes both experiential and preference features, it integrates users' existing experience with their differentiated needs. Then, utilizing the two-layer structure of a study tour knowledge graph, it achieves precise selection of study tour products and differentiated matching of resource allocation based on the integrated profile. Simultaneously, it optimizes the graph parameters through user feedback, achieving a closed-loop improvement in recommendation results. This avoids the problem of users only expressing macro-level needs due to insufficient experience, failing to address their deeper study tour expectations. It also avoids resource waste or inadequacy caused by ignoring individual user differences in resource allocation. This ensures that study tour products and resource allocation are precisely adapted to users' actual experience, preferences, and constraints, improving the user's study tour experience and effectiveness, thereby achieving synergistic optimization of study tour resource utilization efficiency and user satisfaction. Attached Figure Description

[0011] Figure 1 This is a flowchart illustrating a method for jointly matching study tour products and resources, provided in one embodiment of this specification. Figure 2 This is a schematic diagram illustrating one embodiment of a study tour product and study tour theme provided in this specification; Figure 3 This is a schematic diagram showing the interface of a study tour product and study tour theme provided in one embodiment of this specification; Figure 4 This is a schematic diagram of the structure of a research and learning knowledge graph provided in one embodiment of this specification; Figure 5 This is a structural block diagram of a joint matching system for study tour products and resources provided in one embodiment of this specification; Figure 6 This is a structural block diagram of a computing device provided in one embodiment of this specification. Detailed Implementation

[0012] Many specific details are set forth in the following description to provide a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar extensions without departing from the spirit of this specification. Therefore, this specification is not limited to the specific implementations disclosed below.

[0013] The terminology used in one or more embodiments of this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of this specification. The singular forms “a,” “described,” and “the” as used in one or more embodiments of this specification and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0014] It should be understood that although the terms first, second, etc., may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first may also be referred to as second without departing from the scope of one or more embodiments of this specification, and similarly, second may also be referred to as first. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0015] Furthermore, it should be noted that the data involved in one or more embodiments of this specification (including but not limited to data used for analysis, stored data, displayed data, etc.) are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0016] First, the terms and concepts used in one or more embodiments of this specification will be explained.

[0017] Study tour products refer to study tour activity plans designed based on a study tour theme, including teaching objectives and teaching processes. They serve as the content carriers of study tours. For example, study tour products on the theme of "space exploration" could include: visiting a space museum, listening to lectures by space experts, and making simple rocket models.

[0018] Resource allocation refers to the collection of human, material, and venue resources used to support the implementation of study tours. It is the material basis for the normal operation of study tours. For example, for a "space exploration" study tour product, resource allocation may include: 1 aerospace field guide, 2 handicraft instructors, 30 rocket model making material kits, and a space museum visit site, etc.

[0019] Computing devices: Computing devices are computer systems designed to perform one or more specific tasks. Compared to personal mainframes, they are weaker in performance but have significant advantages in terms of size and power consumption. They are commonly used in various electronic and mechanical control systems.

[0020] With the diversification of user demands for teaching models and content, and the development of educational philosophies, study tours, with their integration of knowledge, experience, and fun, can deeply satisfy users' diverse teaching needs. In the field of study tours, accurately matching users with suitable study tour products and resources is a prerequisite for ensuring a good study tour experience.

[0021] However, existing methods for identifying study tour products and resources for specific users have the following problems: On the one hand, existing methods mostly involve building resource libraries where users manually filter products based on preset tags, or users input their needs (such as number of days, budget, and preferences), and the system matches corresponding products based on those needs. However, users who are new to study tours or have little experience with them cannot provide their deeper needs, resulting in insufficient matching between the matched products and the user, affecting the learning effect and experience. For example, users participating in science-themed study tours for the first time can only provide macro-level needs such as "3 days" and "budget of 5,000 yuan," making it difficult to accurately express their interests and knowledge base in the field of science and technology. This leads to the system matching study tour products that are too superficial to stimulate their interest in exploration, or that involve too much prerequisite knowledge, making it difficult for users to keep up with the learning progress. The actual educational effect of the product is seriously inconsistent with the user's expectations.

[0022] On the other hand, existing solutions for determining study tour resources mostly rely on data such as the content, location, and cost of the study tour product, combined with the experience of resource allocation personnel or pre-made resource allocation tables. However, this resource allocation method ignores individual differences among users, failing to provide differentiated resource allocation and leading to waste or inadequacy in resource allocation. For example, for a "study tour to a certain location" product, the allocation plan typically provides a standard duration and general explanation resources. Users with ample physical strength and a strong interest in history and culture may require longer outdoor exploration time and in-depth thematic explanations; while users with weaker physical strength need more frequent rest stops and more interactive explanation methods. Standardized configurations cannot adapt to the differentiated needs of these two types of users, resulting in insufficient resources for some users and redundant resources for others, leading to low overall resource utilization efficiency.

[0023] To address the aforementioned issues, this specification provides a method for jointly matching study tour products and resources. This specification also relates to a system for jointly matching study tour products and resources, a computing device, and a computer-readable storage medium, which will be described in detail in the following embodiments.

[0024] See Figure 1 , Figure 1 The flowchart illustrates a method for jointly matching study tour products and resources according to an embodiment of this specification, including the following specific steps: Step 102: Obtain the number of study tours and study tour needs of users regarding the target study tour topic.

[0025] The target study tour theme can be any study tour theme specified by the user. For example, the target study tour theme can be a specific technology theme (such as "artificial intelligence"), a specific humanities theme (such as "history and culture", "folk customs and intangible cultural heritage"), a specific ecological theme (such as "forest ecology", "marine life"), a specific labor theme (such as "farming experience", "craftsmanship"), etc.

[0026] Indicatively, Figure 2 This specification shows a schematic diagram illustrating the content of study tour products and study tour themes provided in one embodiment.

[0027] like Figure 2 As shown, the study tour product corresponding to Customer 001 is the XX Dragon Boat Race, which belongs to the labor study tour theme; the study tour product corresponding to Customer 002 is "Exploring XX, A Century of Culture", which belongs to the humanities study tour theme; the study tour product corresponding to Customer 003 is "XX Day Trip", which belongs to the humanities study tour theme; and the study tour product corresponding to Customer 004 is "XX Adventure Journey", which belongs to both the humanities study tour theme and the ecological study tour theme.

[0028] Indicatively, Figure 3This specification shows a schematic diagram of the interface of a study tour product and study tour theme provided in one embodiment.

[0029] refer to Figure 3 The interface includes sections for "Product Name (i.e., the name of the study tour product)," "Product Version (i.e., the version of the study tour product)," "Product Theme (i.e., the theme of the study tour)," "Main Title (i.e., the main title of the study tour product)," "Number of Citations (i.e., the number of times the study tour product has been cited)," and "Number of Orders (i.e., the number of orders for the study tour product)." Product names include "Dragon Boat Race Labor Study Tour," "Five-Day Humanities xx," and "Six-Day Humanities, Science, and Ecology," etc. Product themes include "Labor," "Humanities," "Science," and "Ecology." Main titles include [Dragon Boat Race], [xxx Investigation Team], [Exploring the Mysteries of Desertification Control and Building an Ecological Oasis in the Desert], [Walking Through the Corridor of Two Years Ago, Dreaming of a Vast Sky], and [Spring Awakening Plan]. It should be noted that... Figure 2 and Figure 3 The study tour themes and products shown are for illustrative purposes only and are not intended to limit the names of actual study tour themes and products.

[0030] The number of study tours refers to the number of times a user has participated in study tour products covered by the target study tour theme, reflecting the user's existing experience with the target study tour theme. For example, if a user has participated in study tour products on the theme of "space exploration" three times, it indicates that the user has relatively rich experience in this theme. The higher the level of experience accumulation, the more familiar the user is with the study tour content of the target study tour theme.

[0031] Study tour needs refer to the expectations and requirements of users regarding a specific study tour theme. For example, study tour needs may include the desired direction of study tour content, the desired duration of study tour, the desired level of activity intensity, and whether there are any special safety requirements.

