A tourist route recommendation system and method based on big data

Abstract

The present invention relates to the technical field of travel route recommendation, and in particular to a travel route recommendation system and method based on big data. The method comprises the following steps: collecting travel demand data and basic physical sign information of users, and performing hierarchical analysis on the demand to obtain master / slave demand data; then, simulating muscle soreness intensity by combining user physical sign data through intersection analysis of demand behavior, and further estimating the physical fitness decline rate; formulating a rest and recuperation strategy based on the physical fitness decline rate, and optimizing the allocation and adjustment of travel routes; finally, transmitting the adjusted travel route data to the user terminal. The present invention makes the travel route recommendation technology more perfect by improving the processing of the travel route recommendation technology.

Description

Technical Field

[0001] The present invention relates to the technical field of travel route recommendation, and in particular to a travel route recommendation system and method based on big data. Background Art

[0002] With the development of information technology and big data, the tourism industry is undergoing profound changes. More and more tourists are seeking personalized and customized travel experiences, and the previous travel recommendation methods can no longer meet the needs of modern tourists. Big data technology provides unprecedented opportunities for the tourism industry. By collecting and analyzing massive amounts of user data, behavioral data, geographic information, weather information, etc., tourists' preferences and needs can be more accurately grasped. At the same time, as consumers' requirements for the quality of tourism services increase, previous tourism recommendation systems often rely solely on users' basic preferences, ignoring individual differences such as physical fitness and health. The travel route recommendation system based on big data can comprehensively consider tourists' multi-dimensional information, such as physical fitness level, health status, and special needs during travel, through deep learning and data mining technology, so as to provide more detailed and accurate recommendation solutions. This can not only effectively improve tourists' satisfaction, but also optimize the allocation of tourism resources and improve the overall quality of tourism services. However, a traditional travel route recommendation method based on big data has the problem of low accuracy in identifying users' motion status, resulting in large errors in personalized recommendation of travel routes. Summary of the invention

[0003] Based on this, it is necessary to provide a travel route recommendation system and method based on big data to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a method for recommending tourist routes based on big data is provided, the method comprising the following steps:

[0005] Step S1: obtaining a tourism demand data set input by a user and basic physical sign data of the user; performing a master / slave demand hierarchy analysis on the tourism demand data set to obtain tourism master / slave demand hierarchy data;

[0006] Step S2: Performing a master / slave demand behavior intersection analysis on the tourism master / slave demand hierarchy data to obtain master / slave demand behavior intersection data; performing a physical sign muscle soreness intensity simulation estimation on the user's individual basic physical sign data based on the master / slave demand behavior intersection data to obtain physical sign muscle soreness intensity data; performing physical fitness decline rate matching based on the physical sign muscle soreness intensity data to obtain physical fitness decline rate data;

[0007] Step S3: Formulate a node segmentation rest strategy for the intersection data of the master / slave demand behavior based on the physical fitness decline rate data to obtain a tourism-related node segmentation rest strategy; adjust the tourism route allocation according to the tourism-related node segmentation rest strategy to obtain tourism route allocation adjustment data, and send the tourism route allocation adjustment data to the terminal.

[0008] Preferably, step S1 comprises the following steps:

[0009] Step S11: obtaining a travel demand data set input by a user and basic individual vital sign data of the user;

[0010] Step S12: performing data cleaning on the tourism demand data set to obtain a tourism demand cleaned data set;

[0011] Step S13: parsing the context statement logical structure of the tourism demand cleansing data set to obtain the demand context statement logical structure;

[0012] Step S14: performing master / slave demand hierarchy analysis on the tourism demand cleansing data set according to the demand context statement logical structure to obtain tourism master / slave demand hierarchy data.

[0013] Preferably, step S2 comprises the following steps:

[0014] Step S21: performing a master / slave demand behavior intersection analysis on the tourism master / slave demand hierarchy data to obtain master / slave demand behavior intersection data;

[0015] Step S22: Performing exercise intensity evaluation based on the master / slave demand behavior intersection data to obtain travel master / slave demand exercise intensity data;

[0016] Step S23: performing a physical sign muscle soreness intensity simulation estimation on the user's individual basic physical sign data according to the travel master / slave demand exercise intensity data to obtain physical sign muscle soreness intensity data;

[0017] Step S24: Perform physical fitness decline rate matching based on the physical sign muscle soreness intensity data to obtain physical fitness decline rate data.

[0018] Preferably, step S23 includes the following steps:

[0019] Step S231: extracting basic physical sign parameters of the user from the basic physical sign data of the user to obtain basic physical sign parameters of the user, wherein the basic physical sign parameters of the user include height and weight, basic heart rate, body fat rate and basic breathing frequency;

[0020] Step S232: performing exercise intensity quantification processing on the required exercise intensity data of the tourist master / slave to obtain required exercise intensity quantitative data;

[0021] Step S233: performing a metabolic equivalent relative iterative change analysis on the user's basic vital sign parameters according to the required exercise intensity quantitative data to obtain exercise metabolic equivalent iterative change data;

[0022] Step S234: evaluating muscle viscosity kinetic energy based on the exercise metabolism equivalent iterative change data and the required exercise intensity quantitative data to obtain muscle viscosity kinetic energy data;

[0023] Step S235: Perform a simulation estimate of the intensity of physical signs of muscle soreness based on the exercise metabolism equivalent iterative change data and the muscle viscosity kinetic energy data to obtain physical signs of muscle soreness intensity data.

[0024] Preferably, step S3 comprises the following steps:

[0025] Step S31: normalizing the physical fitness decline rate data to obtain normalized physical fitness decline rate data;

[0026] Step S32: extracting tourism-related nodes based on the intersection data of master / slave demand behaviors to obtain a master / slave demand tourism-related node data set;

[0027] Step S33: formulating a node segmentation rest strategy for the master / slave demand tourism-related node data set based on the normalized data of the physical fitness decline rate, and obtaining a tourism-related node segmentation rest strategy;

[0028] Step S34: adjusting the allocation of tourist routes according to the normalized data of the physical fitness decline rate and the segmented rest and recuperation strategy of tourist-related nodes, and obtaining tourist route allocation adjustment data.

[0029] Preferably, step S33 includes the following steps:

[0030] Step S331: Obtain historical playing time statistics of each node according to the master / slave demand tourism associated node data set;

[0031] Step S332: performing a concentrated long-tail skewness analysis on the historical play time statistical data to obtain concentrated long-tail skewness data on the historical play time;

[0032] Step S333: Based on the normalized data of physical energy decay rate, the long-tail skewed data in the historical play time concentration is used to perform a cumulative calculation of the node physical energy consumption to obtain a node cumulative physical energy consumption sequence;

[0033] Step S334: performing piecewise linear fitting on the node cumulative physical energy consumption sequence to obtain piecewise fitting data of node physical energy consumption;

[0034] Step S335: Formulate a node segmented rest and recuperation strategy based on the segmented fitting data of node physical energy consumption to obtain a segmented rest and recuperation strategy for tourism-related nodes.

