Consumer behavior analysis method and system for tourism marketing

Abstract

This invention relates to the field of big data analytics, and discloses a method and system for analyzing consumer behavior in cultural tourism marketing. The invention quantifies the impact of context by introducing context intensity and duration, and dynamically adjusts user preference weights accordingly. This allows the system to more sensitively capture and respond to the user's real needs in the current context, avoiding erroneous recommendations caused by over-reliance on long-term preferences. Furthermore, the progressive weight recovery mechanism effectively avoids the problem of over-correction that may occur after the context ends, ensuring the consistency and stability of recommendations. Therefore, this invention effectively solves the problem in existing cultural tourism marketing systems where the recommendation decision-making mechanism becomes rigid and unable to accurately respond to the user's immediate needs when the user's travel context changes temporarily; it significantly improves the accuracy of cultural tourism marketing recommendations and user satisfaction, optimizes user experience, and improves the utilization efficiency of marketing resources.

Description

Technical Field

[0001] This invention relates to the field of big data analytics, and in particular to a consumer behavior analysis method and system for cultural tourism marketing. Background Technology

[0002] In the field of cultural tourism marketing, accurately grasping consumer behavior patterns and changes in interests, and formulating effective marketing strategies accordingly, is an important issue. Related technologies typically involve deploying a system to collect and analyze various information about tourists to depict their interests and provide personalized information delivery.

[0003] However, there are significant limitations when dealing with temporary changes in consumers' travel contexts. For example, if a user long identified as a history and culture enthusiast suddenly purchases a family package and books a family-themed hotel, this clearly indicates a shift in their travel purpose from individual in-depth travel to family travel. Although the backend data integration portion of existing systems can capture and correlate these contextual changes, the internal judgment mechanism in the recommendation decision-making process often assigns higher weight to long-accumulated preference information, assuming that long-term interests better represent the user's true preferences, while lacking sensitivity to immediate contextual changes. Therefore, even if the system knows the user is currently in a family-oriented context, when generating recommended content, it may still prioritize pushing information such as academic lectures that are inconsistent with the current family needs based on the user's history and culture enthusiast tag. Such erroneous recommendations not only fail to attract users but may also cause interference, leading to user dissatisfaction and even the closure of app notification permissions.

[0004] Furthermore, existing systems may overcorrect upon receiving negative feedback. For example, in the scenario described above, the system might simply treat family travel as a strong signal, drastically reducing the weight of historical and cultural tags while excessively increasing the weight of all tags related to parents and children. This could lead to the system completely ignoring adults' cultural needs during the user's next family trip, instead pushing purely children's entertainment programs. This either-or recommendation strategy prevents the system from satisfying children's entertainment needs while also considering adults' interests, making it difficult to provide truly comprehensive and balanced marketing services that meet the complexities and diverse needs of real-world scenarios. This rigid recommendation logic not only wastes marketing resources but may also lead to user churn due to continuous erroneous recommendations. Summary of the Invention

[0005] The technical problem solved by this invention is to provide a consumer behavior analysis method and system for cultural tourism marketing, in order to solve the problems of existing cultural tourism marketing systems having rigid recommendation decision-making mechanisms when dealing with temporary changes in consumers' travel scenarios, failing to accurately identify and respond to users' immediate needs, and potentially over-correcting after receiving negative feedback, resulting in recommended content that does not match users' actual needs, thereby affecting user experience and marketing effectiveness.

[0006] In a first aspect, the present invention provides a consumer behavior analysis method for cultural tourism marketing, comprising: Acquire user behavior data and identify real-time contextual events that indicate a change in the user's travel context based on the user behavior data; the user behavior data includes ticket purchase records and hotel booking information; Based on the real-time situational events, generate a situation intensity value and a situation duration that characterize the degree of situational impact. Based on the intensity value of the context and the duration of the context, the weight of the user's long-term accumulated preference information in the recommendation decision is adjusted. Based on the adjusted weights and the recommendation decision, marketing recommendation content is generated; After the immediate contextual event ends, the weights are gradually restored to their original weights.

[0007] In one embodiment, the identification of immediate contextual events indicating a change in the user's travel context based on the user behavior data includes: Based on the ticket purchase records, identify ticket combinations that include children's tickets, family packages, or business packages; Based on the hotel booking information, identify family-themed, family room, or business room types; The real-time contextual events are structured data packets that include event type, associated geographic location information, context category, and user identifier; wherein, the event type includes ticket purchase behavior and hotel booking, and the context category includes parent-child context and business context.

[0008] In one embodiment, generating a situation intensity value and situation duration characterizing the degree of situational impact based on the immediate situational event includes: The situation intensity value is calculated by weighting the ticket types and quantities in the ticket package and the hotel type and room type characteristics in the hotel booking information, combined with the preset situation intensity assessment rules. The duration of the scenario is determined based on the travel duration in the ticket purchase record and / or the hotel booking information.

[0009] In one embodiment, adjusting the weight of the user's long-term accumulated preference information in the recommendation decision based on the context intensity value and the context duration includes: Calculate the situation influence factor based on the situation intensity value and the situation duration; When the value of the context influence factor is greater than or equal to the preset influence threshold, a first inhibition coefficient is used, and the weight of the user's long-term accumulated preference information in the recommendation decision is adjusted based on the first inhibition coefficient, the preset weight of the user's long-term accumulated preference information, and the context influence factor. If the value of the contextual influence factor is less than the preset influence threshold, a second inhibition coefficient is used. Based on the second inhibition coefficient, the preset weight of the user's long-term accumulated preference information, and the contextual influence factor, the weight of the user's long-term accumulated preference information in the recommendation decision is adjusted; wherein, the second inhibition coefficient is less than the first inhibition coefficient.

[0010] In one embodiment, generating marketing recommendation content based on the adjusted weights and the recommendation decision includes: Based on the adjusted weights and the recommendation decision, the overall matching degree of each candidate cultural and tourism product is calculated. Based on the overall matching degree of each candidate cultural and tourism product, at least one candidate cultural and tourism product with a high overall matching degree is generated.