[0032] Optionally, one way to obtain a user's number of study tours and study tour needs related to the target study tour theme is to obtain the number of study tours and study tour needs input by the user through an interactive interface on the user's terminal device (such as a mobile phone or computer). Another way is to extract the number of study tours from the user's historical study tour order records and obtain study tour needs through a questionnaire survey. This specification does not limit this approach.

[0033] For example, through the input interface of the study tour matching application running on the terminal, the user enters the number of study tours as 2 under the theme of "Nature Exploration", and fills in "I hope to participate in study tours involving plant knowledge and outdoor hiking, with an activity duration of no more than 10 hours and moderate physical exertion" in the study tour demand input box.

[0034] In step 102, the number of times a user has participated in a study tour and their study tour needs related to the target study tour theme are obtained, which provides a data foundation for building a study tour profile of the user related to the target study tour theme.

[0035] Step 104: Based on the number of study tours and study tour needs, construct a study tour profile for the user regarding the target study tour theme.

[0036] A study tour profile is a multi-dimensional representation of a user's characteristics regarding a target study tour topic. For example, a study tour profile may include features representing the user's level of study tour experience, preferences for study tour content, and study tour constraints.

[0037] Optionally, one way to construct a user profile for a target study tour theme based on the number of study tours and study tour needs is to map the number of study tours to the user's experience level on the target study tour theme, extract keywords and perform semantic understanding on study tour needs to obtain preference directions and constraints, and combine experience level, preference directions, and constraints into a structured study tour profile feature vector. Another way is to input the number of study tours and study tour needs into a pre-trained profile model, and have the model output the study tour profile. This specification does not limit this approach in the embodiments.

[0038] For example, continuing from the previous example, if the user inputs "2 times" for the number of study tours, then the user's study tour experience characteristic is determined to be "moderate familiarity". Keyword extraction is performed on the study tour needs, yielding "plant cognition", "outdoor hiking", "duration no more than 10 hours", and "moderate physical exertion". "Plant cognition" and "outdoor hiking" constitute the content direction in the study tour preference characteristics, "moderate physical exertion" represents the activity intensity preference, and "duration no more than 10 hours" represents the duration constraint. The final study tour profile includes: study tour experience characteristics (familiarity: moderate), study tour preference characteristics (content direction: plant cognition and outdoor hiking; activity intensity: moderate), and constraint characteristics (duration constraint: no more than 10 hours).

[0039] In step 104, based on the number of study tours and study tour needs, a study tour profile of the user about the target study tour theme is constructed. This can transform the user's original input into a structured multi-dimensional feature representation, providing standardized query conditions for subsequent knowledge graph queries.

[0040] Step 106: Based on the study tour profile, query the study tour knowledge graph to obtain candidate study tour products and candidate resource configurations. The study tour knowledge graph includes a study tour product layer and a resource configuration layer. The nodes in the study tour product layer are study tour products, and the nodes in the resource configuration layer are resource configurations. There are same-layer edges between each study tour product and cross-layer edges between any study tour product and each resource configuration.

[0041] The study tour knowledge graph is a graph data structure that contains the relationships between study tour product entities and their associated resource allocation entities.

[0042] Indicatively, Figure 4 A schematic diagram of the structure of a research-based knowledge graph provided in an embodiment of this specification is shown.

[0043] like Figure 4 As shown, the study tour knowledge graph includes a study tour product layer and a resource allocation layer. In the study tour product layer, the study tour product nodes are products (i.e., study tour products) 1 to products 5. In the resource allocation layer, the resource allocation nodes are resources (i.e., resource allocation) 1 to resources 5. There are edges between study tour product nodes at the same level (represented by solid lines), and there are no edges between resource allocation nodes at the same level. Any study tour product node has cross-level edges with multiple resource allocation nodes (only a part is shown in the figure).

[0044] Candidate study tour products are one or more study tour products that are initially selected from the study tour knowledge graph based on the user's study tour profile and have a high degree of matching with the user's needs.

[0045] Candidate resource allocation refers to a preliminary combination of available study tour resources associated with the candidate study tour products.

[0046] The study tour product layer is a layer in the study tour knowledge graph composed of study tour product nodes, with each node representing a study tour product.

[0047] The resource configuration layer is a layer in the research and study knowledge graph composed of resource configuration nodes, with each node representing a resource configuration.

[0048] Same-layer edges are edges that connect two study tour product nodes in the study tour product layer, used to represent the relationship between the two study tour products.

[0049] Cross-layer edges are edges that connect study tour product nodes and resource allocation nodes, used to represent the mapping relationship between study tour products and resource allocation.

[0050] Optionally, one approach to obtaining candidate study tour products and resource configurations by querying the study tour knowledge graph based on the study tour profile can be as follows: Determine the initial activation value of each node in the study tour product layer based on the experience and preference features in the study tour profile; propagate and iteratively update the activation value along the same-layer edges between adjacent nodes; determine candidate study tour products based on the updated activation values; propagate the activation values ​​of the candidate study tour products along cross-layer edges to the resource configuration layer; and filter candidate resource configurations based on constraint features. Another approach can be as follows: Use the study tour profile as a query vector and perform a vector similarity-based retrieval in the study tour knowledge graph to obtain matching study tour product nodes and resource configuration nodes. This specification does not limit the scope of this approach.

[0051] For example, continuing from the previous example, the preference feature in the study tour profile is "plant knowledge and outdoor hiking," the experience feature is "moderate familiarity," and the constraint is "duration not exceeding 10 hours." In the study tour knowledge graph, the study tour product layer includes product nodes such as "Botanical Garden Science Tour," "Forest Park Hiking," and "Geological Museum Visit," with edges between products at the same level (such as association edges between outdoor products). Based on the experience and preference features, the initial activation values ​​of "Botanical Garden Science Tour" and "Forest Park Hiking" are determined to be relatively high. The activation value is propagated along the edges at the same level between adjacent nodes. After multiple rounds of iterative updates, the activation value of the "Botanical Garden Science Tour" node reaches the highest and is identified as a candidate study tour product. The activation value of this candidate study tour product is propagated along the cross-level edges to the resource allocation layer to obtain associated resource allocation nodes, including "Professional Botany Lecturer," "Outdoor Hiking Equipment," and "Park Shuttle Bus," etc. Based on the constraint "duration not exceeding 10 hours", the resource allocation items that exceed the time limit are excluded. The final candidate resource allocation is "professional botany lecturer (4 hours) + outdoor hiking equipment (4 hours) + park shuttle bus (2 hours)".

[0052] In step 106, based on the study tour profile, the study tour knowledge graph is queried to obtain candidate study tour products and candidate resource configurations. The study tour knowledge graph includes a study tour product layer and a resource configuration layer. The nodes in the study tour product layer are study tour products, and the nodes in the resource configuration layer are resource configurations. There are same-layer edges between each study tour product and cross-layer edges between any study tour product and each resource configuration. The two-layer structure of the graph can complete product selection and resource matching respectively, making the determination process of candidate study tour products and candidate resource configurations interpretable and targeted.

[0053] Step 108: Receive user feedback on candidate study tour products and candidate resource configurations, and adjust the parameters of the same-layer edges and cross-layer edges based on the feedback and study tour profile.

[0054] Feedback information consists of user evaluations of the candidate study tour products and resource configurations, representing user satisfaction with the candidate study tour products and resource configurations.

[0055] Optionally, one way to receive user feedback on candidate study tour products and resource configurations is to display a details page of the candidate study tour products and resource configurations on the user's terminal, and receive the ratings and comments submitted by the user on the page. Another way is to receive the user's voice feedback through voice interaction and convert it into text. This specification does not limit this approach.

[0056] The parameters of the same-layer edge are weight values ​​that control the propagation strength of the activation signal between each research and learning product node.

[0057] The parameters of the cross-layer edge are weight values ​​that control the intensity of the activation signal propagation from the study product node to the resource configuration node.

[0058] Optionally, one way to adjust the parameters of the same-layer edge and the cross-layer edge based on feedback information and the research and study profile is to calculate the correlation between the feedback information and the research and study profile, determine the adjustment direction and adjustment range of the same-layer edge parameters and the cross-layer edge parameters according to the magnitude of the correlation, and update the parameter values ​​according to the adjustment direction and adjustment range. Another way is to input the feedback information and the research and study profile into the parameter adjustment model, and have the model output the adjusted same-layer edge parameters and cross-layer edge parameters. The embodiments in this specification do not limit this approach.