[0035] Preferably, the present invention also provides a travel route recommendation system based on big data, which is used to execute the travel route recommendation method based on big data as described above, and the travel route recommendation system based on big data includes:

[0036] The master / slave demand hierarchy parsing module is used to obtain the tourism demand data set and the user's individual basic vital sign data input by the user; perform master / slave demand hierarchy parsing on the tourism demand data set to obtain tourism master / slave demand hierarchy data;

[0037] The physical fitness decline rate matching module is used to perform the intersection analysis of the main / slave demand behavior on the tourism main / slave demand hierarchy data to obtain the main / slave demand behavior intersection data; simulate and estimate the physical fitness muscle soreness intensity of the user's individual basic physical sign data based on the main / slave demand behavior intersection data to obtain the physical fitness muscle soreness intensity data; perform physical fitness decline rate matching based on the physical fitness muscle soreness intensity data to obtain the physical fitness decline rate data;

[0038] The travel route allocation adjustment module is used to formulate a node segmentation rest strategy for the intersection data of the master / slave demand behavior based on the physical fitness decline rate data, and obtain the travel-related node segmentation rest strategy; adjust the travel route allocation according to the travel-related node segmentation rest strategy, obtain the travel route allocation adjustment data, and send the travel route allocation adjustment data to the terminal.

[0039] The beneficial effect of the present invention is that by obtaining the tourism demand data set input by the user and the user's individual basic vital sign data, and performing master / slave demand hierarchical analysis on the tourism demand data set, accurate classification and structured processing of user needs can be achieved. The main demand hierarchical data can clearly express the core intention of the user, and the slave demand hierarchical data supplements the auxiliary demand information. This processing method avoids redundant interference of demand data, improves the efficiency and accuracy of demand analysis, and lays the foundation for subsequent behavior analysis and vital sign simulation. Especially in the personalized tourism recommendation scenario, the master / slave demand hierarchical analysis can greatly improve the accuracy of matching and the pertinence of services. By performing master / slave demand behavior intersection analysis on the tourism master / slave demand hierarchical data, the user's behavioral preferences and action relevance can be effectively extracted, and the simulation estimation of the physical sign muscle soreness intensity can be completed in combination with the user's individual basic vital sign data to form an accurate physical load evaluation index. Physical decline rate matching based on physical sign muscle soreness intensity data can predict the user's physical change trend and fatigue accumulation degree at different tourism nodes. This process realizes the deep combination of demand behavior and individual characteristics, provides a scientific basis for dynamically adjusting tourism planning, and enhances the controllability of health management in personalized tourism. By formulating the node segmentation rest strategy for the intersection data of master / slave demand behavior based on the physical decline rate data, it is possible to reasonably design the rest nodes in the whole process of tourism and balance the physical load of users. The adjustment of tourism route allocation in combination with the segmentation rest strategy of tourism-related nodes can optimize the time and space arrangement of the route, making the user's travel experience more comfortable and efficient. The adjusted tourism route allocation data is sent to the terminal, and updated planning information can be provided to the user in real time to improve the timeliness and interactivity of the service. This closed-loop optimization method effectively improves the scientificity and personalization of tourism planning, while reducing the discomfort or risk caused by physical decline of users, and improving the satisfaction and safety of the overall tourism experience. Therefore, the present invention is an improved processing of a traditional method for recommending tourist routes based on big data, which solves the problem that the traditional method for recommending tourist routes based on big data has low accuracy in identifying the user's motion state, thereby causing large errors in personalized recommendation of tourist routes, improves the accuracy of identifying the user's motion state, and reduces the error of personalized recommendation of tourist routes. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A schematic diagram of the steps of a travel route recommendation method based on big data;

[0041] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0042] Figure 3 for Figure 1Detailed implementation steps of step S3 in FIG. DETAILED DESCRIPTION

[0043] See also Figures 1 to 3 , a tourist route recommendation method based on big data, the method comprising the following steps:

[0044] Step S1: obtaining a tourism demand data set input by a user and basic physical sign data of the user; performing a master / slave demand hierarchy analysis on the tourism demand data set to obtain tourism master / slave demand hierarchy data;

[0045] Step S2: Performing a master / slave demand behavior intersection analysis on the tourism master / slave demand hierarchy data to obtain master / slave demand behavior intersection data; performing a physical sign muscle soreness intensity simulation estimation on the user's individual basic physical sign data based on the master / slave demand behavior intersection data to obtain physical sign muscle soreness intensity data; performing physical fitness decline rate matching based on the physical sign muscle soreness intensity data to obtain physical fitness decline rate data;

[0046] Step S3: Formulate a node segmentation rest strategy for the intersection data of the master / slave demand behavior based on the physical fitness decline rate data to obtain a tourism-related node segmentation rest strategy; adjust the tourism route allocation according to the tourism-related node segmentation rest strategy to obtain tourism route allocation adjustment data, and send the tourism route allocation adjustment data to the terminal.

[0047] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a tourist route recommendation method based on big data of the present invention. In this example, the tourist route recommendation method based on big data includes the following steps:

[0048] Step S1: obtaining a tourism demand data set input by a user and basic physical sign data of the user; performing a master / slave demand hierarchy analysis on the tourism demand data set to obtain tourism master / slave demand hierarchy data;

[0049] In an embodiment of the present invention, an input tourism demand data set and user individual basic physical sign data are received from a user terminal through an interface. Among them, the tourism demand data set includes the user's preferred area, target activity type (such as mountaineering, sightseeing, etc.), acceptable travel time and budget restrictions, etc.; the user's individual basic physical sign data includes basic physiological parameters such as height, weight, basic heart rate, body fat rate, and respiratory rate. The tourism demand data set is preprocessed. First, duplicate, null or abnormal data points are removed by a cleaning algorithm to ensure data integrity and consistency. Then, a logical structure parsing method based on contextual semantic analysis is used to perform word segmentation, part-of-speech tagging, and phrase segmentation on the text content of the demand data set. The user demand is mapped to the logical semantic space in combination with the word vector method to form a semantic tree structure of tourism demand. Based on the hierarchical characteristics of the semantic tree structure, a hierarchical clustering algorithm is used to divide the key attributes of the demand into primary demand and secondary demand, and obtain the tourism primary / secondary demand hierarchical data. Among them, the primary demand is the user's core goal (such as visiting a specific attraction), and the secondary demand is an auxiliary or secondary demand (such as surrounding dining and shopping experience).

[0050] Step S2: Performing a master / slave demand behavior intersection analysis on the tourism master / slave demand hierarchy data to obtain master / slave demand behavior intersection data; performing a physical sign muscle soreness intensity simulation estimation on the user's individual basic physical sign data based on the master / slave demand behavior intersection data to obtain physical sign muscle soreness intensity data; performing physical fitness decline rate matching based on the physical sign muscle soreness intensity data to obtain physical fitness decline rate data;

[0051] In the embodiment of the present invention, the intersection analysis of the master / slave demand hierarchy data generated in step S1 is performed, and the overlap and coupling relationship between the master demand and the slave demand are evaluated based on the behavior feature analysis method. Specifically, the behavior time window, activity range, consumption resources and other elements of each demand item are calculated, and the intersection data of the master / slave demand behavior is obtained by cross-comparison, and the intersection part represents the demand activity area and time period that overlap in time and space. Based on the intersection data of the master / slave demand behavior, combined with the individual basic physical sign data of the user, the simulation estimation of muscle soreness intensity is performed. First, the key indicators of the user's basic physical sign parameters are extracted, for example, the BMI index is calculated based on height and weight, and the user's resting metabolic rate is estimated in combination with the basic heart rate and body fat rate. Then, the corresponding exercise intensity (such as walking distance, climbing slope, etc.) in the travel behavior intersection data is quantified, and the user's muscle metabolic load is dynamically simulated using the exercise metabolism equivalent formula. Here, the linear relationship between the user's breathing frequency and exercise intensity is used, combined with the coupling effect of the breathing increment and the change of body fat rate, and the muscle lactic acid accumulation rate of the user under specific demand conditions is gradually derived, thereby obtaining muscle soreness intensity data. Based on the muscle soreness intensity data and combined with the user's exercise metabolism parameters, the physical decline rate calculation formula is used for calculation. The physical decline rate is matched by the relationship between the cumulative exercise metabolic load and the recovery time. The result data reflects the dynamic trend of the user's physical energy consumption during continuous exercise.

