A method, system, device, and storage medium for recognizing common school behavior.

Abstract

This invention discloses a method, system, device, and storage medium for identifying commuting behavior. The method includes: acquiring mobile phone signaling data and school AOI data, and preprocessing the mobile phone signaling data and school AOI data; determining, based on the mobile phone signaling data and school AOI data, fixed-station objects based on the school and the corresponding types of fixed-station objects; the type being the type corresponding to the geographical grid visited or resided in by the object; and, based on the type and the object's basic information, determining the object's population category, the start point, and the end point of commuting behavior based on a preset classification strategy. This application combines large-scale mobile phone signaling data and school AOI data to determine the object's stationing type; further, it allows for fine-grained segmentation of population categories and commuting behavior, which helps improve the precision of population identification. This method can be widely applied in the field of urban transportation planning technology.

Description

Technical Field

[0001] This invention relates to the field of urban traffic planning technology, and in particular to a method, system, device and storage medium for recognizing commuting behavior. Background Technology

[0002] With the development of big data and mobile communication technologies, mobile signaling data has become an important resource for studying population flow, urban activities and travel patterns. Due to its advantages such as wide coverage, strong real-time performance and relatively low cost, it is widely used in fields such as transportation planning and urban planning.

[0003] Among related technologies, some solutions propose methods for identifying population movement based on mobile phone signaling data, such as identifying unemployed individuals, extracting urban residents' travel chains, identifying population migration, and characterizing travelers' work-residence locations and travel chains. These methods typically involve steps such as data collection, feature extraction, and cluster analysis, using machine learning and data mining techniques to process and analyze the data. However, these technical solutions often focus on a single data source or a single application scenario, lacking comprehensive utilization of multi-source data. Furthermore, in solutions targeting school-residence identification, existing technologies mainly rely on small-data methods such as questionnaires for identification, failing to perform large-scale, wide-ranging identification and analysis, and lacking detailed categorization of school visitors. Summary of the Invention

[0004] The purpose of this invention is to provide a sophisticated method, system, device, and storage medium for recognizing common school behavior.

[0005] On the one hand, this application provides a method for identifying commuting behavior. The method includes: acquiring mobile phone signaling data and school AOI data, and preprocessing the mobile phone signaling data and the school AOI data; determining, based on the mobile phone signaling data and the school AOI data, a fixed-resident object at the school and the type corresponding to the fixed-resident object; the type being the type corresponding to the geographical grid visited or resided in by the object; and, based on the type and the object's basic information, determining the object's population category, the start point, and the end point of the commuting behavior based on a preset classification strategy. This application combines large-scale mobile phone signaling data and school AOI data to determine the object's residence type; thereby, it refines the population category and commuting behavior, which helps improve the precision of population identification.

[0006] Optionally, based on the mobile phone signaling data and the school AOI data, the fixed-resident objects based on the school and their corresponding types are determined, including:

[0007] A tiered buffer zone is set for the school; the tiered buffer zone is used to ensure that the area of ​​the school is greater than the extraction accuracy of mobile signaling.

[0008] Create a geographic grid based on each of the schools and the hierarchical buffer zones;

[0009] By traversing the mobile phone signaling data, the signaling trajectory points are mapped to the geographic grid to obtain the access data of each object to the school;

[0010] Based on the access data, determine the fixed resident object and its type.

[0011] Optionally, determining the fixed-resident object and the type based on the access data includes:

[0012] Based on the access data, within a preset time period, objects whose access and stay frequency to the school is greater than or equal to the preset frequency are selected and designated as fixed-stay objects.

[0013] Based on the access data of the fixed resident objects to the school, it is determined whether the first grid, second grid and third grid corresponding to each object are located within the school area; the first grid, second grid and third grid are the resident characteristics of the fixed resident objects on weekdays and non-weekdays.

[0014] Optionally, based on the type and the basic information of the object, and using a preset classification strategy, the population category, the starting point and the ending point of the commuting behavior of the object are determined, including:

[0015] If the first grid and the second grid are the same, both the first grid and the second grid belong to the school area, and the age of the fixed resident is greater than or equal to the first preset age, then the fixed resident is determined to be a boarding student.

[0016] If the first grid belongs to the school area, the second grid does not belong to the school area, and the age of the fixed resident is greater than or equal to the first preset age, the fixed resident is determined to be a day student.

[0017] If the first grid belongs to the school area, the fixed resident object uses at least one teacher-related application, and the fixed resident object's age is greater than or equal to a second preset age, then the fixed resident object is determined to be a teacher.