[0059] For example, continuing from the previous example, the candidate study tour product "Botanical Garden Science Popularization Tour" and its corresponding resource configuration "Professional Botany Lecturer (4 hours) + Outdoor Hiking Equipment (4 hours) + Park Shuttle Bus (2 hours)" are shown to the user. The user gives the study tour product a positive feedback of 4.5 points (out of 5) and the resource configuration a neutral feedback of 3 points. Based on the first feedback information (product rating of 4.5 points) and the experience characteristics (moderate familiarity) and preference characteristics (plant knowledge and outdoor hiking) in the study tour profile, the first correlation is calculated to be high. Accordingly, the same-layer edge propagation weight parameter between "Botanical Garden Science Popularization Tour" and its adjacent product nodes is increased, so that other outdoor study tour products associated with "Botanical Garden Science Popularization Tour" can obtain a stronger activation signal in subsequent recommendations. Based on the second feedback information (resource allocation score of 3 points) and the constraint characteristics (duration not exceeding 10 hours), the second correlation degree is calculated to be medium to low correlation degree. Accordingly, the mapping weight parameter of the cross-layer edge between "Botanical Garden Science Popularization Tour" and "Professional Botany Lecturer" is appropriately reduced, so that the matching strength of this resource allocation in subsequent recommendations is moderately reduced.

[0060] In step 108, feedback information from users on candidate study tour products and candidate resource configurations is received. Based on the feedback information and study tour profiles, the parameters of the same-layer edges and cross-layer edges are adjusted. The graph parameters can be optimized in a targeted manner using user evaluation opinions. The adjustment of the same-layer edge parameters optimizes the accuracy of subsequent study tour product recommendations, and the adjustment of the cross-layer edge parameters optimizes the accuracy of subsequent resource configuration matching, enabling the system to have adaptive evolution capabilities based on user feedback.

[0061] Step 110: Based on the adjusted parameters of the same-layer edges and cross-layer edges, update the candidate study tour products and candidate resource configurations to obtain the target study tour products and target resource configurations.

[0062] The target study tour product is the study tour product that is finally recommended to the user after the candidate study tour products are updated based on the adjusted parameters.

[0063] The target resource configuration is the resource configuration that is finally recommended to the user after updating the candidate resource configurations based on the adjusted parameters.

[0064] Optionally, one implementation of updating candidate study tour products and candidate resource configurations based on the adjusted parameters of the same-layer edges and cross-layer edges to obtain the target study tour product and target resource configuration can be as follows: recalculate the activation value of each candidate study tour product using the adjusted same-layer edge propagation weight parameters, and select the target study tour product from the candidate study tour products based on the recalculated activation value; recalculate the matching degree between each candidate resource configuration and the target study tour product using the adjusted cross-layer edge mapping weight parameters, and select the target resource configuration from the candidate resource configurations based on the matching degree. Another implementation can be as follows: substitute the adjusted same-layer edge parameters and cross-layer edge parameters into the study tour knowledge graph, re-execute the query process, and use the results obtained from the re-query as the target study tour product and target resource configuration. This specification does not limit the implementation in this way.

[0065] For example, continuing from the previous example, after adjusting the parameters of the same layer, the activation value of "Botanical Garden Science Tour" can be maintained, while the activation value of its adjacent product "Forest Park Hiking" increases. However, considering the user's preference characteristics in the study tour profile (plant knowledge is prioritized over outdoor hiking), "Botanical Garden Science Tour" is determined as the final target study tour product. After adjusting the parameters of the cross-layer, the matching degree of "Professional Botany Lecturer" decreases. After re-evaluating the resource configuration based on the constraint characteristics (duration not exceeding 10 hours), this resource configuration is still the option that best meets the constraint conditions, so it is retained as the final target resource configuration. The system outputs the target study tour product "Botanical Garden Science Tour" and the target resource configuration "Professional Botany Lecturer (4 hours) + Outdoor Hiking Equipment (4 hours) + Park Shuttle Bus (2 hours)" to the user.

[0066] In step 110, based on the adjusted parameters of the same-layer edges and cross-layer edges, the candidate study tour products and candidate resource configurations are updated to obtain the target study tour products and target resource configurations. This enables precise optimization of the recommendation results driven by user feedback, ensuring that the final output study tour products and resource configurations not only meet the user's initial needs but also incorporate the user's evaluation of the preliminary recommendation results, thereby improving the recommendation quality.

[0067] In this embodiment, the following steps are taken: The number of times a user has participated in a study tour on the target study tour theme and their study tour needs are obtained. Based on the number of times and their needs, a study tour profile of the user on the target study tour theme is constructed. Based on the study tour profile, a study tour knowledge graph is queried to obtain candidate study tour products and candidate resource configurations. The study tour knowledge graph includes a study tour product layer and a resource configuration layer. Nodes in the study tour product layer are study tour products, and nodes in the resource configuration layer are resource configurations. There are same-layer edges between study tour products and cross-layer edges between any study tour product and any resource configuration. Feedback information from the user on the candidate study tour products and candidate resource configurations is received. Based on the feedback information and the study tour profile, the parameters of the same-layer edges and the parameters of the cross-layer edges are adjusted. Based on the adjusted parameters of the same-layer edges and the cross-layer edges, the candidate study tour products and candidate resource configurations are updated to obtain the target study tour product and target resource configuration. This system leverages the number of study tours to represent users' existing experience with the target study tour theme and their differentiated needs to represent their specific requirements. By constructing a study tour profile that includes both experiential and preference features, it integrates users' existing experience with their differentiated needs. Then, utilizing the two-layer structure of a study tour knowledge graph, it achieves precise selection of study tour products and differentiated matching of resource allocation based on the integrated profile. Simultaneously, it optimizes the graph parameters through user feedback, achieving a closed-loop improvement in recommendation results. This avoids the problem of users only expressing macro-level needs due to insufficient experience, failing to address their deeper study tour expectations. It also avoids resource waste or inadequacy caused by ignoring individual user differences in resource allocation. This ensures that study tour products and resource allocation are precisely adapted to users' actual experience, preferences, and constraints, improving the user's study tour experience and effectiveness, thereby achieving synergistic optimization of study tour resource utilization efficiency and user satisfaction.

[0068] In one optional embodiment of this specification, the study tour profile includes study tour experience characteristics, study tour preference characteristics, and constraint condition characteristics; Step 104 includes the following specific steps: The characteristics of study tour experience are determined based on the number of study tours, and the characteristics of study tour preferences are determined based on study tour needs. Based on the characteristics of study tour experience and study tour preferences, the characteristics of the constraints are determined; Based on the characteristics of study tour experience, study tour preferences, and constraints, a study tour profile of the user regarding the target study tour theme is constructed.

[0069] Study tour experience characteristics are digital descriptions of a user's level of proficiency in participating in the target study tour topic. For example, study tour experience characteristics can be "proficient" or "basic understanding".

[0070] Study tour preference characteristics are digital descriptions of users' interests and content preferences on target study tour topics. For example, study tour preference characteristics could be "preferring outdoor practical content" or "preferring theoretical inquiry content".

[0071] Constraint features are digital descriptions of the restrictions that users need to meet during study tours. For example, constraint features could be "duration not exceeding 3 hours" or "physical exertion at a moderate level".

[0072] Optionally, one way to determine the learning experience characteristics based on the number of learning trips is to map the number of learning trips to a preset experience level, with a higher number of learning trips resulting in a higher experience level. Another way is to determine the corresponding experience level identifier based on the numerical range of the number of learning trips. This specification does not limit this approach in the embodiments.

[0073] Optionally, one way to determine study tour preference features based on study tour needs is to extract keywords and perform semantic classification on the study tour needs to obtain preference direction and preference intensity. Another way is to input the study tour needs into a pre-trained preference recognition model and output a preference feature vector. This specification does not limit this approach in the embodiments.

