A flow behavior identification method, device, equipment and storage medium

Abstract

The application provides a flow behavior identification method and device, equipment and a storage medium, and relates to the field of big data. The method comprises: mapping user data for representing the space-time behavior characteristics of a user into a bitmap sequence according to a geographical space dimension and a time resolution; generating migration behavior data based on the user data, the migration behavior data being used to represent the migration behavior of the user from a starting point to an ending point within a time period; processing the bitmap sequence based on task information through a bitmap sequence addition algorithm and a residence duration interval filtering algorithm to obtain a processed bitmap sequence, the task information comprising a source area, a target area, a time period and a residence duration; and identifying a user meeting the task information from a plurality of users corresponding to the user data based on the processed bitmap sequence and the migration behavior data. The flexibility of identifying the regional flow behavior of a user is improved.

Description

Technical Field

[0001] This application relates to the field of big data, and in particular to a method, apparatus, device and storage medium for identifying mobile behavior. Background Technology

[0002] In related technologies, regional population mobility identification mainly relies on constructing mobility behavior characteristic indicators through mobile communication network data (MME, Mobility Management Entity), HSS (Home Subscriber Server), or user location trajectory data from user terminals. However, this approach suffers from problems such as small data processing scale, high data storage cost, regional spatial limitations, and insufficient spatiotemporal analysis capabilities, resulting in poor performance in regional population mobility identification. Summary of the Invention

[0003] This application provides a method, apparatus, device, and storage medium for identifying mobile behavior, in order to solve the problem of high false positive rates in existing mobile behavior identification methods.

[0004] To solve the above-mentioned technical problems, this application is implemented as follows:

[0005] In a first aspect, embodiments of this application provide a method for recognizing mobile behavior, the method comprising:

[0006] User data used to characterize users’ spatiotemporal behavioral features are mapped into bitmap sequences according to geospatial dimension and temporal resolution.

[0007] Migration behavior data is generated based on the user data, and the migration behavior data is used to characterize the user's migration behavior from the starting point to the destination within a time period.

[0008] Based on the task information, the bitmap sequence is processed by a bitmap sequence addition algorithm and a dwell time interval filtering algorithm to obtain a processed bitmap sequence. The task information includes the source region, target region, time period, and dwell time.

[0009] Based on the processed bitmap sequence and the migration behavior data, users matching the task information are identified from multiple users corresponding to the user data.

[0010] Secondly, embodiments of this application also provide a mobility behavior recognition device. The mobility behavior recognition device includes:

[0011] The mapping module is used to map user data that characterizes the spatiotemporal behavior of users into a bitmap sequence according to the geospatial dimension and time resolution.

[0012] A generation module is used to generate migration behavior data based on the user data, wherein the migration behavior data is used to characterize the user's migration behavior from the starting point to the destination within a time period;

[0013] The processing module is used to process the bitmap sequence based on task information using a bitmap sequence addition algorithm and a dwell time interval filtering algorithm to obtain a processed bitmap sequence. The task information includes the source region, the target region, the time period, and the dwell time.

[0014] The identification module is used to identify users who match the task information from multiple users corresponding to the user data based on the processed bitmap sequence and the migration behavior data.

[0015] Thirdly, embodiments of this application provide an electronic device, including a transceiver and a processor, wherein the processor is used for:

[0016] User data used to characterize users’ spatiotemporal behavioral features are mapped into bitmap sequences according to geospatial dimension and temporal resolution.

[0017] Migration behavior data is generated based on the user data, and the migration behavior data is used to characterize the user's migration behavior from the starting point to the destination within a time period.

[0018] Based on the task information, the bitmap sequence is processed by a bitmap sequence addition algorithm and a dwell time interval filtering algorithm to obtain a processed bitmap sequence. The task information includes the source region, target region, time period, and dwell time.

[0019] Based on the processed bitmap sequence and the migration behavior data, users matching the task information are identified from multiple users corresponding to the user data.

[0020] Fourthly, embodiments of this application provide an electronic device, including: a processor, a memory, and a program stored in the memory and executable on the processor, wherein when the program is executed by the processor, it implements the steps of the flow behavior recognition method as described in the first aspect above.

[0021] Fifthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the flow behavior recognition method described in the first aspect above.

[0022] In a sixth aspect, embodiments of this application provide a computer program product, including computer instructions, which, when executed by a processor, implement the steps of the flow behavior recognition method as described in the first aspect above.

[0023] The mobile behavior recognition method of this application includes mapping user data, which characterizes the spatiotemporal behavioral features of users, into a bitmap sequence according to geospatial dimension and time resolution; generating migration behavior data based on the user data, the migration behavior data being used to characterize the user's migration behavior from origin to destination within a time period; processing the bitmap sequence based on task information using a bitmap sequence addition algorithm and a dwell time interval filtering algorithm to obtain a processed bitmap sequence, the task information including source region, target region, time period, and dwell time; and identifying users that match the task information from multiple users corresponding to the user data based on the processed bitmap sequence and the migration behavior data. This method achieves compressed storage of user behavior data by mapping user dwell time into a binary bitmap sequence according to a preset spatiotemporal dimension; it facilitates subsequent large-scale data processing through binary bit operations; it supports flexible configuration of multiple granular time precisions (such as minute level, hour level) and multiple types of geographical units (base station, custom area) through a unified bitmap data structure, enhancing the adaptability of the mobility behavior recognition method in this application to different scenarios; and it simplifies the spatiotemporal data processing flow through standardized conversion rules, thereby improving the efficiency of user mobility behavior recognition and reducing implementation costs. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is one of the flowcharts of the flow behavior recognition method provided in the embodiments of this application;

[0026] Figure 2 This is a schematic diagram of a grouped converged bitmap sequence according to a task and time period provided in an embodiment of this application;

[0027] Figure 3 This is a schematic diagram of a user screening algorithm that uses bitmap sequences to specify a dwell time interval, provided in an embodiment of this application.