[0018] If the first grid does not belong to the school area, the time the fixed resident object visits the school is within a preset time period, and the age of the fixed resident object is greater than or equal to a second preset age, the fixed resident object is determined to be a parent; the first grid is the grid where the fixed resident object stays for the longest time during the daytime on weekdays, and the second grid is the grid where the fixed resident object stays for the longest time at night on weekdays.

[0019] Optionally, the identification method of this application further includes:

[0020] The area where the boarding student's third grid is located is determined as their residence, the area where the day student's second or third grid is located is determined as their residence, the area where the teacher's second or third grid is located is determined as their residence, the area where the parent's first grid is located is determined as their workplace, and the area where the parent's second or third grid is located is determined as their residence; the third grid is the grid where the fixed resident subject stays for the longest time on non-working nights.

[0021] Optionally, the identification method of this application further includes:

[0022] The schools are divided according to their school system, and the commuting behavior of the schools under different school systems is displayed based on the start and end points of the commuting behavior of all the subjects.

[0023] Alternatively, the schools can be categorized based on teaching quality, and the commuting behaviors of schools with different teaching qualities can be displayed according to the start and end points of the commuting behaviors of all the subjects.

[0024] Optionally, the mobile phone signaling data and the school AOI data are preprocessed, including:

[0025] Obtain all AOI data for the stated schools and text data related to the stated schools;

[0026] The text data is understood through natural language processing to determine the basic information of the school.

[0027] On the other hand, embodiments of the present invention propose a school behavior recognition system, which includes:

[0028] The first model is used to acquire mobile phone signaling data and school AOI data, and to preprocess the mobile phone signaling data and school AOI data.

[0029] The second model is used to determine the fixed resident objects based on the school and their corresponding types based on the mobile phone signaling data and the school AOI data; the type is the type corresponding to the geographical grid visited or resided by the object.

[0030] The third model is used to determine the population category, the starting point and the ending point of the commuting behavior of the object based on the type and the basic information of the object, and on a preset classification strategy.

[0031] On the other hand, embodiments of the present invention provide a school behavior recognition device, the device comprising:

[0032] At least one processor;

[0033] At least one memory for storing at least one program;

[0034] When the at least one program is executed by the at least one processor, the at least one processor implements the above-described method for recognizing common learning behaviors.

[0035] On the other hand, embodiments of the present invention provide a storage medium storing a processor-executable program, which, when executed by a processor, is used to implement the above-described method for recognizing common learning behaviors.

[0036] The method provided in this invention includes: acquiring mobile phone signaling data and school AOI data, and preprocessing the mobile phone signaling data and school AOI data; determining, based on the mobile phone signaling data and school AOI data, fixed-resident objects and their corresponding types based on the school; the type being the type corresponding to the geographical grid visited or resided in by the object; and determining, based on the type and the object's basic information and a preset classification strategy, the object's population category, the start and end points of its commuting behavior. This application combines large-scale mobile phone signaling data and school AOI data to determine the object's residency type; thereby, it refines the population category and commuting behavior, which helps improve the precision of population identification. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following description is provided with accompanying drawings of the relevant technical solutions in the embodiments of the present invention or the prior art. It should be understood that the accompanying drawings described below are only for the purpose of clearly illustrating some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0038] Figure 1 A flowchart illustrating one embodiment of the school behavior recognition method provided by the present invention;

[0039] Figure 2 A flowchart illustrating another embodiment of the learning behavior recognition method provided by the present invention;

[0040] Figure 3 A flowchart illustrating one embodiment of the crowd segmentation and commuting behavior recognition provided by the present invention;

[0041] Figure 4 A schematic diagram of the structure of one embodiment of the school behavior recognition system provided by the present invention;

[0042] Figure 5 This is a schematic diagram of one embodiment of the school behavior recognition device provided by the present invention. Detailed Implementation

[0043] The embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. The step numbers in the following embodiments are set only for ease of explanation, and there is no limitation on the order between the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0044] With the development of big data and mobile communication technologies, mobile signaling data has become an important resource for studying population flow, urban activities, and travel patterns. Due to its wide coverage, real-time performance, and relatively low cost, it is widely used in fields such as transportation planning, urban planning, and socio-economic analysis. Existing technical solutions include methods for identifying population movement based on mobile signaling data, such as identifying the unemployed, extracting urban residents' travel chains, identifying population migration, and characterizing travelers' work-residence locations and travel chains. These methods typically involve data collection, feature extraction, and cluster analysis, using machine learning and data mining techniques to process and analyze the data. However, these solutions often focus on a single data source or a single application scenario, lacking comprehensive utilization of multi-source data, such as vector geographic information data (AOI, POI), thus limiting their ability to achieve refined identification. Furthermore, existing solutions for school-residence identification primarily rely on small-data methods such as questionnaires, lacking mature methods for large-scale, wide-ranging identification and analysis using multi-source spatiotemporal big data such as mobile phone signaling. They also lack detailed categorization of school visitors (e.g., teachers, parents, day / boarding students, etc.). Specifically, the relevant technologies suffer from the following problems:

[0045] (1) Single data source:

[0046] Existing technologies for identifying crowd movement mainly rely on single mobile phone signaling data, failing to fully utilize other data sources such as school AOI data, resulting in an inability to comprehensively and accurately identify the daily activities and off-campus residences of student groups.