[0011] In one embodiment, the step of progressively restoring the weights to their original weights after the immediate contextual event ends includes: Once the immediate contextual event is detected to have ended, the weights of the user's long-term accumulated preference information are gradually restored to their original weights before adjustment within a preset recovery period.

[0012] In one embodiment, detecting the end of the immediate contextual event includes: The location service detects that the user device has left the area corresponding to the associated geographic location information recorded in the real-time contextual event; and / or The system detects that the current time has reached or exceeded the ticket validity period or hotel stay end time determined based on the aforementioned real-time contextual event.

[0013] In one embodiment, obtaining user behavior data includes: With the user's authorization, new transaction records generated by the user on online travel agency platforms, ticketing systems, or hotel booking systems are obtained. The user behavior data stream in the new transaction record is identified based on preset rules and pattern matching logic.

[0014] On the other hand, the present invention also provides a consumer behavior analysis system for cultural tourism marketing, the system comprising: The data acquisition module is used to acquire user behavior data and identify real-time contextual events that indicate a change in the user's travel context based on the user behavior data; the user behavior data includes ticket purchase records and hotel booking information. The context impact analysis module is used to generate a context intensity value and a context duration that characterize the degree of context impact based on the real-time context event. The first weight adjustment and recovery module is used to adjust the weight of the user's long-term accumulated preference information in the recommendation decision based on the context intensity value and the context duration. The marketing content generation module is used to generate marketing recommendation content based on the adjusted weights and the recommendation decision; The second weight adjustment and recovery module is used to gradually restore the weight to its original weight after the immediate contextual event ends.

[0015] On the other hand, the present invention also provides a computer-readable storage medium storing computer instructions for causing a processor to execute the consumer behavior analysis method for cultural tourism marketing as described above.

[0016] In summary, the consumer behavior analysis method and system for cultural tourism marketing provided by this invention quantifies the impact of context by introducing context intensity and duration, and dynamically adjusts user preference weights accordingly. This allows the system to more sensitively capture and respond to the user's real needs in the current context, avoiding erroneous recommendations caused by over-reliance on long-term preferences. Furthermore, the progressive weight recovery mechanism effectively avoids the problem of over-correction that may occur after the context ends, ensuring the consistency and stability of recommendations. Therefore, this invention effectively solves the problem in existing technologies where the recommendation decision-making mechanism of cultural tourism marketing systems becomes rigid and unable to accurately respond to users' immediate needs when users' travel contexts temporarily change; it significantly improves the accuracy of cultural tourism marketing recommendations and user satisfaction, optimizes user experience, and improves the utilization efficiency of marketing resources. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1This is a flowchart illustrating a consumer behavior analysis method for cultural tourism marketing according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating a consumer behavior analysis method for cultural tourism marketing according to another embodiment of the present invention. Figure 3 This is a flowchart illustrating a consumer behavior analysis method for cultural tourism marketing according to another embodiment of the present invention. Figure 4 This is a flowchart illustrating a consumer behavior analysis method for cultural tourism marketing according to another embodiment of the present invention. Figure 5 This is a flowchart illustrating a consumer behavior analysis method for cultural tourism marketing according to another embodiment of the present invention. Figure 6 This is a schematic diagram of the structure of a consumer behavior analysis system for cultural tourism marketing according to an embodiment of the present invention; Figure 7 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] In the description of this invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0021] In the field of cultural tourism marketing, traditional consumer behavior analysis systems have significant limitations when dealing with temporary changes in consumers' travel scenarios. For example, when a user's travel purpose changes from individual in-depth travel to family travel, existing systems still prioritize recommendations based on their long-accumulated preference information, leading to recommendations that are inconsistent with the current context and causing user dissatisfaction. Furthermore, after receiving negative feedback, the learning mechanism of existing systems may over-correct, resulting in rigid recommendation strategies that fail to address the diverse needs of users in complex situations.

[0022] The following is combined Figures 1 to 7 The following describes embodiments of the present invention.

[0023] According to embodiments of the present invention, such as Figure 1 As shown, on the one hand, a consumer behavior analysis method for cultural tourism marketing is provided, including the following steps: Step S100: Obtain user behavior data and identify real-time contextual events that indicate a change in the user's travel context based on the user behavior data; user behavior data includes ticket purchase records and hotel booking information; Step S200: Based on real-time situational events, generate a situational intensity value and situational duration that characterize the degree of situational impact. Step S300: Based on the context intensity value and the duration of the context, adjust the weight of the user's long-term accumulated preference information in the recommendation decision; Step S400: Based on the adjusted weights and recommendation decisions, generate marketing recommendation content; Step S500: After the immediate contextual event ends, gradually restore the weights to their original weights.

[0024] In this embodiment, user behavior data refers to various recordable information generated by users during cultural and tourism consumption, such as ticket purchase records, hotel booking information, browsing history, search queries, and social media interactions. This data reflects users' interests, preferences, consumption habits, and travel patterns. Immediate contextual events refer to specific behaviors or events that indicate a change in a user's travel context, such as purchasing family packages or booking family-themed hotels. These events typically possess immediacy in time and specificity in context, reflecting the user's current or upcoming specific travel needs. Context strength is a quantitative indicator characterizing the degree of influence of immediate contextual events on user recommendation decisions; a higher value indicates a greater contextual impact. Context duration refers to the expected duration of the context state corresponding to the immediate contextual event. Preference information refers to data accumulated by users over a long period that reflects their stable interests and preferences, such as historical consumption records and long-term favorite cultural and tourism themes. Recommendation decision refers to the process by which the system generates personalized marketing recommendations for users based on user preferences and contextual information. Original weight refers to the default weight of the user's long-term accumulated preference information in the recommendation decision without the influence of immediate contextual events.

