Station passenger travel classification method and device, terminal equipment and storage medium

Abstract

The invention discloses a station passenger travel classification method and device, terminal equipment and a storage medium. The method comprises the steps of extracting residence information of passengers; determining a station range of the target station, and determining a station passenger group based on the station range; according to the target residence information, determining a station passenger group of which the residence position is in a preset area within the station range, and generating a first travel passenger category; determining the travel average speed of the passengers, and removing the passengers corresponding to the travel average speed in a preset threshold range to obtain a second travel passenger category; and identifying mistakenly classified passengers in the second travel passenger category by adopting an unsupervised clustering detection method, and removing the mistakenly classified passengers to obtain a third travel passenger category. According to the method, on the basis of preliminarily screening out the types of the travel passengers, the non-supervision clustering detection method is used for identifying and removing the mistakenly classified passengers, various complex conditions that the passengers get in and out of the station can be comprehensively covered, and the accuracy of passenger travel classification is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of transportation, and particularly to a method for classifying passengers' trips at stations, a device, a terminal device, and a storage medium. Background Art

[0002] High-speed railways have become the core component of modern transportation infrastructure and an efficient way of traveling. With their high-speed driving ability and large passenger capacity, they have had a profound impact on residents' cross-city travel and even the development pattern of cities and regions. In this context, accurately identifying and extracting high-speed rail passengers and their departure travel information is of crucial significance for the development of the TOD model in high-speed railway station areas and the social and economic progress of cities and regions.

[0003] Existing methods for classifying passengers' trips at stations usually establish classification rules based on the knowledge of professionals and the spatio-temporal characteristics of research groups to complete screening and classification. However, due to the complexity of travel choices and the uneven quality of big data, the simply set rules are difficult to comprehensively cover various complex situations, resulting in a low accuracy of classifying passengers' trips at stations. Summary of the Invention

[0004] The present invention provides a method for classifying passengers' trips at stations, a device, and a storage medium, which can solve the technical problem in the prior art that due to the complexity of travel choices and the uneven quality of big data, the simply set rules are difficult to comprehensively cover various complex situations, resulting in a low accuracy of classifying passengers' trips at stations.

[0005] The present invention provides a method for classifying passengers' trips at stations, including: Extracting the residence information of passengers; Determining the station range of the target station, screening out the target residence information with the residence location within the station range from the residence information, and determining the passengers corresponding to the target residence information as the station passenger group; According to the target residence information, determining the station passenger group with the residence location within a preset area within the station range, and generating the first type of traveling passengers; wherein, the first type of traveling passengers includes inbound passengers and outbound passengers; Calculating the distance between the destination of each passenger in the first type of traveling passengers and the nearest station, determining the average travel speed according to the distance and travel duration, and removing the passengers corresponding to the average travel speed within a preset threshold range from the first type of traveling passengers to obtain the second type of traveling passengers; Using an unsupervised clustering detection method to identify the misclassified passengers in the second type of traveling passengers, and removing the misclassified passengers from the second type of traveling passengers to obtain the third type of traveling passengers.

[0006] Further, the station passenger group includes local passengers and non-local passengers. Based on the target residence information, the station passenger group with a residence location within a preset area within the station range is determined, and a first travel passenger category is generated, including: For non-local passengers, if the current non-local passenger has no residence record in the preset area before noon the next day, the current non-local passenger is determined as an inbound passenger; if the current non-local passenger has a residence record in the preset area before noon the next day, the current non-local passenger is determined as an outbound passenger. For local passengers, if the current local passenger has no residence record in the preset area the previous day but has a record in the preset area that evening, the current local passenger is determined as an outbound passenger; if the current local passenger has a residence record in the preset area the previous day but has no residence record in the preset area that evening, the current local passenger is determined as an inbound passenger.

[0007] Further, before determining the station passenger group with a residence location within a preset area within the station range based on the target residence information and generating a first travel passenger category, it further includes: Determine the residence location with the longest residence days for each passenger in the station passenger group. Based on the distribution and inflection point of the residence days at the residence location, determine the residence days threshold. Passengers with residence days corresponding to the residence location greater than or equal to the residence days threshold are determined as local passengers; passengers with residence days corresponding to the residence location less than the residence days threshold are determined as non-local passengers.

[0008] Further, the misclassified passengers include first misclassified passengers. Identifying the misclassified passengers in the second travel passenger category includes: Convert the residence information of the passengers in the second travel passenger category into corresponding travel records, where the travel records include residence duration, travel speed, and travel duration. For the destination of each passenger in the second travel passenger category, select the station with the smallest distance from the destination as the nearest station, and calculate the station distance between the destination and the nearest station. Perform unsupervised clustering detection based on the residence duration, travel speed, travel duration, and station distance to determine multiple first clusters. By analyzing the box plot characteristics of each first cluster, determine the passengers who only pass through the target station, and determine the passengers who only pass through the target station as the first misclassified passengers.

