Method and system for trip mode clustering identification based on multi-source signaling

Abstract

The application provides a travel mode clustering identification method and system based on multi-source signaling. The method comprises: obtaining travel basic data of a preset travel OD of a target user, extracting travel physical parameters and traffic behavior index parameters, obtaining weather types and a preset initial travel physical threshold set, processing to obtain a corrected travel physical threshold set, performing matching processing according to the travel physical parameters and the corrected travel physical threshold set, obtaining a travel mode result, including a clear OD and a fuzzy OD, if the travel mode result is the fuzzy OD, obtaining the membership degrees of each preset clustering category and determining a target clustering category, extracting a maximum membership degree and a second maximum membership degree, if the difference between the two is less than a preset difference threshold, performing path verification and correction processing on the preset travel OD, and obtaining a final travel mode result, thereby realizing travel mode clustering identification based on multi-source signaling.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation technology, and more specifically, to a method and system for travel mode clustering and identification based on multi-source signaling. Background Technology

[0002] Current travel mode identification based on signaling data often suffers from low accuracy due to its reliance on a single-dimensional parameter, fixed thresholds, and failure to consider differences in origin-destination (OD) types. For example, using a uniform initial threshold for travel physical parameters makes it difficult to adapt to travel characteristics in different scenarios, easily leading to clear OD identification bias. For fuzzy ODs, relying solely on cluster membership degrees can easily result in misjudgments if the membership degrees of the maximum and second-largest values ​​are close, and the lack of a path verification and correction mechanism further reduces the reliability of identification.

[0003] Meanwhile, existing methods do not effectively integrate travel physical parameters and traffic behavior indicators, nor do they dynamically adjust thresholds based on weather types, thus failing to meet the needs for accurate identification in various scenarios. Therefore, there is an urgent need for a multi-source signaling travel mode clustering and identification scheme that can dynamically adjust thresholds, combine clustering and path verification, and adapt to different OD types, in order to improve identification accuracy and applicability. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for travel mode clustering and identification based on multi-source signaling. This method involves acquiring the target user's preset travel origin-destination (OD) data, extracting travel physical parameters and traffic behavior indicators, obtaining weather type and a preset initial travel physical threshold set, processing to obtain a corrected travel physical threshold set, and matching the travel physical parameters with the corrected travel physical threshold set to obtain travel mode results, including explicit ODs and fuzzy ODs. If the travel mode result is a fuzzy OD, fuzzy clustering is performed based on the travel physical parameters and traffic behavior indicators to obtain the membership degree of each preset cluster category, determine the target cluster category, and extract the maximum and second-largest membership degrees. If the difference between the maximum and second-largest membership degrees is less than a preset difference threshold, path verification and correction processing is performed on the preset travel OD to obtain the final travel mode result, thus achieving travel mode clustering and identification based on multi-source signaling.

[0005] This application also provides a travel mode clustering and identification method based on multi-source signaling, including the following steps: Obtain the target user's preset travel origin-destination (OD) basic travel data, and extract travel physical parameters and traffic behavior index parameters; Obtain the weather type and the preset initial travel physical threshold set, and process them to obtain the corrected travel physical threshold set; The travel physical parameters are matched with the corrected travel physical threshold set to obtain travel mode results, including explicit OD and fuzzy OD. If the travel mode result is a fuzzy OD, fuzzy clustering is performed based on the travel physical parameters and traffic behavior index parameters to obtain the membership degree of each preset cluster category and determine the target cluster category. Extract the maximum and second-largest membership degrees. If the difference between the maximum and second-largest membership degrees is less than a preset difference threshold, perform path verification and correction processing on the preset travel OD to obtain the final travel mode result.

[0006] Optionally, in the travel mode clustering and identification method based on multi-source signaling described in this application, the step of obtaining the preset travel origin-destination (OD) data of the target user and extracting travel physical parameters and traffic behavior index parameters includes: Acquire the target user's preset travel origin-destination (OD) basic travel data, including mobile signaling data, road network data, and POI data; Extract travel physical parameters and traffic behavior index parameters based on mobile phone signaling data, road network data, and POI data; The travel physical parameters include travel distance, travel duration, travel time period, average speed, and OD area type; The traffic behavior index parameters include trajectory point density, base station switching frequency, speed fluctuation variance, lane trajectory ratio, and parking point matching degree.

[0007] Optionally, in the travel mode clustering and identification method based on multi-source signaling described in this application, the step of obtaining the weather type and a preset initial travel physical threshold set, and processing to obtain the corrected travel physical threshold set, includes: Obtain the weather type and the preset initial travel physical threshold set; The preset initial travel physical threshold set includes a preset initial speed threshold set and a preset initial distance threshold set; Based on the preset initial distance threshold set, each preset threshold is combined with the weather type, travel time, and OD area type, and then processed through a preset travel threshold correction model to obtain the corresponding corrected travel physical threshold set.

[0008] Optionally, in the travel mode clustering and identification method based on multi-source signaling described in this application, the step of matching the travel physical parameters with the corrected travel physical threshold set to obtain the travel mode result includes explicit OD and fuzzy OD, including: The matching results are obtained by matching the travel distance and average speed with the corresponding thresholds of the corrected travel physical threshold set. The travel mode results are obtained based on the matching results, including explicit OD and fuzzy OD.