[0025] User behavior data can be captured in various ways, such as through transaction records generated by users on online travel agency platforms, ticketing systems, or hotel booking systems. After acquiring user behavior data, real-time contextual events can be identified from the user behavior data stream based on preset rules and pattern matching logic. For example, when the system detects that a user has purchased a specific type of ticket package, such as a family package including a child's ticket, or booked a hotel with a specific theme, such as a family-themed hotel, these behaviors can be identified as real-time contextual events.

[0026] Next, the identified immediate contextual events are analyzed to generate context intensity values ​​and context duration, which characterize the degree of contextual impact. The context intensity value can be evaluated based on various factors, such as the type of immediate contextual event, the number of users involved, and the amount spent. For example, the context intensity value of a family ticket purchase including multiple children's tickets may be higher than that of a single business ticket purchase. The context duration can be determined based on the characteristics of the event itself, such as the trip duration in the ticket purchase record or the accommodation duration in the hotel booking information.

[0027] Then, based on the context intensity value and context duration, the weight of the user's long-term accumulated preference information in the recommendation decision is adjusted. When no immediate contextual event occurs, the user's long-term accumulated preference information has a preset original weight in the recommendation decision. When an immediate contextual event occurs, the system calculates a contextual influence factor based on the context intensity value and context duration. This contextual influence factor reflects the degree of temporary influence of the current context on the user's preferences. For example, when the contextual influence factor is high, it means that the current context has a greater impact on the user's decision. In this case, a larger inhibition coefficient can be used to reduce the weight of long-term preference information, thus making the recommended content more likely to match the current context. Conversely, when the contextual influence factor is low, a smaller inhibition coefficient can be used to avoid excessive deviation from the user's long-term preferences.

[0028] After adjusting the weights, marketing recommendations are generated based on the adjusted weights and recommendation decisions. When generating these recommendations, the system comprehensively considers the adjusted weights of long-term preferences and the current needs reflected in immediate contextual events. For example, if a user is currently in a family-oriented situation, even if their long-term preference is history and culture, the system will prioritize recommending cultural and tourism products suitable for family activities, such as children's playgrounds and family-friendly hotel packages. Simultaneously, the system can also calculate the overall matching degree of each candidate cultural and tourism product based on the adjusted weights, prioritizing the output of products with high overall matching degrees, or generating a combination solution containing multiple highly matched products.

[0029] Finally, after the immediate contextual event ends, the weights are gradually restored to their original values. This gradual restoration mechanism is to prevent the system from immediately reverting to the original weights after the context ends, which could cause the recommended content to become disconnected from the user's actual needs again. For example, when the system detects that the user's device has left the area corresponding to the associated geographical location information recorded in the immediate contextual event, or when the current time reaches or exceeds the ticket validity period or hotel stay end time determined based on the immediate contextual event, the system will initiate the weight restoration process. Within a preset restoration period, the weights of the user's long-term accumulated preference information will gradually increase until they are restored to their original weights before adjustment.

[0030] This invention introduces the identification and analysis of immediate contextual events, along with a dynamic weighting mechanism based on contextual intensity and duration. Traditional tourism marketing systems often rely excessively on long-term accumulated user preference information when faced with temporary changes in a user's travel context, leading to a disconnect between recommended content and the user's immediate needs. For example, a user with a long-standing interest in history and culture might still receive recommendations for history lectures instead of family activities during a family trip. This rigid recommendation strategy not only fails to attract users but may also cause disruption.

[0031] This invention captures user behavior data and identifies immediate contextual events, such as family packages in ticket purchase records or family-themed room types in hotel booking information, to accurately perceive temporary changes in a user's travel context. Furthermore, by analyzing these immediate contextual events and generating contextual intensity and duration values, this invention quantifies the degree and duration of a context's impact on user preferences. Based on this, the system dynamically adjusts the weight of long-term accumulated user preference information in recommendation decisions according to the contextual intensity and duration values, enabling recommended content to more accurately match the user's current immediate contextual needs. For example, in a family-oriented context, even if a user has a long-standing preference for history and culture, the system will temporarily reduce the weight of history and culture preferences and increase the weight of family-oriented activity preferences, thereby prioritizing recommendations of cultural and tourism products suitable for family travel.

[0032] Furthermore, this invention introduces a mechanism to gradually restore weights to their original values ​​after the immediate contextual event ends. This contrasts sharply with the over-correction problem that may occur in existing systems after receiving negative feedback. Existing systems may simply and drastically reduce the weight of a particular preference, causing subsequent recommendations to completely ignore other aspects of the user's needs. The gradual restoration mechanism of this invention ensures that long-term user preferences can smoothly resume their effectiveness after the context ends, avoiding excessive fluctuations in the recommendation strategy and thus providing a more balanced and continuous personalized marketing service. Therefore, this invention significantly improves the accuracy of cultural tourism marketing and user satisfaction, effectively solving the problems of inconsistencies between recommended content and the user's immediate context, as well as rigid recommendation strategies in existing technologies.

[0033] Furthermore, in order to capture user behavior data more specifically and identify immediate contextual events that indicate changes in a user's travel context, it is necessary to clarify the specific sources and types of user behavior data.

[0034] Specifically, user behavior data can include ticket purchase records and hotel booking information. Ticket purchase records refer to user purchases of airline tickets, train tickets, attraction tickets, performance tickets, etc., generated through various ticketing platforms, airline websites, railway ticketing platforms, and other channels. These purchase records typically include key information such as the user's travel destination, travel time, number of companions (e.g., children's tickets, family packages), and ticket type. Hotel booking information refers to user booking records generated through online travel agency platforms, hotel websites, and other channels. This information typically includes key information such as the user's check-in and check-out times, hotel type (e.g., family hotel, business hotel), room type (e.g., family room, business suite), and number of guests. By capturing this specific behavioral data, a rich and direct data foundation can be provided for subsequent contextual event identification and analysis.

