Ship-car-person virtual image establishing method and system

Abstract

The application provides a ship-vehicle-person virtual image establishment method and system, which comprises the following steps: obtaining ship, personnel and vehicle data as source data for preprocessing, calculating and mining labels according to the preprocessed source data, and producing ship, personnel and vehicle labels; defining hierarchical labels; defining group labels; extracting basic attributes of the ship, personnel and vehicle; defining personalized labels, adding personalized labels to the ship, personnel and vehicle according to their basic attributes and respective source data; defining correlation relationship labels among the ship, personnel and vehicle; creating a label table in a database, adding the calculated labels into the label table, and then performing correlation analysis on the respective source data of the ship, personnel and vehicle and the label table according to business requirements to obtain index data; and providing the index data to a ship-vehicle-person virtual image system for drawing charts. The application can improve traffic operation efficiency, safety and risk prediction.

Description

Technical Field

[0001] This invention relates to the field of virtual portrait technology, specifically to a method and system for creating virtual portraits of people, vehicles, and ships. Background Technology

[0002] Because the behavioral data of ships, vehicles, and people collected in water monitoring cannot be directly used for data analysis and model training, nor can useful information be directly extracted from the behavioral data, we gain an intuitive understanding of ships, vehicles, and people by labeling their behavioral data. The technology used in this labeling is virtual profiling. Virtual profiling technology is an advanced data analysis technique based on artificial intelligence and big data analytics. It integrates and analyzes information such as behavioral characteristics, preferences, and interests of objects to form a virtual profile. The background of virtual profiling technology mainly includes the following aspects:

[0003] 1. The development of big data technology. Virtual portrait technology mainly relies on big data analysis and processing technology, including various stages such as data collection, storage, cleaning, mining and analysis.

[0004] 2. The development of artificial intelligence and machine learning. The crowd profiling in virtual portrait technology is obtained by integrating and analyzing data from different sources, types, and granularities using artificial intelligence and machine learning algorithms.

[0005] 3. Technical support from cloud computing and distributed computing. Virtual portrait technology requires large-scale computing and storage resources, and cloud computing and distributed computing technologies can provide powerful computing and storage capabilities to meet the needs of virtual portrait technology.

[0006] 4. The widespread adoption of the Internet of Things (IoT) and smart terminals. Virtual profiling technology requires the collection and integration of massive amounts of data. IoT and smart terminals can provide various types of data sources to better support the implementation of virtual profiling technology.

[0007] In summary, virtual portrait technology is based on advanced data analysis techniques and artificial intelligence algorithms. It utilizes big data analysis and processing techniques to integrate and analyze data from different sources, types, and granularities to form a virtual portrait.

[0008] In summary, the existing related technologies have the following shortcomings:

[0009] 1) Inability to fully explore and utilize the relationships between multi-source data. Existing virtual profiling methods are often based on a single model, treating people, vehicles, and ships as three independent virtual profiling systems. While this facilitates analysis of individual targets, it may overlook important information when analyzing the relationships between them, affecting the accuracy of the final behavioral risk prediction.

[0010] 2) There may be privacy risks during data processing and model training. Although encryption and pseudo-naming techniques can protect data privacy to some extent, data security and privacy protection issues still need to be considered when processing massive amounts of data and connecting to multiple data sources.

[0011] 3) Predicting behavioral risks for ships, vehicles, and people may require addressing different scenarios and complexities. Existing technologies may perform well in certain specific scenarios but poorly in others, or struggle with complex problems. Therefore, there is a need to develop more flexible and adaptable virtual profiling methods. Summary of the Invention

[0012] To address the shortcomings of existing technologies, this invention provides a method and system for creating virtual portraits of people, vehicles, and ships.

[0013] According to the present invention, a method and system for creating virtual portraits of people in boats, vehicles, and other vehicles are provided, and the solution is as follows:

[0014] Firstly, a method for creating virtual portraits of people in boats and vehicles is provided, the method comprising:

[0015] Step S1: Obtain ship, personnel, and vehicle data as source data for tag calculation and mining to generate ship, personnel, and vehicle tags. Source data processing employs data cleaning, data transformation, and other methods for data preprocessing.

[0016] Step S2: Define layered labels to divide ships, personnel, and vehicles into multiple non-overlapping parts according to manually defined features, with each part being a layer;

[0017] Step S3: Define cluster labels and cluster ships, personnel, and vehicles at each layer according to more granular, manually defined features;

[0018] Step S4: Extract the basic attributes of ships, personnel, and vehicles;

[0019] Step S5: Define personalized tags. Add personalized tags to ships, personnel, and vehicles based on their basic attributes and respective source data. Generate personalized tags using technologies such as behavior analysis and association rule mining.

[0020] Step S6: Define the relationship labels between ships, personnel, and vehicles, and use methods such as association analysis and time series analysis to mine the relationships;

[0021] Step S7: Create a label table in the database, add the calculated labels to the label table, and then perform correlation analysis between the source data of ships, personnel, and vehicles and the label table according to business needs to obtain indicator data;

[0022] Step S8: Provide the indicator data to the virtual portrait system for boats, vehicles and people to generate charts.

[0023] Preferably, in step S1, the ship data is obtained by radar detection of relevant data including unique ID, MMSI number, ship name, heading, speed, longitude, latitude, time, status, and data source; the personnel data is obtained from a personnel information database, including relevant data including unique ID, name, age, gender, location, occupation, and address; the vehicle data is obtained by BeiDou detection of relevant data including unique ID, VIN code, license plate number, brand, vehicle series, vehicle type, vehicle class, and approved number of seats.

[0024] The source data is processed using big data processing frameworks such as Spark and Hadoop. During data cleaning, outlier detection and handling methods are employed to remove abnormal data and improve data quality. The specific data processing flow is as follows:

[0025] Data cleaning: By deleting (null values, garbled data), correcting (negative numbers, maximum values, minimum values), and filling (average filling, empty string filling), data quality problems are repaired to ensure the integrity and correctness of the source data.

[0026] Data conversion: Converting data of different formats and sources into a unified format to meet the needs of subsequent processing.

[0027] Preferably, step S2, defining the layered labels, includes:

[0028] Ships are categorized according to artificially defined characteristics into four groups: Key Concern, Concern, General Concern, and Ordinary Concern.

[0029] 1) Key vessel behaviors to be monitored: abnormal navigation plans, excessive speed, lingering, and illegal border crossings.

[0030] 2) Vessel behaviors of concern: including vessels exceeding port hours, vessels concentrating in one area, vessels being overloaded, vessels disabling AIS, and vessels frequently operating at night;

[0031] 3) Vessel behaviors of general concern: including vessels staying longer than expected, vessels berthing in inappropriate locations, and vessels berthing in prohibited areas;

[0032] 4) Ordinary ship behavior: related behaviors including normal ship navigation, normal berthing, departure and operation in port;

[0033] Personnel are categorized into four groups based on manually defined characteristics: key focus, focus, general focus, and ordinary focus.

[0034] 1) Key personnel to focus on: those involved in security threats, sensitive information, and important clients or senior executives;

[0035] 2) People of concern: those who enter and exit frequently, frequently change their mobile phone numbers or other personal information, engage in abnormal interpersonal interactions, and have personal security personnel;

[0036] 3) People of general concern: those who spend too much time in crowded places, and those who frequently change their place of residence or workplace.

[0037] 4) Ordinary people: Individuals whose travel history is normal and poses no safety risks, who abide by laws and regulations, and who have no record of misconduct;

[0038] Vehicles are categorized into four groups based on manually defined characteristics: Priority Concern, Concern, General Concern, and Ordinary Concern.

[0039] 1) Vehicles under special attention: Vehicles involved in the transportation of dangerous goods or high-value goods, vehicles that are frequently transferred in ownership, and vehicles with people on board.

[0040] 2) Vehicles under surveillance: Vehicles that frequently change license plates, speed, follow irregular routes, or repeatedly appear at locations involved in illegal activities;

[0041] 3) Vehicles of general concern: Vehicles that stay for too long, appear around sensitive locations, or have frequent changes in ownership or driver;

[0042] 4) Ordinary vehicles: Vehicles that comply with regulations, are parked in designated areas, and have no illegal activities.