[0074] Optionally, one way to determine the constraint features based on research and study experience characteristics and research and study preference characteristics is to: determine content-related constraints based on research and study preference characteristics, and modify the content-related constraints based on research and study experience characteristics to obtain the constraint features. Another way is to: input research and study experience characteristics and research and study preference characteristics into a pre-trained constraint generation model, and have the model output the constraint features. This specification does not limit the specific implementation of this method.

[0075] Optionally, one way to construct a user's study tour profile based on study tour experience characteristics, study tour preference characteristics, and constraint characteristics is to combine these characteristics into a structured feature vector, which serves as the study tour profile. Another way is to encapsulate these three elements into a study tour profile object according to a preset data structure. This specification does not limit this approach.

[0076] For example, continuing from the previous example, the user's number of study tours is "2 times," which is mapped to the study tour experience feature "familiarity level: moderate." The study tour needs "plant cognition," "outdoor hiking," and "moderate physical exertion" are extracted to obtain the study tour preference feature "content direction: plant cognition and outdoor hiking, activity intensity: moderate." Based on the experience feature and preference feature, the constraint feature "duration constraint: no more than 10 hours" is determined. These three features are combined into a structured study tour profile vector to complete the construction of the study tour profile.

[0077] In this embodiment, learning experience characteristics are determined based on the number of learning trips, and learning preference characteristics are determined based on learning needs. Constraint characteristics are determined based on these two characteristics. A learning profile of the user regarding the target learning theme is constructed based on these characteristics, including learning experience, learning preference, and constraint characteristics. Learning experience characteristics represent the user's existing experience accumulation, determining the initial activation intensity of learning product nodes in subsequent graph queries, enabling the system to match learning products appropriate to the user's knowledge level. Learning preference characteristics represent the user's interest direction and content preference, determining the initial direction of activation signal propagation along the same-layer edge in subsequent graph queries, allowing the system to filter learning products that match the user's interests. Simultaneously, constraint characteristics, jointly determined by learning experience and learning preference characteristics, transform the user's existing experience and desired direction into quantifiable and matchable constraints, used to limit the selection range of candidate learning products in subsequent graph queries. By transforming the number of study tours and study tour needs input by users into the three dimensions mentioned above, a logical connection is formed between the features of each dimension in the study tour profile: experience features and preference features characterize the user's state from the two dimensions of "existing experience" and "expected direction" respectively, and constraint features, as the result of the joint deduction of the two, provide standardized query conditions for subsequent querying of study tour products and resource configurations through knowledge graph, so that the query process can accurately locate study tour products and resource configurations that match the user's experience level, interest direction and constraint restrictions under the guidance of constraints.

[0078] In one optional embodiment of this specification, the constraint features include duration constraints and content constraints; Based on the characteristics of study tour experience and study tour preferences, the characteristics of the constraints are determined, including: Based on the characteristics of study tour preferences, determine the depth and direction of study tour content for users on the target study tour theme; Based on the characteristics of study tour experience, determine the user's level of understanding of the target study tour theme; The duration of the study tour will be determined based on the depth of the content and the level of mastery achieved. Content constraints are determined based on the depth and direction of the study tour content.

[0079] The duration constraint is the acceptable time range for study tour activities for users. For example, the duration constraint could be "the total duration of study tour activities shall not exceed 10 hours".

[0080] Content constraints define the scope and level of study tour activities that users can accept. For example, content constraints could be "Study tour activities should focus on plant knowledge while also incorporating outdoor practice."

[0081] The depth of study tour content is a quantitative indicator of the complexity of the study tour content that users expect. For example, the depth of study tour content can be divided into three levels: "shallow cognition", "intermediate understanding" and "deep exploration".

[0082] The research and study content category describes the category to which the user expects the research and study content to belong. For example, the research and study content category can be "plant classification", "ecosystem" or "geology and geomorphology".

[0083] The level of mastery is a quantitative indicator of a user's existing knowledge and skill proficiency on the target research topic. For example, the level of mastery can be expressed as "possessing basic botanical knowledge".

[0084] Optionally, one way to determine the depth and direction of a user's study tour content for a target study tour theme based on study tour preference characteristics is to extract keywords from the preference description text in the study tour preference characteristics and map the keywords to preset content depth levels and content direction categories. Another way is to perform similarity matching between the study tour preference feature vector and preset content depth and direction vectors, and select the depth level and direction category with the highest similarity. This specification does not limit this approach in the embodiments.

[0085] Optionally, one way to determine a user's level of mastery of the target study tour theme based on study tour experience characteristics is to determine the level of mastery based on the numerical range of the number of study tours. Another way is to comprehensively determine the level of mastery based on the number of study tours and the complexity of the study tour needs. This specification does not limit this approach in the embodiments.

[0086] Optionally, one way to determine the duration constraint based on the depth and level of mastery of the learning content is to obtain a preset baseline learning duration, calculate a duration adjustment coefficient based on the content depth level and mastery level, and multiply the baseline learning duration by the adjustment coefficient to obtain the duration constraint. Another way is to pre-establish a mapping table between the content depth level, mastery level, and duration constraint, and determine the duration constraint by looking up the mapping table. This specification does not limit the implementation of this method.

[0087] Optionally, one way to determine content constraints based on the depth and direction of the study tour content is to use the direction of the study tour content as the thematic scope of the content constraint and the depth of the study tour content as the hierarchical requirement of the content constraint, combining them to form content constraint conditions. Another way is to obtain a preset content constraint template, fill the study tour content direction and content depth into the template, and generate content constraints. This specification does not limit this approach.

[0088] For example, continuing from the previous example, the study tour preference characteristics are "content direction: plant cognition and outdoor hiking, activity intensity: moderate." Based on these preference characteristics, the depth of the study tour content is determined to be "moderate understanding," and the direction of the study tour content is "plant cognition." The study tour experience characteristic is "familiarity level: moderate." Based on this experience characteristic, the mastery level is determined to be "possessing basic botanical knowledge." Based on the content depth of "moderate understanding" and the mastery level of "possessing basic botanical knowledge," the time constraint is determined to be "total duration not exceeding 10 hours." Based on the content depth of "moderate understanding" and the content direction of "plant cognition," the content constraint is determined to be "the study tour content focuses on plant cognition, while also taking into account outdoor practical experience."

[0089] In this embodiment, by determining the depth and direction of a user's learning content for a target learning theme based on learning preference characteristics, the system can decompose the user's learning preferences into content depth and direction, respectively representing the user's expectations for the complexity and category of the learning content. This allows the system to accurately pinpoint the user's content needs from both "depth" and "direction" dimensions. Based on learning experience characteristics, the system determines the user's level of mastery of the target learning theme, quantifying these characteristics as mastery levels and reflecting the user's existing proficiency in the target theme. By determining time constraints based on content depth and mastery, the system can adaptively adjust the time constraints according to the complexity of the learning content and the user's proficiency level, ensuring that the time constraint matches the user's actual ability. Finally, by determining content constraints based on content depth and direction, the system ensures that the content constraints align with the user's interests and their knowledge level, providing precise constraints for subsequent selection of learning products from the learning knowledge graph.

[0090] In one optional embodiment of this specification, based on the study tour profile, the study tour knowledge graph is queried to obtain candidate study tour products and candidate resource configurations, including: Based on the characteristics of study tour experience and study tour preferences, determine the initial activation value of each study tour product node in the study tour product layer; Based on the constraint characteristics, the initial activation value of each study tour product node is propagated along the same layer edge to the adjacent study tour product nodes, so that the activation value of each study tour product node is updated iteratively. Candidate study tour products are determined based on the updated activation values ​​of each study tour product node; The activation value of the candidate study tour product is propagated to the resource allocation layer along the cross-layer edge corresponding to the candidate study tour product, resulting in multiple resource allocation nodes associated with the candidate study tour product. Based on the characteristics of constraints, multiple resource configuration nodes associated with candidate study tour products are screened to obtain candidate resource configurations.

[0091] Study tour product nodes are nodes within the study tour product layer of the study tour knowledge graph, with each node corresponding to a study tour product.

[0092] The initial activation value is the starting activation intensity value of the study tour product node, determined based on the user's study tour profile.

[0093] Adjacent study tour product nodes are other study tour product nodes that are directly connected to the target study tour product node through edges on the same layer.