[0035] This invention, by specifically limiting user behavior data to ticket purchase records and hotel booking information, enables the system to more accurately focus on user behaviors directly related to cultural tourism marketing. Ticket purchase records and hotel booking information directly reflect users' travel decisions and behaviors, containing key contextual elements such as the time, location, purpose, and companions of their trip. Based on these clearly defined data sources, the system can efficiently identify real-time contextual events indicating changes in a user's travel context. For example, by analyzing ticket combinations in ticket purchase records or room type characteristics in hotel booking information, it can determine whether a user is traveling with family, on business, or in other contexts. This specific data source limitation lays a solid data foundation for subsequent context intensity assessment, context duration calculation, and preference weight adjustment.

[0036] By acquiring ticket purchase records and hotel booking information, the system can capture user behavior data more effectively, avoiding the processing of irrelevant data and improving the efficiency and accuracy of data capture. Furthermore, since ticket purchase records and hotel booking information directly reflect users' travel intentions and contexts, it can more effectively identify immediate contextual events indicating changes in users' travel circumstances. This provides more accurate contextual awareness for subsequent personalized marketing recommendations, improving the effectiveness of marketing recommendations and user experience.

[0037] like Figure 2 As shown, in one embodiment, step S100 includes the following steps: Step S110: Identify ticket packages based on ticket purchase records. Ticket packages include at least one of child tickets, family packages, or business packages. Step S120: Identify the type and room type characteristics of the booked hotel based on the hotel booking information. The hotel type and room type characteristics include at least one of family-themed room type, family room type, or business room type. Step S130: Identify immediate contextual events that indicate a change in the user's travel context based on ticketing combinations and hotel type and room type features; the immediate contextual event is at least one of a structured data packet including event type, associated geographic location information, context category and user identifier; wherein, the event type includes ticket purchase behavior and hotel booking, and the context category includes family context and business context.

[0038] In this embodiment, ticket purchase records and hotel booking information from user behavior data are considered key sources for identifying real-time contextual events. Specifically, ticket purchase records identify ticket combinations including children's tickets, family packages, or business packages; analyzing ticket types and quantities infers the user's travel purpose and the composition of their travel companions. For example, the presence of children's tickets typically indicates a family travel scenario, while business packages may indicate a business travel scenario. Simultaneously, hotel booking information identifies the booked hotel type and room type characteristics, including family-themed, family-style, or business-style rooms, and further confirms the user's travel scenario through hotel type and room characteristics. Family-themed hotels or family-style rooms clearly indicate a family travel scenario, while business-style rooms tend to indicate a business travel scenario.

[0039] Furthermore, based on ticketing combinations and hotel type / room type characteristics, real-time contextual events indicating changes in a user's travel context are identified, and these identified real-time contextual events are constructed into a structured data package. This structured data package includes an event type, associated geographic location information, a context category, and a user identifier. The event type specifies the specific action that triggered the context change, such as ticket purchase or hotel booking. The associated geographic location information records the geographic location associated with the contextual event, such as the destination of the ticket purchase or the location of the hotel. The context category clarifies the currently identified context type, such as a family context or a business context. The user identifier uniquely identifies the user whose context has changed. By structuring this information, clear and standardized data input can be provided for subsequent contextual analysis and marketing recommendations.

[0040] This invention, through in-depth analysis and pattern matching of ticket purchase records and hotel booking information in user behavior data, can accurately capture key signals indicating changes in a user's travel context from massive amounts of user data. By identifying specific ticket combinations and hotel room type characteristics, the system can effectively distinguish between different contexts such as family travel and business travel, and transform them into standardized real-time contextual event data packages. This context recognition mechanism based on specific behavioral data ensures the objectivity and accuracy of context judgment, laying a solid foundation for subsequent context intensity assessment and personalized recommendations.

[0041] like Figure 3 As shown, in one embodiment, step S200 includes the following steps: Step S210: Based on the ticket types and quantities in the ticket package and the hotel type and room type characteristics in the hotel booking information, and combined with the preset situation intensity assessment rules, a weighted score is calculated to obtain the situation intensity value. Step S220: Determine the duration of the scenario based on the trip duration in the ticket purchase record and / or hotel booking information.

[0042] In this embodiment, the ticket types and quantities in the ticketing package refer to the ticketing composition reflected in the user's ticketing records, such as the number of children's tickets, the number of adult tickets, and the purchase status of family packages or business packages. This information can directly reflect the composition of the travel group and the purpose of the trip. The hotel type and room type characteristics in the hotel booking information refer to the category of the hotel booked by the user (e.g., resort hotel, business hotel, budget hotel) and the selected room type (e.g., family-themed room, family room, business suite, standard room), reflecting the user's accommodation preferences and situational needs. The preset situational intensity assessment rule is a predefined scoring standard or algorithm model used to quantify the degree of influence of different ticketing packages, hotel types, and room type characteristics on the situation. This rule can include multiple weighting factors; for example, children's tickets have a higher weight than adult tickets, and family-themed rooms have a higher weight than standard rooms. Weighted scoring calculation refers to comprehensively evaluating the ticketing and hotel characteristics according to the preset situational intensity assessment rule to obtain a numerical value, namely the situational intensity value; the situational intensity value can objectively quantify the potential influence of the current situation on the user's preferences. The trip duration can be understood as the number of travel days reflected in the user's ticket purchase record or the number of accommodation days reflected in the hotel booking information, used to determine the duration of the immediate situational event.

[0043] This invention addresses the problem of accurately quantifying the impact of contextual events by conducting a deeper analysis of real-time contextual events. Specifically, by leveraging the types and quantities of tickets in ticketing combinations and the hotel type and room type characteristics in hotel booking information, it can more meticulously capture the inherent attributes of a user's travel context. For example, purchasing multiple children's tickets or booking a family-themed room clearly indicates a stronger family-oriented contextual impact compared to purchasing a single adult ticket or booking a standard room. These characteristics are input into preset contextual intensity assessment rules, and through weighted scoring calculations, objective and quantitative contextual intensity values ​​can be generated. Furthermore, by extracting the travel duration from ticketing records or hotel booking information, the duration of the context can be directly obtained. Thus, this invention provides a precise quantitative basis for subsequently adjusting the weight of users' long-accumulated preference information in recommendation decisions, ensuring the rationality and effectiveness of weight adjustments, thereby making marketing recommendations more aligned with users' current specific contextual needs.