[0009] Optionally, in the travel mode clustering and identification method based on multi-source signaling described in this application, if the travel mode result is a fuzzy OD, fuzzy clustering processing is performed based on the travel physical parameters and traffic behavior index parameters to obtain the membership degree of each preset cluster category and determine the target cluster category, including: If the travel mode result is a fuzzy OD, obtain the preset clustering category, including public transportation, taxi, short- and medium-distance self-driving, and short- and medium-distance cycling; Based on the travel physical parameters and traffic behavior index parameters, a preset fuzzy K-means algorithm is used to process them to obtain the membership degree corresponding to each preset cluster category; The data is sorted, and the target cluster category is determined based on the sorting results.

[0010] Optionally, in the travel mode clustering and identification method based on multi-source signaling described in this application, the step of extracting the maximum and second-largest membership values, and if the difference between the maximum and second-largest membership values ​​is less than a preset difference threshold, performing path verification and correction processing on the preset travel OD to obtain the final travel mode result, includes: Extract the membership degree of the maximum and second-largest values, perform statistical processing, and obtain the difference. The difference is compared with a preset difference threshold. If the difference is less than a preset difference threshold, the preset travel OD is processed by a preset ST-DBSCAN algorithm to obtain the time matching degree and ratio matching degree corresponding to each preset cluster category; The path matching comprehensive score corresponding to each preset cluster category is obtained by weighting the time matching degree and the ratio matching degree. The path matching scores are sorted to obtain the maximum path matching score. The preset clustering category corresponding to the maximum path matching comprehensive score is marked as the final travel mode result.

[0011] Secondly, this application provides a travel mode clustering and identification system based on multi-source signaling. The system includes a memory and a processor. The memory includes a program for a travel mode clustering and identification method based on multi-source signaling. When the program for the travel mode clustering and identification method based on multi-source signaling is executed by the processor, it performs the following steps: Obtain the target user's preset travel origin-destination (OD) basic travel data, and extract travel physical parameters and traffic behavior index parameters; Obtain the weather type and the preset initial travel physical threshold set, and process them to obtain the corrected travel physical threshold set; The travel physical parameters are matched with the corrected travel physical threshold set to obtain travel mode results, including explicit OD and fuzzy OD. If the travel mode result is a fuzzy OD, fuzzy clustering is performed based on the travel physical parameters and traffic behavior index parameters to obtain the membership degree of each preset cluster category and determine the target cluster category. Extract the maximum and second-largest membership degrees. If the difference between the maximum and second-largest membership degrees is less than a preset difference threshold, perform path verification and correction processing on the preset travel OD to obtain the final travel mode result.

[0012] Optionally, in the travel mode clustering and identification system based on multi-source signaling described in this application, the step of obtaining the preset travel origin-destination (OD) data of the target user and extracting travel physical parameters and traffic behavior indicator parameters includes: Acquire the target user's preset travel origin-destination (OD) basic travel data, including mobile signaling data, road network data, and POI data; Extract travel physical parameters and traffic behavior index parameters based on mobile phone signaling data, road network data, and POI data; The travel physical parameters include travel distance, travel duration, travel time period, average speed, and OD area type; The traffic behavior index parameters include trajectory point density, base station switching frequency, speed fluctuation variance, lane trajectory ratio, and parking point matching degree.

[0013] Optionally, in the travel mode clustering and identification system based on multi-source signaling described in this application, the step of obtaining the weather type and the preset initial travel physical threshold set, and processing to obtain the corrected travel physical threshold set, includes: Obtain the weather type and the preset initial travel physical threshold set; The preset initial travel physical threshold set includes a preset initial speed threshold set and a preset initial distance threshold set; Based on the preset initial distance threshold set, each preset threshold is combined with the weather type, travel time, and OD area type, and then processed through a preset travel threshold correction model to obtain the corresponding corrected travel physical threshold set.

[0014] Optionally, in the travel mode clustering and identification system based on multi-source signaling described in this application, the step of matching the travel physical parameters with the corrected travel physical threshold set to obtain travel mode results, including explicit OD and fuzzy OD, includes: The matching results are obtained by matching the travel distance and average speed with the corresponding thresholds of the corrected travel physical threshold set. The travel mode results are obtained based on the matching results, including explicit OD and fuzzy OD.

[0015] As can be seen from the above, the travel mode clustering and identification method and system based on multi-source signaling provided in this application obtains the basic travel data of the target user's preset travel OD, extracts travel physical parameters and traffic behavior index parameters, obtains the weather type and preset initial travel physical threshold set, processes to obtain the corrected travel physical threshold set, and performs matching processing based on the travel physical parameters and the corrected travel physical threshold set to obtain the travel mode result, including explicit OD and fuzzy OD. If the travel mode result is fuzzy OD, fuzzy clustering processing is performed based on the travel physical parameters and traffic behavior index parameters to obtain the membership degree of each preset cluster category, and the target cluster category is determined. The maximum membership degree and the second largest membership degree are extracted. If the difference between the maximum membership degree and the second largest membership degree is less than the preset difference threshold, the preset travel OD is subjected to path verification and correction processing to obtain the final travel mode result, thereby realizing travel mode clustering and identification based on multi-source signaling.