[0047] (2) Insufficient spatiotemporal coverage:

[0048] Existing solutions for identifying student groups mainly rely on methods such as on-site questionnaires, which result in limited data coverage and poor timeliness, making it impossible to conduct large-scale, wide-ranging identification and analysis.

[0049] (3) Insufficient recognition accuracy:

[0050] The methods for accurately identifying student groups, especially their home and school locations, are still immature, and there is a lack of in-depth analysis of the characteristics of student groups.

[0051] (4) The classification is not detailed enough:

[0052] The lack of detailed classification of school visitors makes it difficult to distinguish between different groups of people, such as teachers, parents, boarding students, and day students, leading to biases in student group identification.

[0053] This invention aims to overcome the shortcomings of existing technologies by combining mobile phone signaling data and school AOI data to achieve accurate identification of students' home and school locations, as well as refined classification of school visitors. This method not only improves the accuracy and real-time performance of identification but also expands its application scope, enabling more precise data support for fields such as education management, urban planning, and transportation planning. Compared with existing technologies, this invention provides a more comprehensive and in-depth analysis method by integrating multi-source data, particularly in the segmentation of school visitor types and identification of student residences, achieving technological advancement and innovation.

[0054] The following describes in detail the school commuting behavior recognition method and system proposed according to embodiments of the present invention with reference to the accompanying drawings. First, the school commuting behavior recognition method proposed according to embodiments of the present invention will be described with reference to the accompanying drawings.

[0055] Reference Figure 1 This invention provides a method for recognizing common learning behaviors, which mainly includes the following steps:

[0056] S100: Acquire mobile phone signaling data and school AOI data, and preprocess the mobile phone signaling data and school AOI data;

[0057] S200: Based on mobile signaling data and school AOI data, determine the fixed-resident objects based on the school and their corresponding types; the type is the type corresponding to the geographic grid visited or resided in by the object;

[0058] S300: Based on the basic information of the type and object, and using a preset classification strategy, determine the object's population category, the starting point and the ending point of the commuting behavior.

[0059] In some possible implementations, this application uses mobile signaling data and school AOI data to filter and determine fixed-term visitors to each school, and further determines the type corresponding to the geographical grid visited or resided by the visitors. The type can be the type corresponding to the grid where the visitor / user stayed the longest within a preset time period. It is understood that the preset classification strategy in this application can be set according to the actual situation / characteristics of the school, and this application does not impose specific limitations. It should be noted that the visitors in this application can be end users or groups of people visiting the school.

[0060] Optionally, based on mobile signaling data and school AOI data, determine the school-based permanent residents and their corresponding types, including:

[0061] Set up tiered buffer zones for schools; these buffer zones are used to ensure that the area of ​​a school is greater than the extraction accuracy of mobile phone signaling.

[0062] Create a geographic grid based on each school and its hierarchical buffer zone;

[0063] Traverse the mobile signaling data, map the signaling trajectory points to the geographic grid, and obtain the access data of each object to the school;

[0064] Based on the access data, determine the permanent resident objects and their types.

[0065] In some possible implementations, tiered buffers are used to prevent situations where the school area is too small for accurate coverage of mobile phone signaling trajectories. This application filters for fixed-stay objects that meet the requirements for the number of stays / duration based on access data, and determines the type corresponding to each fixed-stay object.

[0066] Optionally, based on the access data, the permanently resident objects and their types can be determined, including:

[0067] Based on the access data, within a preset time period, select objects whose access and stay frequency to the school is greater than or equal to the preset frequency, and designate them as fixed stay objects.

[0068] Based on the access data of fixed residents to the school, determine whether the first grid, second grid and third grid corresponding to each resident are located within the school area; the first grid, second grid and third grid represent the residence characteristics of fixed residents on weekdays and non-weekdays.

[0069] In some possible implementations, the preset duration can be one month, two months, or set according to actual needs; this application does not impose a specific limitation. This application determines the types of the first grid, second grid, and third grid based on access data. It is understood that the first grid, second grid, and third grid represent the number of grids corresponding to different scenarios based on the types generally set by schools in this application. Those skilled in the art can set the number of grids according to actual needs and determine commuting behavior based on the types corresponding to the set grids.