[0044] In some alternative implementations, suppose user A booked a trip through an online travel agency platform. Their ticketing record shows they purchased two adult tickets and two child tickets to a theme park and booked a family suite at a nearby hotel for a three-day stay. The system recognizes that the ticket combination includes child tickets and the hotel booking information includes a family room. Based on preset situational intensity assessment rules—for example, assigning higher weight to child tickets and family suites—a high situational intensity value, such as 0.8 (assuming a range of 0-1), can be obtained through weighted scoring. Simultaneously, based on the length of stay in the hotel booking information, the situation duration is determined to be three days. Therefore, the system can accurately determine that user A is in a high-intensity parent-child situation that will last for three days. In contrast, if user B only purchased one business class ticket and booked one night in a standard room at a business hotel, according to preset rules, their situational intensity value might be lower, such as 0.3, and the situation duration would be one day, indicating a lower-intensity business situation. This quantitative analysis enables the system to adjust recommendation weights differently based on the intensity and duration of different situations.

[0045] like Figure 4 As shown, in one embodiment, step S300 includes the following steps: Specifically, the steps mentioned above for adjusting the weight of users' long-term accumulated preference information in recommendation decisions based on context intensity and context duration include: Step S310: Calculate the situation influence factor based on the situation intensity value and situation duration; Step S320: When the value of the context influence factor is greater than or equal to the preset influence threshold, the first inhibition coefficient is used. Based on the first inhibition coefficient, the preset weight of the user's long-term accumulated preference information and the context influence factor, the weight of the user's long-term accumulated preference information in the recommendation decision is adjusted. Step S330: When the value of the contextual influence factor is less than the preset influence threshold, a second inhibition coefficient is used. Based on the second inhibition coefficient, the preset weight of the user's long-term accumulated preference information, and the contextual influence factor, the weight of the user's long-term accumulated preference information in the recommendation decision is adjusted; wherein, the second inhibition coefficient is less than the first inhibition coefficient.

[0046] In this embodiment, the contextual influence factor can be understood as a quantitative indicator that comprehensively reflects the impact of contextual intensity and duration on user preferences. Its calculation method can be designed according to specific business needs and data models; for example, it can use weighted average, product, or nonlinear functions combined with contextual intensity and duration for calculation. The contextual influence factor can integrate multiple dimensions of contextual information into a unified indicator to facilitate subsequent decision-making.

[0047] The preset influence threshold is a critical value used to distinguish between high and low contextual influence factors. Its setting can be based on historical data analysis, expert experience, or optimized through machine learning models to ensure effective differentiation between contexts with significant and weak influence on user preferences. The first and second inhibition coefficients are parameters used to adjust the weighting of long-term accumulated user preference information in recommendation decisions. The first inhibition coefficient is used when the contextual influence factor is high and is usually set to a relatively large value to more significantly suppress long-term preference weights, allowing immediate contextual preferences to dominate. The second inhibition coefficient is used when the contextual influence factor is low and its value is lower than the first inhibition coefficient, aiming to lessen the suppression of long-term preference weights, avoid over-adjustment, and maintain recommendation stability. In practical applications, these inhibition coefficients can be dynamically adjusted and optimized based on actual marketing strategies and user feedback.

[0048] In some alternative implementations, assume a user has accumulated long-term preference information on a tourism marketing platform, such as a preference for historical and cultural tours and a moderate budget. When this user books tickets for a theme park (two adult tickets and one child ticket) and a family-themed hotel, the system recognizes this as a family-oriented situation. First, based on the ticket type and quantity (child ticket) in the purchase record and the room type (family-themed hotel) in the hotel booking information, the system calculates the situation intensity value and the situation duration based on the trip duration. Next, the system calculates a situation influence factor based on these situation intensity values ​​and situation duration. For example, if the calculated situation influence factor is 0.8, and the preset influence threshold is 0.6, then 0.8 is greater than 0.6. In this case, the system uses a first inhibition coefficient (e.g., 0.7) to adjust the weight of the user's long-term accumulated preference information in the recommendation decision. This means that the weight of the user's long-term preference (historical and cultural tours) will be significantly reduced, while the weight of preferences related to the family-oriented situation (such as restaurants around the theme park, children's activities, family-friendly attractions, etc.) will be relatively increased. Conversely, if a user only books a regular adult ticket at a business-class hotel, the calculated contextual influence factor might be 0.3, less than the preset influence threshold of 0.6. In this case, the system will use a second inhibition coefficient (e.g., 0.5) lower than the first inhibition coefficient to adjust the weights. This results in less suppression of the user's long-term preferences, allowing the recommended content to reflect their long-term interests while also catering to immediate business needs. In this way, the system can flexibly adjust its recommendation strategy based on the actual influence of the context. This ensures that when the user is in a family-oriented situation, family-related cultural and tourism products are prioritized, while in other situations, the recommended content is adjusted according to the strength of the contextual influence, thus providing marketing recommendations that better suit the user's current needs.

[0049] like Figure 5 As shown, in one embodiment, step S400 includes the following steps: Step S410: Based on the adjusted weights and recommendation decisions, calculate the overall matching degree of each candidate cultural and tourism product; Step S420: Based on the overall matching degree of each candidate cultural and tourism product, generate at least one candidate cultural and tourism product with a high overall matching degree.

[0050] In this embodiment, after obtaining the user preference weights adjusted for contextual influences and the recommendation decision based on the user's long-term accumulated preference information, the system evaluates a series of potential cultural and tourism products (i.e., candidate cultural and tourism products). This evaluation process combines the adjusted weights with the recommendation decision, using a pre-defined algorithm model to quantify the degree of fit between each candidate cultural and tourism product and the user's current context and long-term preferences, thereby deriving a comprehensive matching score. This comprehensive matching score is then calculated for each candidate cultural and tourism product. For example, this matching score can be a score between 0 and 1, with a higher score indicating a better matching score.