[0052] Step S3: Formulate a node segmentation rest strategy for the intersection data of the master / slave demand behavior based on the physical fitness decline rate data to obtain a tourism-related node segmentation rest strategy; adjust the tourism route allocation according to the tourism-related node segmentation rest strategy to obtain tourism route allocation adjustment data, and send the tourism route allocation adjustment data to the terminal.

[0053] In an embodiment of the present invention, a segmented rest strategy for tourist nodes is formulated based on the physical energy decay rate data and the intersection data of the master / slave demand behavior. First, the historical play time distribution of the key nodes involved in the intersection data is extracted, and the average stay time and long-tail skewness characteristics of each node are calculated using a statistical analysis method, and the cumulative physical energy consumption of each node is predicted in combination with the physical energy decay rate data. The cumulative physical energy consumption uses the distance between nodes, the difficulty of the road, and the user's motion characteristics as parameters, and a segmented physical energy consumption trend curve is generated by linear regression fitting. On this basis, a piecewise linear fitting algorithm is used to divide the physical energy recovery demand points in different stages, and an optimal allocation plan for the node rest time is formulated. Combined with the segmented rest strategy, the tourist route is dynamically adjusted. The route optimization algorithm is used to reorder the connection relationship between nodes to ensure that users complete the necessary rest before the physical energy decay reaches the limit, while maximizing the coverage of the main demand target node. Specifically, the solution method of the traveling salesman problem is used to combine the normalized data of the physical energy decay rate and the optimized weights between nodes to perform route planning calculations, and finally generate adjusted tourist route allocation data. The adjusted route data includes the order of play, stay time and optimized exercise distribution ratio of each node. Finally, the travel route allocation adjustment data is sent to the terminal device through the interface to provide users with an optimized travel plan.

[0054] Step S1 includes the following steps:

[0055] Step S11: obtaining a travel demand data set input by a user and basic individual vital sign data of the user;

[0056] Step S12: performing data cleaning on the tourism demand data set to obtain a tourism demand cleaned data set;

[0057] Step S13: parsing the context statement logical structure of the tourism demand cleansing data set to obtain the demand context statement logical structure;

[0058] Step S14: performing master / slave demand hierarchy analysis on the tourism demand cleansing data set according to the demand context statement logical structure to obtain tourism master / slave demand hierarchy data.

[0059] In an embodiment of the present invention, a tourism demand data set and individual basic vital signs data of a user inputted by a user are received through a communication interface. The tourism demand data set is composed of structured and unstructured data provided by the user, including the destination name, travel duration, main activity type, budget range, etc., and the data format is a multidimensional key-value pair or a free text input form. The individual basic vital signs data of the user is obtained through measurement or historical records, and includes specific physiological indicators such as the user's height, weight, basic heart rate, body fat rate, and basic breathing frequency. These data are stored in JSON or other standardized formats, and are preliminarily normalized by a data extraction module to ensure that subsequent analysis can be based on a consistent structure. The field names and values ​​in the input data are mapped using a fixed template to eliminate structural deviations caused by inconsistent data sources. After the tourism demand data set is inputted, a data cleaning process is first applied, including operations such as format specification, duplicate data detection, outlier processing, and missing data completion. The format specification uses regular expression-based parsing rules to clean up punctuation, character case, and blank characters for free text input. Duplicate data detection identifies redundant records by calculating hash value comparisons, and retains the latest input according to timestamps. Outlier processing uses statistical methods to determine the reasonable interval of each data item, such as setting upper and lower limits based on the historical distribution of the budget range, and using the median backfill method for values ​​outside the range. For missing value completion, based on the collaborative filtering method, similar data of other users are used as references to generate supplementary data for missing items by weighted average. After cleaning, the output tourism demand cleaning data set is standardized and highly complete structured data. For the free text fields in the tourism demand cleaning data set, natural language processing (NLP) technology is used to analyze the logical structure of contextual sentences. First, the text data is segmented, and the maximum forward matching method based on dictionary matching is used to identify key phrases. At the same time, the word segmentation results are tagged by part-of-speech tagging algorithms (such as CRF or HMM models). Secondly, the semantic dependency relationship diagram of the sentence is constructed using dependency syntactic analysis technology (such as the dependency tree generation method based on the arc-eager algorithm) to clarify the association logic between the subject, predicate, and object. Through rule template matching and word vector embedding methods, user needs are mapped to the predefined demand semantic space, and a logical structure tree is generated in combination with context semantic relations to represent the hierarchy and semantic dependency of user needs. According to the logical structure of the demand context statement constructed in step S13, combined with the master / slave demand hierarchical division algorithm, user needs are hierarchically parsed into master needs and slave needs. The parsing process adopts a hierarchical clustering-based method. First, the feature vector of each demand is extracted, including the semantic vector (generated based on the Word2Vec model) and the demand intensity weight (calculated by the frequency and position of the keyword in the context). Cosine similarity calculation is applied to all demand vectors to generate a similarity matrix, and a distance-based hierarchical clustering algorithm (such as the Ward method) is used to hierarchically group the needs.Based on the clustering results, we select the needs that are directly related to the user's core tourism goals as the main needs (such as sightseeing, core activity participation, etc.), and classify the remaining needs as secondary needs (such as surrounding facilities, auxiliary services). The output tourism main / secondary demand hierarchical data is a structured tree data structure, and each node contains detailed information such as demand description, weight, semantic relationship, etc., which provides input basis for subsequent behavior analysis.

[0060] Step S2 includes the following steps:

[0061] Step S21: performing a master / slave demand behavior intersection analysis on the tourism master / slave demand hierarchy data to obtain master / slave demand behavior intersection data;

[0062] Step S22: Performing exercise intensity evaluation based on the master / slave demand behavior intersection data to obtain travel master / slave demand exercise intensity data;

[0063] Step S23: performing a physical sign muscle soreness intensity simulation estimation on the user's individual basic physical sign data according to the travel master / slave demand exercise intensity data to obtain physical sign muscle soreness intensity data;

[0064] Step S24: Perform physical fitness decline rate matching based on the physical sign muscle soreness intensity data to obtain physical fitness decline rate data.

[0065] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0066] Step S21: performing a master / slave demand behavior intersection analysis on the tourism master / slave demand hierarchy data to obtain master / slave demand behavior intersection data;

[0067] In the embodiment of the present invention, the intersection analysis of the main / slave demand behavior is realized by matching the main demand behavior data and the slave demand behavior data. The specific operation includes the following steps: First, the tourism main demand hierarchy data and the slave demand hierarchy data are parsed, and uniformly encoded according to the time dimension, space dimension, and activity type dimension, and these demand data are converted into quantifiable behavior sequences. Subsequently, the set operation method is used to construct sets for the main demand behavior and the slave demand behavior respectively. The main demand behavior set and the slave demand behavior set are matched and analyzed by the logical operator "and" (intersection operation), and the behavior fragments that meet the main demand and the slave demand conditions are identified. In order to improve the accuracy of the analysis, the dynamic time warping algorithm is used to align the main and slave demand behavior sequences in time, correct the existing time deviation, and ensure the accuracy of the matching results. The analysis results are output as the intersection data of the main / slave demand behavior and saved in the form of a multidimensional matrix, including feature dimensions such as time, activity type, and spatial position.