[0070] Optionally, based on the basic information of the type and object, and using a preset classification strategy, the object's population category, the starting point and ending point of the learning behavior are determined, including:

[0071] If the first grid and the second grid are the same, both the first grid and the second grid belong to the school area, and the age of the fixed resident is greater than or equal to the first preset age, the fixed resident is determined to be a boarding student.

[0072] If the first grid belongs to the school area, the second grid does not belong to the school area, and the age of the fixed resident is greater than or equal to the first preset age, the fixed resident is determined to be a day student.

[0073] If the first grid belongs to a school area, the fixed resident uses at least one teacher-related application, and the fixed resident's age is greater than or equal to the second preset age, then the fixed resident is determined to be a teacher;

[0074] If the first grid does not belong to the school area, the visit time of the fixed resident to the school is within the preset time period, and the age of the fixed resident is greater than or equal to the second preset age, the fixed resident is determined to be a parent; the first grid is the grid where the fixed resident stays for the longest time during the daytime on weekdays, the second grid is the grid where the fixed resident stays for the longest time at night on weekdays, and the third grid is the grid where the fixed resident stays for the longest time at night on non-weekdays.

[0075] In some possible implementations, this application determines the population category of the fixed-resident population through logical judgments of the first grid, the second grid, and the third grid. Of course, those skilled in the art can also classify the population according to other preset classification strategies.

[0076] Optionally, the identification method of this application further includes:

[0077] The area where the third grid of boarding students is located is determined as their residence, the area where the second or third grid of day students is located is determined as their residence, the area where the second or third grid of teachers is located is determined as their residence, the area where the first grid of parents is located is determined as their workplace, and the area where the second or third grid of parents is located is determined as their residence; the third grid is the grid where the fixed residents stay for the longest time on non-working nights.

[0078] In some possible implementations, different areas are identified based on the location of the first, second, and third grids corresponding to different population categories. Specifically, determining the location of a teacher's second or third grid as their residence is understood as follows: if the first and second grids are the same, then the third grid is confirmed as the residence, and the first and second grids as the school; if the first and second grids are different, then the second and / or third grids are confirmed as the residence, and the first grid as the school.

[0079] Optionally, the identification method of this application further includes:

[0080] Schools are divided based on their educational system, and the commuting behavior of students in different educational systems is displayed according to the starting and ending points of their commuting activities.

[0081] Alternatively, schools can be categorized based on teaching quality, and the commuting behaviors of schools with different teaching qualities can be displayed according to the starting and ending points of the commuting behaviors of all subjects.

[0082] In some possible implementations, after identifying different groups of people and their corresponding commuting behaviors, the commuting behaviors of different types of schools can be graphically displayed, allowing for a direct comparison between them. Of course, users can set the content to be displayed on the visual interface, and the corresponding graphical interface will be shown based on the user's settings.

[0083] Optionally, the mobile signaling data and school AOI data are preprocessed, including:

[0084] Obtain all school AOI data and school-related text data;

[0085] By understanding textual data through natural language processing, basic information about the school can be determined.

[0086] In some possible implementations, the school's basic information could include its academic system, teaching quality, etc.

[0087] The following is a detailed description of the school behavior recognition method provided in this application using a specific embodiment:

[0088] The purpose of this invention is to provide a method for accurately identifying student commuting behavior based on mobile phone signaling and school AOI data.

[0089] This invention uses a massive sample of mobile signaling data across the entire city, comprehensively employing methods such as data cleaning, topology correction, and geographic information science. It combines school AOI data with features such as residence distribution, user attributes, and travel behavior during the mobile signaling sampling process to accurately identify user demographics and school-residence information.

[0090] To achieve the above objectives, the present invention proposes a method for accurate identification of commuting behavior based on mobile phone signaling and school AOI data, which includes: preprocessing of mobile phone signaling and school AOI data, identification of fixed resident users at the school, classification of commuting population, and school-residence discrimination.

[0091] Reference Figure 2 As shown, the basic steps of this invention are as follows:

[0092] S1. Preprocessing of mobile signaling and school AOI data: including matching of basic user information in mobile signaling and classifying and organizing school AOI data attributes;

[0093] S11. Prioritize acquiring mobile signaling data within the study area, and then obtain basic attribute information of the corresponding mobile users from the operators, including daily user dwell time, monthly user dwell time, travel behavior, age, number location, and app usage (see Table 1 for the classification of mobile signaling data). The obtained raw mobile user data does not contain any personal privacy information.