[0016] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing embodiments of this application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating the travel mode clustering and identification method based on multi-source signaling provided in this application embodiment; Figure 2 A flowchart illustrating the process of obtaining the corrected travel physical threshold set using a travel mode clustering and identification method based on multi-source signaling, as provided in this application embodiment. Figure 3 A flowchart illustrating the process of obtaining travel mode results using a travel mode clustering and identification method based on multi-source signaling, as provided in this embodiment of the application. Figure 4 This is a flowchart illustrating the determination of the target cluster category in the travel mode clustering identification method based on multi-source signaling provided in the embodiments of this application. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0020] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] Please refer to Figure 1 , Figure 1 This is a flowchart of a travel mode clustering and identification method based on multi-source signaling according to some embodiments of this application. This travel mode clustering and identification method based on multi-source signaling is used in terminal devices, such as computers and mobile terminals. The travel mode clustering and identification method based on multi-source signaling includes the following steps: S11. Obtain the target user's preset travel OD basic data, and extract travel physical parameters and traffic behavior index parameters; S12. Obtain the weather type and the preset initial travel physical threshold set, and process them to obtain the corrected travel physical threshold set. S13. Match the travel physical parameters with the corrected travel physical threshold set to obtain travel mode results, including explicit OD and fuzzy OD. S14. If the travel mode result is a fuzzy OD, perform fuzzy clustering processing based on the travel physical parameters and traffic behavior index parameters to obtain the membership degree of each preset cluster category and determine the target cluster category. S15. Extract the maximum and second-largest membership degrees. If the difference between the maximum and second-largest membership degrees is less than a preset difference threshold, perform path verification and correction processing on the preset travel OD to obtain the final travel mode result.

[0022] It should be noted that existing travel mode identification based on multi-source signaling suffers from several problems, including the lack of dynamic threshold adjustment, the susceptibility to misjudgment due to relying solely on clustering membership for ambiguous origin-destination (OD) identification, and the absence of a path verification and correction mechanism. To address these issues, the following approach is adopted: First, the basic travel data of the target user's preset OD is acquired, including mobile signaling data, road network data, and POI data. Travel physical parameters and traffic behavior indicators are extracted, along with weather type and a preset initial travel physical threshold set. These are then combined with the travel time period and OD area type to obtain a corrected travel physical threshold set. Finally, the travel physical parameters are matched against the corrected travel physical threshold set. The system processes and obtains travel mode results, including explicit origin-destination (OD) and fuzzy OD. If the travel mode result is a fuzzy OD, it obtains preset clustering categories, including public transportation, taxi, short- and medium-distance self-driving, and short- and medium-distance cycling. Fuzzy clustering is performed based on travel physical parameters and traffic behavior index parameters to obtain the membership degree of each preset clustering category. The target clustering category is determined, and the maximum and second-largest membership degrees are extracted. If the difference between the maximum and second-largest membership degrees is less than a preset difference threshold, path verification and correction processing is performed on the preset travel OD to obtain the final travel mode result, thereby achieving travel mode clustering and recognition based on multi-source signaling.

[0023] According to an embodiment of the present invention, the step of obtaining the basic travel data of the target user's preset travel origin-destination (OD) and extracting travel physical parameters and traffic behavior index parameters includes: Acquire the target user's preset travel origin-destination (OD) basic travel data, including mobile signaling data, road network data, and POI data; Extract travel physical parameters and traffic behavior index parameters based on mobile phone signaling data, road network data, and POI data; The travel physical parameters include travel distance, travel duration, travel time period, average speed, and OD area type; The traffic behavior index parameters include trajectory point density, base station switching frequency, speed fluctuation variance, lane trajectory ratio, and parking point matching degree.

[0024] It should be noted that, in order to accurately identify travel modes, it is necessary to first obtain multi-dimensional basic travel data of the target user's preset origin-destination (OD) points, including mobile phone signaling data, road network data, and point-of-interest (POI) data. Based on the above multi-source data, the system can accurately extract travel physical parameters and traffic behavior indicators through data fusion and feature analysis technologies. Among them, mobile phone signaling data provides core clues to the user's movement trajectory, road network data clarifies the road attributes of the travel path, and POI data helps to determine the characteristics of the travel scenario. Travel physical parameters focus on the basic attributes of travel, including travel distance (actual distance traveled between the OD points) and travel duration (from the origin to the destination). The parameters include total arrival time, travel time (the specific time period during which the trip occurs, such as the morning peak from 7:00 to 9:00), average speed (the ratio of total distance to total time), and OD area type (such as commercial area, residential area, industrial area, etc., determined in conjunction with POI data); traffic behavior indicators reflect the behavioral characteristics of users during their movement, including trajectory point density (the number of signaling trajectory points per unit time), base station switching frequency (the number of times the mobile phone switches to a base station while moving), speed fluctuation variance (a stability indicator of driving speed), lane trajectory ratio (the proportion of trajectories that match specific lanes), and parking point matching degree (the degree of fit between the stopping trajectory and parking areas such as POIs).