[0068] Step S22: Performing exercise intensity evaluation based on the master / slave demand behavior intersection data to obtain travel master / slave demand exercise intensity data;

[0069] In an embodiment of the present invention, the exercise intensity assessment is carried out based on the activity type, time and space feature dimensions in the intersection data of the master / slave demand behavior. First, according to the activity types contained in the intersection data, such as mountaineering, hiking, swimming, etc., the exercise intensity of each activity type is numerically represented with reference to the internationally accepted metabolic equivalent (METs) standard. Secondly, the exercise intensity is time-weighted according to the length of the activity time, and the specific operation is to multiply the activity duration by the metabolic equivalent value of the corresponding activity type. For data containing spatial changes, such as altitude changes during hiking, the cumulative values ​​of horizontal distance and vertical distance are calculated using a geographic information system (GIS) tool, and combined with physiological formulas, the additional impact of terrain changes on exercise intensity is estimated. The above method is used to accurately quantify the exercise intensity of each activity unit, and the results are summarized to form the master / slave demand exercise intensity data, which is expressed in the form of a matrix or time series.

[0070] Step S23: performing a physical sign muscle soreness intensity simulation estimation on the user's individual basic physical sign data according to the travel master / slave demand exercise intensity data to obtain physical sign muscle soreness intensity data;

[0071] In an embodiment of the present invention, the simulation estimation of the intensity of physical signs of muscle soreness is carried out based on the joint analysis of the basic physical sign parameters and exercise intensity data of the individual user. The specific implementation steps are: first, extract relevant parameters from the basic physical sign data of the individual user, including height, weight, basic heart rate, body fat rate and respiratory rate, etc. Then, according to the exercise intensity data of the master / slave demand of the tourist, calculate the energy consumption and muscle workload of the user under a specific activity intensity, and use the energy metabolism formula to couple the exercise intensity data with the user's physical sign parameters for analysis. Then, according to the theory of muscle physiology, the lactic acid accumulation model is used to estimate the change in muscle lactic acid concentration under the action of exercise intensity and duration, and the muscle soreness intensity is calculated in combination with the muscle viscosity characteristics. Finally, the physical sign muscle soreness intensity sequence data of the user under different time and activity conditions is generated.

[0072] Step S24: Perform physical fitness decline rate matching based on the physical sign muscle soreness intensity data to obtain physical fitness decline rate data.

[0073] In an embodiment of the present invention, the physical fitness decline rate matching is achieved by analyzing the relationship between the physical sign muscle soreness intensity data and the exercise duration. The specific implementation steps are: first, the physical sign muscle soreness intensity data is converted into a muscle fatigue degree index, and different fatigue stages are divided according to the fatigue threshold theory. Secondly, according to the activity duration and the corresponding fatigue degree, the piecewise linear regression method is used to fit the physical fitness decline rate curve, and the physical fitness decline rate value of each activity stage is derived. In order to further improve the matching accuracy, the nonlinear optimization algorithm is used to dynamically correct the physical fitness decline rate in combination with the heart rate and respiratory rate change rate in the user's individual basic signs, and a complete physical fitness decline rate time series is generated. Finally, the output physical fitness decline rate data is used to guide subsequent travel route optimization and node rest strategy formulation.

[0074] Step S23 includes the following steps:

[0075] Step S231: extracting basic physical sign parameters of the user from the basic physical sign data of the user to obtain basic physical sign parameters of the user, wherein the basic physical sign parameters of the user include height and weight, basic heart rate, body fat rate and basic breathing frequency;

[0076] Step S232: performing exercise intensity quantification processing on the required exercise intensity data of the tourist master / slave to obtain required exercise intensity quantitative data;

[0077] Step S233: performing a metabolic equivalent relative iterative change analysis on the user's basic vital sign parameters according to the required exercise intensity quantitative data to obtain exercise metabolic equivalent iterative change data;

[0078] Step S234: evaluating muscle viscosity kinetic energy based on the exercise metabolism equivalent iterative change data and the required exercise intensity quantitative data to obtain muscle viscosity kinetic energy data;

[0079] Step S235: Perform a simulation estimate of the intensity of physical signs of muscle soreness based on the exercise metabolism equivalent iterative change data and the muscle viscosity kinetic energy data to obtain physical signs of muscle soreness intensity data.

[0080] In an embodiment of the present invention, the extraction of basic vital sign parameters of users is completed by classifying and feature analyzing the vital sign data provided by users. The specific implementation steps include: first, storing the individual vital sign data input by the user in a structured format, such as JSON or CSV table format; secondly, extracting basic vital sign data related to sports physiology according to predetermined parameter requirements, including height, weight, resting heart rate, body fat rate and basic breathing frequency, etc. The extraction process uses a string matching algorithm to parse the input field and identify the target parameter through the key field index. If there is missing data, an interpolation algorithm or a standard value is used instead to ensure data integrity. After the extraction is completed, the basic vital sign parameters of the user are stored in the form of a structured vector for subsequent calculation and analysis. The quantitative processing of exercise intensity takes the master / slave demand exercise intensity data as input and performs numerical characterization for different activity types and environmental characteristics. The specific operation is as follows: first, the activity types in the required exercise intensity data are classified and coded, for example, mountaineering is set as activity code 1, and swimming is set as activity code 2; secondly, the international standard metabolic equivalent (METs) database is consulted according to the activity type, and the corresponding metabolic equivalent value is extracted as the basic exercise intensity index. Subsequently, the activity duration is multiplied by the metabolic equivalent value to calculate the total energy consumption; for activity types involving terrain changes (such as hiking), the digital elevation model (DEM) data is used to calculate the cumulative climbing height, and the exercise intensity value is corrected in combination with the terrain resistance coefficient. Finally, the processed exercise intensity data is normalized to a unit energy consumption value and saved as a required exercise intensity quantitative data matrix for subsequent analysis. The metabolic equivalent relative iterative change analysis simulates the dynamic effects of different exercise intensities on metabolic indicators by coupling the required exercise intensity quantitative data with the user's basic physical sign parameters. First, the user's basic physical sign parameters are substituted into the energy metabolism formula to calculate the resting metabolic rate (RMR). Subsequently, the resting metabolic rate is coupled with the required exercise intensity quantitative data, and the dynamic metabolic consumption change curve is simulated by the segmented integration method. The dynamic change value of metabolic consumption is iteratively optimized to generate equivalent iterative change data of exercise metabolism, and is stored in the form of a time series array, marking the metabolic consumption intensity at each time point. Muscle viscosity kinetic energy assessment is based on muscle kinematic characteristics and physiological energy conversion principles. First, based on the equivalent iterative change data of exercise metabolism, combined with the quantitative data of required exercise intensity, the mechanical kinetic energy of each activity unit is extracted; then, based on the user's weight, the total muscle load force of each activity is calculated, combined with the muscle viscosity coefficient (selected in the empirical value range of 0.1-0.3), the viscosity mechanics formula: muscle viscosity kinetic energy = viscosity coefficient × muscle load force × activity time, the viscosity kinetic energy of each activity unit is calculated in segments, and the muscle viscosity kinetic energy sequence data is generated to mark the change pattern of muscle energy consumption intensity over time. The simulation estimation of physical sign muscle soreness intensity comprehensively considers the impact of metabolic changes and viscosity kinetic energy on muscle soreness.The specific operation is as follows: First, the muscle lactate accumulation model is used to convert the equivalent iterative change data of exercise metabolism into lactate concentration change values, and the lactate-induced muscle stress index is calculated in combination with the user's individual lactate tolerance threshold. Secondly, the viscous kinetic energy data is substituted into the muscle elastic recovery model to calculate the muscle fatigue index caused by the viscosity effect. Finally, the lactate stress index and the muscle fatigue index are weighted and superimposed to generate muscle soreness intensity time series data. This sequence data is annotated according to different activity stages to form a physical sign muscle soreness intensity data set, which is used for subsequent physical decline assessment and tourism route optimization strategy formulation.