[0028] Figure 4 This is a second flowchart of a flow behavior recognition method provided in an embodiment of this application;

[0029] Figure 5 This is a schematic diagram of the operation of each module provided in an embodiment of this application;

[0030] Figure 6 This is a structural diagram of a mobile behavior recognition device provided in an embodiment of this application;

[0031] Figure 7 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0032] 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 some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0033] This application provides a method for recognizing mobile behavior. See also... Figure 1 , Figure 1 This is a flowchart of the flow behavior recognition method provided in the embodiments of this application, such as... Figure 1 As shown, it includes the following steps:

[0034] Step 101: Map the user data used to characterize the spatiotemporal behavioral features of users into a bitmap sequence according to the geospatial dimension and time resolution.

[0035] In this step, the aforementioned user data can be understood as signaling data of the user terminal, which may include spatiotemporal behavioral information such as location updates and base station connection records. The aforementioned geospatial dimension can be understood as the geospatial dimension used to analyze the aforementioned user data (such as grid, base station, district / county, city, province, etc.), and is not limited to the spatial range covered by a base station signal, such as a single base station, an administrative region, or a custom area. The aforementioned time resolution precision can be understood as the smallest granularity used for time analysis, used to characterize the user's behavior of staying in a certain space, such as 1 minute, 5 minutes, or 1 hour at the hour level. The specific value of the time resolution precision is not specifically limited in the embodiments of this application. For example, defining a 5-minute time precision means the user's behavior of staying in a certain space for less than or equal to 5 minutes, which can be rounded up or down.

[0036] The above time series can be understood as an increasing dwell time, where the dwell time equals 2. n-1 * For example, with a time resolution of 5 minutes, the time series sets are 5, 10, 20, 40 minutes, etc. The above bitmap sequence can be understood as a list of bitmaps with multiple bitmaps arranged according to time series size. Each bitmap represents a user set with a certain dwell time. In some optional implementations, the above bitmap can be a sequence of binary bits, where each bit can correspond to a dwell time interval that increases in powers of 2 (e.g., when the time resolution is 5 minutes, the bitmap sequence corresponds to 5 minutes, 10 minutes, 20 minutes, etc.), used to efficiently store user dwell behavior. 1 can be used to indicate that the dwell time is met, and 0 can be used to indicate that it is not met, etc.

[0037] For example, if a user's dwell time in a certain area is greater than or equal to the time resolution precision, the duration is converted into a binary number, and the corresponding bit in the bitmap sequence can be set to 1, for example:

[0038] When the time resolution is set to 5 minutes, the user stays for 15 minutes, which is 3 in decimal and 11 in binary. The first and second bits in the corresponding bitmap sequence are set to 1.

[0039] If the dwell time is less than the time resolution precision but greater than 0, the least bit can be set to 1 (e.g., if the user dwells for 3 minutes and the time resolution precision is set to 5 minutes, the least bit can be set to 1 and the other bits can be set to 0).

[0040] If the user does not stay, all bits are set to 0.

[0041] Step 102: Generate migration behavior data based on the user data. The migration behavior data is used to characterize the user's migration behavior from the starting point to the destination within a time period.

[0042] In this step, the aforementioned migration behavior data can be understood as a set of regional migration paths that meet the regional stay time requirements, arranged chronologically within a given time period, such as a day. This embodiment of the application does not specifically limit the aforementioned time period; it can be 1 hour, 1 day, or 1 week, etc. The aforementioned migration behavior data can include data pairs from origin to destination, such as "from base station A to base station B," "from area A to area B," etc.; it can also include all of the user's movement paths within the time period; and it can also include the user's stay time in the destination area.

[0043] Step 103: Based on the task information, process the bitmap sequence using a bitmap sequence addition algorithm and a dwell time interval filtering algorithm to obtain a processed bitmap sequence. The task information includes the source region, target region, time period, and dwell time.

[0044] In this step, the bitmap sequence addition algorithm can be understood as an algorithm for merging bitmap sequences from different regions or times, and the dwell time interval filtering algorithm can be understood as an algorithm for quickly filtering out user bitmaps that meet the specified dwell time interval.

[0045] Step 104: Based on the processed bitmap sequence and the migration behavior data, identify users who match the task information from multiple users corresponding to the user data.

[0046] In this step, it can be understood that the bitmaps that have been filtered through bitmap sequence addition algorithm and dwell time interval filtering algorithm are combined to record the user's movement path and dwell time within the time period. This is used to verify whether the user's flow direction meets the requirements of the task information, such as whether the movement path is "from the source area preset by the task information to the preset target area" and whether it conforms to the time order represented by the task information.

[0047] In this implementation, by mapping user spatiotemporal behavior data into bitmap sequences, compressed storage and rapid computation of user spatiotemporal data can be achieved. By processing bitmap sequences in parallel using bitmap sequence addition algorithms and dwell time interval filtering algorithms, the processing efficiency for large-scale multi-task data can be improved. Furthermore, by verifying user flow direction using migration behavior data, the accuracy of identifying user flow behavior under custom regions and fine duration conditions is improved, thereby enhancing the flexibility of identifying user regional flow behavior.

[0048] Optionally, user data used to characterize users' spatiotemporal behavioral features is mapped into a bitmap sequence according to geospatial dimension and temporal resolution, including:

[0049] The duration of a user's stay in the first region is determined based on the user data.

[0050] If the dwell time meets the time resolution accuracy, the dwell time is converted into a bitmap sequence of binary numbers based on the time resolution accuracy.

[0051] If the dwell time does not meet the time resolution accuracy and the dwell time is not 0, the dwell time is converted into a bitmap sequence of a binary number with the least significant bit set to 1 and the remaining bits set to 0.

[0052] When the dwell time is 0, the dwell time is converted into a bitmap sequence of all binary numbers with all bits set to 0.

[0053] In the mobility behavior identification method of this application embodiment, the first region can be understood as the geographical unit where the currently analyzed user resides. The residence time can be obtained through user data. A large amount of user data, that is, the user's original signaling data, records the continuous spatiotemporal behavioral characteristics of the user. However, the applicability of directly using the original signaling data for rapid analysis of large-scale user mobility behavior is low. In the implementation of this application, the data can be divided into a time bitmap sequence according to a pre-set geographic spatial dimension and time resolution precision for a time period (such as 1 hour, 1 day, etc.). Then, the residence time of the signaling data of the user within the corresponding time period is summarized and calculated, and the data is mapped to the corresponding geographic space and time sequence bitmap sequence data in binary form. Each bitmap represents the set of users whose residence time within the geographic space includes the time sequence, thereby realizing efficient storage and calculation of user spatiotemporal behavior data.