[0094] Optionally, one way to determine the initial activation value of each study tour product node in the study tour product layer based on study tour experience characteristics and study tour preference characteristics is to combine the study tour experience characteristics and study tour preference characteristics into a feature vector, calculate the similarity between this vector and the attribute vector of each study tour product node, and use the similarity as the initial activation value of each node. Another way is to determine the study tour products that match the preferences based on the study tour preference characteristics, and then perform a weighted adjustment on the matching result based on the study tour experience characteristics to obtain the initial activation value of each node. This specification does not limit the implementation of this method.

[0095] Optionally, based on constraint characteristics, one implementation method for propagating the initial activation value of each study tour product node along the same-layer edges to its neighboring study tour product nodes, so that the activation value of each study tour product node is iteratively updated, is as follows: The initial activation value of each node is used as the first-round activation value. In each iteration, each node sends its current activation value to its neighboring nodes along the same-layer edges. Each node updates its own activation value based on the received activation values ​​from its neighboring nodes, and this process is repeated until convergence. Another implementation method is to use the message passing mechanism of a graph neural network to perform multiple rounds of activation value propagation and aggregation between nodes. After each round of update, it is checked whether the constraint characteristics are satisfied, and the activation value is selectively retained or suppressed based on the constraint verification results. This specification does not limit this implementation method.

[0096] Optionally, one implementation of determining candidate study tour products based on the updated activation values ​​of each study tour product node is to identify one or more study tour product nodes with the highest updated activation values ​​as candidate study tour products. Another implementation is to identify study tour product nodes with updated activation values ​​exceeding a preset threshold as candidate study tour products. This specification does not limit the scope of this implementation.

[0097] The cross-layer edges corresponding to the candidate study tour products are the edges connecting the candidate study tour product nodes and their associated resource configuration nodes.

[0098] The multiple resource configuration nodes associated with the candidate study tour products are each resource configuration node in the resource configuration layer that is connected to the candidate study tour product nodes through cross-layer edges.

[0099] Optionally, one way to filter multiple resource configuration nodes associated with candidate study tour products based on constraint features to obtain candidate resource configurations is to: obtain the attribute information of each resource configuration node, compare the attribute information with the constraint features, and retain the resource configuration nodes that satisfy all constraints. Another way is to: calculate the matching degree between the attributes of each resource configuration node and the constraint features, and select the resource configuration nodes whose matching degree meets preset conditions. This specification does not limit the scope of this method.

[0100] For example, continuing from the previous example, the experience feature in the study tour profile is "Familiarity: Medium," the preference feature is "Content Direction: Plant Recognition and Outdoor Hiking, Activity Intensity: Medium," and the constraint is "Duration not exceeding 10 hours." The study tour product layer of the study tour knowledge graph includes product nodes such as "Botanical Garden Science Popularization Tour," "Forest Park Hiking," and "Geological Museum Visit." Based on the experience and preference features, the initial activation values ​​of the "Botanical Garden Science Popularization Tour" node are calculated to be 0.85, "Forest Park Hiking" to 0.72, and "Geological Museum Visit" to 0.31. Based on the constraint feature, the initial activation values ​​of each node are propagated along the edges of the same layer to adjacent nodes. During the propagation process, the constraint feature verifies the propagation. After multiple rounds of iterative updates, the activation value of the "Botanical Garden Science Popularization Tour" node is updated to 0.91, "Forest Park Hiking" to 0.78, and "Geological Museum Visit" to 0.25. The "Botanical Garden Science Popularization Tour," with the highest activation value, is selected as the candidate study tour product. The activation value of the candidate product is propagated along the cross-layer edge to the resource allocation layer, resulting in multiple resource allocation nodes such as "Professional Botany Lecturer", "Outdoor Hiking Equipment", "Park Shuttle Bus", and "Specimen Preparation Tools". Based on the constraint "duration not exceeding 10 hours", "Specimen Preparation Tools" is excluded (due to its longer duration), resulting in the candidate resource allocation "Professional Botany Lecturer + Outdoor Hiking Equipment + Park Shuttle Bus".

[0101] In this embodiment, the initial activation value of each study tour product node in the study tour product layer is determined based on the study tour experience characteristics and study tour preference characteristics. Based on the constraint characteristics, the initial activation value of each study tour product node is propagated along the same-layer edge to the adjacent study tour product nodes, so that the activation value of each study tour product node is iteratively updated. Candidate study tour products are determined based on the updated activation values ​​of each study tour product node. The activation value of the candidate study tour products is propagated along the cross-layer edge corresponding to the candidate study tour products to the resource allocation layer, resulting in multiple resource allocation nodes associated with the candidate study tour products. Based on the constraint characteristics, the multiple resource allocation nodes associated with the candidate study tour products are filtered to obtain candidate resource allocations. This allows for product association recommendations through same-layer edge propagation and product-to-resource mapping through cross-layer edge propagation. The constraint characteristics act as a gating signal in real time during each iteration in the propagation process and as a filtering signal in the filtering process to filter resource allocation layer nodes, retaining only resource allocation nodes that meet the constraint conditions. Constraint features work synergistically as gating and filtering signals in the propagation and screening stages, respectively, so that candidate study tour products and candidate resource allocation are guided and double-screened by constraints throughout the reasoning stage, ensuring that the final recommendation results meet the user's personalized constraint requirements at both the product and resource levels.

[0102] In one optional embodiment of this specification, based on constraint characteristics, the initial activation value of each study tour product node is propagated along the same-layer edges to iteratively update the activation value of each study tour product node, including: Based on the characteristics of the constraints, target study tour product nodes are selected from each study tour product node; The current activation values ​​of adjacent study tour product nodes connected to each target study tour product node are propagated along the same layer edge to each target study tour product node for aggregation, to obtain the aggregated activation value of each target study tour product node. In the first iteration, the current activation value is the initial activation value. Update the current activation value of the target study tour product node using the aggregated activation value, and return to the step of executing the current activation value of the adjacent study tour product nodes that connect each target study tour product node, until the preset convergence condition is met.

[0103] The target study tour product node is the study tour product node that needs to receive and aggregate the activation values ​​of adjacent nodes in the current iteration round.

[0104] The current activation value is the activation value that the target study tour product node has at the start of the current iteration round.

[0105] The aggregated activation value is the activation value obtained by weighting and summing the current activation values ​​of adjacent nodes of the target study tour product node according to the propagation weight of the same layer edge.

[0106] The convergence condition is the basis for determining when to terminate the iterative update of the activation value of each study tour product node. For example, the convergence condition can be that the change in the activation value of each study tour product node is less than a preset change threshold, or that the number of iterations reaches a preset maximum number of iterations, or a combination of the above two conditions. That is, the iteration is terminated when any one of the conditions is met, so as to balance computational accuracy and computational efficiency.

[0107] Optionally, one way to select target study tour product nodes from various study tour product nodes based on constraint characteristics is to determine study tour product nodes that meet the constraint characteristics as target study tour product nodes, and exclude study tour product nodes that do not meet the constraint characteristics from the propagation process. Another way is to sort each study tour product node from high to low according to the matching degree between the node and the constraint characteristics, and determine the node with the highest matching degree as the target study tour product node. The embodiments in this specification do not limit this approach.

[0108] Optionally, one implementation of aggregating the current activation values ​​of adjacent study tour product nodes connected to each target study tour product node by propagating them along the same-layer edges to each target study tour product node to obtain the aggregated activation value of each target study tour product node can be: obtaining the current activation values ​​of each adjacent node connected to the target node, multiplying each activation value by the corresponding same-layer edge propagation weight parameter, and then summing them to obtain the aggregated activation value. Another implementation can be: obtaining the current activation values ​​of each adjacent node and taking the average to obtain the aggregated activation value. This specification does not limit the implementation to this method.

[0109] Optionally, one way to update the current activation value of the target study tour product node using the aggregated activation value is to directly use the aggregated activation value as the new activation value of the target study tour product node. Another way is to use the aggregated activation value after a non-linear transformation as the new activation value of the target study tour product node. This specification does not limit the implementation of this method in the embodiments.