[0081] Step S233 includes the following steps:

[0082] Conducting exercise frequency fluctuation assessment on the required exercise intensity quantitative data to obtain exercise frequency fluctuation data;

[0083] According to the movement frequency fluctuation data, the respiratory basic frequency in the user's basic vital sign parameters is coupled with the respiratory frequency phase increment to obtain the respiratory frequency phase increment data;

[0084] Performing nonlinear time series iterative processing on the respiratory frequency phase increment data, thereby obtaining the respiratory frequency nonlinear iterative increment data;

[0085] Based on the nonlinear iterative incremental data of the respiratory frequency, the change relationship of the body fat rate in the user's basic physical sign parameters is analyzed to obtain the respiratory increment-body fat rate change relationship data;

[0086] Based on the nonlinear iterative increment data of respiratory frequency and the relationship data of respiratory increment-body fat percentage change, metabolic equivalent relative iterative change analysis was performed to obtain exercise metabolic equivalent iterative change data.

[0087] In an embodiment of the present invention, the fast Fourier transform algorithm is applied to evaluate the movement frequency fluctuation for the quantitative data of the required movement intensity. The specific operation is to regard the quantitative data of the required movement intensity as a time series signal, set the sliding window length to twice the movement period, and the window sliding step to one quarter of the movement period. Perform a fast Fourier transform on the data in each sliding window to obtain a frequency domain representation. Calculate the power spectrum density of the frequency domain signal, find the frequency corresponding to the maximum value of the power spectrum density, and define it as the main movement frequency in the window. Calculate the difference between the main movement frequencies of adjacent windows to obtain the movement frequency fluctuation data. The specific implementation of the fluctuation evaluation depends on the accuracy of the fast Fourier transform algorithm and the setting of the window parameters. For example, assuming that the quantitative data of the required movement intensity is reflected in the time series {2, 4, 6, 8, 6, 4, 2, 4, 6, 8, 6, 4}, the frequency fluctuation is reflected in the position of the maximum value of the power spectrum density, and the degree of deviation of the position of the maximum value of the power spectrum density reflects the amplitude of the frequency fluctuation. Then, according to the movement frequency fluctuation data, the respiratory basic frequency in the user's basic vital signs parameters is coupled with the respiratory frequency in a phased incremental manner. The specific operation is as follows: First, the motion frequency fluctuation data is normalized so that its value range is mapped to between 0 and 1. Then, the normalized motion frequency fluctuation data is multiplied by the respiratory frequency increment factor, and the value of the respiratory frequency increment factor is set to 10% of the respiratory basic frequency to obtain the respiratory frequency phase increment. Finally, the respiratory frequency phase increment is accumulated with the respiratory basic frequency to obtain the respiratory frequency phase increment data. For example, assuming that the respiratory basic frequency is 15 times per minute, the respiratory frequency increment factor is 1.5, and the motion frequency fluctuation data is normalized to 0.6, then the respiratory frequency phase increment is 0.9, and the accumulated respiratory frequency phase increment data is 15.9 times per minute. Subsequently, the respiratory frequency phase increment data is processed nonlinearly in time series iteration to obtain the respiratory frequency nonlinear iterative increment data. The specific operation is to use a long short-term memory network model. First, the respiratory frequency phase increment data is divided into multiple time series segments, and the length of each time series segment is set to 10 time units. Then, these time series segments are input into the long short-term memory network model for training. During the training process, the long short-term memory network model will learn the nonlinear time dependency in the respiratory frequency phase increment data. After the training is completed, the respiratory frequency phase increment data is input into the long short-term memory network model again, and the model will predict the respiratory frequency increment value of the next time unit based on the learned nonlinear time dependency. The predicted respiratory frequency increment value is added to the respiratory frequency phase increment data to obtain the respiratory frequency nonlinear iterative increment data. Then, based on the respiratory frequency nonlinear iterative increment data, the change relationship of the body fat rate in the user's basic physical sign parameters is analyzed.The specific operation is to use the Pearson correlation coefficient to calculate the correlation between the nonlinear iterative increment data of respiratory frequency and the body fat rate. First, collect a large amount of nonlinear iterative increment data and body fat rate data of users. Then, use the Pearson correlation coefficient formula to calculate the correlation coefficient between the two variables. If the correlation coefficient is positive, it indicates that the nonlinear iterative increment of respiratory frequency is positively correlated with the body fat rate; if the correlation coefficient is negative, it indicates that the nonlinear iterative increment of respiratory frequency is negatively correlated with the body fat rate; if the correlation coefficient is close to zero, it indicates that there is no obvious correlation between the nonlinear iterative increment of respiratory frequency and the body fat rate. According to the positive and negative nature of the correlation coefficient, the change relationship between the nonlinear iterative increment of respiratory frequency and the body fat rate is determined. If it is positively correlated, it indicates that the larger the nonlinear iterative increment of respiratory frequency, the higher the body fat rate; if it is negatively correlated, it indicates that the larger the nonlinear iterative increment of respiratory frequency, the lower the body fat rate. This change relationship is expressed as a data table or a function relationship as the respiratory increment-body fat rate change relationship data. Finally, a metabolic equivalent relative iterative change analysis is performed based on the nonlinear iterative increment data of respiratory frequency and the respiratory increment-body fat rate change relationship data. The specific operation is to input the nonlinear iterative increment data of the respiratory frequency into the respiratory increment-body fat rate change relationship data to obtain the corresponding body fat rate change value. Then, use the metabolic equivalent formula to calculate the corresponding metabolic equivalent value. Multiply the metabolic equivalent value by the nonlinear iterative increment data of the respiratory frequency to obtain the metabolic equivalent relative iterative change data. Assuming that the nonlinear iterative increment data of the respiratory frequency is 0.5 times per minute, the respiratory increment-body fat rate change relationship data shows that an increase of 0.5 times per minute in the respiratory frequency will lead to an increase of 0.1% in the body fat rate, and the body fat rate change value is 0.1%. Assuming the metabolic equivalent value is 1.5, the metabolic equivalent relative iterative change data is 0.75.

[0088] Step S234 includes the following steps:

[0089] Perform metabolic equivalent parameter analysis on the exercise metabolism equivalent iterative change data to obtain exercise metabolism equivalent parameters;

[0090] The muscle stretching lactate accumulation is deduced based on the exercise metabolism equivalent parameters and the required exercise intensity quantitative data to obtain the muscle stretching lactate accumulation data;

[0091] The muscle contraction and relaxation efficiency loss is evaluated on the muscle stretching lactic acid accumulation data to obtain the muscle contraction and relaxation efficiency loss data;

[0092] The muscle viscosity kinetic energy is evaluated based on the muscle stretching lactate accumulation data and the muscle contraction and relaxation efficiency loss data to obtain the muscle viscosity kinetic energy data.

