Intelligent tourism service recommendation management system and method based on artificial intelligence

Abstract

The invention relates to the technical field of computers, in particular to a tourism service intelligent recommendation management system and method based on artificial intelligence, in the system, a demand adaptation recommendation module generates a tourism spot travel planning route adaptation result of a user based on a to-be-executed travel planning scenic spot deviation sequence and a travel execution feature set; and extracting dining planning information in the travel route planning route of the scenic spot of the user, locking a dining adaptive area of the user, and generating a space-time matching set of the travel route and the dining reservation. According to the method, user behavior preference characteristics are analyzed in combination with behavior density space-time distribution data of each scenic spot; therefore, the stay behavior vector feature information of the user in the travel planning route of each scenic spot in the to-be-executed travel planning scenic spot deviation sequence is obtained, and data support is provided for the subsequent dynamic generation of the travel spot travel planning route adaptation result of the user and the space-time matching set of the travel route and catering reservation.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and specifically to an intelligent recommendation management system and method for tourism services based on artificial intelligence. Background Art

[0002] With the rapid economic development and the improvement of people's living standards, tourism has become an important part of people's lives. The diverse needs and quality requirements of the general public for tourism are increasing day by day. Group tours are gradually decreasing, and emerging tourism modes such as self-driving tours and free independent travel are becoming increasingly popular.

[0003] When users go on self-driving tours, they usually formulate a travel itinerary planning route for tourist attractions in advance and reserve dining locations in advance. However, during the actual execution process, the execution progress of the user's travel itinerary planning route for tourist attractions is often affected by unexpected events, which may lead to the ineffective execution of the corresponding travel itinerary planning route for tourist attractions and the inability to implement the reserved dining arrangements, etc. Existing intelligent recommendation management systems for tourism services can often only recommend dining locations based on the user's own search, and cannot extract unexpected events and make itinerary adaptation changes, nor can they make adaptive adjustments and planning recommendations for the user's dining services in advance according to the results of itinerary adaptation changes. Therefore, there are significant defects in the existing technology. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent recommendation management system and method for tourism services based on artificial intelligence to solve the problems raised in the existing technology.

[0005] To achieve the above purpose, the present invention provides the following technical solution: An intelligent recommendation management method for tourism services based on artificial intelligence, the method includes:

[0006] Step S100: Obtain the travel itinerary planning route of the user for tourist attractions and the travel itinerary execution progress information of the user for tourist attractions at the current time; identify the itinerary interference events during the execution of the travel itinerary for tourist attractions, analyze the itinerary execution progress deviation value of the user for tourist attractions at the current time, construct an itinerary execution feature set, and perform an itinerary unexpected event adaptation determination operation based on the itinerary execution feature set; the itinerary unexpected event adaptation determination result includes itinerary adaptation change and itinerary remaining unchanged;

[0007] Step S200: If the itinerary unexpected event adaptation determination result is itinerary adaptation change, obtain the scenic spot information involved in the unexecuted travel itinerary planning route of the user at the current time, extract the social public opinion information corresponding to each scenic spot in the historical data and the social public opinion information corresponding to the user's historical tourist attractions through image recognition technology, and obtain the preference sequence of the to-be-executed itinerary planning scenic spots;

[0008] Step S300: Obtain the spatio-temporal distribution data of the behavior density of each tourist attraction in the travel itinerary execution progress information of the user at the current time, and generate a set of user behavior preference features; combine the platform scenic spot information in the historical data to generate the stay behavior vector feature information in the play plan route of each scenic spot in the scenic spot preference sequence based on the to-be-executed itinerary plan of the user;

[0009] Step S400: Generate the adaptation result of the travel itinerary plan route of the user based on the scenic spot preference sequence of the to-be-executed itinerary plan and the itinerary execution feature set; extract the dining plan information in the travel itinerary plan route of the user, lock the dining adaptation area of the user, and generate the spatio-temporal matching set of the itinerary route and the dining reservation;

[0010] Step S500: Real-time feedback the binding result of the adaptation result of the travel itinerary plan route of the user and the spatio-temporal matching set of the itinerary route and the dining reservation to the user.

[0011] Further, the step S100 includes the following steps:

[0012] Step S101: Obtain the travel itinerary plan route of the user, where the travel itinerary plan route includes the stay duration corresponding to each tourist attraction and the play plan route of each tourist attraction;

[0013] Step S102: Obtain the travel itinerary execution progress information of the user at the current time, where the travel itinerary execution progress information of the user at the current time includes the completed tourist attractions and the user's location in the current tourist attraction;

[0014] Step S103: Identify the itinerary interference events during the execution of the travel itinerary. The itinerary interference events include weather information that meets the preset conditions in the database and the situation where the proportion of the abnormal heart rate state duration is greater than the preset value. The abnormal heart rate state is the state where the heart rate is greater than the preset heart rate value;

[0015] Step S104: Construct an itinerary execution feature set, where the itinerary execution feature set includes the itinerary interference event recognition result and the deviation value of the travel itinerary execution progress of the user at the current time; the calculation formula of the deviation value of the travel itinerary execution progress of the user at the current time is as follows:

[0016]

[0017] Among them, DP represents the progress deviation value of the user's tourist attraction itinerary execution at the current time; T1 represents the sum of the actual stay durations corresponding to each completed tourist attraction and the interval duration from the start time of the current tourist attraction to the current time; T2 represents the planned stay duration values corresponding to each completed tourist attraction in the tourist attraction itinerary planning route; T3 represents the planned stay duration value corresponding to the current tourist attraction in the tourist attraction itinerary planning route; L1 represents the total length of the planned play route of the current tourist attraction in the tourist attraction itinerary planning route; L2 represents the length from the user's position in the current tourist attraction to the starting point of the planned play route of the current tourist attraction in the tourist attraction itinerary planning route.

[0018] Step S105: Perform an itinerary emergency adaptation determination operation based on the itinerary execution feature set.

[0019] If DP is greater than or equal to the preset deviation value or the recognition result of the itinerary interference event is not empty, then determine that the itinerary emergency adaptation determination result is an itinerary adaptation change; otherwise, determine that the itinerary emergency adaptation determination result is that the itinerary remains unchanged.

[0020] Further, the step S200 includes the following steps:

[0021] Step S201: Obtain the scenic spot information involved in the tourist attraction itinerary planning route that the user has not executed at the current time, and the scenic spot information includes the scenic spot name and the planned play route of the corresponding scenic spot.

[0022] Step S202: Identify the background scenery types in each scenic spot image published on the social platform through image recognition technology, and the background scenery types are preset in the database; extract the social public opinion information corresponding to each scenic spot in the historical data and the social public opinion information corresponding to the historical tourist attractions of the corresponding user; the social public opinion information includes the proportion of picture data of each background scenery type, the number of pictures of the corresponding scenic spot published in the social platform, and the user collection volume of the corresponding scenic spot in the social platform.