[0054] In some alternative implementations, it can be assumed that the user set is U, the number of users is N, and each user has a unique ID i, then:

[0055]

[0056] Assuming the geospatial dimension of the analysis is predefined as SG and the temporal resolution is TG, then any larger region or range can be represented as a set of multiple SGs. The statement "when the dwell time meets the temporal resolution, convert the dwell time into a bitmap sequence of binary numbers based on the temporal resolution" can be understood as dividing the time bitmap sequence in binary form. One time period DT_PRD consists of K time-series bitmaps, each bitmap representing a duration of 2... n-1 *TG, minuteF represents the function to retrieve minutes, which can be obtained as follows:

[0057]

[0058] For example, assuming the time period is 1 day and the time resolution TG is set to 1 minute, K=11, that is, the time interval of one day is composed of 11 user dwell time sequence bitmaps, and the duration of each bitmap is 1, 2, 4, 8, 16, 32, 64, 128, 256, 512, and 1024 minutes respectively.

[0059] Based on the user's original continuous signaling data, the dwell time of the user mapped on the geospatial SG can be aggregated. Given the set time resolution precision, it is determined that the user's time bitmap sequence contains K bitmap data points. Then, based on the user within the specified time period t... i In location space SG j Duration of stay on ST t,j,iInformation can determine the timing bitmap number of each user in the bitmap sequence, where binaryF is a decimal-to-binary function. In some optional implementations, when the dwell time does not meet the time resolution precision and the dwell time is not 0, the dwell time is converted into a bitmap sequence of binary numbers with the least significant bit set to 1 and the remaining bits set to 0; when the dwell time is 0, the dwell time is converted into a bitmap sequence of binary numbers with all bits set to 0. A bitmap sequence consisting of K user bitmaps can be represented as:

[0060]

[0061] Among them, bitmap sequences t,j,i It consists of K user bitmaps, where each bitmap represents a user that satisfies a dwell time of 2. n-1 *TG's user bitmap (e.g., TG is set to 1 minute, user dwell time ST) t,j,i For a duration of 5 minutes, the decimal-to-binary value is 101. The user's unique ID is mapped to bits 1 in the first and third bitmaps, and bits 0 in the remaining bitmaps. t,j,i It stores the SG geospatial location of the first region and the user residency information with TG time resolution.

[0062] In this embodiment, by mapping user dwell time into a binary bitmap sequence according to a preset spatiotemporal dimension, compressed storage of user behavior data is achieved. This facilitates subsequent large-scale data processing through binary bit operations. The unified bitmap data structure supports flexible configuration of multiple granular time precisions (such as minute-level, hour-level) and multiple types of geographical units (base stations, custom areas), enhancing the adaptability of the mobility behavior recognition method in this embodiment to different scenarios. The standardized conversion rules simplify the spatiotemporal data processing flow, thereby improving the efficiency of user mobility behavior recognition and reducing implementation costs.

[0063] Optionally, generating migration behavior data based on the user data includes:

[0064] Based on the user data, obtain the movement path data from the origin to the destination for each user within the time period and the dwell time of the user at the destination.

[0065] Among them, for multiple moves with the same starting point and destination, only the migration behavior data of the user with the longest stay at the destination is retained.

[0066] In the migration behavior recognition method of this application embodiment, since users may travel back and forth between the same origin and destination multiple times within a time period (such as commuters traveling between home and work daily), directly using the original path data would lead to redundant calculations and affect the accuracy of the analysis. The implementation method of this application can optimize the migration behavior data structure by deduplication and filtering based on the longest dwell time.

[0067] In some alternative implementations, the aforementioned dwell time can be obtained in the following ways:

[0068] Based on user data, specifically the user's base station signaling data associated with the basic spatial unit mapping dimension table, a zipper data of basic spatial units arranged chronologically within the user's corresponding time period can be constructed. The time a user enters the next basic spatial unit is recorded as the departure time of the previous basic spatial unit, thus obtaining the user's dwell time. The user's movement area can be within the basic spatial unit range, meaning user signaling data, such as base station handover records, is bound to the basic spatial units. For example, entering the coverage area of ​​base station A is recorded as "dwelling SU1". Only when the user's movement path is broken down into an ordered sequence of basic spatial units can a structured "start-to-end" pair be generated, thus forming regional movement behavior. The aforementioned basic spatial unit can be understood as an indivisible geographic spatial unit with clear spatial boundaries and a unique identifier, divided according to business needs. For example: in the communications field: the coverage area of ​​a single base station; in the urban planning field: a grid; in the administrative field: the smallest administrative unit such as a community or street. Each unit can correspond to an independent bitmap sequence. For example, the bitmap of SU1 records the duration of a user's stay at the corresponding base station. The duration filtering of mobile behavior, such as "staying in SU1 for ≥30 minutes", can also be achieved through the bitmap sequence of this unit.

[0069] For example, suppose we assume a set of basic spatial unit data for users and a corresponding set of dwell times, representing the user's data within a time period t. i The N basic space units SU passed through j And the corresponding stay time is T j .

[0070] SU ijt ={SU i1t ,SU i2t ,...,SU iNt}T ijt ={T i1t ,T i2t ,...,T iNt}

[0071] The aforementioned "for multiple moves with the same origin and destination, only retain the migration behavior data with the longest stay at the destination" can be understood as constructing migration behavior data based on pairwise permutations and combinations of user basic spatial units. When identical combinations exist, the event with the longest stay at the migration destination is retained, ultimately obtaining the user's basic spatial unit migration details. In some optional implementations, the specific implementation methods can be as follows:

[0072] Generate all basic spatial unit pair combinations (SU) iat ,SU ibt ,T ibt M represents the value from unit SU. iat To unit SU ibt The flow behavior, and the basic spatial unit SU at the destination. ibt The length of stay is T ibt

[0073] M={(SU iat ,SU ibt ,T ibt )|1≤a <b≤N}

[0074] For each basic spatial unit pair (SU) iat ,SU ibt ,T ibt When there are repeated migration behaviors (i.e., the origin and destination of the migration are the same), the SelectMigrateF method selects the pair of users who stay at the destination for the longest time, and finally obtains the user. i The set of migration details D within the corresponding time period t:

[0075]

[0076] Performing the above steps on all users will yield migration behavior data consisting of a complete set of detailed user records.