[0110] For example, continuing from the previous example, based on the constraint "duration not exceeding 10 hours," target study tour product nodes are selected from all study tour product nodes. After constraint verification, "Botanical Garden Science Popularization Tour" and "Forest Park Hiking" both meet the duration constraint and are identified as target study tour product nodes; "Geological Museum Visit" does not meet the duration constraint and is excluded from the propagation process. In the first iteration, the initial activation value of each node is used as the current activation value. The current activation values ​​of the adjacent nodes of "Botanical Garden Science Popularization Tour" (including "Forest Park Hiking," etc.) are propagated along the same layer edge to the "Botanical Garden Science Popularization Tour" node for aggregation, resulting in an aggregated activation value of 0.78. This value is used to update the activation value of "Botanical Garden Science Popularization Tour." At the same time, the current activation values ​​of the adjacent nodes of "Forest Park Hiking" are propagated to the "Forest Park Hiking" node for aggregation, resulting in an aggregated activation value of 0.69. This value is used to update the activation value of "Forest Park Hiking." The next iteration is then executed until the change in the activation value of each node is less than a preset threshold (e.g., 0.01), at which point the iteration ends.

[0111] In this embodiment, target study tour product nodes are selected from each study tour product node based on constraint characteristics. The current activation values ​​of adjacent study tour product nodes connected to each target study tour product node are propagated along the same-layer edge to each target study tour product node for aggregation, resulting in the aggregated activation value of each target study tour product node. The current activation value of the target study tour product node is updated using the aggregated activation value, and execution returns until a preset convergence condition is met. By selecting each study tour product node based on constraint characteristics before each round of iterative propagation, nodes that meet the constraint characteristics are identified as target study tour product nodes and allowed to participate in this round of propagation, while nodes that do not meet the constraint characteristics are excluded from this round of propagation. This ensures that the activation signal is only transmitted between product nodes that match the user's constraints, avoiding deviations in the propagation direction due to the presence of nodes that do not meet the constraints. After the activation value of a target study tour product node is updated along the same-layer edge, execution returns to the next round of selection and propagation, making the selection results dynamically change with the iterative update of the activation values ​​of each node. The update of the node's activation value will change its ranking position in the next round of selection, thereby affecting the set of participating nodes in the next round of propagation. Through dynamic filtering in each iteration, the activation signal is always propagated within the node group that meets the constraints, and the propagation direction is continuously directed towards the product area that best matches the constraints, thus achieving constraint-guided directional propagation and ensuring that the convergence result meets the user's personalized needs.

[0112] In one optional embodiment of this specification, the feedback information includes first feedback information and second feedback information, the parameters of edges at the same layer are propagation weight parameters, and the parameters of edges across layers are mapping weight parameters. Step 108 includes the following specific steps: Receive first feedback from users regarding candidate study tour products, and / or second feedback regarding the configuration of candidate resources; Based on the first feedback information and the research profile, adjust the dissemination weight parameters, and / or, based on the second feedback information and the research profile, adjust the mapping weight parameters; Based on the adjusted parameters of edges within the same layer and edges across layers, the candidate study tour products and candidate resource configurations are updated to obtain the target study tour products and target resource configurations, including: Based on the adjusted propagation weight parameters, the candidate study tour products are updated to obtain the target study tour products, and / or, based on the adjusted mapping weight parameters, the candidate resource configuration is updated to obtain the target resource configuration.

[0113] The first feedback information consists of user evaluations of the candidate study tour products, representing user satisfaction with these products. For example, the first feedback information could be a user rating or satisfaction level for the candidate study tour product. The second feedback information consists of user evaluations of the candidate resource configurations, representing user satisfaction with these resource configurations. For example, the second feedback information could be a user rating or satisfaction level for the candidate resource configurations.

[0114] Optionally, one way to adjust the propagation weight parameters based on the first feedback information and the research profile is to determine the adjustment direction of the propagation weight parameters based on the first feedback information and the adjustment magnitude based on the research profile. Another way is to input the first feedback information and the research profile into the parameter adjustment model, and then output the adjusted propagation weight parameters from the model. This specification does not limit the implementation of this method.

[0115] Similarly, the implementation method of adjusting the mapping weight parameters based on the second feedback information and the research profile can refer to the implementation method of adjusting the propagation weight parameters based on the first feedback information and the research profile, which will not be elaborated here.

[0116] Optionally, one way to update candidate study tour products based on the adjusted propagation weight parameters to obtain the target study tour product is to re-execute the same-layer edge propagation process using the adjusted propagation weight parameters and determine the target study tour product based on the recalculated activation values. Another way is to recalculate the scores of each study tour product node using the adjusted propagation weight parameters and determine the target study tour product based on the recalculated scores. This specification does not limit the implementation of this method.

[0117] Similarly, based on the adjusted mapping weight parameters, the candidate resource configuration is updated to obtain the implementation method of the target resource configuration. This can be referenced from the implementation method of updating the candidate study tour products based on the adjusted propagation weight parameters to obtain the implementation method of the target study tour product, which will not be elaborated here.

[0118] For example, continuing from the previous example, the user provides positive feedback on the candidate study tour product "Botanical Garden Science Popularization Tour" and general feedback on the candidate resource configuration "Professional Botany Lecturer + Outdoor Hiking Equipment + Park Shuttle Bus". Based on the first feedback information and the study tour profile, the propagation weight parameter of the same-layer edge between "Botanical Garden Science Popularization Tour" and adjacent product nodes is adjusted, with positive feedback increasing the propagation weight parameter. Based on the second feedback information and the study tour profile, the mapping weight parameter of the cross-layer edge between "Botanical Garden Science Popularization Tour" and "Professional Botany Lecturer" is adjusted, with general feedback moderately decreasing the mapping weight parameter. The candidate study tour products are updated again using the adjusted propagation weight parameters. The activation value of "Botanical Garden Science Popularization Tour" is maintained, and it is still identified as the target study tour product. The candidate resource configuration is updated again using the adjusted mapping weight parameters. Due to the moderate decrease in the mapping weight parameter, the matching degree of "Professional Botany Lecturer" is slightly reduced, but after re-evaluation, it is still the optimal configuration and is still identified as the target resource configuration.

[0119] In this embodiment, by receiving first feedback information from users regarding candidate study tour products and second feedback information regarding candidate resource configurations; adjusting propagation weight parameters based on the first feedback information and study tour profiles, and adjusting mapping weight parameters based on the second feedback information and study tour profiles; updating candidate study tour products based on the adjusted propagation weight parameters to obtain target study tour products, and updating candidate resource configurations based on the adjusted mapping weight parameters to obtain target resource configurations, the user's feedback information can be applied to both the product-level reasoning and resource-level mapping stages. The first feedback information is used to adjust the propagation weights between products, and the second feedback information is used to adjust the mapping weights between products and resources, achieving decoupled parameter optimization. This ensures that the adjustment of propagation weight parameters reflects both the user's immediate evaluation of the product and is constrained by the user's existing experience and interests, avoiding deviations in product recommendation direction from the user's true study tour expectations due to a single feedback signal; and that the adjustment of mapping weight parameters reflects both the user's evaluation of resource configurations and is constrained by the user's own time and content constraints, avoiding deviations in resource configuration matching from the user's objective limitations due to a single feedback signal.

[0120] In an optional embodiment of this specification, the dissemination weight parameters are adjusted based on the first feedback information and the research profile, including: Calculate the first correlation degree between the first feedback information and the characteristics of study tour experience and study tour preference; Adjust the propagation weight parameters based on the first degree of relevance; Based on the second feedback information and the research profile, the mapping weight parameters are adjusted, including: Calculate the second degree of correlation between the second feedback information and the constraint characteristics; Adjust the mapping weight parameters based on the second degree of correlation.

[0121] The first correlation degree is a quantitative indicator of the consistency or matching degree between the first feedback information and the characteristics of study tour experience and study tour preference.

[0122] The second correlation degree is a quantitative indicator of the consistency or matching degree between the second feedback information and the characteristics of the constraint conditions.

[0123] Optionally, one way to calculate the first correlation between the first feedback information and the study experience features and study preference features is to convert the first feedback information, study experience features, and study preference features into vector forms respectively, calculate the cosine similarity between the vectors, and use this similarity as the first correlation. Another way is to extract the key evaluation elements from the first feedback information, count the number of matches between the key evaluation elements and the study experience features and study preference features, and calculate the correlation based on the number of matches. This specification does not limit the implementation of this method.