[0009] Further, the misclassified passengers further include second misclassified passengers. Identifying the misclassified passengers in the second travel passenger category further includes: After removing the first misclassified passenger from the second travel passenger category, for the destination of each passenger in the second travel passenger category after removing the first misclassified passenger, select the rail line with the smallest distance to the destination, and calculate the line distance between the destination and the rail line; According to the residence duration, travel speed, travel duration, line distance, and the building environment where the destination is located, perform unsupervised clustering detection to determine multiple second clusters; By analyzing the box plot features of each second cluster, determine the passengers who only pass through the target station, and determine the passengers who only pass through the target station as the second misclassified passengers.

[0010] Further, extracting the residence information of the passengers includes: Obtain the mobile signaling data provided by the operator; Extract the residence information in the mobile signaling data, where the residence information includes the user unique number, residence location, and residence duration.

[0011] Further, determining the site range of the target station includes: Determine the usage situation of the target station. When the usage status of the target station is normal, crawl the longitude and latitude information of the target station, convert the longitude and latitude information into the corresponding polygon range, and determine the polygon range as the site range of the target station.

[0012] The present invention also provides a device for classifying the travel of passengers at a station, including: A residence information extraction module for extracting the residence information of passengers; A station passenger group determination module for determining the site range of the target station, screening out the target residence information with the residence location within the site range from the residence information, and determining the passengers corresponding to the target residence information as the station passenger group; A first travel passenger classification module for determining the station passenger group with the residence location within a preset area within the site range according to the target residence information, and generating a first travel passenger category; wherein, the first travel passenger category includes inbound passengers and outbound passengers; A second travel passenger classification module for calculating the distance between the destination of each passenger in the first travel passenger category and the nearest station, determining the travel average speed according to the distance and travel duration, and removing the passengers corresponding to the travel average speed within the preset threshold range from the first travel passenger category to obtain a second travel passenger category; The third travel passenger classification module is used to identify misclassified passengers in the second travel passenger category by using an unsupervised clustering detection method, remove the misclassified passengers from the second travel passenger category, and obtain the third travel passenger category.

[0013] The present invention also provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the station passenger travel classification method as described above is implemented.

[0014] The present invention also provides a computer-readable storage medium, including: a stored computer program, wherein when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the station passenger travel classification method as described above.

[0015] Through steps such as extracting the residence information of passengers, screening target residence information, and calculating the average travel speed, the present invention preliminarily screens out the travel passenger categories, and on this basis, uses an unsupervised clustering detection method to identify and remove misclassified passengers, further optimizing the classification results. It can automatically discover hidden similarities according to the internal structure and pattern of the data, thereby more accurately distinguishing inbound and outbound passengers, comprehensively covering various complex situations, and effectively improving the accuracy of passenger travel classification.

[0016] Furthermore, the present invention identifies the travel intention of passengers by analyzing their spatio-temporal characteristics. For non-local passengers, by judging their residence records in a preset area, misjudgment caused by short stays can be effectively avoided; for local passengers, by analyzing their residence records in the preset area, their travel intention can be more accurately identified. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the present application, the drawings required for implementation will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 is a flowchart showing a method for classifying the travel of passengers at a station provided by an embodiment of the present invention; Figure 2 is a flowchart showing the determination process of the first travel passenger category provided by an embodiment of the present invention; Figure 3 is a flowchart showing the determination process of the third travel passenger category provided by an embodiment of the present invention; Figure 4It is another process schematic diagram of a method for classifying the travel of passengers at a station provided by an embodiment of the present invention; Figure 5 It is a schematic structural diagram of a device for classifying the travel of passengers at a station provided by an embodiment of the present invention. Detailed implementation manners

[0019] To make the objectives, technical solutions, and advantages of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Apparently, the described embodiments are some but not all of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above drawings are intended to cover non-exclusive inclusion.

[0021] In the description of the embodiments of the present application, technical terms such as "first" and "second" are only used to distinguish different objects and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, "a plurality of" means two or more unless otherwise specifically defined.

[0022] Referring to "embodiments" herein means that the specific features, structures, or characteristics described in connection with the embodiments may be included in at least one embodiment of the present application. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein may be combined with other embodiments.

[0023] In the description of the embodiments of the present application, the term "and / or" is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " herein generally represents an "or" relationship between the preceding and following associated objects.

[0024] In the description of the embodiments of the present application, the term "a plurality of" refers to two or more (including two). Similarly, "a plurality of groups" refers to two or more groups (including two groups), and "a plurality of pieces" refers to two or more pieces (including two pieces).

[0025] In the description of the embodiments of the present application, unless otherwise clearly specified and limited, technical terms such as "installation", "connection", "connection", "fixation" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or integrated; it can also be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components or the interaction relationship between two components. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to specific situations.

[0026] See Figure 1 , to solve the technical problem in the prior art that due to the complexity of travel choices and the uneven quality of big data, the simply set rules are difficult to comprehensively cover various complex situations, resulting in a low accuracy of classifying the travel of passengers at stations, an embodiment of the present invention provides a method for classifying the travel of passengers at stations, including: S1. Extract the residence information of passengers; In the embodiments of the present invention, the residence information of passengers can be further extracted by obtaining the mobile phone signaling data of passengers, where the residence information can include the start residence time point, the residence location, and the residence duration, etc.