[0077] In this implementation, the integrity of migration behavior data is ensured by acquiring all user movement paths and dwell times; data deduplication and simplification are achieved by retaining only the longest dwell time record for multiple movements from the same origin to the destination, reducing redundant storage and improving computational efficiency; by focusing on the core behavior of the longest dwell time, interference from short stays is eliminated, the main flow characteristics of users are strengthened, and the accuracy and effectiveness of user flow behavior analysis are improved.

[0078] Optionally, the bitmap sequence addition algorithm is used to merge bitmap sequences from different regions or times by bitwise OR, XOR, and AND operations.

[0079] In the mobility behavior recognition method of this application embodiment, when analyzing user mobility behavior over a larger area or time range, the regional population mobility task can be broken down into fine-grained subtasks for parallel computation. Furthermore, the selected area can be split and calculated according to time periods and geographic spatial combinations. Therefore, the output intermediate bitmap sequence data needs to be added together. The aforementioned time bitmap sequence addition algorithm can be understood as an algorithm for merging different bitmap sequences.

[0080] Optionally, the bitmap sequence addition algorithm includes the following steps:

[0081] Initialize the result bitmap sequence, carry bitmap, and current result bitmap;

[0082] Perform a bitwise XOR operation on the first bit of the bitmap sequence to generate the current result bitmap and store it in the result bitmap sequence. At the same time, generate a carry bitmap by performing a bitwise AND operation.

[0083] For the bitmaps from the 2nd to the Kth bitmap in the bitmap sequence, perform an XOR operation bit by bit to generate the current result bitmap, update the carry bitmap through AND and OR operations and pass it to the next bit;

[0084] After the traversal is completed, the resulting bitmap sequence is the user summary bitmap sequence data of the flow region after multiple time bitmap sequences are aggregated and compressed.

[0085] In this embodiment, the bitmap sequence addition algorithm can be:

[0086] BitmapsAddF(bitmaps1,bitmaps2);

[0087] Specifically, assuming bitmaps 1,1,i and bitmaps 1,2,i Given two bitmap sequences (SG1 and SG2) from different geographic spaces within the same time period t=1 that need to be merged and added, output the resulting bitmap sequence rbitmaps. Given:

[0088]

[0089] Furthermore, the initialization of the result bitmap sequence rbitmaps, the carry bitmap mbitmap, and the result bitmap kbitmap can be represented as:

[0090]

[0091] mbitmap=(0,0,…,0)kbitmap=(0,0,…,0)

[0092] With i = 1 initially, the bitmap can be traversed from 1 to K based on i to perform bit OR (∨), AND (∧), and XOR (⊕) operations to generate a result bitmap kbitmap, which is stored in the result bitmap sequence rbitmaps. The carry bitmap mbitmap passes the carry to the next result bitmap. After the traversal is completed, the merged result bitmap sequence rbitmaps is generated.

[0093] When i = 1, for bitmap1 1,1,i and bitmap1 1,2,i Perform a bitwise XOR operation to generate a kbitmap and store it in rbitmaps. Simultaneously, perform a bitwise AND operation to generate a carry bitmap mbitmap, passing the carry to the next result bitmap. Specifically, this can be represented as:

[0094]

[0095] mbitmap = bitmap1 1,1,i ∧bitmap1 1,2,i

[0096] When i≠1, a bitwise XOR operation needs to be performed with the carry bitmap mbitmap' generated in the previous traversal to update the carry bitmap mbitmap:

[0097]

[0098] mbitmap=(mbitmap'∧(bitmapk 1,1,i ∨bitmapk 1,2,i ))∨(bitmapk 1,1,i ∧bitmapk 1,2,i )

[0099] In this implementation, unique resident users in different regions or time bitmap sequences can be identified through bitwise XOR, AND, and OR operations. Overlapping resident users are merged and carry is generated, achieving efficient and lossless merging of multi-source bitmap sequences. While reducing redundant data storage, it accurately reflects the superposition effect of user residence time, providing underlying data support for large-scale regional flow analysis.

[0100] Optionally, the dwell time interval filtering algorithm includes hierarchical bit operations and recursive binary search, and the dwell time interval filtering algorithm is used to filter out bitmap sequences that meet the preset duration requirements.

[0101] In the flow behavior recognition method of this application embodiment, the screening process is optimized to logarithmic complexity through hierarchical bitwise operations and recursive binary search. The aforementioned hierarchical bitwise operations can be understood as dividing the binary bitmap into different levels according to bit weights (e.g., 5 minutes, 10 minutes, 20 minutes), and directly screening for matching level combinations through bitwise operations. For example, taking a time greater than or equal to 30 minutes as an example:

[0102] First, the key layer can be extracted, and the threshold of 30 minutes needs to cover the 20-minute position (2). 4 ×5=20) and 10-minute position (2 3 ×5=10);

[0103] Next, perform bitwise verification to check if the 3rd bit (20 minutes) and 2nd bit (10 minutes) of the user bitmap are 1; if the 3rd bit of user A is 1 but the 2nd bit is 0, then it does not meet the requirement; if the 3rd bit and the 2nd bit of user B are both 1, then it meets the requirement.

[0104] The above recursive binary search method can be understood as splitting a bitmap sequence in half hierarchically, and then quickly locating the sub-intervals that meet the conditions through recursion. For example, assuming a time resolution precision TG = 5 minutes, if filtering users in the "30-60 minute" interval:

[0105] The first binary search takes the median value of the lower limit of the target interval (30 minutes), which corresponds to the third position of the bitmap, 20 minutes; the higher position (≥20 minutes) and lower position (<20 minutes) are processed recursively.

[0106] When processing the high-level interval, filter the bitmaps with a time greater than or equal to 20 minutes, perform further bisection, take the median value of 40 minutes, which corresponds to the fourth position of the bitmap, and recursively process "20-40 minutes" and "≥40 minutes" respectively.

[0107] When processing the lower range, filter bitmaps with "<20 minutes" (the 3rd bit is 0) and exclude them directly, since the lower limit is 30 minutes.