[0094] S12. Utilize web crawling tools to collect AOI data for all schools within the study area from Baidu Maps. Collect textual data related to schools from relevant websites, school websites, and online comments (such as relevant forums, social apps, and sharing apps). Use natural language processing to construct a sentiment analysis model to process and understand the textual data, and categorize and organize the basic attributes of different school AOIs, including school name, grade level, school quality, and school type. Details are as follows:

[0095] (1) Educational stages: divided into five categories: primary school, junior high school, senior high school, nine-year integrated system, and twelve-year integrated system;

[0096] (2) School quality: divided into four categories: top, excellent, average, and poor;

[0097] (3) School type: divided into two categories: public and private.

[0098] Among these, text records containing attributes such as school name, grade level, and school type are relatively straightforward and can be directly extracted using fixed-vocabulary retrieval. However, the vocabulary related to school quality evaluation is more vague and ambiguous, requiring further analysis using a natural language processing model, followed by refinement based on factors such as peer-to-peer parent reputation rankings. The specific steps are as follows:

[0099] S111. School Name Correction: Based on the government school directory of the study area, the school names in the school AOI data are compared and corrected, and AOIs that do not exist in the government school directory are removed;

[0100] S112. Text data collection: Using school names as keywords, conduct batch searches on government websites, school websites, and other online platforms, and filter out text data that matches the requested keywords to form a text database that corresponds one-to-one with the school names;

[0101] S113. Text Preprocessing: Cleaning and organizing the collected text data, including filtering key semantic information, removing stop words, and word segmentation. Details are as follows:

[0102] (1) Determine whether each piece of text data in the text database contains key semantic information describing basic attributes such as school grade, school quality and school nature. If it exists, retain the text data; otherwise, remove it.

[0103] (2) Stop word removal is performed on each text data item to reduce meaningless statements and alleviate the complexity of subsequent model training. The specific methods for stop word removal are publicly available technologies and will not be elaborated upon in this invention.

[0104] (3) Use the "Jieba" Chinese word segmentation tool to segment the text sentences into shorter phrases.

[0105] S114. Feature Extraction and Word Vector Construction: Since sentiment analysis models cannot directly process raw text data, the text needs to be converted into numerical vectors for model processing. This invention uses the open-source model GloVe developed by Stanford University to train the text sentences, obtaining the final dictionary and word vectors, including positive sentiment words, negative sentiment words, and neutral words. The training process of GloVe is also a publicly available technology, and this invention will not elaborate further.

[0106] S115. Sentiment Analysis Model Selection and Training: This invention employs an LSTM deep network learning model as the sentiment analysis model. This model can capture long-term dependencies in the text, thereby improving the model's accuracy. The word vectors constructed in the previous step are input into the LSTM model. The LSTM model processes the data through its neural network layers and finally outputs a sentiment classification, including positive, negative, and neutral evaluations of school quality. Similarly, sentiment analysis based on the LSTM model is a publicly available and mature technology, and this invention will not elaborate on it here.

[0107] S116. Model Result Evaluation and Correction: Based on online reviews and peer-to-peer parental reputation rankings, the school quality evaluation results output by the sentiment analysis model are corrected and improved; then, by extracting and recording the grade level and school nature attributes through keywords, and combining the school's official website and Baidu Encyclopedia introduction, the missing grade level and school nature attributes are supplemented;

[0108] S117. Match the basic attribute information recorded in the above steps into the school AOI data to form a complete school AOI geospatial dataset containing school grade level, school quality, and school nature.

[0109]

[0110]

[0111] Table 1

[0112] S2. Identification of school fixed-resident users (i.e., fixed-resident objects in this application, the objects of this application may be users, user-corresponding terminal devices, etc., this application does not make specific limitations) based on large-scale mobile signaling data;

[0113] S21. Set up tiered buffers for schools. Considering the limited extraction accuracy of mobile signaling data (approximately 250m × 250m), and that the area of ​​some schools is smaller than the extraction accuracy, it is necessary to set up tiered buffers for school AOI data to ensure that the total area of ​​the schools and the buffers is greater than the extraction accuracy.

[0114] S22. Create a geographic grid. Use the "Create Net" tool in GIS software to grid the study area, mapping all signaling trajectory points into the grid to form a large-scale geographic dataset containing basic attribute information of all users;

[0115] S23. Filter users with fixed residency. Traverse the dataset and count the frequency of user access and residency in the school AOI and its buffer grid. Filter users whose access and residency frequency in the school buffer AOI is ≥8 (i.e., the preset frequency in this application) within one month (i.e., the preset duration in this application). In this step, if the trajectory point only moves within the same AOI area, it is not considered a new trip and is still in the residency process;

[0116] S24. Identify user dwell time types. Based on the fixed dwelling users selected in the previous step, and combined with the fields such as weekday daytime dwell time, weekday nighttime dwell time, and weekend nighttime dwell time in the user monthly dwell time table in the mobile signaling data, the longest weekday daytime dwell time grid, the longest weekday nighttime dwell time grid, and the longest weekend nighttime dwell time grid for each user are selected and defined as three types: A, B, and C (i.e., the first grid, the second grid, and the third grid in this application).