[0023] Step S203: Calculate the execution preference coefficient of each scenic spot information involved in the tourist attraction itinerary planning route that the user has not executed at the current time, and the involved calculation formula is as follows:

[0024]

[0025] Among them, A n represents the execution preference coefficient of the nth scenic spot information involved in the tourist attraction itinerary planning route that the user has not executed at the current time; τp i represents the proportion of picture data of the ith background scenery type in the social public opinion information corresponding to the nth scenic spot in the historical data; B nrepresents the number of pictures posted by the corresponding scenic spot in the social public opinion information corresponding to the nth scenic spot in the historical data; C n represents the user collection volume of the corresponding scenic spot in the social public opinion information corresponding to the nth scenic spot in the historical data; τy i represents the proportion of picture data of the ith background scenery type in the social public opinion information corresponding to the historical tourist scenic spots of the corresponding user; is represents the number of preset background scenery types in the database; μ represents the preset conversion coefficient;

[0026] Step S204: Obtain the to-be-executed itinerary planning scenic spot preference sequence, and the to-be-executed itinerary planning scenic spot preference sequence is obtained by arranging according to the corresponding execution preference coefficients of the corresponding scenic spots from large to small.

[0027] Further, the step S300 includes the following steps:

[0028] Step S301: Obtain the behavior density and time-space distribution data of each tourist scenic spot in the tourist scenic spot itinerary execution progress information of the user at the current time. The behavior density and time-space distribution data is a data pair composed of the mapped point position of each photo-taking position of the user in the corresponding tourist scenic spot on the corresponding play planning route and the corresponding photo-taking time; the mapped point position of each photo-taking position of the user in the corresponding tourist scenic spot on the corresponding play planning route represents the position point closest to the corresponding photo-taking position of the user in the corresponding tourist scenic spot in the corresponding play planning route;

[0029] Step S302: Statistically analyze the correlation between each shooting position point in the obtained behavior density and time-space distribution data of each tourist scenic spot, and divide different shooting position points with an interval distance less than the preset distance between shooting positions into the same array to obtain different user behavior preference feature analysis sets;

[0030] Step S303: Extract the background scenery types in the photos corresponding to each element in each user behavior preference feature analysis set, and summarize the set of the types of background scenery types corresponding to the elements of the same user behavior preference feature analysis set as a user behavior preference feature;

[0031] Step S304: Obtain the behavior preference coefficient of each user behavior preference feature. The behavior preference coefficient of the user behavior preference feature is equal to the ratio of the number of user behavior preference features in the subset of the corresponding user behavior preference feature analysis set that are the same as the user behavior preference feature of the corresponding user behavior preference feature analysis set to the total number of user behavior preference features;

[0032] Step S305: Generate a set of user behavior preference features. The set of user behavior preference features is a set composed of the respective user behavior preference features obtained in Step S303, and each element in the set of user behavior preference features is bound to the corresponding behavior preference coefficient;

[0033] Step S306: According to a preset interval length, divide the corresponding scenic spots into different route sections along the scenic spot tour planning route, and use the set of position points that are the closest to the corresponding section interval in the corresponding scenic spots as the scenic spot division area of the corresponding route section; respectively, use the behavior preference coefficients bound to the elements with the same extraction results of the behavior preference features bound to the photos in the corresponding division area of the same scenic spot in the social platform historical data as the behavior preference coefficients corresponding to the corresponding division area in the corresponding scenic spot; the binding relationship between the photos of the same scenic spot in the social platform historical data and the corresponding division area is preset artificially;

[0034] Step S307: Generate the stop behavior vector feature information in the tour planning route of each scenic spot in the scenic spot preference sequence of the user's to-be-executed itinerary plan;

[0035] The stop behavior vector feature information is composed of the stop behavior vectors corresponding to each position point in the corresponding scenic spot tour planning route. The modulus of the stop behavior vector is equal to the behavior preference coefficient of the division area to which the corresponding position point belongs, and the direction of the stop behavior vector is the direction from the corresponding position point to the subsequent adjacent position point in the corresponding scenic spot tour planning route.

[0036] Further, in the process of generating the adaptation result of the user's tourist scenic spot itinerary plan in Step S400, predict the itinerary plan mapping duration based on the itinerary execution feature set, and the calculation formula is as follows:

[0037]

[0038] Among them, TY represents the prediction result of the itinerary plan mapping duration based on the itinerary execution feature set; TW represents the predicted value of the duration required to complete the user's subsequent tourist scenic spot itinerary plan; TS represents the sum of the planned stop duration values corresponding to each tourist scenic spot other than the current scenic spot in the scenic spot preference sequence of the to-be-executed itinerary plan;

[0039] Remove the scenic spot corresponding to the current time from the to-be-executed itinerary planning scenic spot preference sequence, obtain the updated to-be-executed itinerary planning scenic spot preference sequence, and sequentially obtain the itinerary planning durations corresponding to each element in the updated to-be-executed itinerary planning scenic spot preference sequence; extract the summary set of the first k elements and the scenic spot corresponding to the current time in the updated to-be-executed itinerary planning scenic spot preference sequence, denoted as the route analysis set; arrange the elements in the route analysis set in the order of the scenic spots in the user's travel scenic spot itinerary planning route, and splice the scenic spot play planning routes corresponding to adjacent elements in the arrangement result to obtain the user's travel scenic spot itinerary planning route adaptation result; the k satisfies the condition: TY k ≤TY<TY k+1 ,TY k represents the sum of the itinerary planning durations corresponding to the first k elements respectively in the updated to-be-executed itinerary planning scenic spot preference sequence; TY k+1 represents the sum of the itinerary planning durations corresponding to the first k + 1 elements respectively in the updated to-be-executed itinerary planning scenic spot preference sequence.

[0040] Further, when locking the user's dining adaptation area in step S400, preset the user's itinerary position at the minimum time point of the dining planning information in the user's travel scenic spot itinerary planning route as the distance from the initial point of the non-traveled route section to the g-th scenic spot in the user's travel scenic spot itinerary planning route adaptation result to be LC; preset the user's itinerary position at the maximum time point of the dining planning information in the user's travel scenic spot itinerary planning route as the distance from the initial point of the non-traveled route section to the g1-th scenic spot in the user's travel scenic spot itinerary planning route adaptation result to be LC1; satisfy the following equation:

[0041]

[0042] TR1 represents the time interval duration from the minimum time point in the planned dining time interval of the dining planning information to the current time; TR2 represents the time interval duration from the maximum time point in the planned dining time interval of the dining planning information to the current time;