[0025] Please refer to Figure 2 , Figure 2 This is a flowchart illustrating the process of obtaining a corrected travel physical threshold set using a travel mode clustering and identification method based on multi-source signaling in some embodiments of this application. According to embodiments of the present invention, obtaining the weather type and a preset initial travel physical threshold set, and then processing to obtain the corrected travel physical threshold set, includes: S21. Obtain the weather type and the preset initial travel physical threshold set; S22, The preset initial travel physical threshold set includes a preset initial speed threshold set and a preset initial distance threshold set; S23. Based on the preset thresholds of the preset initial distance threshold set, the weather type, travel time period and OD area type are respectively processed by the preset travel threshold correction model to obtain the corresponding corrected travel physical threshold set.

[0026] It should be noted that, to improve the accuracy of travel mode recognition, two core basic information types need to be obtained first: one is the real-time and predicted weather type, such as sunny, rainy, snowy, and foggy; the other is a preset initial travel physical threshold set. This preset initial travel physical threshold set mainly includes a preset initial speed threshold set and a preset initial distance threshold set. The former covers the speed range benchmarks for different travel modes, while the latter clarifies the distance division standards for various travel scenarios. Since weather, time of day, and region significantly affect travel characteristics—for example, rain reduces travel speed, urban roads are prone to congestion during morning rush hour, and travel patterns differ greatly between commercial and residential areas—fixed thresholds are difficult to adapt to complex scenarios. Therefore, a preset travel threshold correction model is needed to fuse the various thresholds of the preset initial distance threshold set with weather type, travel time, and OD region type, respectively. The model dynamically adjusts each initial threshold by learning the correlation patterns between multiple factors and travel thresholds of various travel modes in historical data, ultimately outputting the corresponding corrected travel physical threshold set, including the corrected speed threshold set and the corrected distance threshold set.

[0027] Please refer to Figure 3 , Figure 3 This is a flowchart illustrating the process of obtaining travel mode results using a multi-source signaling-based travel mode clustering and identification method as described in some embodiments of this application. According to embodiments of the present invention, the step of matching the travel physical parameters with a modified travel physical threshold set to obtain travel mode results includes explicit OD and fuzzy OD, comprising: S31. Match the travel distance and average speed with the corresponding thresholds of the corrected travel physical threshold set to obtain the matching results; S32. Obtain travel mode results based on the matching results, including explicit OD and fuzzy OD.

[0028] It should be noted that after extracting and correcting the travel physical parameters, the core step in determining the mode of travel is to accurately match the two key indicators—travel distance and average speed—with the corresponding standards of the corrected travel physical threshold set. The matching process employs a dual-indicator collaborative verification rule: the actual travel distance of the target user is compared one by one with the corrected distance threshold set, while the actual average speed is cross-matched with the corrected speed threshold set. For example, if the travel distance is within the walking distance threshold range and the average speed meets the walking speed standard, a preliminary matching result for walking is formed. Based on the matching results, two types of travel mode results can be directly distinguished: explicit OD refers to a high degree of consistency between the two indicator matching results, both pointing to a single mode of travel, such as distance and speed matching the bicycle threshold range simultaneously, without ambiguity; fuzzy OD refers to conflicting two indicator matching results or both falling within the intersection of multiple mode thresholds, such as distance matching the electric vehicle range, but speed falling between bicycle and electric vehicle, requiring further analysis.

[0029] Please refer to Figure 4 , Figure 4 This is a flowchart illustrating the determination of a target cluster category in a travel mode clustering identification method based on multi-source signaling, as described in some embodiments of this application. According to an embodiment of the present invention, if the travel mode result is a fuzzy OD (Original Distance), fuzzy clustering processing is performed based on the travel physical parameters and traffic behavior index parameters to obtain the membership degree of each preset cluster category, and the target cluster category is determined, including: S41. If the travel mode result is a fuzzy OD, obtain the preset clustering category, including public transportation, taxi, short- and medium-distance self-driving, and short- and medium-distance cycling. S42. Based on the travel physical parameters and traffic behavior index parameters, the membership degree corresponding to each preset cluster category is obtained by processing them using a preset fuzzy K-means algorithm. S43. Perform sorting processing and determine the target cluster category based on the sorting results.

[0030] It should be noted that when the travel mode result is determined to be fuzzy OD, cluster analysis is required for further precise identification. First, predefined cluster categories are established, categorized into four types based on the commonalities and modes of short- and medium-distance travel: public transport, taxi, short- and medium-distance driving, and short- and medium-distance cycling, covering possible travel options for mainstream fuzzy scenarios. Next, to improve clustering accuracy, travel physical parameters and traffic behavior indicators are used as inputs and fused using a predefined fuzzy K-means algorithm. This algorithm overcomes the limitations of traditional clustering's binary choice approach, reflecting the degree of association between samples and multiple categories, thus fitting the characteristics of fuzzy OD. During the calculation, the algorithm combines the weights of each parameter (e.g., speed fluctuation variance has a higher distinguishing effect between driving and public transport) and iteratively optimizes the objective function, outputting the membership degree corresponding to each predefined cluster, i.e., the probability that the target travel sample belongs to that category; the closer the value is to 1, the stronger the association. Then, all membership degrees are sorted in descending order, and the cluster category corresponding to the first value in the sorted list is the target cluster category after integrating the features of all parameters.