[0027] S2. Determine the site range of the target station, screen out the target residence information with the residence location within the site range from the residence information, and determine the passengers corresponding to the target residence information as the station passenger group; In the embodiments of the present invention, the usage situation of the site can be determined according to the official railway website, and then the target station can be determined, and further the site range of the target station can be determined.

[0028] S3. According to the target residence information, determine the station passenger group within the preset area with the residence location within the site range, and generate the first travel passenger category; among them, the first travel passenger category includes inbound passengers and outbound passengers; In the embodiments of the present invention, the preset area can be the key area for analysis or research within the site range. For example, this area can be an area where the density of hotels or inns is higher than the preset density threshold, or an area where the number of commercial entities reaches a certain amount.

[0029] S4. Calculate the distance between the destination of each passenger in the first travel passenger category and the nearest station, determine the travel average speed according to the distance and the travel duration, and remove the passengers corresponding to the travel average speed within the preset threshold range from the first travel passenger category to obtain the second travel passenger category; In an embodiment of the present invention, the average travel speed is determined based on the distance and travel duration, and compared with the average travel speed within a preset threshold range, so as to determine whether the travel speed of the passenger related to the target station is an unconventional travel speed. If so, the passenger corresponding to the travel record is excluded. The unconventional travel speed in the embodiment of the present invention can be 0 or higher than the preset threshold range. When the travel speed is 0, the passenger may enter and exit at the same station. When the travel speed is higher than the preset threshold range, it is determined that there is a data error. The passengers corresponding to the travel records with these unconventional travel speeds are noise data. By reducing the influence of these noise data, the embodiment of the present invention can effectively improve the accuracy of passenger travel classification.

[0030] S5. Identify the misclassified passengers in the second travel passenger category by using an unsupervised clustering detection method, and remove the misclassified passengers from the second travel passenger category to obtain the third travel passenger category.

[0031] In the embodiment of the present invention, by extracting the residence information of passengers, screening the target residence information, calculating the average travel speed and other steps, the travel passenger category is initially screened. On this basis, the unsupervised clustering detection method is used to identify and remove the misclassified passengers, further optimizing the classification result. It can automatically discover hidden similarities according to the internal structure and pattern of the data, so as to more accurately distinguish the inbound and outbound passengers, comprehensively cover various complex situations, and effectively improve the accuracy of passenger travel classification.

[0032] In one embodiment, step S1. Extract the residence information of passengers, including: S11. Obtain the mobile phone signaling data provided by the operator; S12. Extract the residence information in the mobile phone signaling data. The residence information includes the user unique number, residence location and residence duration.

[0033] In the embodiment of the present invention, the mobile phone signaling data can record the assistant information according to each mobile phone for each communication event and the uniquely corresponding base station (Base Station, BS). Among them, the communication events include sending and receiving text messages, making and receiving calls, etc. The mobile phone signaling data in the embodiment of the present invention includes the user unique number, residence duration, corresponding location ID and start residence time point, and their expression forms are as follows: <uid, datetime, grid_id, date>.

[0034] In the embodiment of the present invention, the specific residence information can be marked according to all the residence information of each user in the current month. Among them, the specific residence information can include the residence days at each residence location and the residence location activity type. The residence location activity type is the activity type of the passenger at the residence location, including the residence type, work type and other types.

[0035] Please refer to Table 1, which is a schematic table of residence information provided by an embodiment of the present invention.

[0036] Table 1: Schematic Table of Residence Information In the embodiment of the present invention, for all residence locations where each user travels in a month, the total number of days that the user appears at each residence location in that month is counted and marked as days_poi. Secondly, the activity types of the user's residence locations are distinguished. They can be classified into three types: residence, work, and others, corresponding to 1, 2, and 0 respectively, and marked as ptype. Among them, for the residence type (ptype = 1), it is marked according to the longest residence time of the user from 9 pm to 8 am the next morning within that month; for the work type (ptype = 2), it is marked according to the longest residence time of the user from 9 am to 5 pm within that month and different from the residence location; the remaining residence types (ptype = 0) are all defined as others.

[0037] In the embodiment of the present invention, based on the residence information obtained above, the residence information can be further converted into corresponding travel records. According to the same user ID and the same date, in the order of residence time from early to late, the end time in each residence information is selected as the departure time, the start time of the next residence information is selected as the arrival time, and the numbers of the two residence records are used as the start and end position marks, obtaining preliminary travel records and converting them into the following data structure: <uid, s_time, e_time, o_grid_id, d_grid_id, date>.

[0038] In the embodiment of the present invention, the travel duration can be obtained by subtracting the departure time from the arrival time of each trip, and the residence duration of each residence record can be calculated by subtracting the start time of the residence from the end time of the residence and added to the travel record, finally obtaining the basic travel record data: <uid, s_time, e_time, o_grid_id, d_grid_id, duration, staytime, date>.

[0039] Please refer to Table 2, which is a schematic table of travel records provided by an embodiment of the present invention. In the embodiment of the present invention, passenger individual identification and residence information extraction are realized through the mobile phone signaling data provided by the operator, which can effectively reduce the data collection cost, efficiently expand the coverage range, and through the extraction of corresponding data from the mobile phone information data, the accuracy and reliability of the data can be effectively ensured.