[0043] TSG (2，g-1) represents the sum of the predicted values of the durations required for the itinerary planning routes of the 2nd to the g - 1th scenic spots in the user's travel scenic spot itinerary planning route adaptation result; F (g，m) represents the modulus length of the stay behavior vector corresponding to the position point when the distance from the initial point of the non-traveled route section to the g-th scenic spot in the user's travel scenic spot itinerary planning route adaptation result is m; Q g represents the integral value of the modulus lengths of the stay behavior vectors corresponding to each position point in the non-traveled route section of the g-th scenic spot in the user's travel scenic spot itinerary planning route adaptation result; TSG (g-1，g)Denotes the predicted value of the duration required for the g-th scenic spot itinerary planning route in the tourist scenic spot itinerary planning route adaptation result of the user; TSG (2，g1-1) Denotes the sum of the predicted values of the durations required for the 2nd to the g1-1th scenic spot itinerary planning routes in the tourist scenic spot itinerary planning route adaptation result of the user; F (g1，m) Denotes the norm of the stay behavior vector corresponding to the position point when the distance from the initial point of the un-traveled route section in the g1-th scenic spot in the tourist scenic spot itinerary planning route adaptation result of the user is m; Q g1 Denotes the integral value of the norms of the stay behavior vectors corresponding to each position point in the un-traveled route section in the g1-th scenic spot in the tourist scenic spot itinerary planning route adaptation result of the user; TSG (g1-1，g1) Denotes the predicted value of the duration required for the g1-th scenic spot itinerary planning route in the tourist scenic spot itinerary planning route adaptation result of the user;

[0044] The user's dining adaptation area is the planar area within a preset unit radius around the play planning route section formed by each travel position of the user within the dining planning information of the user's tourist scenic spot itinerary planning route in the tourist scenic spot itinerary planning route adaptation result of the user;

[0045] The spatio-temporal matching set of the itinerary route and the dining reservation is the set composed of the names of the dining stores located within the locked user's dining adaptation area and having dining reservation quotas in the dining planning information in the user's tourist scenic spot itinerary planning route.

[0046] An intelligent recommendation management system for tourism services based on artificial intelligence, the system includes a user demand change dynamic determination module, an itinerary planning preference analysis module, a behavior vector feature analysis module, a demand adaptation recommendation module, and an adaptation information feedback module;

[0047] The user demand change dynamic determination module obtains the user's tourist scenic spot itinerary planning route and the tourist scenic spot itinerary execution progress information of the user at the current time; identifies the itinerary interference events during the execution of the tourist scenic spot itinerary, analyzes the itinerary execution progress deviation value at the current time, constructs an itinerary execution feature set, and performs an itinerary emergency adaptation determination operation based on the itinerary execution feature set; the itinerary emergency adaptation determination result includes itinerary adaptation change and itinerary remaining unchanged;

[0048] If the itinerary emergency adaptation determination result of the itinerary planning preference analysis module is itinerary adaptation change, it obtains the scenic spot information involved in the unexecuted tourist scenic spot itinerary planning route of the user at the current time, extracts the social public opinion information corresponding to each scenic spot in the historical data and the social public opinion information corresponding to the user's historical tourist scenic spots through image recognition technology, and obtains the pending itinerary planning scenic spot preference sequence;

[0049] The behavior vector feature analysis module obtains the spatio-temporal distribution data of the behavior density of each tourist attraction in the tourist attraction itinerary execution progress information of the user at the current time, and generates a set of user behavior preference features; combines the platform attraction information in the historical data to generate the stay behavior vector feature information in the play planning route of each attraction in the tourist attraction bias sequence based on the to-be-executed itinerary planning of the user.

[0050] The demand adaptation recommendation module generates the adaptation result of the tourist attraction itinerary planning route of the user based on the tourist attraction bias sequence of the to-be-executed itinerary planning and the itinerary execution feature set; extracts the dining planning information in the tourist attraction itinerary planning route of the user, locks the dining adaptation area of the user, and generates the spatio-temporal matching set of the itinerary route and the dining reservation.

[0051] The adaptation information feedback module feeds back the adaptation result of the tourist attraction itinerary planning route of the user and the binding result of the spatio-temporal matching set of the itinerary route and the dining reservation to the user in real time.

[0052] Further, the user demand change dynamic determination module includes an itinerary execution progress acquisition unit, an itinerary execution feature set construction unit, and an adaptation determination unit;

[0053] The itinerary execution progress acquisition unit obtains the tourist attraction itinerary planning route of the user and the tourist attraction itinerary execution progress information of the user at the current time;

[0054] The itinerary execution feature set construction unit identifies the itinerary interference events during the execution of the tourist attraction itinerary, analyzes the itinerary execution progress deviation value of the user at the current time, and constructs an itinerary execution feature set;

[0055] The adaptation determination unit performs an itinerary emergency adaptation determination operation based on the itinerary execution feature set.

[0056] Further, the behavior vector feature analysis module includes a user behavior preference feature analysis unit and a stay behavior vector acquisition unit,

[0057] The user behavior preference feature analysis unit obtains the spatio-temporal distribution data of the behavior density of each tourist attraction in the tourist attraction itinerary execution progress information of the user at the current time, and generates a set of user behavior preference features;

[0058] The stay behavior vector acquisition unit combines the platform attraction information in the historical data to generate the stay behavior vector feature information in the play planning route of each attraction in the tourist attraction bias sequence based on the to-be-executed itinerary planning of the user.

[0059] Further, the demand adaptation recommendation module includes a route adaptation unit and a catering service adaptation unit,

[0060] The route adaptation unit generates a route adaptation result for the user's tourist attraction itinerary plan based on the planned scenic spot preference sequence to be executed and the itinerary execution feature set;

[0061] The catering service adaptation unit extracts the dining plan information in the user's tourist attraction itinerary plan route, locks the user's dining adaptation area, and generates a spatio-temporal matching set of the itinerary route and catering reservation.

[0062] Compared with the prior art, the beneficial effects of the present invention are:

[0063] (1) The present invention takes into account the impact of sudden events during the execution of the itinerary plan route, analyzes the deviation value of the execution progress of the user's tourist attraction itinerary at the current time, and performs an adaptation determination operation for itinerary emergencies; and combines image recognition technology to extract the social public opinion information of the scenic spots, analyzes the user's preference for the planned scenic spots to be executed, and provides data reference for obtaining the route adaptation result of the user's tourist attraction itinerary plan subsequently;

[0064] (2) The present invention combines the spatio-temporal distribution data of the behavior density of each tourist attraction to analyze the user's behavior preference characteristics; and then obtains the stay behavior vector feature information in the play plan route of each scenic spot in the planned scenic spot preference sequence to be executed by the user, providing data support for dynamically generating the route adaptation result of the user's tourist attraction itinerary plan and the spatio-temporal matching set of the itinerary route and catering reservation subsequently;

[0065] (3) The present invention can dynamically lock the dining adaptation area corresponding to the dining plan information in the user's tourist attraction itinerary plan route through the route adaptation result of the user's tourist attraction itinerary plan and the stay behavior vector feature information in the play plan route of each scenic spot, recommend the spatio-temporal matching set of the itinerary route and catering reservation, and realize the effective management of intelligent recommendation data for tourism services. Brief Description of the Drawings