[0117] S3. Classification of Students Who Attend Schools: Refer to Figure 3 As shown, the fixed users in the previous step are finely divided into user categories based on their location, age, and app usage.

[0118] S31. Determine whether grid type A belongs to the school buffer zone;

[0119] S32. Record the user's app usage type and age attribute;

[0120] S33. Categorize the audience. The categorization results are as follows: four categories (other visitors in category 5 are excluded):

[0121] (1) Boarding students: A = B, that is, A and B belong to the same grid and are both located within the school buffer zone, and the user's age is ≥6 years old (i.e., the first preset age in this application);

[0122] (2) Day students: A≠B, that is, A is within the school's buffer zone but B is not, and the user's age is ≥6 years old;

[0123] (3) Teachers: A is within the school's buffer zone, and the user uses at least one teacher-related app (such as: Class Optimization Master, Teacher Training, Teacher Recruitment, etc.) and is ≥18 years old (i.e., the second preset age in this application);

[0124] (4) Parents: A is not within the school buffer zone, and these users access the school buffer zone during weekday school hours (6:30-9:00, 16:00-19:00, i.e. the preset time period in this application), and are ≥18 years old;

[0125] (5) Other visitors: Excluded based on fields such as start time of stay (excluding school hours, such as 10:00 AM), duration of stay (less than 2 hours), and frequency of stay (less than 8 times per month).

[0126] S4. School-Residence Determination: Combining the above analysis steps, further determine the school and residence of mobile phone users to form a precise commuting OD that includes different school levels, school quality, and school type.

[0127] S41. Identify school-residence grids based on the mobility characteristics of four population groups. Details are as follows:

[0128] (1) For boarding students: The place where this group spends the longest time during the day and night on weekdays is the school. They may return to their residence on weekends. Therefore, the grid where they spend the longest time during the day on weekdays (A) and the grid where they spend the longest time at night on weekdays (B) are determined to be the school location, and the grid where they spend the longest time at night on weekends (C) is determined to be their residence.

[0129] (2) For day students: The place where this group spends the longest time during the weekday is the school, and they return to their residence on weekday nights and weekends. Therefore, the grid where they spend the longest time during the weekday (A) is determined to be the school location, and the grid where they spend the longest time at night on weekdays (B) and the grid where they spend the longest time at night on weekends (C) are determined to be their residence.

[0130] (3) For teachers: The place where this group spends the most time during the day on weekdays is the school. In most cases, the school will provide teachers with dormitory accommodation. Therefore, it is necessary to further determine whether the longest daytime grid (A) and the longest nighttime grid (B) of teachers are the same. If they are the same, then similar to day students, A and B are determined to be the school location, and the longest nighttime grid (C) on weekends is determined to be the residence location. If they are different, then A is determined to be the school location, and B and C are determined to be the residence location.

[0131] (4) For parents: This group's visits to schools are mainly concentrated during weekday school hours (6:30-9:00, 16:00-19:00), with a frequency of more than 8 visits per month. The longest daytime stay on weekdays is not at the school, and they return to their residence on weekday nights. Therefore, the grid with the longest daytime stay on weekdays (A) is determined as their work location, and the grid with the longest nighttime stay on weekdays (B) and the grid with the longest nighttime stay on weekends (C) are determined as their residence location.

[0132] S42. Visualization of commuting OD flow under different attributes. Combining the basic attributes of schools and the school-residence discrimination results, the residence grid is defined as the commuting start point O, and the school grid is defined as the commuting end point D, to visualize commuting behavior under different attributes.

[0133] In summary, the method provided in this application includes: acquiring mobile phone signaling data and school AOI data, and preprocessing the mobile phone signaling data and school AOI data; determining, based on the mobile phone signaling data and school AOI data, fixed-resident objects based on the school and the corresponding types of fixed-resident objects; the type is the type corresponding to the geographical grid visited or resided by the object; and, based on the type and the object's basic information, determining the object's population category, the start point and end point of commuting behavior based on a preset classification strategy. This application combines large-scale mobile phone signaling data and school AOI data to determine the object's residency type; and further refines the population category and commuting behavior, which is beneficial to improving the precision of population identification.

[0134] Secondly, refer to the appendix Figure 4 A school behavior recognition system according to an embodiment of the present invention is described, the system specifically comprising:

[0135] The first model 310 is used to acquire mobile phone signaling data and school AOI data, and to preprocess the mobile phone signaling data and school AOI data.