[0031] According to an embodiment of the present invention, the step of extracting the maximum and second-largest membership degrees, and if the difference between the maximum and second-largest membership degrees is less than a preset difference threshold, performing path verification and correction processing on the preset travel OD to obtain the final travel mode result, includes: Extract the membership degree of the maximum and second-largest values, perform statistical processing, and obtain the difference. The difference is compared with a preset difference threshold. If the difference is less than a preset difference threshold, the preset travel OD is processed by a preset ST-DBSCAN algorithm to obtain the time matching degree and ratio matching degree corresponding to each preset cluster category; The path matching comprehensive score corresponding to each preset cluster category is obtained by weighting the time matching degree and the ratio matching degree. The path matching scores are sorted to obtain the maximum path matching score. The preset clustering category corresponding to the maximum path matching comprehensive score is marked as the final travel mode result.

[0032] It should be noted that, to further improve the reliability of fuzzy OD travel mode identification, the membership degrees obtained from clustering need to be further processed: First, the maximum and second-largest membership degrees are extracted, and the statistical difference between the two is obtained through difference calculation. This difference directly reflects the distinguishing degree of the target travel sample to the top 2 cluster categories. This difference is compared with a preset difference threshold. If the difference is less than the threshold, it means that the probability of the sample belonging to the two clusters is close, and the membership degree alone cannot accurately determine the travel mode. A path verification mechanism needs to be introduced. At this time, the preset travel OD is processed by the preset ST-DBSCAN algorithm. This algorithm integrates time and space dimensions and can accurately mine the trajectory time series features. After the calculation, the time matching degree (the degree of fit between the trajectory time and the typical time consumption of the mode) and the ratio matching degree (the degree of fit between the trajectory parameter ratio and the category features) of each cluster category are output. The weights are assigned according to the importance of the two matching degrees and the weighted calculation is performed to obtain the path matching comprehensive score. After sorting the comprehensive scores in descending order, the cluster category corresponding to the maximum value is taken as the final travel mode result after path verification to ensure the accuracy of the judgment.

[0033] Secondly, the present invention also discloses a travel mode clustering and identification system based on multi-source signaling, including a memory and a processor. The memory includes a travel mode clustering and identification method program based on multi-source signaling. When the travel mode clustering and identification method program based on multi-source signaling is executed by the processor, it performs the following steps: Obtain the target user's preset travel origin-destination (OD) basic travel data, and extract travel physical parameters and traffic behavior index parameters; Obtain the weather type and the preset initial travel physical threshold set, and process them to obtain the corrected travel physical threshold set; The travel physical parameters are matched with the corrected travel physical threshold set to obtain travel mode results, including explicit OD and fuzzy OD. If the travel mode result is a fuzzy OD, fuzzy clustering is performed based on the travel physical parameters and traffic behavior index parameters to obtain the membership degree of each preset cluster category and determine the target cluster category. Extract the maximum and second-largest membership degrees. If the difference between the maximum and second-largest membership degrees is less than a preset difference threshold, perform path verification and correction processing on the preset travel OD to obtain the final travel mode result.

[0034] It should be noted that existing travel mode identification based on multi-source signaling suffers from several problems, including the lack of dynamic threshold adjustment, the susceptibility to misjudgment due to relying solely on clustering membership for ambiguous origin-destination (OD) identification, and the absence of a path verification and correction mechanism. To address these issues, the following approach is adopted: First, the basic travel data of the target user's preset OD is acquired, including mobile signaling data, road network data, and POI data. Travel physical parameters and traffic behavior indicators are extracted, along with weather type and a preset initial travel physical threshold set. These are then combined with the travel time period and OD area type to obtain a corrected travel physical threshold set. Finally, the travel physical parameters are matched against the corrected travel physical threshold set. The system processes and obtains travel mode results, including explicit origin-destination (OD) and fuzzy OD. If the travel mode result is a fuzzy OD, it obtains preset clustering categories, including public transportation, taxi, short- and medium-distance self-driving, and short- and medium-distance cycling. Fuzzy clustering is performed based on travel physical parameters and traffic behavior index parameters to obtain the membership degree of each preset clustering category. The target clustering category is determined, and the maximum and second-largest membership degrees are extracted. If the difference between the maximum and second-largest membership degrees is less than a preset difference threshold, path verification and correction processing is performed on the preset travel OD to obtain the final travel mode result, thereby achieving travel mode clustering and recognition based on multi-source signaling.