[0043] Preferably, defining the grouping labels in step S3 includes:

[0044] Ships at each level are grouped according to more granular, artificially defined characteristics, with grouping based on activity frequency and region, including:

[0045] 1) Vessels that frequently operate in specific areas: Vessels that frequently enter or leave a certain port or stay in a certain sea area, or are included in the area of ​​interest pre-marked by the platform;

[0046] 2) Vessels that frequently stay in fixed areas: Vessels that frequently stay in a certain dock, waterway, or sea area, with a fixed range of activity;

[0047] 3) Vessels that sail irregularly but operate in limited areas: including vessels that navigate in coastal waters or fixed sea areas;

[0048] 4) Vessels that frequently sail but whose operating areas are not fixed: Vessels that sail in different sea areas or ports;

[0049] 5) Vessels that remain in port or dock for extended periods: vessels that are unable to navigate normally;

[0050] Individuals at each level are grouped according to activity frequency and region based on more granular, manually defined characteristics, including:

[0051] 1) People who frequently operate in a specific area: Crew members who frequently operate in a certain port, dock or on a specific shipping route, as well as related personnel including fishermen or tourists;

[0052] 2) People who frequently stay in a fixed area: People on ships that are moored at a certain dock or port for a long time;

[0053] 3) People who go out irregularly but whose activity area is limited: including crew members or fishermen whose activity range is limited to a certain sea area or river;

[0054] 4) People who frequently travel but whose activity areas are not fixed: crew members of commercial vessels that frequently sail between different waters or passengers on yachts;

[0055] 5) People who stay on board for extended periods: Crew members or captains and other relevant personnel who work and live at sea or inland waters year-round;

[0056] The vehicles at each layer are grouped according to activity frequency and region based on more granular, manually defined characteristics, including:

[0057] 1) Vehicles that frequently operate in specific sea areas: Vehicles that are used for land transport, operations, or are permanently parked in a specific sea area;

[0058] 2) Vehicles that are frequently parked in specific areas: Vehicles that are parked for extended periods at a specific dock, parking area, or berthing area;

[0059] 3) Vehicles that travel irregularly but have limited activity areas: including private cars and official vehicles that operate in specific sea areas or rivers;

[0060] 4) Vehicles that frequently travel but whose activity areas are not fixed: related vehicles including trucks and buses that frequently travel between different shores and docks;

[0061] 5) Vehicles parked in fixed locations for extended periods: scrapped cars, antique cars, exhibition cars, and other vehicles that are parked on the shore or in specific areas for extended periods.

[0062] Preferably, in step S4, the basic attributes of the ship include AIS number, ship name, ship type, ship size, ship tonnage, deadweight, ship construction year, hull material, ship owner, and number of crew members.

[0063] The basic attributes of an individual include name, gender, age, ID number, occupation, contact information, employer, work location, IP address, and criminal record.

[0064] The basic attributes of a vehicle include its license plate number, vehicle type, vehicle color, brand and model, and vehicle owner.

[0065] Preferably, step S5, defining the personalized label, includes:

[0066] Personalized ship tags are generated through the ship's basic attributes or by analyzing source ship data. Personalized ship tags are as follows:

[0067] Speed ​​tag: corresponds to the ship's maximum speed, cruising speed, and other related attributes;

[0068] Load capacity label: corresponds to the maximum load capacity of the vessel;

[0069] Usage Environment Label: Relevant attributes of the vessel, including its usage environment, applicable routes, and sea conditions;

[0070] Construction Year Tag: Corresponds to the ship's construction year and related attributes, including whether it is an old ship;

[0071] Ship type tag: Relevant attributes including the ship type and whether it meets specific navigation conditions;

[0072] Vessel category label: corresponds to the vessel type;

[0073] Maintenance status label: Relevant attributes including the maintenance status of the vessel and whether it undergoes regular maintenance;

[0074] Fuel type label: corresponds to the type of fuel used on the ship;

[0075] Destination label: The destination of the vessel's voyage;

[0076] Safety labels: These correspond to the ship's safety level and whether it complies with certain safety regulations.

[0077] Personalized personnel tags are generated through the generation of personnel's basic attributes or by analyzing personnel source data. Personalized personnel tags are as follows:

[0078] Age group tags: Relevant personnel including youth, middle-aged, and elderly;

[0079] Work experience tags: financial professionals, IT professionals, maintenance personnel, and related personnel;

[0080] Regional tags: Relevant personnel including residents of East China and people from rural areas in southern China;

[0081] Education level tags: relevant personnel including doctoral students and junior college graduates;

[0082] Navigation habit tags: Personnel including those who sail year-round and those who sail occasionally;

[0083] Violation tags: Users who violate marine environmental protection regulations, and related personnel suspected of illegal fishing;

[0084] Crew information tags: professional crew members, hired temporary workers, and related personnel;

[0085] Risk level labels: Individuals including those active in high-risk areas and those active in low-risk areas;

[0086] Personalized vehicle tags are generated through the vehicle's basic attributes or by analyzing source vehicle data. Personalized vehicle tags are as follows:

[0087] Port transportation tags: related vehicles including port trucks and port employee shuttle buses;

[0088] Vehicle type tags: Related vehicles including large trucks and light vans;

[0089] Violation tags: Vehicles including those speeding and overloaded;

[0090] Transport volume tags: Related vehicles including high transport volume vehicles and low transport volume vehicles;

[0091] Goods type label: Related vehicles including liquid chemical transport vehicles and grain transport vehicles;

[0092] Loading time tags: related vehicles including vehicles loading during peak hours and vehicles loading at night;

[0093] Vehicle speed tags: Related vehicles including high-speed vehicles and low-speed vehicles;

[0094] Route tags: Related vehicles including those traveling along the river and those traveling on the cross-sea bridge.

[0095] Personalized tags are generated using techniques such as behavioral analysis and association rule mining. For example, by analyzing ship, vehicle, and pedestrian behavior data, behavioral preferences of ships, personnel, and vehicles can be mined to generate personalized tags. The specific processing flow is as follows:

[0096] Data collection: By collecting data on the behavior of ships, vehicles, and people, we can obtain data on the behavioral preferences of ships, people, and vehicles.

[0097] Data preprocessing: Cleaning, transforming, and other preprocessing of behavioral data to meet the needs of subsequent analysis.

[0098] Pattern mining: Using techniques such as association rule mining and clustering, behavioral data is analyzed to obtain association rules, information such as ship, vehicle, and person categories, in order to generate personalized tags.

[0099] Preferably, in step S6, the association tags between ships, personnel, and vehicles are mainly defined manually based on actual business needs, as follows:

[0100] Personnel and vehicle relationship tags: related tags including passenger tags and ownership tags;

[0101] Personnel-Vessel Relationship Labels: Related labels including passenger label, ownership label, and employment label;

[0102] Ship and vehicle labels: related labels including transport labels and illegal docking labels.

[0103] Furthermore, methods such as association analysis and time series analysis are used to mine relationships in source data of highly correlated ships, vehicles, and people. For example, association rule mining algorithms are used to discover frequent association patterns between ships, personnel, and vehicles, identifying highly correlated relationship tags. The specific processing flow is as follows:

[0104] Data collection: Collect interaction data between ships, personnel, and vehicles to obtain a dataset.

[0105] Association analysis: By using techniques such as association rule mining and frequent itemset mining, we analyze datasets to identify frequent association patterns between ships, people, and vehicles.

[0106] Time series analysis: Through techniques such as time series analysis and sequence pattern mining, analyze datasets to identify time series patterns among ships, personnel, and vehicles.

[0107] Preferably, in step S7, the tag table is designed with five fields: subject, object, feature type, feature value, and detailed description. An empty object field indicates that only the subject field is recorded. Subsequent tags added as the business expands can be directly added as row data. When the data volume is too large, hierarchical or clustered tags are used to partition the table. Finally, the source data of ships, personnel, and vehicles are associated with the tag data, and the indicator data is analyzed according to business needs.

[0108] Secondly, a virtual portrait creation system for boats, vehicles, and people is provided, the system comprising:

[0109] Module M1: Acquires ship, personnel, and vehicle data as source data for tag calculation and mining, and generates ship, personnel, and vehicle tags. The source data processing employs data cleaning, data transformation, and other methods for data preprocessing.

[0110] Module M2: Defines hierarchical labels to divide ships, personnel, and vehicles into multiple non-overlapping parts according to manually defined features, with each part being a layer;

[0111] Module M3: Defines cluster labels to group ships, personnel, and vehicles at each level according to more granular, manually defined features;

[0112] Module M4: Extracts basic attributes of ships, personnel, and vehicles;

[0113] Module M5: Define personalized tags. Add personalized tags to ships, personnel, and vehicles based on their basic attributes and respective source data. Generate personalized tags using technologies such as behavior analysis and association rule mining.