[0040] In one embodiment, step S2, determining the site range of the target station, includes: Determine the usage situation of the target station. When the usage status of the target station is normal, crawl the longitude and latitude information of the target station, convert the longitude and latitude information into the corresponding polygon range, and determine the polygon range as the site range of the target station.

[0041] In the embodiments of the present invention, the station includes a railway station and a high-speed railway station. It can be understood that railway tracks with a speed of more than 200 / h are defined as high-speed railways. Among them, high-speed railways cover trains with the prefixes C, G, and D, which serve both the connection between regions and the demand for the flow of people within regions.

[0042] In the embodiments of the present invention, the usage situation of the site can be determined according to the official railway APP. After screening and determining the research object of the high-speed railway station, the map web service can be used according to its name to search for the site, so as to crawl the longitude and latitude information of the spatial range in the web page, convert it into the corresponding polygon range in QGIS, and finally convert the coordinate system to WGS84.

[0043] In the embodiments of the present invention, the land use information within the target station can be summarized and marked. For example, the Sentinel-2 satellite remote sensing interpretation image data provided by ESRI is used, which is in the WGS-84 coordinate system and has a precision of 10 meters. Select the land use type of urban built-up area to reflect the built environment characteristics of the travel destination, and count the number of this land use type in the grid unit where the travel end point belongs.

[0044] The embodiments of the present invention can accurately define the site range of the station by crawling the longitude and latitude information of the target station and converting it into a polygon range, so as to accurately judge whether the position of the passenger is within the station range, and further more accurately identify whether the passenger is in the inbound or outbound state, effectively improving the accuracy of passenger travel classification.

[0045] In one embodiment, before step S3, according to the target residence information, determining the station passenger group whose residence location is within the preset area within the site range and generating the first travel passenger category, further includes: Determine the residence location with the longest residence days for each passenger in the station passenger group; In the embodiments of the present invention, the preset research area is an area determined according to the residential area within the site range. For example, the residential area can be determined as the preset research area.

[0046] Local passengers live in the preset research area of the target station for a long time, while non-local passengers are the opposite.

[0047] Embodiments of the present invention can divide local passengers and non-local passengers based on the days_poi tag in the passenger's descriptive information. Specifically, it can be: count the longest number of days (the maximum value of days_poi) among all stay locations of each passenger in the current month and define it as max_day.

[0048] Determine the stay days threshold according to the distribution and inflection point of the stay days at the stay location; In embodiments of the present invention, the stay days threshold k can be determined according to the distribution and inflection point of max_day to distinguish local and non-local passengers.

[0049] Passengers whose stay days corresponding to the stay location are greater than or equal to the stay days threshold are determined as local passengers; passengers whose stay days corresponding to the stay location are less than the stay days threshold are determined as non-local passengers.

[0050] In embodiments of the present invention, if a passenger's max_day is greater than or equal to k, it means that the passenger has a relatively stable stay pattern in the preset research area and is classified as a local passenger with a fixed residence; if a passenger's max_day is less than k, it means that the passenger has almost no relatively stable stay record in the preset research area and is classified as a non-local passenger who enters the preset research area by high-speed rail and visits.

[0051] Embodiments of the present invention determine the stay days threshold by analyzing the distribution and inflection point of the passenger's stay days, can accurately identify local passengers and non-local passengers, and based on the data-driven classification method, can avoid classification errors caused by subjective judgment, thereby more accurately identifying the travel characteristics of passengers.

[0052] In one embodiment, in step S3, the station passenger group includes local passengers and non-local passengers. According to the target stay information, determine the station passenger group within the preset area within the station range of the stay location, and generate the first travel passenger category, including: S31. For non-local passengers, if the current non-local passenger has no stay record in the preset area before noon the next day, then determine the current non-local passenger as an inbound passenger; if the current non-local passenger has a stay record in the preset area before noon the next day, then determine the current non-local as an outbound passenger; S32. For local passengers, if the current local passenger had no stay record in the preset area the previous day and has a record in the preset area that night, then determine the current local passenger as an outbound passenger; if the current local passenger had a stay record in the preset area the previous day and has no stay record in the preset area that night, then determine the current local passenger as an inbound passenger.

[0053] In the embodiments of the present invention, the travel intention of passengers is identified by analyzing the spatio-temporal characteristics of passengers. For non-local passengers, by judging their stay records in the preset area, misjudgment caused by short-term stays can be effectively avoided; for local passengers, by analyzing their stay records in the preset area, their travel intention can be more accurately identified.

[0054] Please refer to Figure 2 , which is a schematic diagram of the first travel passenger category determination process provided by the embodiments of the present invention.

[0055] In one embodiment, in step S4 of calculating the first travel passenger category, the distance between the destination of each passenger and the nearest station is calculated, and the average travel speed is determined according to the distance and travel duration. Passengers corresponding to the average travel speed within the preset threshold range are removed from the first travel passenger category to obtain the second travel passenger category. The distribution and mean of the average travel speed of all passengers are statistically analyzed and compared with the average travel speed within the preset range to determine travel records with unconventional travel attributes. For example, if the average travel speed is higher than the preset threshold, data anomalies may occur, and if the average travel speed is 0, it may mean that the passenger gets on and off at the same station, etc. For travel records belonging to unconventional ones, the travel records corresponding to these travel speeds are determined as noise data, and the passengers corresponding to the noise data are removed from the first travel passenger category to update the travel passenger category and obtain the second travel passenger category.