[0066] Figure 1 It is a schematic structural diagram of an intelligent recommendation management system for tourism services based on artificial intelligence;

[0067] Figure 2 It is a schematic diagram of the steps of an intelligent recommendation management method for tourism services based on artificial intelligence. Detailed Embodiment

[0068] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0069] Example: As Figures 1 to 2 shown, the present invention provides an intelligent recommendation management system for tourism services based on artificial intelligence. The system includes a dynamic determination module for user demand changes, a travel plan preference analysis module, a behavior vector feature analysis module, a demand adaptation recommendation module, and an adaptation information feedback module;

[0070] The dynamic determination module for user demand changes includes a travel execution progress acquisition unit, a travel execution feature set construction unit, and an adaptation determination unit;

[0071] The travel execution progress acquisition unit acquires the travel route of tourist attractions planned by the user and the travel execution progress information of tourist attractions of the user at the current time;

[0072] The travel execution feature set construction unit identifies travel interference events during the execution of the travel route of tourist attractions, analyzes the deviation value of the travel execution progress of the user's travel route of tourist attractions at the current time, and constructs a travel execution feature set;

[0073] The adaptation determination unit performs an adaptation determination operation for travel emergencies based on the travel execution feature set;

[0074] If the adaptation determination result of the travel plan preference analysis module for travel emergencies is a travel adaptation change, the module acquires the scenic spot information involved in the unexecuted travel route of tourist attractions of the user at the current time, extracts the social public opinion information corresponding to each scenic spot in the historical data and the social public opinion information corresponding to the user's historical tourist attractions through image recognition technology, and obtains a sequence of biases of scenic spots in the to-be-executed travel plan;

[0075] The behavior vector feature analysis module includes a user behavior preference feature analysis unit and a stay behavior vector acquisition unit,

[0076] The user behavior preference feature analysis unit acquires the spatio-temporal distribution data of the behavior density of each tourist attraction in the travel execution progress information of the user's travel route of tourist attractions at the current time, and generates a set of user behavior preference features;

[0077] The stay behavior vector acquisition unit combines the platform scenic spot information in the historical data to generate the stay behavior vector feature information in the play plan route of each scenic spot in the sequence of biases of scenic spots in the to-be-executed travel plan for the user;

[0078] The demand adaptation recommendation module includes a route adaptation unit and a catering service adaptation unit,

[0079] The route adaptation unit generates a travel route adaptation result of tourist attractions for the user based on the sequence of biases of scenic spots in the to-be-executed travel plan and the travel execution feature set;

[0080] The catering service adaptation unit extracts the dining planning information in the user's travel itinerary planning route for tourist attractions, locks the user's dining adaptation area, and generates a spatio-temporal matching set of the itinerary route and catering reservation;

[0081] The adaptation information feedback module feeds back the adaptation result of the user's travel itinerary planning route for tourist attractions and the binding result of the spatio-temporal matching set of the itinerary route and catering reservation to the user in real time.

[0082] In this embodiment, an intelligent recommendation management method for tourism services based on artificial intelligence, the method includes:

[0083] Step S100: Obtain the user's travel itinerary planning route for tourist attractions and the travel itinerary execution progress information of the user at the current time; identify the itinerary interference events during the execution of the travel itinerary for tourist attractions, analyze the itinerary execution progress deviation value of the user at the current time, construct an itinerary execution feature set, and perform an itinerary emergency adaptation determination operation based on the itinerary execution feature set; the itinerary emergency adaptation determination result includes itinerary adaptation change and itinerary remaining unchanged;

[0084] The step S100 includes the following steps:

[0085] Step S101: Obtain the user's travel itinerary planning route for tourist attractions, where the travel itinerary planning route for tourist attractions includes the stay duration corresponding to each tourist attraction and the play planning route for each tourist attraction;

[0086] Step S102: Obtain the travel itinerary execution progress information of the user at the current time, where the travel itinerary execution progress information of the user at the current time includes the completed tourist attractions and the user's location in the current tourist attraction;

[0087] Step S103: Identify the itinerary interference events during the execution of the travel itinerary for tourist attractions, where the itinerary interference events include weather information that meets the preset conditions in the database and the situation where the proportion of the abnormal heart rate state duration is greater than the preset value, and the abnormal heart rate state is the state where the heart rate is greater than the preset heart rate value;

[0088] Step S104: Construct an itinerary execution feature set, where the itinerary execution feature set includes the itinerary interference event recognition result and the itinerary execution progress deviation value of the user at the current time; the calculation formula of the itinerary execution progress deviation value of the user at the current time is as follows:

[0089]

[0090] Wherein, DP represents the progress deviation value of the user's tourist attraction itinerary execution at the current time; T1 represents the sum of the actual stay duration corresponding to each completed tourist attraction and the duration from the start time of the current tourist attraction to the current time; T2 represents the planned stay duration value corresponding to each completed tourist attraction in the planned route of the tourist attraction itinerary; T3 represents the planned stay duration value corresponding to the current tourist attraction in the planned route of the tourist attraction itinerary; L1 represents the total length of the planned play route of the current tourist attraction in the planned route of the tourist attraction itinerary; L2 represents the length from the user's position in the current tourist attraction to the starting point of the planned play route of the current tourist attraction in the planned route of the tourist attraction itinerary.

[0091] Step S105: Perform an adaptation determination operation for itinerary emergencies based on the itinerary execution feature set.

[0092] If DP is greater than or equal to the preset deviation value or the recognition result of the itinerary interference event is not empty, then determine that the adaptation determination result of the itinerary emergency is an itinerary adaptation change; otherwise, determine that the adaptation determination result of the itinerary emergency is that the itinerary remains unchanged.

[0093] Step S200: If the adaptation determination result of the itinerary emergency is an itinerary adaptation change, obtain the scenic spot information involved in the unexecuted tourist attraction itinerary plan of the user at the current time, extract the social public opinion information corresponding to each scenic spot in the historical data and the social public opinion information corresponding to the user's historical tourist attractions through image recognition technology, and obtain the deviation sequence of the planned scenic spots of the to-be-executed itinerary.

[0094] The step S200 includes the following steps:

[0095] Step S201: Obtain the scenic spot information involved in the unexecuted tourist attraction itinerary plan of the user at the current time, and the scenic spot information includes the scenic spot name and the planned play route of the corresponding scenic spot.

[0096] Step S202: Identify the background scenery type in each scenic spot image published on the social platform through image recognition technology, and the background scenery type is preset in the database; extract the social public opinion information corresponding to each scenic spot in the historical data and the social public opinion information corresponding to the user's historical tourist attractions; the social public opinion information includes the proportion of picture data of each background scenery type, the number of pictures of the corresponding scenic spot published in the social platform, and the user collection volume of the corresponding scenic spot in the social platform.