[0114] Module M6: Defines the relationship tags between ships, personnel, and vehicles, and uses methods such as association analysis and time series analysis to mine the relationships;

[0115] Module M7: Creates a label table in the database, adds the calculated labels to the label table, and then performs correlation analysis between the source data of ships, personnel, and vehicles and the label table according to business needs to obtain indicator data;

[0116] Module M8: Provides the indicator data to the virtual portrait system for ships, vehicles, and people to generate charts.

[0117] Preferably, in module M1, the ship data is obtained through radar detection, including relevant data such as unique ID, MMSI number, ship name, heading, speed, longitude, latitude, time, status, and data source; the personnel data is obtained from a personnel information database, including relevant data such as unique ID, name, age, gender, location, occupation, and address; the vehicle data is obtained through BeiDou detection, including relevant data such as unique ID, VIN code, license plate number, brand, vehicle series, vehicle type, vehicle class, and approved number of seats. Source data processing employs big data processing frameworks such as Spark and Hadoop. During data cleaning, outlier detection and processing methods are used to remove abnormal data and improve data quality. The specific data processing flow is as follows:

[0118] Data cleaning: By deleting (null values, garbled data), correcting (negative numbers, maximum values, minimum values), and filling (average filling, empty string filling), data quality problems are repaired to ensure the integrity and correctness of the source data;

[0119] Data conversion: Converting data of different formats and sources into a unified format to meet the needs of subsequent processing;

[0120] The module M2 defines hierarchical tags including:

[0121] Ships are categorized according to artificially defined characteristics into four groups: Key Concern, Concern, General Concern, and Ordinary Concern.

[0122] 1) Key vessel behaviors to be monitored: abnormal navigation plans, excessive speed, lingering, and illegal border crossings.

[0123] 2) Vessel behaviors of concern: including vessels exceeding port hours, vessels concentrating in one area, vessels being overloaded, vessels disabling AIS, and vessels frequently operating at night;

[0124] 3) Vessel behaviors of general concern: including vessels staying longer than expected, vessels berthing in inappropriate locations, and vessels berthing in prohibited areas;

[0125] 4) Ordinary ship behavior: related behaviors including normal ship navigation, normal berthing, departure and operation in port;

[0126] Personnel are categorized into four groups based on manually defined characteristics: key focus, focus, general focus, and ordinary focus.

[0127] 1) Key personnel to focus on: those involved in security threats, sensitive information, and important clients or senior executives;

[0128] 2) People of concern: those who enter and exit frequently, frequently change their mobile phone numbers or other personal information, engage in abnormal interpersonal interactions, and have personal security personnel;

[0129] 3) People of general concern: those who spend too much time in crowded places, and those who frequently change their place of residence or workplace.

[0130] 4) Ordinary people: Individuals whose travel history is normal and poses no safety risks, who abide by laws and regulations, and who have no record of misconduct;

[0131] Vehicles are categorized into four groups based on manually defined characteristics: Priority Concern, Concern, General Concern, and Ordinary Concern.

[0132] 1) Vehicles under special attention: Vehicles involved in the transportation of dangerous goods or high-value goods, vehicles that are frequently transferred in ownership, and vehicles with people on board.

[0133] 2) Vehicles under surveillance: Vehicles that frequently change license plates, speed, follow irregular routes, or repeatedly appear at locations involved in illegal activities;

[0134] 3) Vehicles of general concern: Vehicles that stay for too long, appear around sensitive locations, or have frequent changes in ownership or driver;

[0135] 4) Ordinary vehicles: Vehicles that comply with regulations, are parked in designated areas, and have no illegal activities;

[0136] The clustering labels defined in module M3 include:

[0137] Ships at each level are grouped according to more granular, artificially defined characteristics, with grouping based on activity frequency and region, including:

[0138] 1) Vessels that frequently operate in specific areas: Vessels that frequently enter or leave a certain port or stay in a certain sea area, or are included in the area of ​​interest pre-marked by the platform;

[0139] 2) Vessels that frequently stay in fixed areas: Vessels that frequently stay in a certain dock, waterway, or sea area, with a fixed range of activity;

[0140] 3) Vessels that sail irregularly but operate in limited areas: including vessels that navigate in coastal waters or fixed sea areas;

[0141] 4) Vessels that frequently sail but whose operating areas are not fixed: Vessels that sail in different sea areas or ports;

[0142] 5) Vessels that remain in port or dock for extended periods: vessels that are unable to navigate normally;

[0143] Individuals at each level are grouped according to activity frequency and region based on more granular, manually defined characteristics, including:

[0144] 1) People who frequently operate in a specific area: Crew members who frequently operate in a certain port, dock or on a specific shipping route, as well as related personnel including fishermen or tourists;

[0145] 2) People who frequently stay in a fixed area: People on ships that are moored at a certain dock or port for a long time;

[0146] 3) People who go out irregularly but whose activity area is limited: including crew members or fishermen whose activity range is limited to a certain sea area or river;

[0147] 4) People who frequently travel but whose activity areas are not fixed: crew members of commercial vessels that frequently sail between different waters or passengers on yachts;

[0148] 5) People who stay on board for extended periods: Crew members or captains and other relevant personnel who work and live at sea or inland waters year-round;

[0149] The vehicles at each layer are grouped according to activity frequency and region based on more granular, manually defined characteristics, including:

[0150] 1) Vehicles that frequently operate in specific sea areas: Vehicles that are used for land transport, operations, or are permanently parked in a specific sea area;

[0151] 2) Vehicles that are frequently parked in specific areas: Vehicles that are parked for extended periods at a specific dock, parking area, or berthing area;

[0152] 3) Vehicles that travel irregularly but have limited activity areas: including private cars and official vehicles that operate in specific sea areas or rivers;

[0153] 4) Vehicles that frequently travel but whose activity areas are not fixed: related vehicles including trucks and buses that frequently travel between different shores and docks;

[0154] 5) Vehicles parked in fixed locations for extended periods: scrapped cars, antique cars, exhibition cars, and other vehicles parked on the shore or in specific areas for extended periods;

[0155] In module M4, the basic attributes of a ship include AIS number, ship name, ship type, ship size, ship tonnage, deadweight, year of construction, hull material, ship owner, and number of crew members.

[0156] The basic attributes of an individual include name, gender, age, ID number, occupation, contact information, employer, work location, IP address, and criminal record.

[0157] The basic attributes of a vehicle include its license plate number, vehicle type, vehicle color, brand and model, and vehicle owner.

[0158] The personalized tags defined in module M5 include:

[0159] Personalized ship tags are generated through the ship's basic attributes or by analyzing source ship data. Personalized ship tags are as follows:

[0160] Speed ​​tag: corresponds to the ship's maximum speed, cruising speed, and other related attributes;

[0161] Load capacity label: corresponds to the maximum load capacity of the vessel;

[0162] Usage Environment Label: Relevant attributes of the vessel, including its usage environment, applicable routes, and sea conditions;

[0163] Construction Year Tag: Corresponds to the ship's construction year and related attributes, including whether it is an old ship;

[0164] Ship type tag: Relevant attributes including the ship type and whether it meets specific navigation conditions;

[0165] Vessel category label: corresponds to the vessel type;

[0166] Maintenance status label: Relevant attributes including the maintenance status of the vessel and whether it undergoes regular maintenance;

[0167] Fuel type label: corresponds to the type of fuel used on the ship;

[0168] Destination label: The destination of the vessel's voyage;

[0169] Safety labels: These correspond to the ship's safety level and whether it complies with certain safety regulations.

[0170] Personalized personnel tags are generated through the generation of personnel's basic attributes or by analyzing personnel source data. Personalized personnel tags are as follows:

[0171] Age group tags: Relevant personnel including youth, middle-aged, and elderly;

[0172] Work experience tags: financial professionals, IT professionals, maintenance personnel, and related personnel;

[0173] Regional tags: Relevant personnel including residents of East China and people from rural areas in southern China;

[0174] Education level tags: relevant personnel including doctoral students and junior college graduates;

[0175] Navigation habit tags: Personnel including those who sail year-round and those who sail occasionally;

[0176] Violation tags: Users who violate marine environmental protection regulations, and related personnel suspected of illegal fishing;

[0177] Crew information tags: professional crew members, hired temporary workers, and related personnel;

[0178] Risk level labels: Individuals including those active in high-risk areas and those active in low-risk areas;

[0179] Personalized vehicle tags are generated through the vehicle's basic attributes or by analyzing source vehicle data. Personalized vehicle tags are as follows:

[0180] Port transportation tags: related vehicles including port trucks and port employee shuttle buses;

[0181] Vehicle type tags: Related vehicles including large trucks and light vans;

[0182] Violation tags: Vehicles including those speeding and overloaded;

[0183] Transport volume tags: Related vehicles including high transport volume vehicles and low transport volume vehicles;

[0184] Goods type label: Related vehicles including liquid chemical transport vehicles and grain transport vehicles;

[0185] Loading time tags: related vehicles including vehicles loading during peak hours and vehicles loading at night;

[0186] Vehicle speed tags: Related vehicles including high-speed vehicles and low-speed vehicles;

[0187] Route tags: Related vehicles including those traveling along the river and those traveling on the cross-sea bridge;

[0188] Personalized tags are generated using techniques such as behavioral analysis and association rule mining. For example, by analyzing ship, vehicle, and pedestrian behavior data, behavioral preferences of ships, personnel, and vehicles can be mined to generate personalized tags. The specific processing flow is as follows:

[0189] Data collection: By collecting data on the behavior of ships, vehicles, and people, we can obtain data on the behavioral preferences of ships, people, and vehicles.