[0056] In one embodiment, the misclassified passengers include the first misclassified passengers. Step S5 of identifying the misclassified passengers in the second travel passenger category includes: S501: Convert the stay information of the passengers in the second travel passenger category into corresponding travel records, where the travel records include the stay duration, travel speed, and travel duration. In the embodiments of the present invention, the misclassified passengers are the group of passengers who are only passing through the target station or railway and are misidentified as departing passengers. That is, the travel of this part of the misclassified passengers is not a departing travel, but a stop at a certain station, and then continue to travel by high-speed rail and single trips formed due to base station identification during the operation of the high-speed rail within the city.

[0057] S502: For the destination of each passenger in the second travel passenger category, select the station with the smallest distance from the destination as the nearest station, and calculate the station distance between the destination and the nearest station. S503: Perform unsupervised clustering detection based on the stay duration, travel speed, travel duration, and station distance to determine multiple first clusters. In the embodiments of the present invention, the travel record of each passenger in the second travel passenger category can be used as a sample, and each sample is regarded as a single cluster. Clustering is performed by iteratively merging according to the distance between different classes until all data points form a large cluster.

[0058] During the clustering process, the normalized Manhattan distance can be used as the distance for calculation. The formula for the Manhattan distance is as follows: Where is the distance between two samples, p and q are two samples (two data points), k is the number of each attribute, there are n attributes in total, and pk and qk are the values of the k attribute of samples p and q.

[0059] In the embodiments of the present invention, a fully connected layer can be selected to calculate the distance between two clusters. The distance between the least similar samples in the two clusters is used as the distance between the clusters, that is, the longest distance. The formula is: x and y are sample points in categories Ci and Cj respectively, and the Manhattan distance is used to measure the distance between sample points x and y.

[0060] In the embodiments of the present invention, the steps for determining the number of clusters include: S51. Determine the number interval by observing the branch structure and node connection of the dendrogram; In the embodiments of the present invention, when the branches of the dendrogram are relatively clear and the distances between different branches are large, it may mean that there are significantly different clusters. Based on this, a suitable number interval is determined according to actual needs; S52. Based on the determined number interval, set different categories for clustering; In the embodiments of the present invention, two indicators, the Silhouette Coefficient and the Within-Cluster Sum of Average Distance, can be selected to evaluate the classification effect.

[0061] In the embodiments of the present invention, SC scores are defined. SC scores are the average value of the silhouette coefficients of all samples in the set. It comprehensively considers the within-cluster compactness and between-cluster separation. The value range is from -1 to 1. The closer it is to 1, the better the division effect, and the more obvious the difference between classes. Therefore, the local maximum point is the potential optimal number. WSD is the sum of the average values of the distances between points within each cluster. As the number of clusters increases, WSD will gradually decrease. More clusters mean that the data points within each cluster are closer. Then, when it increases to a certain critical point, its decline rate slows down significantly. This critical point is the "elbow", and the number with less fluctuation after this point is the potential optimal number. In the embodiments of the present invention, the optimal number of categories can be comprehensively determined according to the changing trends of SC scores and WSD.

[0062] Among them, SC is the silhouette coefficient, N is the number of all samples, is the average distance from sample point i to the remaining sample points in its affiliated category, is the average distance from sample point i to the remaining sample points in its affiliated category.

[0063] Among them, among them, WSD k is the within-cluster distance, d manhattan (x, y) is the Manhattan distance between sample points x and y, k is the number of all categories, is one of the categories and has sample points x, y.

[0064] S504. By analyzing the box plot characteristics of each first cluster, determine the passengers who only pass through the target station, and determine the passengers who only pass through the target station as the first misclassified passengers.

[0065] In the embodiment of the present invention, the characteristics of the first misclassified passengers are set as follows: 1. There are abnormalities in their own travel attributes; 2. The spatial distance from the railway facilities is relatively close; 3. The number of this classification group is relatively small compared to the total group, and the proportion is lower than the preset proportion threshold. In the embodiment of the present invention, in combination with the above set characteristics, by analyzing the box plot characteristics of each first cluster, the first misclassified passengers are determined.

[0066] In the embodiment of the present invention, through unsupervised clustering detection and box plot feature analysis, it is possible to identify passenger groups with similar travel characteristics, and screen out the passengers who only pass through the target station as misclassified passengers from them, making full use of the internal structure and distribution characteristics of the data, avoiding misjudgments that may be caused by a single feature (such as residence duration or travel speed), and thus more accurately identifying the passenger travel categories.

[0067] In one embodiment, the misclassified passengers further include second misclassified passengers. Step S5 of identifying the misclassified passengers in the second travel passenger category further includes: S511. After removing the first misclassified passengers from the second travel passenger category, for the destination of each passenger in the second travel passenger category after removing the first misclassified passengers, select the rail line with the smallest distance from the destination, and calculate the line distance between the destination and the rail line; S512. According to the residence duration, travel speed, travel duration, line distance, and the building environment where the destination is located, perform unsupervised clustering detection to determine multiple second clusters; S513. By analyzing the box plot characteristics of each second cluster, determine the passengers who only pass through the target station, and determine the passengers who only pass through the target station as the second misclassified passengers.