[0035] According to an embodiment of the present invention, the step of obtaining the basic travel data of the target user's preset travel origin-destination (OD) and extracting travel physical parameters and traffic behavior index parameters includes: Acquire the target user's preset travel origin-destination (OD) basic travel data, including mobile signaling data, road network data, and POI data; Extract travel physical parameters and traffic behavior index parameters based on mobile phone signaling data, road network data, and POI data; The travel physical parameters include travel distance, travel duration, travel time period, average speed, and OD area type; The traffic behavior index parameters include trajectory point density, base station switching frequency, speed fluctuation variance, lane trajectory ratio, and parking point matching degree.

[0036] It should be noted that, in order to accurately identify travel modes, it is necessary to first obtain multi-dimensional basic travel data of the target user's preset origin-destination (OD) points, including mobile phone signaling data, road network data, and point-of-interest (POI) data. Based on the above multi-source data, the system can accurately extract travel physical parameters and traffic behavior indicators through data fusion and feature analysis technologies. Among them, mobile phone signaling data provides core clues to the user's movement trajectory, road network data clarifies the road attributes of the travel path, and POI data helps to determine the characteristics of the travel scenario. Travel physical parameters focus on the basic attributes of travel, including travel distance (actual distance traveled between the OD points) and travel duration (from the origin to the destination). The parameters include total arrival time, travel time (the specific time period during which the trip occurs, such as the morning peak from 7:00 to 9:00), average speed (the ratio of total distance to total time), and OD area type (such as commercial area, residential area, industrial area, etc., determined in conjunction with POI data); traffic behavior indicators reflect the behavioral characteristics of users during their movement, including trajectory point density (the number of signaling trajectory points per unit time), base station switching frequency (the number of times the mobile phone switches to a base station while moving), speed fluctuation variance (a stability indicator of driving speed), lane trajectory ratio (the proportion of trajectories that match specific lanes), and parking point matching degree (the degree of fit between the stopping trajectory and parking areas such as POIs).

[0037] According to an embodiment of the present invention, the step of obtaining the weather type and a preset initial travel physical threshold set, and processing to obtain a corrected travel physical threshold set, includes: Obtain the weather type and the preset initial travel physical threshold set; The preset initial travel physical threshold set includes a preset initial speed threshold set and a preset initial distance threshold set; Based on the preset initial distance threshold set, each preset threshold is combined with the weather type, travel time, and OD area type, and then processed through a preset travel threshold correction model to obtain the corresponding corrected travel physical threshold set.

[0038] It should be noted that, to improve the accuracy of travel mode recognition, two core basic information types need to be obtained first: one is the real-time and predicted weather type, such as sunny, rainy, snowy, and foggy; the other is a preset initial travel physical threshold set. This preset initial travel physical threshold set mainly includes a preset initial speed threshold set and a preset initial distance threshold set. The former covers the speed range benchmarks for different travel modes, while the latter clarifies the distance division standards for various travel scenarios. Since weather, time of day, and region significantly affect travel characteristics—for example, rain reduces travel speed, urban roads are prone to congestion during morning rush hour, and travel patterns differ greatly between commercial and residential areas—fixed thresholds are difficult to adapt to complex scenarios. Therefore, a preset travel threshold correction model is needed to fuse the various thresholds of the preset initial distance threshold set with weather type, travel time, and OD region type, respectively. The model dynamically adjusts each initial threshold by learning the correlation patterns between multiple factors and travel thresholds of various travel modes in historical data, ultimately outputting the corresponding corrected travel physical threshold set, including the corrected speed threshold set and the corrected distance threshold set.

[0039] According to an embodiment of the present invention, the step of matching the travel physical parameters with the modified travel physical threshold set to obtain travel mode results includes explicit OD and fuzzy OD, comprising: The matching results are obtained by matching the travel distance and average speed with the corresponding thresholds of the corrected travel physical threshold set. The travel mode results are obtained based on the matching results, including explicit OD and fuzzy OD.

[0040] It should be noted that after extracting and correcting the travel physical parameters, the core step in determining the mode of travel is to accurately match the two key indicators—travel distance and average speed—with the corresponding standards of the corrected travel physical threshold set. The matching process employs a dual-indicator collaborative verification rule: the actual travel distance of the target user is compared one by one with the corrected distance threshold set, while the actual average speed is cross-matched with the corrected speed threshold set. For example, if the travel distance is within the walking distance threshold range and the average speed meets the walking speed standard, a preliminary matching result for walking is formed. Based on the matching results, two types of travel mode results can be directly distinguished: explicit OD refers to a high degree of consistency between the two indicator matching results, both pointing to a single mode of travel, such as distance and speed matching the bicycle threshold range simultaneously, without ambiguity; fuzzy OD refers to conflicting two indicator matching results or both falling within the intersection of multiple mode thresholds, such as distance matching the electric vehicle range, but speed falling between bicycle and electric vehicle, requiring further analysis.

[0041] According to an embodiment of the present invention, if the travel mode result is a fuzzy OD, fuzzy clustering is performed based on the travel physical parameters and traffic behavior index parameters to obtain the membership degree of each preset cluster category and determine the target cluster category, including: If the travel mode result is a fuzzy OD, obtain the preset clustering category, including public transportation, taxi, short- and medium-distance self-driving, and short- and medium-distance cycling; Based on the travel physical parameters and traffic behavior index parameters, a preset fuzzy K-means algorithm is used to process them to obtain the membership degree corresponding to each preset cluster category; The data is sorted, and the target cluster category is determined based on the sorting results.