[0190] Data preprocessing: Cleaning, transforming, and other preprocessing of behavioral data to meet the needs of subsequent analysis;

[0191] Pattern mining: Using techniques such as association rule mining and clustering, behavioral data is analyzed to obtain association rules, information such as ship, vehicle, and person categories, in order to generate personalized tags;

[0192] In module M6, the association tags between ships, personnel, and vehicles are defined manually based on actual business needs, as follows:

[0193] Personnel and vehicle relationship tags: related tags including passenger tags and ownership tags;

[0194] Personnel-Vessel Relationship Labels: Related labels including passenger label, ownership label, and employment label;

[0195] Ship and vehicle labels: related labels including transport labels and illegal docking labels;

[0196] Furthermore, methods such as association analysis and time series analysis are used to mine relationships in source data of highly correlated ships, vehicles, and people. For example, association rule mining algorithms are used to discover frequent association patterns between ships, personnel, and vehicles, identifying highly correlated relationship tags. The specific processing flow is as follows:

[0197] Data collection: Collect interaction data between ships, personnel, and vehicles to obtain a dataset;

[0198] Association analysis: Through techniques such as association rule mining and frequent itemset mining, the dataset is analyzed to identify frequent association patterns between ships, personnel, and vehicles;

[0199] Time series analysis: Through techniques such as time series analysis and sequence pattern mining, datasets are analyzed to identify time series patterns among ships, personnel, and vehicles;

[0200] In module M7, the tag table is designed with five fields: subject, object, feature type, feature value, and detailed description. An empty object field indicates that only the subject field is recorded. Subsequent tags added as the business expands can be directly added as row data. When the data volume is too large, the table is partitioned using hierarchical or grouped tags. Finally, the source data of ships, personnel, and vehicles are associated with the tag data, and the indicator data is analyzed according to business needs.

[0201] Compared with the prior art, the present invention has the following beneficial effects:

[0202] 1. This invention achieves comprehensive monitoring and management of ships, personnel, and vehicles by establishing virtual profiles of ships, vehicles, and people, thereby improving the effectiveness of ship, personnel, and vehicle behavior recognition and safety performance.

[0203] 2. This invention adopts a hierarchical, grouped, and personalized tagging approach, which makes the information on ships, personnel, and vehicles more detailed. It can quickly extract and mine valuable information according to actual business needs, thereby improving data utilization and business decision-making efficiency.

[0204] 3. The label table design of this invention is flexible, and new labels can be added as business expands. At the same time, the structure is clear, which facilitates the maintenance and query of data in the later stage, and improves the convenience and scalability of data management.

[0205] 4. By defining the relationship tags between ships, personnel, and vehicles, this invention enables the effective mining of relationships between different objects, which helps to discover potential risks and problems and improves the level of industry management.

[0206] Other beneficial effects of the present invention will be explained in detail through the introduction of specific technical features and technical solutions in specific embodiments. Those skilled in the art should be able to understand the beneficial technical effects brought about by these technical features and technical solutions through the introduction of these technical features and technical solutions. Attached Figure Description

[0207] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0208] Figure 1 This is a schematic diagram of the overall process of the present invention;

[0209] Figure 2 This is a flowchart of the data acquisition process.

[0210] Figure 3 A classification diagram for data on ships, vehicles, and people;

[0211] Figure 4 This is a diagram showing the relationship between personalized tags and related tags. Detailed Implementation

[0212] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0213] This invention provides a method for creating virtual profiles of ships, vehicles, and personnel, using a tagging approach to construct virtual profiles of ships, personnel, and vehicles, referring to... Figure 1 As shown, the method specifically includes:

[0214] Step S1: Obtain ship, personnel, and vehicle data as source data for tag calculation and mining to generate ship, personnel, and vehicle tags. Source data processing employs methods such as data cleaning and data transformation for data preprocessing.

[0215] Ship data is obtained through radar detection and includes unique ID, MMSI number, ship name, heading, speed, longitude, latitude, time, status, data source, etc. Personnel data is obtained from a personnel information database and includes unique ID, name, age, gender, location, occupation, address, etc. Vehicle data is obtained through BeiDou detection and includes unique ID, VIN code, license plate number, brand, vehicle series, type (sedan, truck, agricultural vehicle, etc.), class (imported car, domestic car, joint venture car), and number of seats, etc. Source data processing utilizes big data processing frameworks such as Spark and Hadoop. During data cleaning, outlier detection and handling methods are employed to remove abnormal data and improve data quality. The specific data processing flow is as follows:

[0216] Data cleaning: By deleting (null values, garbled data), correcting (negative numbers, maximum values, minimum values), and filling (average filling, empty string filling), data quality problems are repaired to ensure the integrity and correctness of the source data.

[0217] Data conversion: Converting data of different formats and sources into a unified format to meet the needs of subsequent processing.

[0218] Step S2: Define layered labels to divide ships, personnel, and vehicles into multiple non-overlapping parts according to manually defined features, with each part being a layer.

[0219] Ships are categorized into four levels based on artificially defined characteristics: key concern, concern, general concern, and ordinary concern.

[0220] Key vessel behaviors to monitor include: abnormal navigation plans, excessive speed, lingering, and illegal border crossings.

[0221] Vessel behaviors of concern include: vessels staying in port for extended periods, vessels concentrating in one area, vessels being overloaded, vessels disabling their AIS, and vessels frequently operating at night.

[0222] Commonly concerned vessel behaviors include: vessels staying at port for longer than expected, vessels berthing in inappropriate locations, and vessels berthing in restricted areas.

[0223] Ordinary ship behavior: normal ship navigation, normal berthing, departure, port operations, etc.

[0224] Personnel are categorized into four levels based on manually defined characteristics: key focus, focus, general focus, and ordinary.

[0225] Key individuals to focus on: those involved in security threats, sensitive information, important clients, or senior executives.

[0226] People to watch: those who enter and exit frequently, frequently change their phone numbers or other personal information, exhibit abnormal interpersonal behavior, or have personal security personnel, etc.

[0227] People who are generally of concern include those who spend too much time in crowded places and those who frequently change their place of residence or workplace.

[0228] Ordinary people: have normal travel patterns and no safety hazards, abide by laws and regulations, and have no record of bad behavior, etc.

[0229] Vehicles are categorized into four levels based on manually defined characteristics: key concern, concern, general concern, and ordinary.

[0230] Vehicles requiring special attention include those involved in the transportation of dangerous or high-value goods, those with frequent ownership transfers, and those with passengers on board.

[0231] Vehicles to watch: those that frequently change license plates, speed, follow irregular routes, or repeatedly appear in locations involved in illegal activities.

[0232] Vehicles that are generally of concern include those that stay for too long, appear near sensitive locations, or have frequent changes in ownership or driver.

[0233] Ordinary vehicles: vehicles that comply with regulations, are parked in designated areas, and have no illegal activities, etc.

[0234] Step S3: Define cluster labels and cluster ships, personnel, and vehicles at each layer according to more granular, manually defined features.

[0235] The ships at each level are grouped according to more granular, artificially defined characteristics. Here, we choose to group them based on the frequency and region of their activities, mainly into the following five groups:

[0236] Vessels that frequently operate in specific areas: frequently entering and leaving a certain port and staying in a certain sea area, or areas of interest marked in advance by the platform, etc.

[0237] Vessels that frequently stay in fixed areas: They often stay at a certain dock, waterway, or sea area, and their range of activity is relatively fixed.

[0238] Vessels that sail irregularly but operate in a limited area: those that navigate in coastal waters or fixed sea areas.

[0239] Vessels that frequently sail but operate in different areas: sailing in different sea areas or ports, etc.

[0240] Vessels that remain in port or dock for an extended period: vessels that are unable to navigate normally.