[0093] In an embodiment of the present invention, the implementation process of performing metabolic equivalent parameter analysis on the equivalent iterative change data of exercise metabolism is: using the principal component analysis method to perform dimensionality reduction processing on the equivalent iterative change data of exercise metabolism. Construct a data covariance matrix and extract the main information through the eigenvalue decomposition algorithm. The specific operation is: first, the original data is centered, that is, the mean of each dimension is subtracted. Calculate the covariance matrix of the data matrix, and perform eigenvalue decomposition using the Jacobi eigenvalue iteration algorithm. Sort by eigenvalue size and select the principal component with a cumulative contribution rate of more than 85%. Perform an orthogonal transformation on the principal component to construct a metabolic equivalent parameter vector. Map high-dimensional data to low-dimensional space through linear transformation to extract the core features of metabolic equivalent parameters. The implementation process of performing muscle stretching lactate accumulation deduction based on the exercise metabolism equivalent parameters and the quantitative data of the required exercise intensity is: establish a lactate accumulation prediction system based on a kinetic model. Use differential equations to construct a muscle metabolism kinetic model to simulate the lactate production and diffusion process. The specific steps are: first, discretize the exercise metabolism equivalent parameters and construct a state transfer matrix. A set of dynamic differential equations describing the change of lactate concentration was established, including lactate production rate, diffusion rate and metabolic rate. The Runge-Kutta fourth-order numerical integration method was used to solve the differential equations and simulate the lactate accumulation process under different exercise intensities. The diffusion coefficient and metabolic coefficient were introduced to construct a multidimensional nonlinear dynamic system to accurately deduce the muscle stretching lactate accumulation data. The implementation process of evaluating the loss of muscle contraction and relaxation efficiency of muscle stretching lactate accumulation data is as follows: signal processing and spectrum analysis methods are used to evaluate muscle function loss. A muscle contraction efficiency evaluation model based on Fourier transform is constructed. The specific operation is as follows: first, a discrete Fourier transform is performed on the lactate accumulation data to convert the time domain signal into a frequency domain representation. The power spectral density of the signal is calculated to extract the key frequency components. The time-frequency characteristics of the signal are extracted using wavelet transform to analyze the energy decay characteristics during muscle contraction. A muscle efficiency loss index based on Shannon entropy is constructed to quantify the loss of muscle contraction and relaxation efficiency by calculating the signal complexity and energy decay rate. The implementation process of muscle viscosity kinetic energy evaluation based on muscle stretching lactate accumulation data and muscle contraction and relaxation efficiency loss data is: establish a multi-dimensional coupled muscle viscosity kinetic energy evaluation model. Use grey correlation analysis and entropy weight method to build a comprehensive evaluation system. The specific steps are: first, standardize the lactate accumulation data and efficiency loss data. Calculate the grey correlation between each data indicator and construct a correlation coefficient matrix. Use information entropy theory to calculate the weight coefficient of each indicator and establish a weighted comprehensive evaluation model. Determine the indicator weights by entropy weight method and construct a multi-dimensional quantitative indicator of muscle viscosity kinetic energy. Use the weighted comprehensive evaluation method to calculate the comprehensive score of muscle viscosity kinetic energy and obtain the final muscle viscosity kinetic energy data.

[0094] Step S3 includes the following steps:

[0095] Step S31: normalizing the physical fitness decline rate data to obtain normalized physical fitness decline rate data;

[0096] Step S32: extracting tourism-related nodes based on the intersection data of master / slave demand behaviors to obtain a master / slave demand tourism-related node data set;

[0097] Step S33: formulating a node segmentation rest strategy for the master / slave demand tourism-related node data set based on the normalized data of the physical fitness decline rate, and obtaining a tourism-related node segmentation rest strategy;

[0098] Step S34: adjusting the allocation of tourist routes according to the normalized data of the physical fitness decline rate and the segmented rest and recuperation strategy of tourist-related nodes, and obtaining tourist route allocation adjustment data.

[0099] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:

[0100] Step S31: normalizing the physical fitness decline rate data to obtain normalized physical fitness decline rate data;

[0101] In an embodiment of the present invention, a minimum-maximum normalization method is used to perform a linear transformation on the physical fitness decline rate data. The specific operation is: first calculate the minimum and maximum values ​​of the physical fitness decline rate data set. Construct a normalized linear transformation formula to map the original data to the [0, 1] interval. Perform a linear transformation on each physical fitness decline rate value in the data set one by one. Introduce an error control threshold to truncate abnormal data points. Use the standard deviation calculation method to remove extreme data points that exceed 3 standard deviations. Finally, generate a normalized physical fitness decline rate data set to ensure that the data is distributed within the [0, 1] standard interval.

[0102] Step S32: extracting tourism-related nodes based on the intersection data of master / slave demand behaviors to obtain a master / slave demand tourism-related node data set;

[0103] In an embodiment of the present invention, the implementation process of extracting tourism-related nodes based on the intersection data of master / slave demand behaviors is: using the clustering algorithm in graph theory to perform correlation analysis on tourism nodes. The specific steps are: first, construct an adjacency matrix based on the intersection data of master / slave demand behaviors. Use the spectral clustering algorithm to perform cluster analysis on the nodes and calculate the similarity between the nodes. By constructing a Laplace matrix, perform eigenvalue decomposition on the node similarity. Select the first k main eigenvectors in the eigenvector and map the nodes to a low-dimensional space. Use the K-means clustering algorithm to cluster the mapped nodes and extract a set of tourism nodes with high correlation. Calculate the correlation strength between the nodes and construct a node correlation weight matrix. Finally, a master / slave demand tourism-related node data set is generated.

[0104] Step S33: formulating a node segmentation rest strategy for the master / slave demand tourism-related node data set based on the normalized data of the physical fitness decline rate, and obtaining a tourism-related node segmentation rest strategy;

[0105] In an embodiment of the present invention, the implementation process of formulating a node segmentation rest strategy for a master / slave demand tourism-related node data set based on the normalized data of the physical decline rate is: constructing a multidimensional constraint optimization model and formulating a tourism node rest strategy. The specific operation is: first, establish a linear programming model with the physical decline rate as a constraint condition. Use the simplex method to solve the optimal combination of node rest. Construct an objective function to optimize the allocation of rest time between nodes. Introduce the normalized data of the physical decline rate as a constraint condition and establish an inequality constraint model. Solve the constrained optimization problem through an iterative algorithm to determine the optimal rest time for each node. Use the Lagrange multiplier method to process the constraints and convert the multidimensional constraint problem into an equivalent unconstrained optimization problem. Finally, a detailed tourism-related node segmentation rest strategy is generated.

[0106] Step S34: adjusting the allocation of tourist routes according to the normalized data of the physical fitness decline rate and the segmented rest and recuperation strategy of tourist-related nodes, and obtaining tourist route allocation adjustment data.

[0107] In an embodiment of the present invention, the implementation process of adjusting the allocation of tourist routes according to the normalized data of the physical decline rate and the segmented rest strategy of the tourist-related nodes is: constructing a dynamic programming optimization model for tourist routes. The specific steps are: first, establish a state transfer matrix to describe the conversion relationship between the tourist route nodes. Use a dynamic programming algorithm to solve the optimal route allocation plan. Construct a state transfer function, and use the physical decline rate as the key parameter of the state transfer. Calculate the optimal stay time and transfer path of each node by a recursive method. Introduce the segmented rest strategy of the tourist-related nodes as a constraint condition to establish a multi-stage decision model. Use the Bellman optimality principle to solve the optimal substructure from bottom to top. Calculate the cumulative optimal path for each node to generate a global optimal tourist route allocation plan. Finally, obtain the tourist route allocation adjustment data.