[0108] Optionally, the hierarchical bitwise operation includes a user bitwise hierarchy calculation method, which is used to merge or remove users whose corresponding bitmap bit value is 1 within a specified hierarchy range in the bitmap sequence, and remove users whose bitmap bit value is 1 and is higher than the first hierarchy; the recursive binary search method includes a user segmented bitmap calculation method, which is used to divide the bitmap sublist into two parts according to the base bit hierarchy, until the number of sublists satisfies 2 to the power of the base bit hierarchy, at which point the user bitwise hierarchy calculation method is called.

[0109] In this implementation, the user filtering algorithm for a specified dwell time interval of a user in the bitmap sequence can be defined as TimeRangeSelectF(bitmaps,valueStart,valueEnd), where bitmaps is the bitmap sequence to be filtered.

[0110] in,

[0111]

[0112] valueStart and valueEnd represent the users to be filtered based on the dwell time interval [valueStart, valueEnd]. The specific algorithm for filtering users based on dwell time intervals can be as follows:

[0113] Determine the binary dwell time interval [vmin, vmax] to be calculated based on valueStart, valueEnd, and time resolution precision TG. Let:

[0114]

[0115] The initialization results include the user bitmap rbitmap, the bitmap constant MAX_BITMAP, the bit level variable highLevel = -1, and an empty VBS bitmap list, as shown below:

[0116] rbitmap=(0,0,…,0)MAX_BITMAP=(1,1,…,1)

[0117] Furthermore, a user bit hierarchy calculation method, ValueLevelBitmapF(bitmaps, vbitmap, baseLevel, highLevel), can be defined. This method represents the ability to directly merge or remove all users whose bitmap values ​​within the specified level range [baseLevel, highLevel] of the bitmap sequence `bitmaps` are 1, and whose higher-level bitmap values ​​are 0. Here, `vbitmap` is a K-bit binary number, specifically, as follows:

[0118] Initial result bitmap kbitmap = MAX_BITMAP;

[0119] Initially, i = baseLevel. Iterate through each level of [baseLevel, highLevel], merging or removing bitmaps corresponding to the i-th bitmap. i All users:

[0120] ① When bit i of bitmap vbitmap is not 1, execute kbitmap = kbitmap ∧ (bitmap i ⊕MAX_BITMAP)

[0121] ② Otherwise, execute kbitmap = kbitmap ∧ bitmap i

[0122] Initially, i = highLevel + 1. Iterate through all high-level bitmaps from highLevel + 1 to K and beyond. Remove all users whose bit is 1 in the i-th bitmap corresponding to the high-level bitmap vbitmap in the kbitmap result bitmap. Specifically, it can be done as follows:

[0123] kbitmap = kbitmap ∧ (bitmaps i ⊕MAX_BITMAP);

[0124] Returns the final bitmap result (kbitmap).

[0125] Furthermore, a user-defined segmented bitmap calculation method, ValueRangeBitmapF(bitmaps, vbitmaps, baseLevel, highLevel), can be defined. This method indicates that when the number of bitmaps in vbitmaps does not meet the requirements, vbitmaps can perform binary segmentation based on whether the baseLevel bit is 0 or 1, until the number of bitmaps after binary segmentation equals 2. baseLevel If the value is not in the specified range, call ValueLevelBitmapF to calculate the user bitmap; otherwise, decrement baseLevel by 1 and continue with the binary segmentation process. The steps are as follows:

[0126] When the baselevel is 0, ValueLevelBitmapF is called directly to return the user bitmap result;

[0127] When the base level is not equal to 0, vbitmaps performs binary segmentation based on whether the base level bit is 0 or 1, resulting in two sublists of bitmaps, vbs0 and vbs1.

[0128] Furthermore, the following judgments can be made on the two bitmap sublists, vbs0 and vbs1, respectively:

[0129] ① When the number of bitmap sublists is equal to 2 baseLevel When calling ValueLevelBitmap(bitmaps, vbs) i0 (,baseLevel,highLevel), where vbs i0This represents the first element of the bitmap sublist, yielding the user's result bitmap kb1 or kb0.

[0130] ② Otherwise, if the number of bitmaps is greater than 0, decrement baseLevel by 1 and recursively call ValueRangeBitmapF(bitmaps, vbs) i Continue the binary segmentation process (baseLevel-1, highLevel), repeat the above steps, and finally obtain the user result bitmap kb1 or kb0.

[0131] Returns the final bitmap result: kbitmap = kb1 ∨ kb0;

[0132] Furthermore, v can be iterated from vmin to vmax:

[0133] Convert v to binary number vb, and let vb = binaryF(v) = (b1, b2, ..., bb) k () is a K-bit binary number, where b is the number of bits. k It can take the value 0 or 1;

[0134] When v = vmin, the highest bit index of vb (where vb is 1) is assigned to highLevel for updating. Otherwise, when the index is greater than the highLevel value, the corresponding methods in steps 4.3 and 4.4 are called according to the number of bits in the vbs bitmap, and the rbitmap result bitmap is updated. Specifically, it can be done as follows:

[0135] When the number of VBS bitmaps is equal to 2 highLevel At that time, you can execute it using the first element of VBScript, vbs0:

[0136] rbitmap=rbitmap∨ValueLevelBitmapF(bitmaps,vbs0,highLevel,highLevel)

[0137] Furthermore, when the number of VBS bitmaps is not equal to 2 highLevel At that time, execute:

[0138] rbitmap=rbitmap∨ValueRangeBitmapF(bitmaps,vbs,highLevel-1,highLevel)

[0139] Clear the VBScript and update HighLevel.

[0140] Add the binary number to the VBS bitmap list and update it;

[0141] When v reaches the last value vmax, re-evaluate the number of vbs bitmaps and execute the above steps to update the rbitmap result bitmap.

[0142] Returns the final user result bitmap rbitmap.

[0143] In this implementation, hierarchical bitwise operations and recursive binary search are used to achieve fast and accurate filtering of users with specified dwell time intervals in the bitmap sequence, improving the filtering efficiency from linear to logarithmic. At the same time, it supports flexible configuration of arbitrary time intervals, improving the analysis efficiency and adaptability for analyzing large-scale spatiotemporal data.