[0241] Finally, grouping ships based on activity frequency and region is crucial because it reflects their operational patterns and behavioral habits. Activity frequency reflects how often a ship enters or leaves a port or stays in a particular sea area within a given timeframe, while region reflects the scope and routes of their activities. Grouping ships according to their activity frequency and region allows for a better understanding of their behavioral patterns and habits, facilitating the identification of potential safety risks and violations. For example, ships frequently active within a certain area may pose safety hazards, such as operating too close to sensitive areas or interfering with the routes of other vessels. By distinguishing these ships from those in different categories, these potential safety risks can be monitored and managed more accurately. Furthermore, grouping ships based on activity frequency and region also improves monitoring and management efficiency. For instance, during ship behavior monitoring, priority can be given to ships with higher activity frequencies or larger activity areas to better understand their behavior and risk profiles, allowing for timely and necessary measures.

[0242] The personnel at each level were grouped according to activity frequency and region based on more granular, manually defined characteristics, resulting in the following five groups:

[0243] People who frequently operate in a specific area: such as sailors, fishermen, or tourists who frequently operate in a certain port, dock, or on a specific shipping route.

[0244] People who frequently stay in a fixed area: People on ships that are moored at a certain dock or port for a long time.

[0245] People who go out irregularly but whose activity area is limited: such as sailors or fishermen whose activity range is limited to a certain sea area or river.

[0246] People who frequently travel but whose activity areas are not fixed: crew members of commercial ships or passengers on yachts that frequently sail between different waters.

[0247] People who stay on board for extended periods: crew members or captains who work and live at sea or inland waters year-round.

[0248] The vehicles at each level are grouped according to activity frequency and region based on more granular, manually defined characteristics, resulting in the following five groups:

[0249] Vehicles that frequently operate in specific sea areas: Vehicles that are used for land transportation, operations, or are permanently stationed in a specific sea area, such as trucks and container trucks in ports.

[0250] Vehicles that are frequently parked in specific areas: Vehicles that are parked for extended periods at a particular dock, parking area, or berthing area, such as transport trucks and shuttle buses.

[0251] Vehicles that travel irregularly but have limited activity areas: private cars, official vehicles, etc. that operate in specific sea areas or rivers.

[0252] Vehicles that frequently travel but whose activity areas are not fixed: trucks, buses, etc. that frequently travel between different shores and docks.

[0253] Vehicles that are parked in a fixed location for a long period of time: scrapped cars, antique cars, exhibition cars, etc., that are parked on the shore or in a specific area for a long period of time.

[0254] Step S4: Extract the basic attributes of ships, personnel, and vehicles.

[0255] The basic attributes of a ship include AIS number, ship name, ship type, ship size, ship tonnage, deadweight, year of construction, hull material, ship owner, and number of crew members. The basic attributes of a person include name, gender, age, ID number, occupation, contact information, employer, workplace, IP address, and criminal records. The basic attributes of a vehicle include license plate number, vehicle type, vehicle color, brand and model, and vehicle owner.

[0256] Step S5: Define personalized tags. Add personalized tags to ships, personnel, and vehicles based on their basic attributes and respective source data. Generate personalized tags using techniques such as behavioral analysis and association rule mining.

[0257] Personalized ship tags are generated through the ship's basic attributes or by analyzing source ship data. Examples of personalized ship tags are as follows:

[0258] Speed ​​tag: corresponds to the ship's maximum speed, cruising speed, and other attributes.

[0259] Load capacity label: corresponds to the maximum load capacity of the vessel.

[0260] Usage Environment Label: Corresponds to the vessel's usage environment, applicable routes, sea conditions, and other attributes.

[0261] Construction Year Tag: Corresponds to the construction year of the ship, whether it is an old ship, and other attributes.

[0262] Ship type label: corresponds to the ship type and whether it meets specific navigation conditions (construction materials, propulsion method, design form) and other attributes.

[0263] Ship category label: corresponds to the ship type, such as cargo ship, passenger ship, tugboat, etc.

[0264] Maintenance status label: This label indicates the maintenance status of the corresponding vessel and whether it undergoes regular maintenance.

[0265] Fuel type label: corresponds to the type of fuel used by the ship, such as diesel, natural gas, etc.

[0266] Destination tag: corresponds to the ship's destination, such as Asia, Europe, Africa, etc.

[0267] Safety label: corresponds to the ship's safety level, whether it complies with certain safety regulations, and other attributes.

[0268] Personalized personnel tags are generated through the analysis of personnel's basic attributes or source data. Examples of personalized personnel tags are as follows:

[0269] Age group labels: youth, middle-aged, elderly, etc.

[0270] Work experience tags: financial professional, IT professional, maintenance personnel, etc.

[0271] Regional tags: Residents of East China, people from rural areas in the south, etc.

[0272] Education level tags: PhD graduate, junior college graduate, etc.

[0273] Sailing habits tags: year-round sailing, occasional sailing, etc.

[0274] Violation tags: Users who violate marine environmental protection regulations, suspected of illegal fishing, etc.

[0275] Crew information tags: professional crew members, hired temporary workers, etc.

[0276] Risk level labels: high-risk area activists, low-risk area activists, etc.

[0277] Personalized vehicle tags are generated through the vehicle's basic attributes or by analyzing source vehicle data. Examples of personalized vehicle tags are listed below:

[0278] Port transportation labels: port trucks, port employee shuttle buses, etc.

[0279] Vehicle type tags: large trucks, small vans, etc.

[0280] Violation tags: speeding vehicles, overloaded vehicles, etc.

[0281] Transport volume labels: high transport volume vehicles, low transport volume vehicles, etc.

[0282] Goods type label: vehicles transporting liquid chemicals, vehicles transporting grain, etc.

[0283] Loading time tags: peak-hour loading vehicles, night-time loading vehicles, etc.

[0284] Vehicle speed labels: high-speed vehicles, low-speed vehicles, etc.

[0285] Route tags: vehicles traveling along the river, vehicles traveling on the cross-sea bridge, etc.

[0286] Personalized tags are generated using techniques such as behavioral analysis and association rule mining. For example, by analyzing ship, vehicle, and pedestrian behavior data, behavioral preferences of ships, personnel, and vehicles can be mined to generate personalized tags. The specific processing flow is as follows:

[0287] Data collection: By collecting data on the behavior of ships, vehicles, and people, we can obtain data on the behavioral preferences of ships, people, and vehicles.

[0288] Data preprocessing: Cleaning, transforming, and other preprocessing of behavioral data to meet the needs of subsequent analysis.

[0289] Pattern mining: Using techniques such as association rule mining and clustering, behavioral data is analyzed to obtain association rules, information such as ship, vehicle, and person categories, in order to generate personalized tags.

[0290] Step S6: Define the relationship labels between ships, personnel, and vehicles, and use methods such as association analysis and time series analysis to mine the relationships.

[0291] The association tags between ships, personnel, and vehicles are mainly defined manually based on actual business needs. Examples include the following:

[0292] Personnel and vehicle relationship tags: riding tag, ownership tag, etc.

[0293] Personnel-ship relationship tags: passenger tag, ownership tag, employment tag, etc.

[0294] Ship and vehicle tags: transport tags, illegal docking tags (docking on unknown shores), etc.

[0295] Furthermore, methods such as association analysis and time series analysis are used to mine relationships in source data of highly correlated ships, vehicles, and people. For example, association rule mining algorithms are used to discover frequent association patterns between ships, personnel, and vehicles, identifying highly correlated relationship tags. The specific processing flow is as follows:

[0296] Data collection: Collect interaction data between ships, personnel, and vehicles to obtain a dataset.

[0297] Association analysis: By using techniques such as association rule mining and frequent itemset mining, we analyze datasets to identify frequent association patterns between ships, people, and vehicles.

[0298] Time series analysis: Through techniques such as time series analysis and sequence pattern mining, analyze datasets to identify time series patterns among ships, personnel, and vehicles.

[0299] Step S7: Create a label table in the database, add the calculated labels to the label table, and then perform correlation analysis between the source data of ships, personnel, and vehicles and the label table according to business needs to obtain indicator data.

[0300] The tag table is designed with five fields: subject, object, feature type, feature value, and detailed description. The object field can be empty, which means that only the subject field is recorded. As the business expands, additional tags can be added directly as rows of data. When the data volume is too large, the table can be partitioned using hierarchical or clustered tags. Finally, the source data of ships, personnel, and vehicles are associated with the tag data, and the indicator data is analyzed according to business needs.

[0301] Step S8: Provide the indicator data to the virtual portrait system for boats, vehicles and people to generate charts.