[0051] In practical applications, after calculating the overall matching degree of all candidate cultural and tourism products, the system will sort them according to these matching degrees and select the top-ranked cultural and tourism products with matching degrees reaching a certain threshold as the recommendation results to present to users. This prioritizes outputting candidate cultural and tourism products with high overall matching degrees, ensuring that the recommended content is highly relevant to the user's current situation and potential needs.

[0052] Furthermore, the system not only recommends individual products, but also packages or combines multiple highly relevant cultural and tourism products based on the user's potential needs and context, forming a more attractive overall solution. For example, for family-friendly scenarios, it can combine family-friendly hotels, children's playground tickets, and family restaurant coupons to provide a one-stop solution, thereby offering a more comprehensive and convenient cultural and tourism experience and increasing the likelihood of user choices and spending.

[0053] Suppose the system captures user behavior data and identifies that a user recently purchased an airline ticket including a child's ticket and booked a family-themed hotel, thus determining that the user is currently in a family-oriented situation. Based on this situation, the system adjusts the weight of the user's long-term accumulated preference information in the recommendation decision based on the situation intensity and duration, significantly increasing the weight of family-related cultural and tourism products. When generating marketing recommendations, the system calculates the overall matching degree for a series of candidate cultural and tourism products, such as a children's amusement park ticket, a family-friendly restaurant package, and a family resort accommodation. For example, a children's amusement park ticket highly matches the adjusted family preference weights and the user's long-term preference for entertainment activities, thus calculating a high overall matching degree (e.g., 0.9). However, a business conference hotel would have a very low matching degree (e.g., 0.1). The system will prioritize outputting products with high matching degrees, such as the children's amusement park ticket. Furthermore, the system can combine a children's amusement park ticket with a family-friendly restaurant package into a family-friendly day trip package and recommend it to the user.

[0054] In one embodiment, step S500 includes: after the immediate contextual event is detected to have ended, gradually restoring the weight of the user's long-term accumulated preference information to its original weight before adjustment within a preset recovery period.

[0055] Specifically, the system determines the end of a user's current travel scenario through a mechanism, such as when the user has completed their trip or left a specific location. The preset recovery period can be understood as a pre-defined time frame during which the weights are gradually restored, rather than instantaneously. Gradual recovery indicates that the weight adjustment is a smooth and continuous process. For example, linear, exponential, or logarithmic function models can be used to gradually increase the weight of preference information within the recovery period until it reaches the original weight before adjustment. This avoids abrupt changes in recommended content after the scenario ends, maintaining the consistency of recommendations and the stability of the user experience.

[0056] Furthermore, assuming a user has booked a hotel and flight for business travel, the system identifies the business context and adjusts its preference weights accordingly, making it more inclined to recommend business-related cultural and tourism products. When the user completes their business trip and returns to their place of residence, the system determines the immediate end of the contextual event by detecting that the user's device has left the area corresponding to the geographical location information associated with the business context. At this point, the system does not immediately remove the weight influence of the business context completely, but instead initiates a preset recovery period, such as 7 days. During this recovery period, the system will gradually reduce the influence of the business context on the recommendation weights daily or at regular intervals, according to a preset recovery function (e.g., linear decrease), while gradually increasing the weights of the user's long-accumulated preferences for leisure, family, etc. For example, 10% recovery on the first day, 20% on the second day, until fully restored to the original weights on the seventh day. During this process, the recommended content received by the user will gradually transition from business-oriented to leisure or family-oriented, avoiding sudden changes in recommended content and making the user experience more consistent and natural.

[0057] In one embodiment, detecting the end of an instant context event includes: detecting, via location services, that the user device has left the area corresponding to the associated geographic location information recorded in the instant context event; and detecting that the current time has reached or exceeded the ticket validity period or hotel stay end time determined based on the instant context event.

[0058] In this embodiment, location service monitoring utilizes various positioning technologies such as Global Positioning System (GPS), Wi-Fi positioning, and base station positioning to acquire the real-time geographic location information of the user's device. The associated geographic location information recorded in the contextual event refers to the geographic area directly related to the event when it is identified, such as the location of a tourist attraction or performance venue in the user's ticket purchase record, or the hotel address explicitly stated in the hotel booking information. When the system detects that the real-time location of the user's device has exceeded the preset range corresponding to the associated geographic location information, it determines that the contextual event has ended physically. This range can be a geofence set according to the actual scenario, such as a circular area with a specific radius centered on the attraction or hotel, or a complex polygonal area defined based on the actual site boundaries.

[0059] Furthermore, detecting that the current time has reached or exceeded the ticket validity period or hotel accommodation end time determined based on the real-time context event means that the system automatically calculates or extracts the expected end time of the event based on ticketing information (such as the valid usage date of the ticket, the performance end time) or hotel booking information (such as check-in date, check-out date) contained in the real-time context event. For example, for events involving tickets, the end time can be determined based on the last valid usage period of the ticket; for hotel accommodation events, it can be determined based on the booked check-out time. When the system detects that the current system time has reached or exceeded these preset end times, it considers the real-time context event to have ended as planned.

[0060] Furthermore, suppose user A books a family package for a theme park through an online travel agency platform and also books a family-friendly hotel near the park, with check-out time at 12:00 noon on the third day. The system recognizes this as a real-time situational event involving a family and records the associated geographical location information of the theme park and the family-friendly hotel.

[0061] In some alternative implementations, the system may use one of the following two methods, or a combination thereof, to monitor the end of the immediate contextual event: 1. Location-based monitoring: The system continuously acquires location service data from user A's mobile device (e.g., smartphone). When user A leaves the theme park area at 10:00 AM on the third day and subsequently leaves the geofence of the family hotel at 11:00 AM, the system detects that the user's device has left the area corresponding to the associated geographical location information, thus determining that the family-friendly event has ended.