[0136] The second model 320 is used to determine the fixed-resident objects based on the school and the corresponding types of the fixed-resident objects based on mobile phone signaling data and school AOI data; the type is the type corresponding to the geographical grid visited or resided by the object.

[0137] The third model 330 is used to determine the population category, the starting point and the ending point of the student's learning behavior based on the basic information of the type and the object, and a preset classification strategy.

[0138] Optionally, the identification system provided in this application further includes a fourth module for:

[0139] Schools are divided based on their educational system, and the commuting behavior of students in different educational systems is displayed according to the starting and ending points of their commuting activities.

[0140] Alternatively, schools can be categorized based on teaching quality, and the commuting behaviors of schools with different teaching qualities can be displayed according to the starting and ending points of the commuting behaviors of all subjects.

[0141] It is evident that the content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0142] Reference Figure 5 This invention provides a school behavior recognition device, which includes:

[0143] At least one processor 410;

[0144] At least one memory 420 is used to store at least one program;

[0145] When the at least one program is executed by the at least one processor 410, the at least one processor 410 implements the general learning behavior recognition method.

[0146] Similarly, the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0147] This invention also provides a computer-readable storage medium storing a processor-executable program, which, when executed by a processor, is used to perform the aforementioned common behavior recognition method.

[0148] Similarly, the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0149] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this invention are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is altered and sub-operations described as part of a larger operation are executed independently.

[0150] Furthermore, although the invention has been described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the invention. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional skill of an engineer. Therefore, those skilled in the art can implement the invention as set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of the invention, which is determined by the full scope of the appended claims and their equivalents.

[0151] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several programs to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0152] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequential list of executable programs for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, a program execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can retrieve and execute a program from or in conjunction with such a program execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can mean any means that can contain, store, communicate, propagate, or transmit a program for use by or in conjunction with a program execution system, apparatus, or device.

[0153] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0154] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable program execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0155] In the foregoing description of this specification, references to terms such as "one embodiment," "another embodiment," or "some embodiments" indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of the present invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0156] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

[0157] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of the present invention.

Claims

1. A method for recognizing common school behavior, characterized in that, The method includes the following steps: Acquire the target's mobile phone signaling data and school AOI data, and preprocess the mobile phone signaling data and school AOI data; Based on the mobile phone signaling data and the school AOI data, determine the fixed-resident objects based on the school and the types corresponding to the fixed-resident objects; the type is the type corresponding to the geographical grid visited or resided by the object; Based on the type and basic information of the object, and using a preset classification strategy, determine the object's population category, the starting point and the ending point of its commuting behavior; Specifically, based on the mobile phone signaling data and the school AOI data, the fixed-resident objects based on the school and their corresponding types are determined, including: A tiered buffer zone is set for the school; the tiered buffer zone is used to ensure that the area of ​​the school is greater than the extraction accuracy of mobile phone signaling. Create a geographic grid based on each of the schools and the hierarchical buffer zones; By traversing the mobile phone signaling data, the signaling trajectory points are mapped to the geographic grid to obtain the access data of each object to the school; Based on the access data, determine the fixed resident object and its type; Based on the type and basic information of the object, and using a preset classification strategy, the object's population category, the starting point and ending point of its commuting behavior are determined, including: If the first grid and the second grid are the same, both the first grid and the second grid belong to the school area, and the age of the fixed resident is greater than or equal to the first preset age, then the fixed resident is determined to be a boarding student. If the first grid belongs to the school area, the second grid does not belong to the school area, and the age of the fixed resident is greater than or equal to the first preset age, the fixed resident is determined to be a day student. If the first grid belongs to the school area, the fixed resident object uses at least one teacher-related application, and the fixed resident object's age is greater than or equal to a second preset age, then the fixed resident object is determined to be a teacher. If the first grid does not belong to the school area, the visit time of the fixed resident object to the school is within a preset time period, and the age of the fixed resident object is greater than or equal to a second preset age, the fixed resident object is determined to be a parent; the first grid is the grid in which the fixed resident object stays for the longest time during the daytime on weekdays, and the second grid is the grid in which the fixed resident object stays for the longest time at night on weekdays; Among them, determining the fixed resident object and its type based on the access data includes: Based on the access data, within a preset time period, objects whose access and stay frequency to the school is greater than or equal to the preset frequency are selected and designated as fixed-stay objects. Based on the access data of the fixed resident objects to the school, determine whether the first grid, second grid, and third grid corresponding to each object are located within the school area; the first grid, second grid, and third grid represent the resident characteristics of the fixed resident objects on weekdays and non-weekdays; Preprocessing of the mobile phone signaling data and the school AOI data includes: Acquire all AOI data of the schools and text data related to the schools; use a large-scale sample of mobile signaling data covering the entire city, and employ data cleaning, topology correction and geographic information science methods, combined with the characteristics of residence distribution, user attributes and travel behavior in the sampling process of school AOI data and mobile signaling; The text data is understood through natural language processing to determine the basic information of the school; a sentiment analysis model is constructed using natural language processing to process and understand the text data, and to classify and organize the basic attributes of different school AOIs, including school name, grade level, school quality, and school nature.