[0144] Optionally, identifying users matching the task information from multiple users corresponding to the user data based on the processed bitmap sequence and the migration behavior data includes:

[0145] Based on the migration behavior data, determine whether the processed bitmap sequence meets a first condition, the first condition including: the user moves from the source region to the target region in chronological order;

[0146] Bitmap sequences that do not meet the first condition are removed from the processed bitmap sequences;

[0147] Users matching the task information are determined based on the bitmap sequence after the elimination process.

[0148] In the mobility behavior identification method of this application embodiment, after filtering out users who meet the dwell time through bitmap sequence, the consistency of the time sequence and direction of their movement path is verified by combining migration behavior data, which can avoid misjudging users who "stay but do not move" or "move in the opposite direction" as valid mobility users.

[0149] Optionally, the step of processing the bitmap sequence based on task information using a bitmap sequence addition algorithm and a dwell time interval filtering algorithm to obtain a processed bitmap sequence includes:

[0150] The task of identifying multiple regions of movement is divided into multiple sub-tasks according to the source region, target region and time period. The number of groups is dynamically determined based on the total number of sub-tasks, the maximum number of groups and the minimum number of tasks in each group, and the groups are dynamically adjusted according to the sub-task load of each group.

[0151] The split subtasks are assigned to each group for parallel computation. Each group aggregates the bitmap sequences corresponding to its subtasks using the bitmap sequence addition algorithm and filters out user bitmaps that meet the dwell time requirements in the task information using the dwell time interval filtering algorithm.

[0152] The calculation results of each group are reduced and summarized to obtain the processed bitmap sequence.

[0153] In this implementation, the aforementioned sub-tasks can be understood as the refined tasks of the "regional flow identification task" after being split into "source region, target region, and time period"; the aforementioned maximum number of groups (maxPa) can be understood as the upper limit of the maximum number of groups allowed for calculation; the minimum number of tasks per group (minTask) can be understood as the minimum number of sub-tasks allocated to a single group; the aforementioned reduction and aggregation can be understood as the process of merging the intermediate calculation results of each group into the final result, that is, reducing and aggregating the intermediate bitmap sequence results output by each group according to the task and time period to obtain the final bitmap sequence data.

[0154] In some optional implementations, the task can be first broken down into fine-grained subtasks for parallel computation. The selected source region is first divided into tasks based on time period and geospatial dimension. Subtasks are sorted according to time period and assigned to computation groups according to a distribution rule. This distribution rule assumes a total number of subtasks of T, a maximum group size of maxPa, and a minimum number of tasks per group of minTask. The number of groups P can be calculated using the following formula:

[0155]

[0156] When the number of subtasks in the group is T p When adding new groups dynamically, the following rules apply:

[0157]

[0158] This distribution rule can dynamically adjust the number of groups, make reasonable use of computing resources, and avoid the processing performance bottleneck caused by too many tasks in a single group.

[0159] Furthermore, querying the user dwell time bitmap sequence data yields the corresponding bitmap sequence for each subtask group. These bitmap sequences can then be grouped and aggregated by task and time period. The intermediate bitmap sequence results output from each group are then reduced and aggregated according to task and time period to obtain the final bitmap sequence data for each task within its corresponding time period. This can be referenced. Figure 2 .

[0160] Furthermore, you can refer to Figure 3For each time period of the task, the user bitmap sequence data is processed according to the dwell time requirement. A user filtering algorithm is used to match the object index bitmap within a specified value range based on the dwell time interval of the bitmap sequence, thereby identifying the user bitmap set bitmapA that meets the requirements from the source region. Similarly, the same processing method is applied to the target region to obtain the user bitmap set bitmapB that meets the dwell time requirement in the target region. A bitwise AND operation is then performed on the two user bitmap sets to obtain the user bitmap set bitmapC.

[0161] bitmapC = bitmapA ∧ bitmapB

[0162] Based on the user migration data of the basic spatial unit, the users who flowed from the source region to the target region in the user bitmap set bitmapC are identified, and finally the result data of the personnel in the regional flow task are identified.

[0163] In some optional implementations, when initiating the calculation of spatiotemporal population flow tasks, the set of users who flow from the source region to the target region and meet a certain stay duration within a single or multiple consecutive time periods is identified. A method of fine-grained automatic sub-task decomposition for parallel computation is employed, combined with bitmap operations to efficiently identify the complete process of regional mobile users meeting the stay behavior conditions. This can be achieved as follows: Figure 4 As shown. Specifically, after the task is started, the source or target area is broken down into sub-tasks and assigned to groups in a fine-grained manner; further, within each group, bitmap sequence data is aggregated according to task and time period based on the bitmap sequence data; further, the bitmap sequence data is aggregated again by grouping indirect tasks and time periods; further, the set of user bitmaps that meet the requirements within each time period is selected according to the task area residence time requirements; further, the user bitmaps of the task source area and target area are intersected to obtain the intersection user bitmap C; further, user bitmap C can confirm the user area flow direction based on the migration data of the basic spatial units; further, the results of the area flow salvage task can be output; finally, the task ends.

[0164] In the mobility behavior recognition method of this application embodiment, a dual mechanism of "dwelling time filtering" and "mobility direction verification" ensures that the recognition results not only meet the spatiotemporal dwelling conditions but also follow the actual movement logic. Combining the calculation of binary bitmaps with the temporal verification of user movement paths improves the accuracy of identifying personnel on mobile tasks within a region.

[0165] In some alternative implementations, such as Figure 5The system can construct user dwell time bitmap sequence data and basic spatial unit migration data in the data processing module; it can construct algorithms for adding bitmap sequences and filtering users within dwell time intervals in the bitmap sequence analysis algorithm module; and it can realize efficient calculation of regional population flow tasks and output recognition results in the multi-task management allocation calculation module.

[0166] See Figure 6 , Figure 6 This is a structural diagram of a mobile behavior recognition device provided in another embodiment of this application.

[0167] like Figure 6 As shown, the mobile behavior recognition device 600 includes:

[0168] The mapping module 601 is used to map user data that characterizes the spatiotemporal behavioral features of users into a bitmap sequence according to the geospatial dimension and time resolution.

[0169] The generation module 602 is used to generate migration behavior data based on the user data, wherein the migration behavior data is used to characterize the user's migration behavior from the starting point to the destination within a time period.