[0062] 2. Time-based monitoring: The system monitors the current time. If the current time reaches or exceeds 12:00 noon on the third day (i.e., the end time of the hotel stay), the system determines that the parent-child situation event has ended.

[0063] By monitoring location services and time points, the system can accurately determine that user A's family travel scenario has ended, thereby promptly initiating a gradual recovery process of the user's long-accumulated preference information weights, providing a more accurate basis for subsequent cultural and tourism marketing recommendations for user A.

[0064] In one embodiment, step S100 further includes: obtaining new transaction records generated by the user on an online travel agency platform, ticketing system, or hotel booking system after user authorization; and identifying user behavior data streams in the new transaction records based on preset rules and pattern matching logic.

[0065] In this embodiment, before acquiring a user's personal behavior data, the system clearly informs the user of the purpose and scope of data use and obtains the user's explicit consent to ensure the legality of data acquisition and the protection of user privacy. Then, new transaction records generated by the user are acquired from channels carrying a large amount of transaction data directly reflecting the user's travel intentions and context, such as online travel agency platforms, major airlines, railway passenger transport, scenic spot ticket sales systems, or official reservation systems of chain hotels, to obtain information on the user's cultural and tourism-related consumption behavior. Furthermore, based on preset rules and pattern matching logic, the system identifies the user behavior data stream in the new transaction records to better identify real-time contextual events.

[0066] According to embodiments of the present invention, such as Figure 6 As shown, on the other hand, a consumer behavior analysis system for cultural tourism marketing is also provided, including: The data acquisition module 100 is used to acquire user behavior data and identify real-time contextual events that indicate changes in the user's travel context based on the user behavior data; the user behavior data includes ticket purchase records and hotel booking information; The context impact analysis module 200 is used to generate context intensity values ​​and context duration based on real-time context events, which characterize the degree of context impact. The first weight adjustment and recovery module 300 is used to adjust the weight of the user's long-term accumulated preference information in the recommendation decision based on the context intensity value and the context duration. The marketing content generation module 400 is used to generate marketing recommendation content based on the adjusted weights and recommendation decisions; The second weight adjustment and recovery module 500 is used to gradually restore the weights to their original weights after the immediate contextual event ends.

[0067] In one embodiment, the data acquisition module 100 includes: The first identification unit is used to identify ticket combinations based on ticket purchase records, wherein the ticket combination includes at least one of child tickets, family packages, or business packages; The second identification unit is used to identify the type of hotel and room type characteristics based on the hotel booking information. The hotel type and room type characteristics include at least one of family-themed room type, family room type, or business room type. The third identification unit is used to identify real-time contextual events that indicate a change in the user's travel context based on ticketing combinations and hotel type and room type characteristics; the real-time contextual event is at least one of the structured data packets including event type, associated geographic location information, context category and user identifier; wherein, the event type includes ticket purchase behavior and hotel booking, and the context category includes parent-child context and business context.

[0068] In one embodiment, the context impact analysis module 200 includes: The first calculation unit is used to calculate the situation intensity value by weighting the score based on the ticket type and quantity in the ticket combination and the hotel type and room type characteristics in the hotel booking information, combined with the preset situation intensity assessment rules. The first acquisition unit is used to determine the duration of the scenario based on the trip duration in the ticket purchase record and / or hotel booking information.

[0069] In one embodiment, the first weight adjustment and recovery module 300 includes: The second calculation unit is used to calculate the situation influence factor based on the situation intensity value and the situation duration. The first adjustment unit is used to adjust the weight of the user's long-term accumulated preference information in the recommendation decision based on the first inhibition coefficient, the preset weight of the user's long-term accumulated preference information, and the contextual influence factor when the value of the contextual influence factor is greater than or equal to the preset influence threshold. The second adjustment unit is used to adjust the weight of the user's long-term accumulated preference information in the recommendation decision based on the second inhibition coefficient, the preset weight of the user's long-term accumulated preference information, and the contextual influence factor when the value of the contextual influence factor is less than the preset influence threshold; wherein the second inhibition coefficient is less than the first inhibition coefficient.

[0070] In one embodiment, the marketing content generation module 400 includes: The third calculation unit is used to calculate the overall matching degree of each candidate cultural and tourism product based on the adjusted weights and recommendation decisions; The generation unit is used to generate at least one candidate cultural tourism product with a high overall matching degree based on the overall matching degree of each candidate cultural tourism product.

[0071] In one embodiment, the second weight adjustment and recovery module 500 includes: The control unit is used to gradually restore the weights of the user's long-term accumulated preference information to their original weights within a preset recovery period after the immediate contextual event is detected to have ended. Detecting the end of the immediate contextual event includes detecting, via location services, that the user's device has left the area corresponding to the associated geographical location information recorded in the immediate contextual event; and / or detecting that the current time has reached or exceeded the ticket validity period or hotel stay end time determined based on the immediate contextual event.

[0072] In one embodiment, the data acquisition module 100 further includes: The second acquisition unit is used to acquire new transaction records generated by the user on online travel agency platforms, ticketing systems, or hotel booking systems after authorization by the user; The fourth identification unit is used to identify user behavior data streams in new transaction records based on preset rules and pattern matching logic.

[0073] Figure 7 The diagram shows a structural schematic of an embodiment of an electronic device provided by the present invention. The specific embodiments of the present invention do not limit the specific implementation of the electronic device.

[0074] like Figure 7 As shown, the electronic device may include: a processor 1002, a communications interface 1004, a memory 1006, and a communications bus 1008.

[0075] The processor 1002, communication interface 1004, and memory 1006 communicate with each other via communication bus 1008. Communication interface 1004 is used to communicate with other network elements such as clients or other servers. The processor 1002 executes program 1010, specifically performing the relevant steps in the above-described embodiment of the consumer behavior analysis method for cultural tourism marketing.