[0108] Step S33 includes the following steps:

[0109] Step S331: Obtain historical playing time statistics of each node according to the master / slave demand tourism associated node data set;

[0110] Step S332: performing a concentrated long-tail skewness analysis on the historical play time statistical data to obtain concentrated long-tail skewness data on the historical play time;

[0111] Step S333: Based on the normalized data of physical energy decay rate, the long-tail skewed data in the historical play time concentration is used to perform a cumulative calculation of the node physical energy consumption to obtain a node cumulative physical energy consumption sequence;

[0112] Step S334: performing piecewise linear fitting on the node cumulative physical energy consumption sequence to obtain piecewise fitting data of node physical energy consumption;

[0113] Step S335: Formulate a node segmented rest and recuperation strategy based on the segmented fitting data of node physical energy consumption to obtain a segmented rest and recuperation strategy for tourism-related nodes.

[0114] In the embodiment of the present invention, the master / slave demand tourism associated node data set is first preprocessed to remove abnormal and duplicate data. A node timestamp mapping matrix is ​​established to record the historical play start and end time of each node. A sliding window algorithm is used to perform statistical analysis on continuous time periods. A node play time calculation function is constructed to accurately calculate the residence time of each node. A time granularity conversion mechanism is introduced to standardize time data of different precisions. The node play time is stored in a hash table to establish a fast mapping relationship between node identification and play time. Finally, a node historical play time statistical data set is generated. The implementation process of performing a long-tail skewness analysis on the historical play time statistical data is: using probability distribution fitting and statistical moment estimation methods to analyze data distribution characteristics. The specific steps are: first calculate the basic statistics of the historical play time data set, including mean, variance, skewness and kurtosis. Construct the probability density function of the lognormal distribution and the Pareto distribution. Use the maximum likelihood estimation method to estimate the distribution parameters of the play time data. Calculate the skewness coefficient and the kurtosis coefficient to quantify the asymmetry and peak degree of the data distribution. Logarithmic transformation and Box-Cox transformation are introduced to perform nonlinear transformation on the original data to eliminate the influence of extreme values. The best fitting distribution model is determined by the chi-square goodness of fit test. Finally, the long-tail skewed data in the historical play time concentration is generated. The implementation process of calculating the cumulative physical energy consumption of nodes based on the normalized data of physical energy decay rate for the long-tail skewed data in the historical play time concentration is: constructing a multi-dimensional coupled physical energy consumption evaluation model. The specific operation is: first establish a nonlinear mapping function between physical energy decay rate and play time. Use the exponential decay model to simulate the physical energy consumption process and construct the physical energy loss rate equation. Introduce the normalized data of physical energy decay rate as the decay parameter to establish a dynamic physical energy consumption model. Solve the cumulative process of physical energy consumption through differential equations and construct a physical energy loss integral model. Use the numerical integration method to discretize the physical energy consumption integral equation. Construct a physical energy consumption cumulative function for each node to calculate the cumulative physical energy loss during the play process. Finally, generate a node cumulative physical energy consumption sequence. The implementation process of piecewise linear fitting of the node cumulative physical energy consumption sequence is: use the piecewise linear regression algorithm to process the physical energy consumption data. The specific steps are as follows: First, use the piecewise function approximation algorithm to divide the node cumulative energy consumption sequence into multiple linear subintervals. Construct a linear fitting model based on the least squares method and calculate the linear fitting parameters of each subinterval. Introduce a split-merge strategy to dynamically adjust the linear segmentation points. Use the Bayesian Information Criterion (BIC) to evaluate the complexity and goodness of fit of the segmentation model. Use an iterative optimization algorithm to find the optimal segmentation points and linear fitting parameters. Calculate the slope and intercept of each linear subinterval and construct a piecewise linear fitting model. Finally, generate the node energy consumption segmented fitting data. The implementation process of formulating node segmented rest strategy based on node energy consumption segmented fitting data is: construct a multi-objective optimization rest strategy model.The specific operation is as follows: First, a linear programming model with segmented fitting data of physical energy consumption as a constraint condition is established. The goal programming method is used to solve the multi-objective optimization problem to balance physical energy consumption and rest time. The objective function is constructed to optimize the node rest time and physical recovery efficiency at the same time. The Lagrangian duality theory is introduced to transform the multi-objective optimization problem into an equivalent single-objective optimization problem. The convex optimization algorithm is used to solve the optimal solution of the rest strategy. The optimal rest time and rest intensity of each node are calculated by an iterative method. Finally, a segmented rest strategy for tourism-related nodes is generated.

[0115] Step S34 includes the following steps:

[0116] Step S341: extracting the required rest time intervals between different nodes according to the segmented rest strategy of the tourism-related nodes to obtain the required rest time intervals between different nodes;

[0117] Step S342: performing linear programming of node routes according to the normalized data of the physical fitness decline rate and the required rest time interval to obtain linear programming data of node routes;

[0118] Step S343: adjusting the travel route allocation based on the node route linear programming data to obtain travel route allocation adjustment data.

[0119] In an embodiment of the present invention, the implementation process of extracting the required rest time intervals between different nodes according to the segmented rest strategy of the tourism-related nodes is: constructing a node rest time extraction algorithm based on graph theory. The specific operation is: first, the tourism-related nodes are constructed as a weighted undirected graph, the nodes represent tourist attractions, and the edge weights represent the transfer costs between nodes. The shortest path algorithm is used, and the Floyd-Warshall algorithm is specifically used to calculate the optimal path between any two nodes. A rest time extraction function is constructed to dynamically calculate the required rest time according to the transfer distance and physical energy consumption degree between nodes. A state transfer matrix is ​​introduced to record the rest time constraints of each node. The node rest time is sequentially constrained by a topological sorting algorithm to ensure the reasonable allocation of rest time. The node rest time allocation is optimized using a greedy algorithm to minimize the time cost of the overall tourist route. Finally, a data set of required rest time intervals between different nodes is generated. The implementation process of performing linear programming processing of node routes according to the normalized data of physical energy decline rate and the required rest time interval is: constructing a linear programming optimization model with multiple constraints. The specific steps are: first, a linear programming model with physical energy decline rate as the objective function is established. The simplex method is used to solve the optimal configuration of node routes. The constraint matrix is ​​constructed, and the normalized data of physical energy decay rate and the rest time interval are introduced as constraint parameters. The linear constraint equation group of node route transfer is established, including time constraint, physical energy constraint and path connectivity constraint. The dual simplex method is used to deal with linear programming problems and solve the optimal route configuration. The Lagrange duality theory is introduced to transform multiple constraints into equivalent unconstrained optimization problems. The optimal node route configuration is solved by an iterative algorithm, minimizing the overall physical energy consumption and route transfer cost. Finally, the node route linear programming data is generated. The implementation process of adjusting the tourism route allocation based on the node route linear programming data is: constructing a tourism route optimization algorithm for dynamic path planning. The specific operation is: first, a state transition network is established, and the node route planning data is converted into a directed acyclic graph. The dynamic programming algorithm is used, specifically the Bellman-Ford algorithm is used to solve the optimal tourism route. The route allocation objective function is constructed, taking into account physical energy consumption, rest time and tourism experience. The state transition probability matrix is ​​introduced to simulate the transfer path selection between nodes. The optimal path of each node is calculated from bottom to top through the reverse recursive method. The optimal substructure theorem is used to decompose the complex route allocation problem into sub-problems for solution. Route allocation evaluation indicators are constructed, including physical energy consumption rate, rest efficiency and tourism satisfaction. The optimization effect of the tourist route is comprehensively evaluated through the multi-dimensional weighted summation method. Finally, the tourist route allocation adjustment data is generated to achieve dynamic optimization of the tourist route.