[0068] In the embodiment of the present invention, after removing the first misclassified passenger, based on the same technical concept as in step S4, unsupervised clustering detection is performed according to the residence duration, travel speed, travel duration, line distance, and the building environment where the destination is located to determine multiple second clusters, thereby determining the second misclassified passengers.

[0069] On the basis of removing the first misclassification, the embodiment of the present invention further performs unsupervised clustering detection according to the residence duration, travel speed, travel duration, line distance, and the building environment where the destination is located to determine and remove the second misclassified passengers, which can further identify passengers deviating from the normal travel pattern, thereby effectively improving the accuracy and reliability of passenger travel classification.

[0070] Please refer to Figure 3 , which is a schematic diagram of the determination process of the third passenger travel category provided by the embodiment of the present invention. In Figure 3 , Scenario 1 is that the next residence point of the passenger's travel is near another station, that is, the first misclassified passenger; Scenario 2 is that the next residence point of the passenger's travel is near the railway track inside the city, that is, the second misclassified passenger. The embodiment of the present invention obtains the final result of the off-station travel extraction by sequentially identifying the passengers in Scenario 1 and Scenario 2 and sequentially removing the corresponding misclassified passengers, which is the third passenger travel category.

[0071] In one embodiment, after step S5, it further includes verifying the accuracy of the passenger travel classification at the station, including: S6. Obtain the train arrival time information of each station passing through the preset research area on the same day, and the start time information of the high-speed rail passengers' off-station travel. Two time series of train arrivals and passenger off-stations are sorted according to different time windows (2, 3, 4... 15, 20, 30, 60 min), and the off-station travel records occurring during the early morning period are excluded (this part is the passengers who arrived the previous night and stayed in the station for a long time); S7. Use the DTW method for evaluation. The DTW method is a method for comparing the similarity of time series with different lengths or different time alignments. Specifically: find the optimal alignment path between the two sequences to minimize their matching cost, and finally evaluate the similarity of the two sequences by calculating the normalized distance. When the Normalized Distance < 0.1, it indicates a high degree of consistency. The specific calculation process is as follows: Among them, DTW(A, B) is the dynamic time warping distance between time series A and B, represents in time series A and B and The local distance in the embodiments of the present invention is set as the Euclidean distance, while w is the alignment path with constraint conditions.

[0072] In the embodiments of the present invention, the cumulative distance matrix C is defined through dynamic programming: Among them, C[i, j] is the element in the cumulative distance matrix, and the final DTW is the last element of the cumulative matrix, that is , where N and M respectively represent the lengths of time series A and time series B. To eliminate the influence of the sequence length, the embodiments of the present invention use Normalized Distance (ND) for evaluation; Among them, ND is the normalized distance, represents the length of w.

[0073] S8. Calculate the ND index of the two sequences of train arrivals and departures for trips under different time windows, and compare it with 0.1. If it is less than 0.1 under all windows, it means that the two sequences are highly similar, that is, the identified and extracted passengers and departure trips have high credibility.

[0074] In the embodiments of the present invention, after the accuracy verification is completed, to reduce the data volume and match other research subjects, the extracted passengers and their departure trips can be aggregated according to stations and grids, and information such as average travel distance, speed, start stay time point, etc. can be calculated, where the grid can be defined according to research needs, such as 500 meters, 1 kilometer, etc.

[0075] After obtaining the third type of outbound passengers in the embodiments of the present invention, a time series of train arrivals and departures for trips is constructed by means of time series comparison, and the DTW algorithm is used to compare the similarity to verify the reliability of the classification results and ensure the accuracy and reliability of passenger trip classification.

[0076] Please refer to Figure 4 , which is another process schematic diagram of a method for classifying station passenger trips provided by an embodiment of the present invention. Figure 4 In, after data collection and preprocessing, heuristic rules are used to initially determine high-speed rail passengers and trip types, and then in combination with heuristic rules and data-driven clustering methods, noise and interference records are removed, and finally accuracy verification and data aggregation are performed to achieve passenger trip classification and verification.

[0077] Implementing the embodiments of the present invention has the following beneficial effects: In the embodiment of the present invention, by extracting the residence information of passengers, screening the target residence information, calculating the average travel speed, etc., the types of outbound passengers are initially screened. On this basis, the unsupervised clustering detection method is used to identify and remove misclassified passengers, further optimizing the classification results. It can automatically discover hidden similarities according to the internal structure and pattern of the data, so as to more accurately distinguish inbound and outbound passengers, comprehensively cover various complex situations, and effectively improve the accuracy of passenger travel classification.

[0078] Furthermore, in the embodiment of the present invention, the travel intention of passengers is identified by analyzing the spatio-temporal characteristics of passengers. For non-local passengers, by judging their residence records within the preset area, the misjudgment caused by short stays can be effectively avoided; for local passengers, by analyzing their residence records in the preset area, their travel intention can be more accurately identified.