[0076] Specifically, program 1010 may include program code, which includes computer-executable instructions.

[0077] The processor 1002 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The electronic device includes one or more processors, which may be processors of the same type, such as one or more CPUs; or they may be processors of different types, such as one or more CPUs and one or more ASICs.

[0078] Memory 1006 is used to store program 1010. Memory 1006 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0079] Specifically, program 1010 can be called by processor 1002 to cause the electronic device to execute the relevant steps in the above embodiment of the consumer behavior analysis method for cultural tourism marketing.

[0080] Those skilled in the art will understand that Figure 7 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned device. For example, electronic devices may also include components that are more... Figure 7 The more or fewer components shown, or having the same Figure 7 The different configurations shown.

[0081] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0082] In the specific implementation of the above embodiments, the technical features can be combined in any non-contradictory way. For the sake of brevity, not all possible combinations of the above technical features are described. However, as long as the combination of these technical features is not contradictory, it should be considered to be within the scope of this specification.

[0083] The specific embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A consumer behavior analysis method for cultural tourism marketing, characterized in that, include: Acquire user behavior data and identify real-time contextual events that indicate a change in the user's travel context based on the user behavior data; The user behavior data includes ticket purchase records and hotel booking information; Based on the real-time situational events, generate a situation intensity value and a situation duration that characterize the degree of situational impact. Based on the intensity value of the context and the duration of the context, the weight of the user's long-term accumulated preference information in the recommendation decision is adjusted. Based on the adjusted weights and the recommendation decision, marketing recommendation content is generated; After the immediate contextual event ends, the weights are gradually restored to their original weights.

2. The consumer behavior analysis method for cultural tourism marketing according to claim 1, characterized in that, The instantaneous contextual events that indicate a change in the user's travel context based on the user behavior data include: Based on the ticket purchase records, ticket combinations are identified, including at least one of child tickets, family packages, or business packages; Based on the hotel booking information, the type of hotel and room type characteristics of the booked hotel are identified. The hotel type and room type characteristics include at least one of family-themed room type, family room type, or business room type. Based on the ticketing combination and the hotel type and room type features, an instantaneous contextual event is identified to indicate a change in the user's travel context; the instantaneous contextual event is at least one of the following: an event type, associated geographic location information, context category, and user identifier; wherein, the event type includes ticket purchase behavior and hotel booking, and the context category includes family context and business context.

3. The consumer behavior analysis method for cultural tourism marketing according to claim 2, characterized in that, The process of generating a situation intensity value and situation duration based on the immediate situational event includes: The situation intensity value is calculated by weighting the ticket types and quantities in the ticket package and the hotel type and room type characteristics in the hotel booking information, combined with the preset situation intensity assessment rules. The duration of the scenario is determined based on the travel duration in the ticket purchase record and / or the hotel booking information.

4. The consumer behavior analysis method for cultural tourism marketing according to claim 1, characterized in that, The step of adjusting the weight of user-accumulated preference information in recommendation decisions based on the context intensity value and the context duration includes: Calculate the situation influence factor based on the situation intensity value and the situation duration; When the value of the context influence factor is greater than or equal to the preset influence threshold, a first inhibition coefficient is used, and the weight of the user's long-term accumulated preference information in the recommendation decision is adjusted based on the first inhibition coefficient, the preset weight of the user's long-term accumulated preference information, and the context influence factor. If the value of the contextual influence factor is less than the preset influence threshold, a second inhibition coefficient is used. Based on the second inhibition coefficient, the preset weight of the user's long-term accumulated preference information, and the contextual influence factor, the weight of the user's long-term accumulated preference information in the recommendation decision is adjusted; wherein, the second inhibition coefficient is less than the first inhibition coefficient.

5. The consumer behavior analysis method for cultural tourism marketing according to claim 1, characterized in that, The process of generating marketing recommendation content based on the adjusted weights and the recommendation decision includes: Based on the adjusted weights and the recommendation decision, the overall matching degree of each candidate cultural and tourism product is calculated. Based on the overall matching degree of each candidate cultural and tourism product, at least one candidate cultural and tourism product with a high overall matching degree is generated.

6. The consumer behavior analysis method for cultural tourism marketing according to claim 1, characterized in that, The step of gradually restoring the weights to their original weights after the immediate contextual event ends includes: Once the immediate contextual event is detected to have ended, the weights of the user's long-term accumulated preference information are gradually restored to their original weights before adjustment within a preset recovery period.

7. The consumer behavior analysis method for cultural tourism marketing according to claim 6, characterized in that, The detection of the end of the immediate contextual event includes: The location service detects that the user device has left the area corresponding to the associated geographic location information recorded in the real-time contextual event; and / or The system detects that the current time has reached or exceeded the ticket validity period or hotel stay end time determined based on the aforementioned real-time contextual event.

8. The consumer behavior analysis method for cultural tourism marketing according to claim 1, characterized in that, The acquisition of user behavior data includes: With the user's authorization, new transaction records generated by the user on online travel agency platforms, ticketing systems, or hotel booking systems are obtained. The user behavior data stream in the new transaction record is identified based on preset rules and pattern matching logic.

9. A consumer behavior analysis system for cultural tourism marketing, characterized in that, The system includes: The data acquisition module is used to acquire user behavior data and identify real-time contextual events that indicate a change in the user's travel context based on the user behavior data. The context impact analysis module is used to generate a context intensity value and a context duration that characterize the degree of context impact based on the real-time context event. The first weight adjustment and recovery module is used to adjust the weight of the user's long-term accumulated preference information in the recommendation decision based on the context intensity value and the context duration. The marketing content generation module is used to generate marketing recommendation content based on the adjusted weights and the recommendation decision. The second weight adjustment and recovery module is used to gradually restore the weight to its original weight after the immediate contextual event ends.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the consumer behavior analysis method for cultural tourism marketing as described in any one of claims 1-8.

Images