[0124] Optionally, one way to adjust the propagation weight parameter based on the first correlation degree is to: preset a correlation degree threshold; increase the propagation weight parameter when the first correlation degree is greater than the threshold, and decrease the propagation weight parameter when the first correlation degree is less than the threshold, with the adjustment magnitude being positively correlated with the degree to which the correlation degree deviates from the threshold. Another way to implement this is to: map the first correlation degree to the rate of change of the propagation weight parameter, and update the propagation weight parameter according to the rate of change. This specification does not limit the implementation of this method in the embodiments.

[0125] For example, continuing from the previous example, the user gave a positive feedback of 4.8 points (out of 5) to the candidate study tour product "Botanical Garden Science Popularization Tour". The first feedback information is converted into a feedback vector, and the cosine similarity between it and the study tour experience feature vector and the study tour preference feature vector is calculated. The weighted average yields a first correlation coefficient of 0.92 (out of 1.0), indicating that the user's positive feedback is highly consistent with its experience and preference features. The preset correlation coefficient threshold is 0.7. The first correlation coefficient of 0.92 is greater than this threshold, so the propagation weight parameter of the same-layer edge between "Botanical Garden Science Popularization Tour" and its adjacent product nodes is increased by +20%. The user gave a feedback of 2.5 points (out of 5) to the candidate resource configuration "Professional Botany Lecturer + Outdoor Hiking Equipment + Park Shuttle Bus". The second feedback information is converted into a vector, and the cosine similarity between it and the constraint feature vector is calculated, yielding a second correlation coefficient of 0.45. Since the second degree of correlation is less than the threshold of 0.7, the mapping weight parameter of the cross-layer edge between "Botanical Garden Science Popularization Tour" and "Professional Botany Lecturer" is reduced by -15%.

[0126] In this embodiment, a first correlation degree is calculated between the first feedback information and the characteristics of study tour experience and study tour preference, and the propagation weight parameter is adjusted based on the first correlation degree; a second correlation degree is calculated between the second feedback information and the constraint condition characteristics, and the mapping weight parameter is adjusted based on the second correlation degree. This allows for joint analysis of the first feedback information with the characteristics of study tour experience and study tour preference, ensuring that the adjustment of the propagation weight parameter simultaneously considers the user's satisfaction with the current recommendation result as well as the user's existing experience level and interests. Similarly, by jointly analyzing the second feedback information with the constraint condition characteristics in the study tour profile, the adjustment of the mapping weight parameter simultaneously considers the user's satisfaction with the current resource allocation as well as the user's own time and content constraints. By jointly analyzing the feedback information with the study tour profile, the parameter adjustment reflects both the user's immediate evaluation and is constrained by the user's objective profile characteristics, avoiding deviations from the user's actual needs due to subjective user feedback alone.

[0127] Corresponding to the aforementioned method for jointly matching study tour products and resources, this specification also provides a system for jointly matching study tour products and resources in its embodiments. (Illustratively,) Figure 5 This specification illustrates a structural block diagram of a joint matching system for study tour products and resources, as provided in an embodiment. The system includes: The acquisition module 502 is configured to acquire the number of times a user has participated in a study tour and their study tour needs related to the target study tour topic; Module 504 is configured to build a user profile for the target study tour theme based on the number of study tours and study tour needs. The query module 506 is configured to query the study tour knowledge graph based on the study tour profile to obtain candidate study tour products and candidate resource configurations. The study tour knowledge graph includes a study tour product layer and a resource configuration layer. The nodes in the study tour product layer are study tour products, and the nodes in the resource configuration layer are resource configurations. There are same-layer edges between each study tour product and cross-layer edges between any study tour product and each resource configuration. The update module 508 is configured to receive user feedback on candidate study tour products and candidate resource configurations, and adjust the parameters of the same-layer edge and the cross-layer edge based on the feedback information and study tour profile. Based on the adjusted parameters of the same-layer edges and cross-layer edges, the candidate study tour products and candidate resource configurations are updated to obtain the target study tour products and target resource configurations.

[0128] In this embodiment, the following steps are taken: First, the number of times a user has participated in a study tour on the target study tour theme and their study tour needs are obtained. Based on these numbers and needs, a study tour profile of the user on the target study tour theme is constructed. Then, based on this profile, a study tour knowledge graph is queried to obtain candidate study tour products and candidate resource configurations. The study tour knowledge graph includes a study tour product layer and a resource configuration layer. Nodes in the study tour product layer represent study tour products, and nodes in the resource configuration layer represent resource configurations. There are same-layer edges between study tour products and cross-layer edges between any study tour product and any resource configuration. Feedback from the user on the candidate study tour products and candidate resource configurations is received. Based on the feedback and the study tour profile, the parameters of the same-layer edges and the parameters of the cross-layer edges are adjusted. Finally, based on the adjusted parameters of the same-layer edges and the cross-layer edges, the candidate study tour products and candidate resource configurations are updated to obtain the target study tour product and target resource configuration. This system leverages the number of study tours to represent users' existing experience with the target study tour theme and their differentiated needs to represent their specific requirements. By constructing a study tour profile that includes both experiential and preference features, it integrates users' existing experience with their differentiated needs. Then, utilizing the two-layer structure of a study tour knowledge graph, it achieves precise selection of study tour products and differentiated matching of resource allocation based on the integrated profile. Simultaneously, it optimizes the graph parameters through user feedback, achieving a closed-loop improvement in recommendation results. This avoids the problem of users only expressing macro-level needs due to insufficient experience, failing to address their deeper study tour expectations. It also avoids resource waste or inadequacy caused by ignoring individual user differences in resource allocation. This ensures that study tour products and resource allocation are precisely adapted to users' actual experience, preferences, and constraints, improving the user's study tour experience and effectiveness, thereby achieving synergistic optimization of study tour resource utilization efficiency and user satisfaction.

[0129] The above is an illustrative scheme of a joint matching system for study tour products and resources according to this embodiment. It should be noted that the technical solution of this joint matching system for study tour products and resources belongs to the same concept as the technical solution of the aforementioned joint matching method for study tour products and resources. Details not described in detail in the technical solution of the joint matching system for study tour products and resources can be found in the description of the technical solution of the aforementioned joint matching method for study tour products and resources. Furthermore, the components in the embodiment of the joint matching system for study tour products and resources should be understood as functional modules necessary to implement each step of the program flow or each step of the method; these functional modules are not actual functional divisions or separations. The claims of the joint matching system for study tour products and resources defined by such a set of functional modules should be understood as a functional module architecture, and not as a physical client primarily implemented through hardware.

[0130] Figure 6A structural block diagram of a computing device according to one embodiment of this specification is shown. The components of the computing device 600 include, but are not limited to, a memory 610 and a processor 620. The processor 620 is connected to the memory 610 via a bus 630, and a content database 650 is used to store content data.

[0131] The computing device 600 also includes an access device 640, which enables the computing device 600 to communicate via one or more networks 660. Examples of these networks include a Public Switched Telephone Network (PSTN), a Local Area Network (LAN), a Wide Area Network (WAN), a Personal Area Network (PAN), or a combination of communication networks such as the Internet. The access device 640 may include one or more of any type of wired or wireless network interface (e.g., a Network Interface Controller (NIC)), such as an IEEE 802.11 Wireless Local Area Network (WLAN) interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, or a Near Field Communication (NFC) interface.

[0132] In one embodiment of this specification, the above-described components of the computing device 600 and Figure 6 Other components, not shown, can also be connected to each other, for example, via a bus. It should be understood that... Figure 6 The block diagram of the computing device shown is for illustrative purposes only and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0133] The computing device 600 can be any type of stationary or mobile computing device, including mobile computers or mobile computing devices (e.g., tablet computers, personal digital assistants, laptop computers, content-based computers, netbooks, etc.), mobile phones (e.g., smartphones), wearable computing devices (e.g., smartwatches, smart glasses, etc.) or other types of mobile devices, or stationary computing devices such as desktop computers or personal computers (PCs). The computing device 600 can also be a mobile or stationary server.