[0042] It should be noted that when the travel mode result is determined to be fuzzy OD, cluster analysis is required for further precise identification. First, predefined cluster categories are established, categorized into four types based on the commonalities and modes of short- and medium-distance travel: public transport, taxi, short- and medium-distance driving, and short- and medium-distance cycling, covering possible travel options for mainstream fuzzy scenarios. Next, to improve clustering accuracy, travel physical parameters and traffic behavior indicators are used as inputs and fused using a predefined fuzzy K-means algorithm. This algorithm overcomes the limitations of traditional clustering's binary choice approach, reflecting the degree of association between samples and multiple categories, thus fitting the characteristics of fuzzy OD. During the calculation, the algorithm combines the weights of each parameter (e.g., speed fluctuation variance has a higher distinguishing effect between driving and public transport) and iteratively optimizes the objective function, outputting the membership degree corresponding to each predefined cluster, i.e., the probability that the target travel sample belongs to that category; the closer the value is to 1, the stronger the association. Then, all membership degrees are sorted in descending order, and the cluster category corresponding to the first value in the sorted list is the target cluster category after integrating the features of all parameters.

[0043] According to an embodiment of the present invention, the step of extracting the maximum and second-largest membership degrees, and if the difference between the maximum and second-largest membership degrees is less than a preset difference threshold, performing path verification and correction processing on the preset travel OD to obtain the final travel mode result, includes: Extract the membership degree of the maximum and second-largest values, perform statistical processing, and obtain the difference. The difference is compared with a preset difference threshold. If the difference is less than a preset difference threshold, the preset travel OD is processed by a preset ST-DBSCAN algorithm to obtain the time matching degree and ratio matching degree corresponding to each preset cluster category; The path matching comprehensive score corresponding to each preset cluster category is obtained by weighting the time matching degree and the ratio matching degree. The path matching scores are sorted to obtain the maximum path matching score. The preset clustering category corresponding to the maximum path matching comprehensive score is marked as the final travel mode result.

[0044] It should be noted that, to further improve the reliability of fuzzy OD travel mode identification, the membership degrees obtained from clustering need to be further processed: First, the maximum and second-largest membership degrees are extracted, and the statistical difference between the two is obtained through difference calculation. This difference directly reflects the distinguishing degree of the target travel sample to the top 2 cluster categories. This difference is compared with a preset difference threshold. If the difference is less than the threshold, it means that the probability of the sample belonging to the two clusters is close, and the membership degree alone cannot accurately determine the travel mode. A path verification mechanism needs to be introduced. At this time, the preset travel OD is processed by the preset ST-DBSCAN algorithm. This algorithm integrates time and space dimensions and can accurately mine the trajectory time series features. After the calculation, the time matching degree (the degree of fit between the trajectory time and the typical time consumption of the mode) and the ratio matching degree (the degree of fit between the trajectory parameter ratio and the category features) of each cluster category are output. The weights are assigned according to the importance of the two matching degrees and the weighted calculation is performed to obtain the path matching comprehensive score. After sorting the comprehensive scores in descending order, the cluster category corresponding to the maximum value is taken as the final travel mode result after path verification to ensure the accuracy of the judgment.

[0045] This invention discloses a travel mode clustering and identification method and system based on multi-source signaling. It acquires the basic travel data of a target user's preset travel origin (OD), extracts travel physical parameters and traffic behavior index parameters, obtains weather type and a preset initial travel physical threshold set, processes these to obtain a corrected travel physical threshold set, and performs matching processing between the travel physical parameters and the corrected travel physical threshold set to obtain travel mode results, including explicit ODs and fuzzy ODs. If the travel mode result is a fuzzy OD, fuzzy clustering processing is performed based on the travel physical parameters and traffic behavior index parameters to obtain the membership degree of each preset cluster category, determine the target cluster category, and extract the maximum and second-largest membership degrees. If the difference between the maximum and second-largest membership degrees is less than a preset difference threshold, path verification and correction processing is performed on the preset travel OD to obtain the final travel mode result, thereby achieving travel mode clustering and identification based on multi-source signaling.

[0046] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0047] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0048] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0049] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0050] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

Claims

1. A method for identifying trip mode clustering based on multi-source signaling, characterized in that, Includes the following steps: The process involves acquiring basic travel data for the target user's preset destination (OD) and extracting travel physical parameters and traffic behavior indicators. Specifically, this includes: acquiring basic travel data for the target user's preset OD, including mobile signaling data, road network data, and point-of-purchase (POI) data; extracting travel physical parameters and traffic behavior indicators based on the mobile signaling data, road network data, and POI data; the travel physical parameters include travel distance, travel duration, travel time period, average speed, and OD area type; the traffic behavior indicators include trajectory point density, base station handover frequency, speed fluctuation variance, lane trajectory ratio, and parking spot matching degree. The process of obtaining weather type and preset initial travel physical threshold set, and processing to obtain a corrected travel physical threshold set, specifically includes: obtaining weather type and preset initial travel physical threshold set; the preset initial travel physical threshold set includes preset initial speed threshold set and preset initial distance threshold set; processing according to each preset threshold of the preset initial travel physical threshold set in combination with the weather type, travel time period and OD area type through a preset travel threshold correction model to obtain the corresponding corrected travel physical threshold set; The travel physical parameters are matched with the corrected travel physical threshold set to obtain travel mode results, including explicit OD and fuzzy OD. If the travel mode result is a fuzzy OD, fuzzy clustering is performed based on the travel physical parameters and traffic behavior index parameters to obtain the membership degree of each preset cluster category and determine the target cluster category. Extract the maximum and second-largest membership degrees. If the difference between the maximum and second-largest membership degrees is less than a preset difference threshold, perform path verification and correction processing on the preset travel OD to obtain the final travel mode result.