[0170] Processing module 603 is used to process the bitmap sequence based on task information using a bitmap sequence addition algorithm and a dwell time interval filtering algorithm to obtain a processed bitmap sequence. The task information includes source region, target region, time period and dwell time.

[0171] The identification module 604 is used to identify users who match the task information from multiple users corresponding to the user data based on the processed bitmap sequence and the migration behavior data.

[0172] Optionally, the mapping module 601 can also be used for:

[0173] The duration of a user's stay in the first region is determined based on the user data.

[0174] If the dwell time meets the time resolution accuracy, the dwell time is converted into a bitmap sequence of binary numbers based on the time resolution accuracy.

[0175] If the dwell time does not meet the time resolution accuracy and the dwell time is not 0, the dwell time is converted into a bitmap sequence of a binary number with the least significant bit set to 1 and the remaining bits set to 0.

[0176] When the dwell time is 0, the dwell time is converted into a bitmap sequence of all binary numbers with all bits set to 0.

[0177] Optionally, the generation module 602 can also be used for:

[0178] Based on the user data, obtain the movement path data from the origin to the destination for each user within the time period and the dwell time of the user at the destination.

[0179] Among them, for multiple moves with the same starting point and destination, only the migration behavior data of the user with the longest stay at the destination is retained.

[0180] Optionally, the bitmap sequence addition algorithm is used to merge bitmap sequences from different regions or times by bitwise OR, XOR, and AND operations.

[0181] Optionally, the dwell time interval filtering algorithm includes hierarchical bit operations and recursive binary search, and the dwell time interval filtering algorithm is used to filter out bitmap sequences that meet the preset duration requirements.

[0182] Optionally, the identification module 604 can also be used for:

[0183] Based on the migration behavior data, determine whether the processed bitmap sequence meets a first condition, the first condition including: the user moves from the source region to the target region in chronological order;

[0184] Bitmap sequences that do not meet the first condition are removed from the processed bitmap sequences;

[0185] Users matching the task information are determined based on the bitmap sequence after the elimination process.

[0186] For details, see Figure 7 As shown in the figure, this application embodiment also provides an electronic device, including a bus 701, a transceiver 702, an antenna 703, a bus interface 704, a processor 705, and a memory 706.

[0187] Processor 705, used for:

[0188] User data used to characterize users’ spatiotemporal behavioral features are mapped into bitmap sequences according to geospatial dimension and temporal resolution.

[0189] Migration behavior data is generated based on the user data, and the migration behavior data is used to characterize the user's migration behavior from the starting point to the destination within a time period.

[0190] Based on the task information, the bitmap sequence is processed by a bitmap sequence addition algorithm and a dwell time interval filtering algorithm to obtain a processed bitmap sequence. The task information includes the source region, target region, time period, and dwell time.

[0191] Based on the processed bitmap sequence and the migration behavior data, users matching the task information are identified from multiple users corresponding to the user data.

[0192] exist Figure 7 In this document, a bus architecture (represented by bus 701) is used. Bus 701 can include any number of interconnected buses and bridges, linking various circuits including one or more processors represented by processor 705 and memory represented by memory 706. Bus 701 can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. Bus interface 704 provides an interface between bus 701 and transceiver 702. Transceiver 702 can be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by processor 705 is transmitted over a wireless medium via antenna 703, which further receives data and transmits data to processor 705.

[0193] Processor 705 manages bus 701 and general processing, and also provides various functions, including timing, peripheral interface, voltage regulation, power management, and other control functions. Memory 706 can be used to store data used by processor 705 during operation.

[0194] Optionally, the processor 705 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a complex programmable logic device (CPLD).

[0195] Optionally, the monitoring information includes distance information and speed information corresponding to the at least one target object, and the processor 705 is specifically used for:

[0196] The duration of a user's stay in the first region is determined based on the user data.

[0197] If the dwell time meets the time resolution accuracy, the dwell time is converted into a bitmap sequence of binary numbers based on the time resolution accuracy.

[0198] If the dwell time does not meet the time resolution accuracy and the dwell time is not 0, the dwell time is converted into a bitmap sequence of a binary number with the least significant bit set to 1 and the remaining bits set to 0.

[0199] When the dwell time is 0, the dwell time is converted into a bitmap sequence of all binary numbers with all bits set to 0.

[0200] Optionally, the bitmap sequence addition algorithm is used to merge bitmap sequences from different regions or times by bitwise OR, XOR, and AND operations.

[0201] Optionally, the dwell time interval filtering algorithm includes hierarchical bit operations and recursive binary search, and the dwell time interval filtering algorithm is used to filter out bitmap sequences that meet the preset duration requirements.

[0202] Optionally, the processor 705 is specifically used for:

[0203] Based on the migration behavior data, determine whether the processed bitmap sequence meets a first condition, the first condition including: the user moves from the source region to the target region in chronological order;

[0204] Bitmap sequences that do not meet the first condition are removed from the processed bitmap sequences;

[0205] Users matching the task information are determined based on the bitmap sequence after the elimination process.

[0206] It should be noted that the electronic device provided in this application embodiment is a device capable of executing the above-described flow behavior recognition method. Therefore, all implementation methods in the above-described flow behavior recognition method embodiments are applicable to this electronic device and can achieve the same or similar beneficial effects. To avoid repetition, this embodiment will not elaborate further.