[0134] The processor 620 is used to execute the following computer program / instructions, which, when executed by the processor, implement the steps of the above-mentioned method for jointly matching research and study products and resources.

[0135] The above is an illustrative scheme of a computing device according to this embodiment. It should be noted that the technical solution of this computing device and the technical solution of the above-mentioned method for jointly matching research and study products and resources belong to the same concept. For details not described in detail in the technical solution of the computing device, please refer to the description of the technical solution of the above-mentioned method for jointly matching research and study products and resources.

[0136] An embodiment of this specification also provides a computer-readable storage medium storing a computer program / instructions that, when executed by a processor, implement the steps of the above-described method for jointly matching research and study products and resources.

[0137] The above is an illustrative scheme of a computer-readable storage medium according to this embodiment. It should be noted that the technical solution of this storage medium belongs to the same concept as the technical solution of the above-described method for jointly matching research and learning products and resources. For details not described in detail in the technical solution of the storage medium, please refer to the description of the technical solution of the above-described method for jointly matching research and learning products and resources.

[0138] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0139] The computer instructions include computer program code, which may be in the form of source code, object code, executable file, or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium may be appropriately added or removed according to the requirements of patent practice. For example, in some regions, according to patent practice, computer-readable media may not include electrical carrier signals and telecommunication signals.

[0140] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments in this specification are not limited to the described order of actions, because according to the embodiments in this specification, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the embodiments in this specification.

[0141] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0142] The preferred embodiments disclosed above are merely illustrative of this specification. The optional embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the embodiments described herein. These embodiments are selected and specifically described in this specification to better explain the principles and practical applications of the embodiments, thereby enabling those skilled in the art to better understand and utilize this specification. This specification is limited only by the claims and their full scope and equivalents.

Claims

1. A method for jointly matching study tour products and resources, characterized in that, include: Obtain the number of study tours and the study tour needs of users regarding the target study tour topic; Based on the number of study tours and study tour needs, construct a study tour profile for the user regarding the target study tour theme; Based on the study tour profile, the study tour knowledge graph is queried to obtain candidate study tour products and candidate resource configurations. The study tour knowledge graph includes a study tour product layer and a resource configuration layer. The nodes of the study tour product layer are study tour products, and the nodes of the resource configuration layer are resource configurations. There are same-layer edges between each study tour product and cross-layer edges between any study tour product and each resource configuration. Receive feedback from the user regarding the candidate study tour products and the candidate resource configurations, and adjust the parameters of the same-layer edges and the cross-layer edges based on the feedback and the study tour profile; Based on the adjusted parameters of the same-layer edges and cross-layer edges, the candidate study tour products and candidate resource configurations are updated to obtain the target study tour products and target resource configurations.

2. The method according to claim 1, characterized in that, The study tour profile includes study tour experience characteristics, study tour preference characteristics, and constraint condition characteristics; The process of constructing a user's study tour profile for the target study tour theme based on the number of study tours and study tour needs includes: The study tour experience characteristics are determined based on the number of study tours, and the study tour preference characteristics are determined based on the study tour needs. Based on the characteristics of the study tour experience and the characteristics of the study tour preference, the characteristics of the constraint conditions are determined; Based on the study tour experience characteristics, study tour preference characteristics, and constraint characteristics, a study tour profile of the user regarding the target study tour topic is constructed.

3. The method according to claim 2, characterized in that, The constraints include duration constraints and content constraints; The determination of the constraint characteristics based on the study tour experience characteristics and the study tour preference characteristics includes: Based on the aforementioned study tour preference characteristics, the depth and direction of the study tour content for the target study tour theme are determined. Based on the characteristics of the study tour experience, determine the user's level of mastery of the target study tour topic; The duration constraint is determined based on the depth of the research and study content and the level of mastery. The content constraints are determined based on the depth and direction of the research and study content.

4. The method according to claim 2, characterized in that, The step of querying the study tour knowledge graph based on the study tour profile to obtain candidate study tour products and candidate resource configurations includes: Based on the study tour experience characteristics and study tour preference characteristics, determine the initial activation value of each study tour product node in the study tour product layer; Based on the aforementioned constraint characteristics, the initial activation value of each study tour product node is propagated along the same layer edge to the adjacent study tour product nodes, so that the activation value of each study tour product node is iteratively updated. The candidate study tour products are determined based on the updated activation values ​​of each study tour product node; The activation value of the candidate study tour product is propagated along the cross-layer edge corresponding to the candidate study tour product to the resource configuration layer, thereby obtaining multiple resource configuration nodes associated with the candidate study tour product; Based on the aforementioned constraint characteristics, multiple resource configuration nodes associated with the candidate study tour products are filtered to obtain the candidate resource configurations.

5. The method according to claim 4, characterized in that, Based on the aforementioned constraint characteristics, the initial activation value of each study tour product node is propagated along the same-layer edges to iteratively update the activation value of each study tour product node, including: Based on the aforementioned constraint characteristics, target study tour product nodes are selected from each study tour product node; The current activation values ​​of adjacent study tour product nodes connected to each target study tour product node are propagated along the same layer edge to each target study tour product node for aggregation, to obtain the aggregated activation value of each target study tour product node, wherein the current activation value in the first iteration is the initial activation value; The current activation value of the target study tour product node is updated using the aggregated activation value. The process then returns to the step of executing the current activation value of the adjacent study tour product nodes that connect each target study tour product node, until the preset convergence condition is met.

6. The method according to any one of claims 2-5, characterized in that, The feedback information includes first feedback information and second feedback information, the parameters of the same-layer edge are propagation weight parameters, and the parameters of the cross-layer edge are mapping weight parameters; The step of receiving feedback information from the user regarding the configuration of the candidate study tour products and the candidate resources, and adjusting the parameters of the same-layer edges and the cross-layer edges based on the feedback information and the study tour profile, includes: Receive the user's first feedback information on the candidate study tour products, and / or the second feedback information on the configuration of the candidate resources; Based on the first feedback information and the research profile, the dissemination weight parameter is adjusted, and / or, based on the second feedback information and the research profile, the mapping weight parameter is adjusted; The process of updating the candidate study tour products and candidate resource configurations based on the adjusted parameters of the same-layer edges and cross-layer edges to obtain the target study tour products and target resource configurations includes: Based on the adjusted propagation weight parameters, the candidate study tour products are updated to obtain the target study tour products, and / or, based on the adjusted mapping weight parameters, the candidate resource configuration is updated to obtain the target resource configuration.

7. The method according to claim 6, characterized in that, The step of adjusting the dissemination weight parameters based on the first feedback information and the research profile includes: Calculate the first correlation degree between the first feedback information and the research experience characteristics and research preference characteristics; The propagation weight parameter is adjusted based on the first correlation degree; The step of adjusting the mapping weight parameters based on the second feedback information and the research profile includes: Calculate the second correlation degree between the second feedback information and the constraint feature; The mapping weight parameters are adjusted based on the second correlation.

8. A joint matching system for study tour products and resources, characterized in that, include: The acquisition module is configured to acquire the number of times a user has participated in a study tour and their study tour needs related to the target study tour topic; The construction module is configured to build a study profile of the user about the target study theme based on the number of study tours and study tour needs; The query module is configured to query the study tour knowledge graph based on the study tour profile to obtain candidate study tour products and candidate resource configurations. The study tour knowledge graph includes a study tour product layer and a resource configuration layer. The nodes of the study tour product layer are study tour products, and the nodes of the resource configuration layer are resource configurations. There are same-layer edges between each study tour product and cross-layer edges between any study tour product and each resource configuration. The update module is configured to receive feedback information from the user regarding the candidate study tour products and the candidate resource configurations, and adjust the parameters of the same-layer edge and the cross-layer edge based on the feedback information and the study tour profile. Based on the adjusted parameters of the same-layer edges and cross-layer edges, the candidate study tour products and candidate resource configurations are updated to obtain the target study tour products and target resource configurations.

9. A computing device, characterized in that, include: Memory and processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions, which, when executed by the processor, implement the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores a computer program / instructions that, when executed by a processor, implement the steps of the method according to any one of claims 1 to 7.

Images