2. The method for recognizing school behavior according to claim 1, characterized in that, The method further includes: The area where the boarding student's third grid is located is determined as their residence, the area where the day student's second or third grid is located is determined as their residence, the area where the teacher's second or third grid is located is determined as their residence, the area where the parent's first grid is located is determined as their workplace, and the area where the parent's second or third grid is located is determined as their residence; the third grid is the grid where the fixed resident subject stays for the longest time on non-working nights.

3. The method for recognizing school behavior according to claim 1, characterized in that, The method further includes: The schools are divided according to their school system, and the commuting behavior of the schools under different school systems is displayed based on the start and end points of the commuting behavior of all the subjects. Alternatively, the schools can be categorized based on teaching quality, and the commuting behaviors of schools with different teaching qualities can be displayed according to the start and end points of the commuting behaviors of all the subjects.

4. A school commuting behavior recognition system, characterized in that, The system includes: The first module is used to acquire mobile phone signaling data and school AOI data, and to preprocess the mobile phone signaling data and school AOI data. The second module is used to determine, based on the mobile phone signaling data and the school AOI data, the fixed resident objects based on the school and the type corresponding to the fixed resident objects; the type is the type corresponding to the geographical grid visited or resided by the object. The third module is used to determine the population category, the starting point and the ending point of the commuting behavior of the object based on the type and the basic information of the object, and on a preset classification strategy. Specifically, based on the mobile phone signaling data and the school AOI data, the fixed-resident objects based on the school and their corresponding types are determined, including: A tiered buffer zone is set for the school; the tiered buffer zone is used to ensure that the area of ​​the school is greater than the extraction accuracy of mobile phone signaling. Create a geographic grid based on each of the schools and the hierarchical buffer zones; By traversing the mobile phone signaling data, the signaling trajectory points are mapped to the geographic grid to obtain the access data of each object to the school; Based on the access data, determine the fixed resident object and its type; Based on the type and basic information of the object, and using a preset classification strategy, the object's population category, the starting point and ending point of its commuting behavior are determined, including: If the first grid and the second grid are the same, both the first grid and the second grid belong to the school area, and the age of the fixed resident is greater than or equal to the first preset age, then the fixed resident is determined to be a boarding student. If the first grid belongs to the school area, the second grid does not belong to the school area, and the age of the fixed resident is greater than or equal to the first preset age, the fixed resident is determined to be a day student. If the first grid belongs to the school area, the fixed resident object uses at least one teacher-related application, and the fixed resident object's age is greater than or equal to a second preset age, then the fixed resident object is determined to be a teacher. If the first grid does not belong to the school area, the visit time of the fixed resident object to the school is within a preset time period, and the age of the fixed resident object is greater than or equal to a second preset age, the fixed resident object is determined to be a parent; the first grid is the grid in which the fixed resident object stays for the longest time during the daytime on weekdays, and the second grid is the grid in which the fixed resident object stays for the longest time at night on weekdays; Among them, determining the fixed resident object and its type based on the access data includes: Based on the access data, within a preset time period, objects whose access and stay frequency to the school is greater than or equal to the preset frequency are selected and designated as fixed-stay objects. Based on the access data of the fixed resident objects to the school, determine whether the first grid, second grid, and third grid corresponding to each object are located within the school area; the first grid, second grid, and third grid represent the resident characteristics of the fixed resident objects on weekdays and non-weekdays; Preprocessing of the mobile phone signaling data and the school AOI data includes: Acquire all AOI data of the schools and text data related to the schools; use a large-scale sample of mobile signaling data covering the entire city, and employ data cleaning, topology correction and geographic information science methods, combined with the characteristics of residence distribution, user attributes and travel behavior in the sampling process of school AOI data and mobile signaling; The text data is understood through natural language processing to determine the basic information of the school; a sentiment analysis model is constructed using natural language processing to process and understand the text data, classify and organize the basic attributes of different school AOIs, including school name, grade level, school quality and school nature, and use an LSTM deep network learning model as the sentiment analysis model for sentiment analysis.

5. A school behavior recognition device, characterized in that, The device includes: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the school behavior recognition method as described in any one of claims 1 to 3.

6. A computer-readable storage medium storing a processor-executable program, characterized in that, The processor-executable program, when executed by the processor, is used to implement the commutation behavior recognition method as described in any one of claims 1 to 3.

Images