[0097] In this example, when extracting the social public opinion information corresponding to each scenic spot information in the historical data and the social public opinion information corresponding to the historical tourist scenic spots of the corresponding users through image recognition technology, the image recognition model used is the gold standard dataset for the Places365 scene classification task, and a high-precision scenic recognition model is quickly constructed by combining a pre-trained model (such as ResNet-Places365); the Clarifai pre-set natural scene recognition model can also be directly called to extract the social public opinion information corresponding to each scenic spot information in the historical data and the social public opinion information corresponding to the historical tourist scenic spots of the corresponding users.

[0098] Step S203: Calculate the execution preference coefficient of each scenic spot information involved in the unexecuted tourist scenic spot itinerary planning route of the user at the current time. The involved calculation formula is as follows:

[0099]

[0100] where A n represents the execution preference coefficient of the nth scenic spot information involved in the unexecuted tourist scenic spot itinerary planning route of the user at the current time; τp i represents the proportion of picture data of the ith background scenery type in the social public opinion information corresponding to the nth scenic spot in the historical data; B n represents the number of pictures posted of the corresponding scenic spot in the social platform in the social public opinion information corresponding to the nth scenic spot in the historical data; C n represents the user collection volume of the corresponding scenic spot in the social platform in the social public opinion information corresponding to the nth scenic spot in the historical data; τy i represents the proportion of picture data of the ith background scenery type in the social public opinion information corresponding to the historical tourist scenic spots of the corresponding user; is represents the number of preset background scenery types in the database; μ represents a preset conversion coefficient.

[0101] Step S204: Obtain the to-be-executed itinerary planning scenic spot preference sequence, and the to-be-executed itinerary planning scenic spot preference sequence is obtained according to the order from large to small of the corresponding execution preference coefficients of the corresponding scenic spots.

[0102] Step S300: Obtain the spatio-temporal distribution data of the behavior density of each tourist scenic spot in the tourist scenic spot itinerary execution progress information of the user at the current time, and generate a user behavior preference feature set; combine the platform scenic spot information in the historical data to generate the stop behavior vector feature information in the play planning route of each scenic spot based on the to-be-executed itinerary planning scenic spot preference sequence of the user.

[0103] The step S300 includes the following steps:

[0104] Step S301: Obtain the spatio-temporal distribution data of the density of behaviors at each tourist attraction in the execution progress information of the tourist attraction itinerary of the user at the current time. The spatio-temporal distribution data of the density of behaviors is a data pair composed of the mapped point position of each photo-taking position of the user at the corresponding tourist attraction on the corresponding planned tour route and the corresponding photo-taking time. The mapped point position of each photo-taking position of the user at the corresponding tourist attraction on the corresponding planned tour route represents the position point closest to the corresponding photo-taking position of the user at the corresponding tourist attraction in the corresponding planned tour route.

[0105] Step S302: Statistically analyze the correlation between each shooting position point in the obtained spatio-temporal distribution data of the density of behaviors at each tourist attraction. Different shooting position points with an interval distance less than the preset distance between shooting positions are divided into the same array to obtain different user behavior preference feature analysis sets.

[0106] Step S303: Extract the background scenery types in the photos corresponding to each element in each user behavior preference feature analysis set. The aggregated set of the types of background scenery corresponding to each element in the same user behavior preference feature analysis set is used as a user behavior preference feature.

[0107] Step S304: Obtain the behavior preference coefficient of each user behavior preference feature. The behavior preference coefficient of the user behavior preference feature is equal to the ratio of the number of user behavior preference features in the subset of the corresponding user behavior preference feature analysis set that are the same as the user behavior preference feature of the corresponding user behavior preference feature to the total number of user behavior preference features.

[0108] Step S305: Generate a user behavior preference feature set. The user behavior preference feature set is the set composed of each user behavior preference feature obtained in Step S303, and each element in the user behavior preference feature set is bound to the corresponding behavior preference coefficient.

[0109] Step S306: Divide the corresponding scenic spot into different route sections along the planned tour route of the scenic spot according to the preset interval length. The position set composed of the position points closest to the corresponding section interval in the corresponding scenic spot is used as the scenic spot division area of the corresponding route section. Respectively, the behavior preference coefficients bound to the elements with the same extraction results of the behavior preference features of the photos bound to the corresponding division area in the same scenic spot in the social platform historical data are used as the behavior preference coefficients corresponding to the corresponding division area in the corresponding scenic spot. The binding relationship between the photos of the same scenic spot in the social platform historical data and the corresponding division area is preset manually.

[0110] Step S307: Generate the stay behavior vector feature information of the user in each scenic spot tour plan route in the tourist attraction bias sequence based on the to-be-executed itinerary plan.

[0111] The residence behavior vector feature information is composed of residence behavior vectors corresponding to each position point in the corresponding scenic spot tour planning route. The modulus of the residence behavior vector is equal to the behavior preference coefficient of the division area to which the corresponding position point belongs, and the direction of the residence behavior vector is the direction from the corresponding position point to the subsequent adjacent position point in the corresponding scenic spot tour planning route.

[0112] Step S400: Generate an adaptation result of the user's tourist scenic spot itinerary planning route based on the to-be-executed itinerary planning scenic spot preference sequence and the itinerary execution feature set; extract the dining planning information in the user's tourist scenic spot itinerary planning route, lock the user's dining adaptation area, and generate a spatio-temporal matching set of the itinerary route and dining reservation.

[0113] In the process of generating the adaptation result of the user's tourist scenic spot itinerary planning route in the step S400, predict the itinerary planning mapping duration based on the itinerary execution feature set, and the calculation formula is as follows:

[0114]

[0115] Among them, TY represents the prediction result of the itinerary planning mapping duration based on the itinerary execution feature set; TW represents the predicted value of the duration required to complete the subsequent tourist scenic spot itinerary planning route of the user; TS represents the sum of the planned residence durations corresponding to each tourist scenic spot other than the current scenic spot in the to-be-executed itinerary planning scenic spot preference sequence in the tourist scenic spot itinerary planning route.

[0116] Exclude the scenic spot to which the current time belongs in the to-be-executed itinerary planning scenic spot preference sequence to obtain an updated to-be-executed itinerary planning scenic spot preference sequence, and sequentially obtain the itinerary planning durations corresponding to each element in the updated to-be-executed itinerary planning scenic spot preference sequence; extract the summary set of the first k elements and the scenic spot to which the current time belongs in the updated to-be-executed itinerary planning scenic spot preference sequence, denoted as the route analysis set; arrange the elements in the route analysis set in the order of the scenic spots in the user's tourist scenic spot itinerary planning route, and splice the scenic spot tour planning routes corresponding to adjacent elements in the arrangement result to obtain the adaptation result of the user's tourist scenic spot itinerary planning route; the k satisfies the condition: TY k ≤TY<TY k+1 ,TY k represents the sum of the itinerary planning durations corresponding to the first k elements respectively in the updated to-be-executed itinerary planning scenic spot preference sequence; TY k+1 represents the sum of the itinerary planning durations corresponding to the first k + 1 elements respectively in the updated to-be-executed itinerary planning scenic spot preference sequence.