[0079] Please refer to Figure 5 , based on the same inventive concept as the above embodiment, the present invention also provides a device for classifying the travel of station passengers, including: A residence information extraction module 10 for extracting the residence information of passengers; A station passenger group determination module 20 for determining the site range of the target station, screening out the target residence information with the residence location within the site range from the residence information, and determining the passengers corresponding to the target residence information as the station passenger group; A first outbound passenger classification module 30 for determining the station passenger group within the preset area with the residence location within the site range according to the target residence information, and generating a first type of outbound passengers; wherein, the first type of outbound passengers includes inbound passengers and outbound passengers; A second outbound passenger classification module 40 for calculating the distance between the destination of each passenger in the first type of outbound passengers and the nearest station, determining the average travel speed according to the distance and travel duration, and removing the passengers corresponding to the average travel speed within the preset threshold range from the first type of outbound passengers to obtain a second type of outbound passengers; A third outbound passenger classification module 50 for using the unsupervised clustering detection method to identify the misclassified passengers in the second type of outbound passengers, and removing the misclassified passengers from the second type of outbound passengers to obtain a third type of outbound passengers.

[0080] In one embodiment, the first outbound passenger classification module 30 is further configured to: For non-local passengers, if the current non-local passenger has no residence record within the preset area before noon the next day, the current non-local passenger is determined as an inbound passenger; if the current non-local passenger has a residence record within the preset area before noon the next day, the current non-local is determined as an outbound passenger; For local passengers, if the current local passenger has no residence record in the preset area the previous day but has a record in the preset area on the same night, the current local passenger is determined as an outbound passenger; if the current local passenger has a residence record in the preset area the previous day but has no residence record in the preset area on the same night, the current local passenger is determined as an inbound passenger.

[0081] In one embodiment, it further includes a passenger classification module for: Determine the residence position with the longest residence days for each passenger in the station passenger group; Determine the residence days threshold according to the distribution and inflection points of the residence days at the residence positions; Passengers whose residence days corresponding to the residence positions are greater than or equal to the residence days threshold are determined as local passengers; passengers whose residence days corresponding to the residence positions are less than the residence days threshold are determined as non-local passengers.

[0082] In one embodiment, the third travel passenger classification module 50 is further used for: Convert the residence information of the passengers in the second travel passenger category into corresponding travel records, where the travel records include residence duration, travel speed, and travel duration; For the destination of each passenger in the second travel passenger category, select the station with the smallest distance from the destination as the nearest station, and calculate the station distance between the destination and the nearest station; Perform unsupervised clustering detection based on the residence duration, travel speed, travel duration, and station distance to determine multiple first clusters; By analyzing the box plot features of each first cluster, determine the passengers who only pass through the target station, and determine the passengers who only pass through the target station as the first misclassified passengers.

[0083] In one embodiment, the third travel passenger classification module 50 is further used for: After removing the first misclassified passengers from the second travel passenger category, for the destination of each passenger in the second travel passenger category after removing the first misclassified passengers, select the rail line with the smallest distance from the destination, and calculate the line distance between the destination and the rail line; Perform unsupervised clustering detection based on the residence duration, travel speed, travel duration, line distance, and the building environment where the destination is located to determine multiple second clusters; By analyzing the box plot features of each second cluster, determine the passengers who only pass through the target station, and determine the passengers who only pass through the target station as the second misclassified passengers.

[0084] In one embodiment, the residence information extraction module 10 is further used for: Obtain the mobile signaling data provided by the operator; Extract the residence information from the mobile phone signaling data, where the residence information includes the user's unique identifier, residence location, and residence duration.

[0085] In one embodiment, the station passenger group determination module 20 is further configured to: Determine the usage situation of the target station. When the usage status of the target station is normal, crawl the longitude and latitude information of the target station, convert the longitude and latitude information into a corresponding polygon range, and determine the polygon range as the station range of the target station.

[0086] It can be understood that the above device item embodiments correspond to the method item embodiments of the present invention, and they can implement the station passenger travel classification method provided by any one of the above method item embodiments of the present invention.

[0087] It should be noted that the device embodiments described above are merely illustrative. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that they have a communication connection, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0088] Based on the above embodiments of the station passenger travel classification method, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the station passenger travel classification method of any embodiment of the present invention.

[0089] Exemplarily, in this embodiment, the computer program can be divided into one or more modules. The one or more modules are stored in the memory and executed by the processor to complete the present invention. The one or more module elements can be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program in the terminal device.

[0090] The terminal device can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The terminal device may include, but is not limited to, a processor and a memory.

[0091] The so-called processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the terminal device and connects all parts of the entire terminal device through various interfaces and circuits.

[0092] Based on the above method embodiment, another embodiment of the present invention provides a computer-readable storage medium, including a stored computer program, wherein when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the station passenger travel classification method described in any one of the above method embodiments of the present invention.

[0093] Among them, if the modules / units integrated in the device / terminal device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above method embodiments of the present invention, it can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0094] The above is the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements are also regarded as the protection scope of the present invention.