2. The travel mode clustering and identification method based on multi-source signaling according to claim 1, characterized in that, The matching process based on the travel physical parameters and the corrected travel physical threshold set yields travel mode results, including explicit origin-destination (OD) and fuzzy origin-destination (OD), including: The matching results are obtained by matching the travel distance and average speed with the corresponding thresholds of the corrected travel physical threshold set. The travel mode results are obtained based on the matching results, including explicit OD and fuzzy OD.

3. The travel mode clustering and identification method based on multi-source signaling according to claim 1, characterized in that, If the travel mode result is a fuzzy OD (Original Direction of Travel), fuzzy clustering is performed based on the travel physical parameters and traffic behavior index parameters to obtain the membership degree of each preset cluster category and determine the target cluster category, including: If the travel mode result is a fuzzy OD, obtain the preset clustering category, including public transportation, taxi, short- and medium-distance self-driving, and short- and medium-distance cycling; Based on the travel physical parameters and traffic behavior index parameters, a preset fuzzy K-means algorithm is used to process them to obtain the membership degree corresponding to each preset cluster category; The data is sorted, and the target cluster category is determined based on the sorting results.

4. The travel mode clustering and identification method based on multi-source signaling according to claim 3, characterized in that, The extraction of the maximum and second-largest membership degrees, if the difference between the maximum and second-largest membership degrees is less than a preset difference threshold, involves path verification and correction processing of the preset travel OD to obtain the final travel mode result, including: Extract the membership degree of the maximum and second-largest values, perform statistical processing, and obtain the difference. The difference is compared with a preset difference threshold. If the difference is less than a preset difference threshold, the preset travel OD is processed by a preset ST-DBSCAN algorithm to obtain the time matching degree and ratio matching degree corresponding to each preset cluster category; The path matching comprehensive score corresponding to each preset cluster category is obtained by weighting the time matching degree and the ratio matching degree. The path matching scores are sorted to obtain the maximum path matching score. The preset clustering category corresponding to the maximum path matching comprehensive score is marked as the final travel mode result.

5. A travel mode clustering and identification system based on multi-source signaling, characterized in that, The system includes a memory and a processor. The memory contains a program for a travel mode clustering and identification method based on multi-source signaling. When the program for the travel mode clustering and identification method based on multi-source signaling is executed by the processor, it performs the following steps: The process involves acquiring basic travel data for the target user's preset destination (OD) and extracting travel physical parameters and traffic behavior indicators. Specifically, this includes: acquiring basic travel data for the target user's preset OD, including mobile signaling data, road network data, and point-of-purchase (POI) data; extracting travel physical parameters and traffic behavior indicators based on the mobile signaling data, road network data, and POI data; the travel physical parameters include travel distance, travel duration, travel time period, average speed, and OD area type; the traffic behavior indicators include trajectory point density, base station handover frequency, speed fluctuation variance, lane trajectory ratio, and parking spot matching degree. The process of obtaining weather type and preset initial travel physical threshold set, and processing to obtain a corrected travel physical threshold set, specifically includes: obtaining weather type and preset initial travel physical threshold set; the preset initial travel physical threshold set includes preset initial speed threshold set and preset initial distance threshold set; processing according to each preset threshold of the preset initial travel physical threshold set in combination with the weather type, travel time period and OD area type through a preset travel threshold correction model to obtain the corresponding corrected travel physical threshold set; The travel physical parameters are matched with the corrected travel physical threshold set to obtain travel mode results, including explicit OD and fuzzy OD. If the travel mode result is a fuzzy OD, fuzzy clustering is performed based on the travel physical parameters and traffic behavior index parameters to obtain the membership degree of each preset cluster category and determine the target cluster category. Extract the maximum and second-largest membership degrees. If the difference between the maximum and second-largest membership degrees is less than a preset difference threshold, perform path verification and correction processing on the preset travel OD to obtain the final travel mode result.

6. The travel mode clustering and identification system based on multi-source signaling according to claim 5, characterized in that, The matching process based on the travel physical parameters and the corrected travel physical threshold set yields travel mode results, including explicit origin-destination (OD) and fuzzy origin-destination (OD), including: The matching results are obtained by matching the travel distance and average speed with the corresponding thresholds of the corrected travel physical threshold set. The travel mode results are obtained based on the matching results, including explicit OD and fuzzy OD.

Images