[0302] The present invention also provides a system for creating virtual portraits of boats, vehicles and people. The system can be implemented by executing the process steps of the method for creating virtual portraits of boats, vehicles and people. That is, those skilled in the art can understand the method for creating virtual portraits of boats, vehicles and people as a preferred embodiment of the system.

[0303] Module M1: Acquires ship, personnel, and vehicle data as source data for tag calculation and mining, generating ship, personnel, and vehicle tags. Source data processing employs methods such as data cleaning and data transformation for data preprocessing.

[0304] Ship data is obtained through radar detection and includes unique ID, MMSI number, ship name, heading, speed, longitude, latitude, time, status, data source, etc. Personnel data is obtained from a personnel information database and includes unique ID, name, age, gender, location, occupation, address, etc. Vehicle data is obtained through BeiDou detection and includes unique ID, VIN code, license plate number, brand, vehicle series, type (sedan, truck, agricultural vehicle, etc.), class (imported car, domestic car, joint venture car), and number of seats, etc. Source data processing utilizes big data processing frameworks such as Spark and Hadoop. During data cleaning, outlier detection and handling methods are employed to remove abnormal data and improve data quality. The specific data processing flow is as follows:

[0305] Data cleaning: By deleting (null values, garbled data), correcting (negative numbers, maximum values, minimum values), and filling (average filling, empty string filling), data quality problems are repaired to ensure the integrity and correctness of the source data.

[0306] Data conversion: Converting data of different formats and sources into a unified format to meet the needs of subsequent processing.

[0307] Module M2: Defines hierarchical labels, dividing ships, personnel, and vehicles into multiple non-overlapping parts according to manually defined characteristics, with each part being a layer.

[0308] Ships are categorized into four levels based on artificially defined characteristics: key concern, concern, general concern, and ordinary concern.

[0309] Key vessel behaviors to monitor include: abnormal navigation plans, excessive speed, lingering, and illegal border crossings.

[0310] Vessel behaviors of concern include: vessels staying in port for extended periods, vessels concentrating in one area, vessels being overloaded, vessels disabling their AIS, and vessels frequently operating at night.

[0311] Commonly concerned vessel behaviors include: vessels staying at port for longer than expected, vessels berthing in inappropriate locations, and vessels berthing in restricted areas.

[0312] Ordinary ship behavior: normal ship navigation, normal berthing, departure, port operations, etc.

[0313] Personnel are categorized into four levels based on manually defined characteristics: key focus, focus, general focus, and ordinary.

[0314] Key individuals to focus on: those involved in security threats, sensitive information, important clients, or senior executives.

[0315] People to watch: those who enter and exit frequently, frequently change their phone numbers or other personal information, exhibit abnormal interpersonal behavior, or have personal security personnel, etc.

[0316] People who are generally of concern include those who spend too much time in crowded places and those who frequently change their place of residence or workplace.

[0317] Ordinary people: have normal travel patterns and no safety hazards, abide by laws and regulations, and have no record of bad behavior, etc.

[0318] Vehicles are categorized into four levels based on manually defined characteristics: key concern, concern, general concern, and ordinary.

[0319] Vehicles requiring special attention include those involved in the transportation of dangerous or high-value goods, those with frequent ownership transfers, and those with passengers on board.

[0320] Vehicles to watch: those that frequently change license plates, speed, follow irregular routes, or repeatedly appear in locations involved in illegal activities.

[0321] Vehicles that are generally of concern include those that stay for too long, appear near sensitive locations, or have frequent changes in ownership or driver.

[0322] Ordinary vehicles: vehicles that comply with regulations, are parked in designated areas, and have no illegal activities, etc.

[0323] Module M3: Defines cluster labels, and groups ships, personnel and vehicles at each layer according to more granular, manually defined features.

[0324] The ships at each level are grouped according to more granular, artificially defined characteristics. Here, we choose to group them based on the frequency and region of their activities, mainly into the following five groups:

[0325] Vessels that frequently operate in specific areas: frequently entering and leaving a certain port and staying in a certain sea area, or areas of interest marked in advance by the platform, etc.

[0326] Vessels that frequently stay in fixed areas: They often stay at a certain dock, waterway, or sea area, and their range of activity is relatively fixed.

[0327] Vessels that sail irregularly but operate in a limited area: those that navigate in coastal waters or fixed sea areas.

[0328] Vessels that frequently sail but operate in different areas: sailing in different sea areas or ports, etc.

[0329] Vessels that remain in port or dock for an extended period: vessels that are unable to navigate normally.

[0330] Finally, grouping ships based on activity frequency and region is crucial because it reflects their operational patterns and behavioral habits. Activity frequency reflects how often a ship enters or leaves a port or stays in a particular sea area within a given timeframe, while region reflects the scope and routes of their activities. Grouping ships according to their activity frequency and region allows for a better understanding of their behavioral patterns and habits, facilitating the identification of potential safety risks and violations. For example, ships frequently active within a certain area may pose safety hazards, such as operating too close to sensitive areas or interfering with the routes of other vessels. By distinguishing these ships from those in different categories, these potential safety risks can be monitored and managed more accurately. Furthermore, grouping ships based on activity frequency and region also improves monitoring and management efficiency. For instance, during ship behavior monitoring, priority can be given to ships with higher activity frequencies or larger activity areas to better understand their behavior and risk profiles, allowing for timely and necessary measures.

[0331] The personnel at each level were grouped according to activity frequency and region based on more granular, manually defined characteristics, resulting in the following five groups:

[0332] People who frequently operate in a specific area: such as sailors, fishermen, or tourists who frequently operate in a certain port, dock, or on a specific shipping route.

[0333] People who frequently stay in a fixed area: People on ships that are moored at a certain dock or port for a long time.

[0334] People who go out irregularly but whose activity area is limited: such as sailors or fishermen whose activity range is limited to a certain sea area or river.

[0335] People who frequently travel but whose activity areas are not fixed: crew members of commercial ships or passengers on yachts that frequently sail between different waters.

[0336] People who stay on board for extended periods: crew members or captains who work and live at sea or inland waters year-round.

[0337] The vehicles at each level are grouped according to activity frequency and region based on more granular, manually defined characteristics, resulting in the following five groups:

[0338] Vehicles that frequently operate in specific sea areas: Vehicles that are used for land transportation, operations, or are permanently stationed in a specific sea area, such as trucks and container trucks in ports.

[0339] Vehicles that are frequently parked in specific areas: Vehicles that are parked for extended periods at a particular dock, parking area, or berthing area, such as transport trucks and shuttle buses.

[0340] Vehicles that travel irregularly but have limited activity areas: private cars, official vehicles, etc. that operate in specific sea areas or rivers.

[0341] Vehicles that frequently travel but whose activity areas are not fixed: trucks, buses, etc. that frequently travel between different shores and docks.

[0342] Vehicles parked in a fixed location for a long period of time: such as vehicles parked on the shore or in a specific area for an extended period of time.

[0343] Module M4: Extracts basic attributes of ships, personnel, and vehicles.

[0344] The basic attributes of a ship include AIS number, ship name, ship type, ship size, ship tonnage, deadweight, year of construction, hull material, ship owner, and number of crew members. The basic attributes of a person include name, gender, age, ID number, occupation, contact information, employer, workplace, IP address, and criminal records. The basic attributes of a vehicle include license plate number, vehicle type, vehicle color, brand and model, and vehicle owner.

[0345] Module M5: Defines personalized tags, adding personalized tags to ships, personnel, and vehicles based on their basic attributes and respective source data. Personalized tags are generated using techniques such as behavioral analysis and association rule mining.

[0346] Personalized ship tags are generated through the ship's basic attributes or by analyzing source ship data. Examples of personalized ship tags are as follows:

[0347] Speed ​​tag: corresponds to the ship's maximum speed, cruising speed, and other attributes.

[0348] Load capacity label: corresponds to the maximum load capacity of the vessel.

[0349] Usage Environment Label: Corresponds to the vessel's usage environment, applicable routes, sea conditions, and other attributes.

[0350] Construction Year Tag: Corresponds to the construction year of the ship, whether it is an old ship, and other attributes.

[0351] Ship type label: corresponds to the ship type and whether it meets specific navigation conditions (construction materials, propulsion method, design form) and other attributes.

[0352] Ship category label: corresponds to the ship type, such as cargo ship, passenger ship, tugboat, etc.

[0353] Maintenance status label: This label indicates the maintenance status of the corresponding vessel and whether it undergoes regular maintenance.

[0354] Fuel type label: corresponds to the type of fuel used by the ship, such as diesel, natural gas, etc.