[0117] When locking the dining adaptation area of the user in step S400, the travel position of the user at the minimum time point of the dining planning information in the travel scenic spot itinerary planning route is preset as the distance from the starting point of the un-traveled route section to the LC at the g-th scenic spot in the travel scenic spot itinerary planning route adaptation result of the user; the travel position of the user at the maximum time point of the dining planning information in the travel scenic spot itinerary planning route is preset as the distance from the starting point of the un-traveled route section to the LC1 at the g1-th scenic spot in the travel scenic spot itinerary planning route adaptation result of the user; the following equation is satisfied:

[0118]

[0119] TR1 represents the time interval from the minimum time point in the planned dining time interval of the dining planning information to the current time; TR2 represents the time interval from the maximum time point in the planned dining time interval of the dining planning information to the current time;

[0120] TSG (2，g-1) represents the sum of the predicted duration values of the travel itinerary planning routes of the 2nd to the (g - 1)-th scenic spots in the travel scenic spot itinerary planning route adaptation result of the user; F (g，m) represents the modulus of the stay behavior vector corresponding to the position point when the distance from the starting point of the un-traveled route section to the m at the g-th scenic spot in the travel scenic spot itinerary planning route adaptation result of the user; Q g represents the integral value of the modulus of the stay behavior vector corresponding to each position point in the un-traveled route section at the g-th scenic spot in the travel scenic spot itinerary planning route adaptation result of the user; TSG (g-1，g) represents the predicted duration value of the travel itinerary planning route of the g-th scenic spot in the travel scenic spot itinerary planning route adaptation result of the user; TSG (2，g1-1) represents the sum of the predicted duration values of the travel itinerary planning routes of the 2nd to the (g1 - 1)-th scenic spots in the travel scenic spot itinerary planning route adaptation result of the user; F (g1，m) represents the modulus of the stay behavior vector corresponding to the position point when the distance from the starting point of the un-traveled route section to the m at the g1-th scenic spot in the travel scenic spot itinerary planning route adaptation result of the user; Q g1 represents the integral value of the modulus of the stay behavior vector corresponding to each position point in the un-traveled route section at the g1-th scenic spot in the travel scenic spot itinerary planning route adaptation result of the user; TSG (g1-1，g1) represents the predicted duration value of the travel itinerary planning route of the g1-th scenic spot in the travel scenic spot itinerary planning route adaptation result of the user;

[0121] The user dining adaptation area is the planar area within the preset unit radius around the play planning route section formed by each travel position of the user within the dining planning information of the travel scenic spot itinerary planning route in the travel scenic spot itinerary planning route adaptation result of the user;

[0122] The set of spatio-temporal matches between the travel route and the dining reservation is a set composed of the names of dining establishments that are located within the locked user dining adaptation area and for which there are dining reservation slots in the dining planning information within the user's travel scenic spot itinerary planning route.

[0123] Step S500: Real-time feedback to the user of the adaptation result of the user's travel scenic spot itinerary planning route and the binding result of the set of spatio-temporal matches between the travel route and the dining reservation.

[0124] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be embraced within the present invention. Any reference signs in the claims should not be construed as limiting the claims involved.

Claims

1. An intelligent recommendation management method for tourism services based on artificial intelligence, characterized in that: The method includes: Step S100: Obtain the travel itinerary planning route of the user and the progress information of the user's travel itinerary execution at the current time; identify the itinerary interference events during the execution of the travel itinerary, analyze the progress deviation value of the user's travel itinerary execution at the current time, construct an itinerary execution feature set, and perform an itinerary emergency adaptation determination operation based on the itinerary execution feature set; the itinerary emergency adaptation determination result includes itinerary adaptation change and itinerary remaining unchanged; Step S200: If the itinerary emergency adaptation determination result is itinerary adaptation change, obtain the scenic spot information involved in the unexecuted travel itinerary planning route of the user at the current time, extract the social public opinion information corresponding to each scenic spot in the historical data and the social public opinion information corresponding to the user's historical travel scenic spots through image recognition technology, and obtain the to-be-executed itinerary planning scenic spot preference sequence; Step S300: Obtain the behavior density and time-space distribution data of each travel scenic spot in the progress information of the user's travel itinerary execution at the current time, and generate a user behavior preference feature set; combine the platform scenic spot information in the historical data to generate the stay behavior vector feature information in the play planning route of each scenic spot in the to-be-executed itinerary planning scenic spot preference sequence for the user; Step S400: Generate the adaptation result of the user's travel itinerary planning route based on the to-be-executed itinerary planning scenic spot preference sequence and the itinerary execution feature set; extract the dining planning information in the user's travel itinerary planning route, lock the dining adaptation area of the user, and generate the time-space matching set of the itinerary route and the dining reservation; Step S500: Real-time feedback the adaptation result of the user's travel itinerary planning route and the binding result of the time-space matching set of the itinerary route and the dining reservation to the user.

2. The intelligent recommendation management method for tourism services based on artificial intelligence according to claim 1, characterized in that: The step S100 includes the following steps: Step S101: Obtain the travel itinerary planning route of the user, where the travel itinerary planning route includes the stay duration corresponding to each travel scenic spot and the play planning route of each travel scenic spot; Step S102: Obtain the progress information of the user's travel itinerary execution at the current time, where the progress information of the user's travel itinerary execution at the current time includes the completed travel scenic spots and the user's location in the current travel scenic spot; Step S103: Identify the itinerary interference events during the execution of the travel itinerary, where the itinerary interference events include weather information that meets the preset conditions in the database and the situation where the proportion of the abnormal heart rate state duration is greater than the preset value, and the abnormal heart rate state is the state where the heart rate is greater than the preset heart rate value; Step S104: Construct an itinerary execution feature set, where the itinerary execution feature set includes the itinerary interference event recognition result and the progress deviation value of the user's travel itinerary execution at the current time; the calculation formula of the progress deviation value of the user's travel itinerary execution at the current time is as follows: Wherein, DP represents the deviation value of the execution progress of the user's tourist attraction itinerary at the current time; T1 represents the sum of the actual stay duration corresponding to each completed tourist attraction and the interval duration from the start time of the current tourist attraction to the current time; T2 represents the planned stay duration corresponding to each completed tourist attraction in the planned route of the tourist attraction itinerary; T3 represents the planned stay duration corresponding to the current tourist attraction in the planned route of the tourist attraction itinerary; L1 represents the total length of the planned play route of the current tourist attraction in the planned route of the tourist attraction itinerary; L2 represents the length from the user's position in the current tourist attraction to the starting point of the planned play route of the current tourist attraction in the planned route of the tourist attraction itinerary. Step S105: Perform an itinerary emergency adaptation determination operation based on the itinerary execution feature set. If DP is greater than or equal to the preset deviation value or the recognition result of the itinerary interference event is not empty, it is determined that the itinerary emergency adaptation determination result is itinerary adaptation change; otherwise, it is determined that the itinerary emergency adaptation determination result is itinerary unchanged.