[0207] This invention also provides an electronic device, including: a processor, a memory, and a program stored in the memory and executable on the processor. When the program is executed by the processor, it implements the various processes of the above-described flow behavior recognition method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0208] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the various processes of the above-described flow behavior recognition method embodiments and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0209] This application also provides a computer program product, including computer instructions. When executed by a processor, the computer instructions implement the various processes of the above-described flow behavior recognition method embodiments and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0210] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0211] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0212] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for recognizing mobile behavior, characterized in that, The method includes: User data used to characterize users’ spatiotemporal behavioral features are mapped into bitmap sequences according to geospatial dimension and temporal resolution. Migration behavior data is generated based on the user data, and the migration behavior data is used to characterize the user's migration behavior from the starting point to the destination within a time period. Based on the task information, the bitmap sequence is processed by a bitmap sequence addition algorithm and a dwell time interval filtering algorithm to obtain a processed bitmap sequence. The task information includes the source region, target region, time period, and dwell time. Based on the processed bitmap sequence and the migration behavior data, identify users who match the task information from multiple users corresponding to the user data; The bitmap sequence addition algorithm includes the following steps: Initialize the result bitmap sequence, carry bitmap, and current result bitmap; Perform a bitwise XOR operation on the first bit of the bitmap sequence to generate the current result bitmap and store it in the result bitmap sequence. At the same time, generate a carry bitmap by performing a bitwise AND operation. For the bitmaps from the 2nd to the Kth bitmap in the bitmap sequence, perform an XOR operation bit by bit to generate the current result bitmap, update the carry bitmap through AND and OR operations and pass it to the next bit; After the traversal is completed, the resulting bitmap sequence is the user summary bitmap sequence data of the flow region after multiple time bitmap sequences are aggregated and compressed; The dwell time interval filtering algorithm includes hierarchical bit operations and recursive binary search. The dwell time interval filtering algorithm is used to filter out bitmap sequences that meet the preset duration requirements. The hierarchical bitwise operations include a user bitwise hierarchy calculation method, which is used to merge or remove users whose corresponding bitmap bit value is 1 within a specified hierarchy range in the bitmap sequence, and remove users whose bitmap bit value is 1 and is higher than the first hierarchy. The recursive binary search method includes a user segmented bitmap calculation method, which is used to divide the bitmap sublist into two parts according to the base bit hierarchy, until the number of sublists satisfies 2 to the power of the base bit hierarchy, at which point the user bitwise hierarchy calculation method is called.

2. The method according to claim 1, characterized in that, User data used to characterize users' spatiotemporal behavioral features is mapped into a bitmap sequence according to geospatial dimension and temporal resolution, including: The duration of a user's stay in the first region is determined based on the user data. If the dwell time meets the time resolution accuracy, the dwell time is converted into a bitmap sequence of binary numbers based on the time resolution accuracy. If the dwell time does not meet the time resolution accuracy and the dwell time is not 0, the dwell time is converted into a bitmap sequence of a binary number with the least significant bit set to 1 and the remaining bits set to 0. When the dwell time is 0, the dwell time is converted into a bitmap sequence of all binary numbers with all bits set to 0.

3. The method according to claim 1, characterized in that, The generation of migration behavior data based on the user data includes: Based on the user data, obtain the movement path data from the origin to the destination for each user within the time period and the dwell time of the user at the destination. Among them, for multiple moves with the same starting point and destination, only the migration behavior data of the user with the longest stay at the destination is retained.

4. The method according to any one of claims 1 to 3, characterized in that, The bitmap sequence addition algorithm is used to merge bitmap sequences from different regions or times by bitwise OR, XOR, and AND operations.

5. The method according to claim 1, characterized in that, The step of identifying users matching the task information from multiple users corresponding to the user data based on the processed bitmap sequence and the migration behavior data includes: Based on the migration behavior data, determine whether the processed bitmap sequence meets a first condition, the first condition including: the user moves from the source region to the target region in chronological order; Bitmap sequences that do not meet the first condition are removed from the processed bitmap sequences; Users matching the task information are determined based on the bitmap sequence after the elimination process.

6. The method according to claim 1, characterized in that, The bitmap sequence is processed based on task information using a bitmap sequence addition algorithm and a dwell time interval filtering algorithm to obtain a processed bitmap sequence, including: The task of identifying multiple regions of movement is divided into multiple sub-tasks according to the source region, target region and time period. The number of groups is dynamically determined based on the total number of sub-tasks, the maximum number of groups and the minimum number of tasks in each group, and the groups are dynamically adjusted according to the sub-task load of each group. The split subtasks are assigned to each group for parallel computation. Each group aggregates the bitmap sequences corresponding to its subtasks using the bitmap sequence addition algorithm and filters out user bitmaps that meet the dwell time requirements in the task information using the dwell time interval filtering algorithm. The calculation results of each group are reduced and summarized to obtain the processed bitmap sequence.

7. A mobile behavior recognition device, characterized in that, The device includes: The mapping module is used to map user data that characterizes the spatiotemporal behavior of users into a bitmap sequence according to the geospatial dimension and time resolution. A generation module is used to generate migration behavior data based on the user data, wherein the migration behavior data is used to characterize the user's migration behavior from the starting point to the destination within a time period; The processing module is used to process the bitmap sequence based on task information using a bitmap sequence addition algorithm and a dwell time interval filtering algorithm to obtain a processed bitmap sequence. The task information includes the source region, the target region, the time period, and the dwell time. The identification module is used to identify users who match the task information from multiple users corresponding to the user data based on the processed bitmap sequence and the migration behavior data; The processing module is used for: Initialize the result bitmap sequence, carry bitmap, and current result bitmap; Perform a bitwise XOR operation on the first bit of the bitmap sequence to generate the current result bitmap and store it in the result bitmap sequence. At the same time, generate a carry bitmap by performing a bitwise AND operation. For the bitmaps from the 2nd to the Kth bitmap in the bitmap sequence, perform an XOR operation bit by bit to generate the current result bitmap, update the carry bitmap through AND and OR operations and pass it to the next bit; After the traversal is completed, the resulting bitmap sequence is the user summary bitmap sequence data of the flow region after multiple time bitmap sequences are aggregated and compressed; The dwell time interval filtering algorithm includes hierarchical bit operations and recursive binary search. The dwell time interval filtering algorithm is used to filter out bitmap sequences that meet the preset duration requirements. The hierarchical bitwise operations include a user bitwise hierarchy calculation method, which is used to merge or remove users whose corresponding bitmap bit value is 1 within a specified hierarchy range in the bitmap sequence, and remove users whose bitmap bit value is 1 and is higher than the first hierarchy. The recursive binary search method includes a user segmented bitmap calculation method, which is used to divide the bitmap sublist into two parts according to the base bit hierarchy, until the number of sublists satisfies 2 to the power of the base bit hierarchy, at which point the user bitwise hierarchy calculation method is called.

8. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the flow behavior recognition method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the flow behavior recognition method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, It includes computer instructions that, when executed by a processor, implement the steps of the flow behavior recognition method as described in any one of claims 1 to 6.

Images