[0355] Destination tag: corresponds to the ship's destination, such as Asia, Europe, Africa, etc.

[0356] Safety label: corresponds to the ship's safety level, whether it complies with certain safety regulations, and other attributes.

[0357] Personalized personnel tags are generated through the analysis of personnel's basic attributes or source data. Examples of personalized personnel tags are as follows:

[0358] Age group labels: youth, middle-aged, elderly, etc.

[0359] Work experience tags: financial professional, IT professional, maintenance personnel, etc.

[0360] Regional tags: Residents of East China, people from rural areas in the south, etc.

[0361] Education level tags: PhD graduate, junior college graduate, etc.

[0362] Sailing habits tags: year-round sailing, occasional sailing, etc.

[0363] Violation tags: Users who violate marine environmental protection regulations, suspected of illegal fishing, etc.

[0364] Crew information tags: professional crew members, hired temporary workers, etc.

[0365] Risk level labels: high-risk area activists, low-risk area activists, etc.

[0366] Personalized vehicle tags are generated through the vehicle's basic attributes or by analyzing source vehicle data. Examples of personalized vehicle tags are listed below:

[0367] Port transportation labels: port trucks, port employee shuttle buses, etc.

[0368] Vehicle type tags: large trucks, small vans, etc.

[0369] Violation tags: speeding vehicles, overloaded vehicles, etc.

[0370] Transport volume labels: high transport volume vehicles, low transport volume vehicles, etc.

[0371] Goods type label: vehicles transporting liquid chemicals, vehicles transporting grain, etc.

[0372] Loading time tags: peak-hour loading vehicles, night-time loading vehicles, etc.

[0373] Vehicle speed labels: high-speed vehicles, low-speed vehicles, etc.

[0374] Route tags: vehicles traveling along the river, vehicles traveling on the cross-sea bridge, etc.

[0375] Personalized tags are generated using techniques such as behavioral analysis and association rule mining. For example, by analyzing ship, vehicle, and pedestrian behavior data, behavioral preferences of ships, personnel, and vehicles can be mined to generate personalized tags. The specific processing flow is as follows:

[0376] Data collection: By collecting data on the behavior of ships, vehicles, and people, we can obtain data on the behavioral preferences of ships, people, and vehicles.

[0377] Data preprocessing: Cleaning, transforming, and other preprocessing of behavioral data to meet the needs of subsequent analysis.

[0378] Pattern mining: Using techniques such as association rule mining and clustering, behavioral data is analyzed to obtain association rules, information such as ship, vehicle, and person categories, in order to generate personalized tags.

[0379] Module M6: Defines the relationship labels between ships, personnel, and vehicles. Relationship analysis and time-series analysis are used to uncover these relationships.

[0380] The association tags between ships, personnel, and vehicles are mainly defined manually based on actual business needs. Examples include the following:

[0381] Personnel and vehicle relationship tags: riding tag, ownership tag, etc.

[0382] Personnel-ship relationship tags: passenger tag, ownership tag, employment tag, etc.

[0383] Ship and vehicle tags: transport tags, illegal docking tags (docking on unknown shores), etc.

[0384] Furthermore, methods such as association analysis and time series analysis are used to mine relationships in source data of highly correlated ships, vehicles, and people. For example, association rule mining algorithms are used to discover frequent association patterns between ships, personnel, and vehicles, identifying highly correlated relationship tags. The specific processing flow is as follows:

[0385] Data collection: Collect interaction data between ships, personnel, and vehicles to obtain a dataset.

[0386] Association analysis: By using techniques such as association rule mining and frequent itemset mining, we analyze datasets to identify frequent association patterns between ships, people, and vehicles.

[0387] Time series analysis: Through techniques such as time series analysis and sequence pattern mining, analyze datasets to identify time series patterns among ships, personnel, and vehicles.

[0388] Module M7: Creates a label table in the database, adds the calculated labels to the label table, and then performs correlation analysis between the source data of ships, personnel, and vehicles and the label table according to business needs to obtain indicator data.

[0389] The tag table is designed with five fields: subject, object, feature type, feature value, and detailed description. The object field can be empty, which means that only the subject field is recorded. As the business expands, additional tags can be added directly as rows of data. When the data volume is too large, the table can be partitioned using hierarchical or clustered tags. Finally, the source data of ships, personnel, and vehicles are associated with the tag data, and the indicator data is analyzed according to business needs.

[0390] Module M8: Provides the indicator data to the virtual portrait system for ships, vehicles, and people to generate charts.

[0391] The present invention will now be described in more detail.

[0392] The virtual profiles of ships, vehicles, and people will be developed from the following aspects:

[0393] 1. Data collection for virtual portraits of ships, vehicles, and people, such as... Figure 2 As shown, the ship collects data through radar and AIS signal base stations, and people and vehicles access the data through their respective information databases. The collected data is cleaned and processed before being stored in the database.

[0394] 2. Classify ships, vehicles, and people according to business needs, using hierarchical and cluster labels to classify the received data, such as... Figure 3 As shown, the hierarchical tags are divided into four levels according to the degree of attention: key attention, attention, general attention, and ordinary attention. The group tags are used to group people by activity frequency and region.

[0395] 3. After receiving the behavioral data of boats, vehicles and people, extract their basic attributes and store them in three tables: ship, vehicle and people. Each table will record a unique ID and the basic attribute fields of the boat, vehicle and people. A boat, a vehicle and a person will correspond to a row of data in ship, vehicle and people.

[0396] 4. For example Figure 4As shown, personalized tags are defined for ships, vehicles, and people. Personalized tags are obtained by analyzing the source data of ships, vehicles, and people, and can also be obtained through correlation analysis. The analyzed tag data is then stored in a tag table. The tag table in the database has five fields: subject, object, feature type, feature value, and detailed description. The object field in the tag table for each ship, vehicle, and person is empty. For correlation tags, tags such as "own" and "employ" will appear, all relative to the subject. For example, a subject owns an object, or a subject employs an object. As the number of unique tags and correlation tags for ships, vehicles, and people increases, the number of rows in the tag table will also increase. The tag table can be sharded or partitioned according to hierarchical tags and group tags, depending on the actual data volume. Once the tag data is stored in the tag table, indicator analysis can be performed according to actual business needs. Analysis can be performed on a specific feature, a specific ship, a specific vehicle, or a specific person. The analyzed indicator data is stored in a separate table. For indicators with small data volumes, real-time analysis can be performed during interactive visualization.

[0397] 5. Visualization and interactive design of virtual portraits of ships, vehicles and people: Using data visualization and interactive design techniques, the results of virtual portraits are displayed in the form of images, charts, animations, etc. The data sources are indicator tables and analysis during interaction.

[0398] 6. Application scenarios of virtual profiles of ships, vehicles, and people: In practical applications, virtual profiles of ships, vehicles, and people can be applied to traffic management, security monitoring, intelligent transportation, and autonomous driving. By analyzing the virtual profiles of ships, vehicles, and people, traffic operation efficiency, safety, and risk prediction can be improved.

[0399] In summary, this invention provides a method and system for creating virtual profiles of ships, vehicles, and people. It designs a virtual profile system that comprehensively analyzes the behavioral data of ships, vehicles, and people, enabling the creation of associated tags and the performance of correlation analysis. By mining the correlations between data from different sources, types, and granularities, the accuracy of behavioral risk prediction is improved.

[0400] Strengthen data security and privacy protection measures. In addition to encryption and pseudo-naming technologies, differential privacy and secure multi-party computation techniques can be used to protect data privacy and security in distributed computing environments, thereby effectively preventing data leaks.

[0401] By employing a multi-model fusion and adaptive algorithm strategy, the virtual portrait system can adapt to the needs of different scenarios and complexities. Through continuous model updates and optimization, the accuracy and robustness of virtual portraits in various application scenarios are improved.