3. The intelligent recommendation management method for tourism services based on artificial intelligence according to claim 2, characterized in that: The step S200 includes the following steps: Step S201: Obtain the scenic spot information involved in the unexecuted tourist attraction itinerary plan of the user at the current time, and the scenic spot information includes the scenic spot name and the planned play route of the corresponding scenic spot. Step S202: Identify the background scenery type in each scenic spot image published on the social platform through image recognition technology, and the background scenery type is preset in the database; extract the social public opinion information corresponding to each scenic spot in the historical data and the social public opinion information corresponding to the historical tourist attractions of the corresponding user; the social public opinion information includes the picture data proportion of each background scenery type, the number of pictures of the corresponding scenic spot published in the social platform, and the user collection volume of the corresponding scenic spot in the social platform. Step S203: Calculate the execution preference coefficient of each scenic spot information involved in the unexecuted tourist attraction itinerary plan of the user at the current time, and the involved calculation formula is as follows: Among them, A n represents the execution preference coefficient of the nth scenic spot information involved in the travel itinerary planned by the user but not executed at the current time; τp i represents the proportion of picture data of the ith background scenery type in the social public opinion information corresponding to the nth scenic spot in the historical data; B n represents the number of pictures posted of the corresponding scenic spot in the social public opinion information corresponding to the nth scenic spot in the historical data on the social platform; C n represents the user collection volume of the corresponding scenic spot in the social public opinion information corresponding to the nth scenic spot in the historical data on the social platform; τy i represents the proportion of picture data of the ith background scenery type in the social public opinion information corresponding to the historical scenic spots of the corresponding user; is represents the number of preset background scenery types in the database; μ represents the preset conversion coefficient; Step S204: Obtain the to-be-executed itinerary plan scenic spot preference sequence, and the to-be-executed itinerary plan scenic spot preference sequence is obtained according to the order of the corresponding scenic spot corresponding execution preference coefficients from large to small.

4. The intelligent recommendation management method for tourism services based on artificial intelligence according to claim 3, characterized in that: The step S300 includes the following steps: Step S301: Obtain the behavior density and time-space distribution data of each tourist attraction in the tourist attraction itinerary execution progress information of the user at the current time, and the behavior density and time-space distribution data is a data pair composed of the mapped point position of each photo-taking position of the user in the corresponding tourist attraction on the corresponding planned play route and the corresponding photo-taking time; the mapped point position of each photo-taking position of the user in the corresponding tourist attraction on the corresponding planned play route represents the position point closest to the corresponding photo-taking position of the user in the corresponding tourist attraction in the corresponding planned play route. Step S302: Statistically analyze the correlation between each shooting position point in the obtained behavior density and time-space distribution data of each tourist attraction, and divide different shooting position points with an interval distance less than the preset distance between shooting positions into the same array to obtain different user behavior preference feature analysis sets. Step S303: Extract the background scenery types in the photos corresponding to each element in each user behavior preference feature analysis set, and summarize the set of the types of the background scenery types corresponding to the elements in the same user behavior preference feature analysis set as a user behavior preference feature; Step S304: Obtain the behavior preference coefficient of each user behavior preference feature, where the behavior preference coefficient of the user behavior preference feature is equal to the ratio of the number of user behavior preference features in the subset of the corresponding user behavior preference feature analysis set that are the same as the user behavior preference feature to the total number of user behavior preference features; Step S305: Generate a user behavior preference feature set, where the user behavior preference feature set is the set composed of the various user behavior preference features obtained in Step S303, and each element in the user behavior preference feature set is bound to the corresponding behavior preference coefficient; Step S306: Divide the corresponding scenic spots into different route sections along the scenic spot tour planning route according to a preset interval length, and use the set of position points in the corresponding scenic spots that are closest to the corresponding road section interval as the scenic spot division area of the corresponding route section; respectively, use the behavior preference coefficients bound to the elements with the same extraction results of the behavior preference features of the photos bound to the corresponding division area of the same scenic spot in the social platform historical data as the behavior preference coefficients corresponding to the corresponding division area in the corresponding scenic spot; the binding relationship between the photos of the same scenic spot in the social platform historical data and the corresponding division area is preset artificially; Step S307: Generate the stay behavior vector feature information in the tour planning route of each scenic spot in the user's scenic spot preference sequence based on the to-be-executed itinerary; The stay behavior vector feature information is composed of the stay behavior vectors corresponding to each position point in the corresponding scenic spot tour planning route. The modulus length of the stay behavior vector is equal to the behavior preference coefficient of the division area to which the corresponding position point belongs, and the direction of the stay behavior vector is the direction from the corresponding position point to the subsequent adjacent position point in the corresponding scenic spot tour planning route.

5. The intelligent recommendation management method for tourism services based on artificial intelligence according to claim 2, characterized in that: In the process of generating the travel scenic spot itinerary planning route adaptation result of the user in Step S400, the itinerary planning mapping duration is predicted based on the itinerary execution feature set, and the calculation formula is as follows: Among them, TY represents the prediction result of the itinerary planning mapping duration based on the itinerary execution feature set; TW represents the predicted value of the duration required to complete the user's subsequent travel scenic spot itinerary planning route; TS represents the sum of the planned stay duration values corresponding to each of the travel scenic spots other than the current scenic spot in the to-be-executed itinerary scenic spot preference sequence in the travel scenic spot itinerary planning route; Remove the scenic spot corresponding to the current time from the to-be-executed itinerary planning scenic spot preference sequence to obtain an updated to-be-executed itinerary planning scenic spot preference sequence, and sequentially obtain the itinerary planning durations corresponding to each element in the updated to-be-executed itinerary planning scenic spot preference sequence; extract the summary set of the first k elements and the scenic spot corresponding to the current time in the updated to-be-executed itinerary planning scenic spot preference sequence, denoted as the route analysis set; arrange the elements in the route analysis set in the order of the scenic spots in the user's tourist scenic spot itinerary planning route, and splice the scenic spot play planning routes corresponding to adjacent elements in the arrangement result to obtain the user's tourist scenic spot itinerary planning route adaptation result; the k satisfies the condition: TY k ≤TY<TY k+1 , TY k represents the sum of the itinerary planning durations corresponding to the first k elements in the updated to-be-executed itinerary planning scenic spot preference sequence respectively; TY k+1 represents the sum of the itinerary planning durations corresponding to the first k + 1 elements in the updated to-be-executed itinerary planning scenic spot preference sequence respectively.