Claims

1. A method for classifying the travel of passengers at a station, characterized in that, Including: Extracting the residence information of passengers; Determining the station range of the target station, screening out the target residence information with the residence location within the station range from the residence information, and determining the passengers corresponding to the target residence information as the station passenger group; According to the target residence information, determining the station passenger group within the preset area within the station range of the residence location, and generating the first travel passenger category; wherein, the first travel passenger category includes inbound passengers and outbound passengers; Calculating the distance between the destination of each passenger in the first travel passenger category and the nearest station, determining the average travel speed according to the distance and travel duration, and removing the passengers corresponding to the average travel speed within the preset threshold range from the first travel passenger category to obtain the second travel passenger category; Using the unsupervised clustering detection method to identify the misclassified passengers in the second travel passenger category, and removing the misclassified passengers from the second travel passenger category to obtain the third travel passenger category.

2. The station passenger travel classification method according to claim 1, characterized in that, The station passenger group includes local passengers and non-local passengers. The determining the station passenger group within the preset area within the station range of the residence location according to the target residence information and generating the first travel passenger category includes: For non-local passengers, if the current non-local passenger has no residence record in the preset area before noon the next day, then determining the current non-local passenger as an inbound passenger; if the current non-local passenger has a residence record in the preset area before noon the next day, then determining the current non-local as an outbound passenger; For local passengers, if the current local passenger has no residence record in the preset area the previous day and has a record in the preset area that evening, then determining the current local passenger as an outbound passenger; if the current local passenger has a residence record in the preset area the previous day and has no residence record in the preset area that evening, then determining the current local passenger as an inbound passenger.

3. The method for classifying the travel of passengers at a station according to claim 2, wherein, Before determining the station passenger group within the preset area within the station range of the residence location according to the target residence information and generating the first travel passenger category, it further includes: Determining the residence location with the longest residence days for each passenger in the station passenger group; Determining the residence days threshold according to the distribution and inflection point of the residence days of the residence location; Determining the passengers with the residence days corresponding to the residence location greater than or equal to the residence days threshold as local passengers; and determining the passengers with the residence days corresponding to the residence location less than the residence days threshold as non-local passengers.

4. The method for classifying the travel of passengers at a station according to claim 1, wherein, The misclassified passengers include the first misclassified passengers. The identifying the misclassified passengers in the second travel passenger category includes: Converting the residence information of the passengers in the second travel passenger category into corresponding travel records, where the travel records include residence duration, travel speed, and travel duration; For the destination of each passenger in the second travel passenger category, selecting the station with the smallest distance from the destination as the nearest station, and calculating the station distance between the destination and the nearest station; Perform unsupervised clustering detection based on the residence duration, travel speed, travel duration, and station distance to determine multiple first clusters. By analyzing the box plot features of each first cluster, determine the passengers who only pass through the target station, and identify the passengers who only pass through the target station as the first misclassified passengers.

5. The method for classifying the travel of passengers at a station according to claim 4, wherein, The misclassified passengers also include the second misclassified passengers. Identifying the misclassified passengers in the second category of traveling passengers further includes: After removing the first misclassified passengers from the second category of traveling passengers, for the destination of each passenger in the second category of traveling passengers after removing the first misclassified passengers, select the rail line with the minimum distance from the destination, and calculate the line distance between the destination and the rail line. Perform unsupervised clustering detection based on the residence duration, travel speed, travel duration, line distance, and the building environment where the destination is located to determine multiple second clusters. By analyzing the box plot features of each second cluster, determine the passengers who only pass through the target station, and identify the passengers who only pass through the target station as the second misclassified passengers.

6. The method for classifying the travel of passengers at a station according to claim 1, characterized in that, The extraction of the residence information of the passengers includes: Obtain the mobile signaling data provided by the operator. Extract the residence information from the mobile signaling data, where the residence information includes the user unique identifier, residence location, and residence duration.

7. The method for classifying the travel of passengers at a station according to claim 1, characterized in that, The determination of the station range of the target station includes: Determine the usage situation of the target station. When the usage status of the target station is normal, crawl the longitude and latitude information of the target station, convert the longitude and latitude information into the corresponding polygon range, and determine the polygon range as the station range of the target station.

8. A device for classifying the travel of passengers at a station, characterized in that, Includes: A residence information extraction module for extracting the residence information of the passengers. A station passenger group determination module for determining the station range of the target station, screening out the target residence information with the residence location within the station range from the residence information, and determining the passengers corresponding to the target residence information as the station passenger group. A first traveling passenger classification module for determining the station passenger group with the residence location within the preset area within the station range according to the target residence information, and generating the first category of traveling passengers; wherein, the first category of traveling passengers includes inbound passengers and outbound passengers. A second traveling passenger classification module for calculating the distance between the destination of each passenger in the first category of traveling passengers and the nearest station, determining the average travel speed according to the distance and travel duration, and removing the passengers corresponding to the average travel speed within the preset threshold range from the first category of traveling passengers to obtain the second category of traveling passengers. A third traveling passenger classification module for identifying the misclassified passengers in the second category of traveling passengers by using the unsupervised clustering detection method, and removing the misclassified passengers from the second category of traveling passengers to obtain the third category of traveling passengers.

9. A terminal device, characterized in that, Includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the station passenger travel classification method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, Includes: A stored computer program, wherein when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the station passenger travel classification method according to any one of claims 1-7.

Images