[0120] The present invention also provides a tourist route recommendation system based on big data, which is used to execute the tourist route recommendation method based on big data as described above. The tourist route recommendation system based on big data includes:

[0121] The master / slave demand hierarchy parsing module is used to obtain the tourism demand data set and the user's individual basic vital sign data input by the user; perform master / slave demand hierarchy parsing on the tourism demand data set to obtain tourism master / slave demand hierarchy data;

[0122] The physical fitness decline rate matching module is used to perform the intersection analysis of the main / slave demand behavior on the tourism main / slave demand hierarchy data to obtain the main / slave demand behavior intersection data; simulate and estimate the physical fitness muscle soreness intensity of the user's individual basic physical sign data based on the main / slave demand behavior intersection data to obtain the physical fitness muscle soreness intensity data; perform physical fitness decline rate matching based on the physical fitness muscle soreness intensity data to obtain the physical fitness decline rate data;

[0123] The travel route allocation adjustment module is used to formulate a node segmentation rest strategy for the intersection data of the master / slave demand behavior based on the physical fitness decline rate data, and obtain the travel-related node segmentation rest strategy; adjust the travel route allocation according to the travel-related node segmentation rest strategy, obtain the travel route allocation adjustment data, and send the travel route allocation adjustment data to the terminal.

[0124] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A tourist route recommendation method based on big data, characterized in that: The following steps are involved: Step S1: obtaining the tourism demand data set and individual basic vital sign data of the user input; Perform master / slave demand hierarchy analysis on the tourism demand data set to obtain tourism master / slave demand hierarchy data; Step S2 includes: Step S21: performing a master / slave demand behavior intersection analysis on the tourism master / slave demand hierarchy data to obtain master / slave demand behavior intersection data; Step S22: Performing exercise intensity evaluation based on the master / slave demand behavior intersection data to obtain travel master / slave demand exercise intensity data; Wherein, step S23 comprises: Step S231: extracting basic physical sign parameters of the user from the basic physical sign data of the user to obtain basic physical sign parameters of the user, wherein the basic physical sign parameters of the user include height and weight, basic heart rate, body fat rate and basic breathing frequency; Step S232: performing exercise intensity quantification processing on the required exercise intensity data of the tourist master / slave to obtain required exercise intensity quantitative data; Wherein, step S233 includes: Conducting exercise frequency fluctuation assessment on the required exercise intensity quantitative data to obtain exercise frequency fluctuation data; According to the movement frequency fluctuation data, the respiratory basic frequency in the user's basic vital sign parameters is coupled with the respiratory frequency phase increment to obtain the respiratory frequency phase increment data; Performing nonlinear time series iterative processing on the respiratory frequency phase increment data, thereby obtaining the respiratory frequency nonlinear iterative increment data; Based on the nonlinear iterative incremental data of the respiratory frequency, the change relationship of the body fat rate in the user's basic physical sign parameters is analyzed to obtain the respiratory increment-body fat rate change relationship data; According to the nonlinear iterative increment data of respiratory frequency and the relationship data of respiratory increment-body fat percentage change, the metabolic equivalent relative iterative change analysis is performed to obtain the exercise metabolic equivalent iterative change data; Wherein, step S234 includes: Perform metabolic equivalent parameter analysis on the exercise metabolism equivalent iterative change data to obtain exercise metabolism equivalent parameters; The muscle stretching lactate accumulation is deduced based on the exercise metabolism equivalent parameters and the required exercise intensity quantitative data to obtain the muscle stretching lactate accumulation data; The muscle contraction and relaxation efficiency loss is evaluated on the muscle stretching lactic acid accumulation data to obtain the muscle contraction and relaxation efficiency loss data; The muscle viscosity kinetic energy is evaluated based on the muscle stretching lactic acid accumulation data and the muscle contraction and relaxation efficiency loss data to obtain the muscle viscosity kinetic energy data; Step S235: performing a simulation estimation of the intensity of physical sign muscle soreness according to the exercise metabolism equivalent iterative change data and the muscle viscosity kinetic energy data to obtain physical sign muscle soreness intensity data; Step S24: performing physical fitness decline rate matching based on the physical sign muscle soreness intensity data to obtain physical fitness decline rate data; Step S3: Formulate a node segmentation rest strategy for the intersection data of the master / slave demand behavior based on the physical fitness decline rate data to obtain a tourism-related node segmentation rest strategy; adjust the tourism route allocation according to the tourism-related node segmentation rest strategy to obtain tourism route allocation adjustment data, and send the tourism route allocation adjustment data to the terminal.

2. The method for recommending tourist routes based on big data according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: obtaining a travel demand data set input by a user and basic individual vital sign data of the user; Step S12: performing data cleaning on the tourism demand data set to obtain a tourism demand cleaned data set; Step S13: parsing the context statement logical structure of the tourism demand cleansing data set to obtain the demand context statement logical structure; Step S14: performing master / slave demand hierarchy analysis on the tourism demand cleansing data set according to the demand context statement logical structure to obtain tourism master / slave demand hierarchy data.

3. The method for recommending tourist routes based on big data according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: normalizing the physical fitness decline rate data to obtain normalized physical fitness decline rate data; Step S32: extracting tourism-related nodes based on the intersection data of master / slave demand behaviors to obtain a master / slave demand tourism-related node data set; Step S33: formulating a node segmentation rest strategy for the master / slave demand tourism-related node data set based on the normalized data of the physical fitness decline rate, and obtaining a tourism-related node segmentation rest strategy; Step S34: adjusting the allocation of tourist routes according to the normalized data of the physical fitness decline rate and the segmented rest and recuperation strategy of tourist-related nodes, and obtaining tourist route allocation adjustment data.

4. The method for recommending tourist routes based on big data according to claim 3, characterized in that: Step S33 includes the following steps: Step S331: Obtain historical playing time statistics of each node according to the master / slave demand tourism associated node data set; Step S332: performing a concentrated long-tail skewness analysis on the historical play time statistical data to obtain concentrated long-tail skewness data on the historical play time; Step S333: Based on the normalized data of physical energy decay rate, the long-tail skewed data in the historical play time concentration is used to perform a cumulative calculation of the node physical energy consumption to obtain a node cumulative physical energy consumption sequence; Step S334: performing piecewise linear fitting on the node cumulative physical energy consumption sequence to obtain piecewise fitting data of node physical energy consumption; Step S335: Formulate a node segmented rest and recuperation strategy based on the segmented fitting data of node physical energy consumption to obtain a segmented rest and recuperation strategy for tourism-related nodes.

5. A travel route recommendation system based on big data, characterized in that: For executing the tourist route recommendation method based on big data as claimed in claim 1, the tourist route recommendation system based on big data comprises: The master / slave demand hierarchy parsing module is used to obtain the tourism demand data set and the user's individual basic vital sign data input by the user; perform master / slave demand hierarchy parsing on the tourism demand data set to obtain tourism master / slave demand hierarchy data; The physical fitness decline rate matching module is used to perform the intersection analysis of the main / slave demand behavior on the tourism main / slave demand hierarchy data to obtain the main / slave demand behavior intersection data; simulate and estimate the physical fitness muscle soreness intensity of the user's individual basic physical sign data based on the main / slave demand behavior intersection data to obtain the physical fitness muscle soreness intensity data; perform physical fitness decline rate matching based on the physical fitness muscle soreness intensity data to obtain the physical fitness decline rate data; The travel route allocation adjustment module is used to formulate a node segmentation rest strategy for the intersection data of the master / slave demand behavior based on the physical fitness decline rate data, and obtain the travel-related node segmentation rest strategy; adjust the travel route allocation according to the travel-related node segmentation rest strategy, obtain the travel route allocation adjustment data, and send the travel route allocation adjustment data to the terminal.

Images