6. The intelligent recommendation management method for tourism services based on artificial intelligence according to claim 4, characterized in that: When locking the dining adaptation area of the user in Step S400, preset the itinerary position of the user at the minimum time point of the dining planning information in the travel scenic spot itinerary planning route as the distance from the initial point of the un-traveled route section to LC in the g-th scenic spot in the travel scenic spot itinerary planning route adaptation result of the user; Preset the travel location of the user at the maximum time point of the dining plan information in the travel scenic spot itinerary planning route as the distance LC1 from the starting point of the un-traveled route section to the g1-th scenic spot in the travel scenic spot itinerary planning route adaptation result of the user; satisfy the following equation: TR1 represents the time interval from the minimum time point in the planned dining time interval of the dining plan information to the current time; TR2 represents the time interval from the maximum time point in the planned dining time interval of the dining plan information to the current time; TSG (2，g-1) It represents the sum of the predicted durations of the itinerary planning routes for the 2nd to the (g - 1)th scenic spots in the adaptation result of the user's travel itinerary planning route for scenic spots; F (g，m) represents the modulus of the stay behavior vector corresponding to the position point when the distance from the starting point of the un-traveled route section in the g-th scenic spot in the travel scenic spot itinerary planning route adaptation result of the user is m; Q g represents the integral value of the modulus of the stay behavior vector corresponding to each position point in the un-traveled route section in the g-th scenic spot in the travel scenic spot itinerary planning route adaptation result of the user; TSG (g-1，g) It represents the predicted value of the required duration of the g-th scenic spot itinerary planning route in the adaptation result of the user's scenic spot itinerary planning route. TSG (2，g1-1) It represents the sum of the predicted duration values of the scenic spot itinerary planning routes from the 2nd to the (g1 - 1)th in the adaptation result of the user's scenic spot itinerary planning route. F (g1，m) represents the modulus of the stay behavior vector corresponding to the position point when the distance from the initial point of the non-traveled route section to the g1-th scenic spot in the travel scenic spot itinerary planning route adaptation result of the user is m; Q g1 represents the integral value of the modulus of the stay behavior vector corresponding to each position point in the non-traveled route section of the g1-th scenic spot in the travel scenic spot itinerary planning route adaptation result of the user; TSG (g1-1，g1) represents the predicted value of the required duration of the travel scenic spot itinerary planning route of the g1-th scenic spot in the travel scenic spot itinerary planning route adaptation result of the user; The user dining adaptation area is a planar area within the travel scenic spot itinerary planning route adaptation result of the user, within the dining plan information of the travel scenic spot itinerary planning route of the user, and within a preset unit radius around the play plan route section formed by each travel location of the user; The spatio-temporal matching set of the travel route and dining reservation is a set composed of the names of dining stores located within the locked user dining adaptation area and having dining reservation quotas within the dining plan information in the travel scenic spot itinerary planning route of the user; 7. An intelligent recommendation management system for tourism services based on artificial intelligence, which applies the intelligent recommendation management method for tourism services based on artificial intelligence according to any one of claims 1-6, characterized in that: The system includes a user demand change dynamic determination module, a travel plan preference analysis module, a behavior vector feature analysis module, a demand adaptation recommendation module, and an adaptation information feedback module; The user demand change dynamic determination module obtains the travel scenic spot itinerary planning route of the user and the travel scenic spot itinerary execution progress information of the user at the current time; identifies itinerary interference events during the execution of the travel scenic spot itinerary, analyzes the itinerary execution progress deviation value of the user at the current time, constructs an itinerary execution feature set, and performs an itinerary emergency adaptation determination operation based on the itinerary execution feature set; the itinerary emergency adaptation determination result includes itinerary adaptation change and itinerary remaining unchanged; If the itinerary emergency adaptation determination result of the itinerary plan preference analysis module is itinerary adaptation change, it obtains the scenic spot information involved in the unexecuted travel scenic spot itinerary plan route of the user at the current time, extracts the social public opinion information corresponding to each scenic spot in the historical data and the social public opinion information corresponding to the user's historical travel scenic spots through image recognition technology, and obtains the pending travel plan scenic spot preference sequence; The behavior vector feature analysis module obtains the spatio-temporal distribution data of the behavior density of each travel scenic spot in the travel scenic spot itinerary execution progress information of the user at the current time, and generates a user behavior preference feature set; combines the platform scenic spot information in the historical data to generate the stay behavior vector feature information of the user in the play plan route of each scenic spot in the pending travel plan scenic spot preference sequence; The demand adaptation recommendation module generates the travel scenic spot itinerary planning route adaptation result of the user based on the pending travel plan scenic spot preference sequence and the itinerary execution feature set; extracts the dining plan information in the travel scenic spot itinerary planning route of the user, locks the user dining adaptation area, and generates the spatio-temporal matching set of the travel route and dining reservation; The adaptation information feedback module real-time feeds back the binding result of the travel scenic spot itinerary planning route adaptation result of the user and the spatio-temporal matching set of the travel route and dining reservation to the user.

8. The intelligent recommendation management system and method for tourism services based on artificial intelligence according to claim 7, characterized in that: The user demand change dynamic determination module includes a travel execution progress acquisition unit, a travel execution feature set construction unit, and an adaptation determination unit; The travel execution progress acquisition unit acquires the travel route of the user's tourist attractions itinerary plan and the travel execution progress information of the user's tourist attractions at the current time; The travel execution feature set construction unit identifies travel interference events during the execution of the tourist attractions itinerary, analyzes the travel execution progress deviation value of the user's tourist attractions at the current time, and constructs a travel execution feature set; The adaptation determination unit performs a travel emergency adaptation determination operation based on the travel execution feature set.

9. The intelligent recommendation management system and method for tourism services based on artificial intelligence according to claim 7, characterized in that: The behavior vector feature analysis module includes a user behavior preference feature analysis unit and a stay behavior vector acquisition unit, The user behavior preference feature analysis unit acquires the behavior density spatio-temporal distribution data of each tourist attraction in the travel execution progress information of the user's tourist attractions at the current time, and generates a user behavior preference feature set; The stay behavior vector acquisition unit combines the platform attraction information in the historical data to generate the stay behavior vector feature information in the play plan route of each attraction in the user's stay behavior vector feature information based on the to-be-executed itinerary plan attraction preference sequence.

10. The intelligent recommendation management system and method for tourism services based on artificial intelligence according to claim 7, characterized in that: The demand adaptation recommendation module includes a route adaptation unit and a catering service adaptation unit, The route adaptation unit generates a travel route adaptation result of the user's tourist attractions itinerary plan based on the to-be-executed itinerary plan attraction preference sequence and the travel execution feature set; The catering service adaptation unit extracts the dining plan information in the travel route of the user's tourist attractions itinerary plan, locks the user's dining adaptation area, and generates a spatio-temporal matching set of the travel route and catering reservation.

Images