[0402] Those skilled in the art will understand that, besides implementing the system and its various devices, modules, and units provided by this invention in the form of purely computer-readable program code, the same functions can be achieved entirely through logical programming of the method steps, making the system and its various devices, modules, and units of this invention function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, the system and its various devices, modules, and units provided by this invention can be considered as a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered as structures within the hardware component; alternatively, the devices, modules, and units for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0403] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A method for creating virtual portraits of people in boats, vehicles, and other vehicles, characterized in that, include: Step S1: Obtain ship, personnel, and vehicle data as source data, perform preprocessing using relevant methods including data cleaning and data transformation, calculate and mine tags based on the preprocessed source data, and generate ship, personnel, and vehicle tags. Step S2: Define layered labels to divide ships, personnel, and vehicles into multiple non-overlapping parts according to manually defined features, with each part being a layer; Step S3: Define cluster labels and cluster ships, personnel, and vehicles at each layer according to more granular, manually defined features; Step S4: Extract the basic attributes of ships, personnel, and vehicles; Step S5: Define personalized tags. Add personalized tags to ships, personnel, and vehicles based on their basic attributes and respective source data. Generate personalized tags using relevant technologies, including behavioral analysis and association rule mining. Step S6: Define the relationship labels between ships, personnel, and vehicles, and use relevant methods, including association analysis and time series analysis, to mine the relationships; Step S7: Create a label table in the database, add the calculated labels to the label table, and then perform correlation analysis between the source data of ships, personnel, and vehicles and the label table according to business needs to obtain indicator data; Step S8: Provide the indicator data to the virtual portrait system for ships, vehicles, and people to generate charts; In step S6, the association tags between ships, personnel, and vehicles are mainly defined manually based on actual business needs, as follows: Personnel and vehicle relationship tags: related tags including passenger tags and ownership tags; Personnel-Vessel Relationship Labels: Related labels including passenger label, ownership label, and employment label; Ship and vehicle labels: related labels including transport labels and illegal docking labels; For highly correlated source data on ships, vehicles, and people, association analysis and time series analysis methods are used to mine the relationships. The specific processing flow is as follows: Data collection: Collect interaction data between ships, personnel, and vehicles to obtain a dataset; Association analysis: By using association rule mining and frequent itemset mining techniques, we analyze datasets to identify frequent association patterns between ships, people, and vehicles. Time series analysis: Through time series analysis and sequence pattern mining techniques, analyze datasets to identify time series patterns among ships, personnel, and vehicles; In step S7, the tag table is designed with five fields: subject, object, feature type, feature value, and detailed description. An empty object field indicates that only the subject field is recorded. Subsequent tags added as the business expands can be directly added as row data. When the data volume is too large, hierarchical or clustered tags are used to partition the table. Finally, the source data of ships, personnel, and vehicles are associated with the tag data, and the indicator data is analyzed according to business needs.

2. The method for establishing virtual portraits of boats, vehicles, and people according to claim 1, characterized in that, In step S1, the ship data is detected by radar, including relevant data such as unique ID, MMSI number, ship name, heading, speed, longitude, latitude, time, status, and data source; the personnel data is obtained from the personnel information database, including relevant data such as unique ID, name, age, gender, location, occupation, and address; the vehicle data is detected by Beidou, including relevant data such as unique ID, VIN code, license plate number, brand, vehicle series, vehicle type, vehicle class, and approved number of seats. The source data preprocessing uses the big data processing frameworks Spark and Hadoop. During data cleaning, outlier detection and handling methods are employed to remove abnormal data. The specific data processing flow is as follows: Data cleaning: By removing null values ​​and garbled data, correcting negative numbers, maximum values, and minimum values, and filling in average values ​​and empty strings, we can fix data quality problems and ensure the integrity and correctness of the source data. Data conversion: Converting data of different formats and sources into a unified format to meet the needs of subsequent processing.

3. The method for establishing virtual portraits of boats, vehicles, and people according to claim 1, characterized in that, Step S2, which defines the hierarchical labels, includes: Ships are categorized according to artificially defined characteristics into four groups: Key Concern, Concern, General Concern, and Ordinary Concern. 1) Key vessel behaviors to be monitored: abnormal navigation plans, excessive speed, lingering, and illegal border crossings. 2) Vessel behaviors of concern: including vessels exceeding port hours, vessels concentrating in one area, vessels being overloaded, vessels disabling AIS, and vessels frequently operating at night; 3) Vessel behaviors of general concern: including vessels staying longer than expected, vessels berthing in inappropriate locations, and vessels berthing in prohibited areas; 4) Ordinary ship behavior: related behaviors including normal ship navigation, normal berthing, departure and operation in port; Personnel are categorized into four groups based on manually defined characteristics: key focus, focus, general focus, and ordinary focus. 1) Key personnel to focus on: those involved in security threats, sensitive information, and important clients or senior executives; 2) People of concern: those who enter and exit frequently, frequently change their mobile phone numbers or other personal information, engage in abnormal interpersonal interactions, and have personal security personnel; 3) People of general concern: those who spend too much time in crowded places, and those who frequently change their place of residence or workplace; 4) Ordinary people: Individuals whose travel history is normal and poses no safety risks, who abide by laws and regulations, and who have no record of misconduct; Vehicles are categorized into four groups based on manually defined characteristics: Priority Concern, Concern, General Concern, and Ordinary Concern. 1) Vehicles under special attention: Vehicles involved in the transportation of dangerous goods or high-value goods, vehicles that are frequently transferred in ownership, and vehicles with people on board; 2) Vehicles under surveillance: Vehicles that frequently change license plates, speed, follow irregular routes, or repeatedly appear at locations involved in illegal activities; 3) Vehicles of general concern: Vehicles that stay for too long, appear around sensitive locations, or have frequent changes in ownership or driver; 4) Ordinary vehicles: Vehicles that comply with regulations, are parked in designated areas, and have no illegal activities.

4. The method for establishing virtual portraits of boats, vehicles, and people according to claim 1, characterized in that, The grouping labels defined in step S3 include: Ships at each level are grouped according to more granular, artificially defined characteristics, with grouping based on activity frequency and region, including: 1) Vessels that frequently operate in specific areas: Vessels that frequently enter or leave a certain port or stay in a certain sea area, or are included in the area of ​​interest pre-marked by the platform; 2) Vessels that frequently stay in fixed areas: Vessels that frequently stay in a certain dock, waterway, or sea area, with a fixed range of activity; 3) Vessels that sail irregularly but operate in limited areas: including vessels that navigate in coastal waters or fixed sea areas; 4) Vessels that frequently sail but whose operating areas are not fixed: Vessels that sail in different sea areas or ports; 5) Vessels that remain in port or dock for extended periods: vessels that are unable to navigate normally; Individuals at each level are grouped according to activity frequency and region based on more granular, manually defined characteristics, including: 1) People who frequently operate in a specific area: Crew members who frequently operate in a certain port, dock or on a specific shipping route, as well as related personnel including fishermen or tourists; 2) People who frequently stay in a fixed area: People on ships that are moored at a certain dock or port for a long time; 3) People who go out irregularly but whose activity area is limited: including crew members or fishermen whose activity range is limited to a certain sea area or river; 4) People who frequently travel but whose activity areas are not fixed: crew members of commercial vessels that frequently sail between different waters or passengers on yachts; 5) People who stay on board for extended periods: Crew members or captains and other relevant personnel who work and live at sea or inland waters year-round; The vehicles at each layer are grouped according to activity frequency and region based on more granular, manually defined characteristics, including: 1) Vehicles that frequently operate in specific sea areas: Vehicles that are used for land transport, operations, or are permanently parked in a specific sea area; 2) Vehicles that are frequently parked in specific areas: Vehicles that are parked for extended periods at a specific dock, parking area, or berthing area; 3) Vehicles that travel irregularly but have limited activity areas: including private cars and official vehicles that operate in specific sea areas or rivers; 4) Vehicles that frequently travel but whose activity areas are not fixed: related vehicles including trucks and buses that frequently travel between different shores and docks; 5) Vehicles parked in fixed locations for extended periods: Vehicles parked on the shore or in specific areas for extended periods.

5. The method for establishing virtual portraits of boats, vehicles, and people according to claim 1, characterized in that, In step S4, the basic attributes of the ship include AIS number, ship name, ship type, ship size, ship tonnage, deadweight, year of ship construction, hull material, ship owner, and number of crew members. The basic attributes of an individual include name, gender, age, ID number, occupation, contact information, employer, work location, IP address, and criminal record. The basic attributes of a vehicle include its license plate number, vehicle type, vehicle color, brand and model, and vehicle owner.

6. The method for establishing virtual portraits of boats, vehicles, and people according to claim 1, characterized in that, Step S5, which defines personalized tags, includes: Personalized ship tags are generated through the ship's basic attributes or by analyzing source ship data. Personalized ship tags are as follows: Speed ​​tag: corresponds to the ship's maximum speed, cruising speed, and other related attributes; Load capacity label: corresponds to the maximum load capacity of the vessel; Usage Environment Label: Relevant attributes of the vessel, including its usage environment, applicable routes, and sea conditions; Construction Year Tag: Corresponds to the ship's construction year and related attributes, including whether it is an old ship; Ship type tag: Relevant attributes including the ship type and whether it meets specific navigation conditions; Vessel category label: corresponds to the vessel type; Maintenance status label: Relevant attributes including the maintenance status of the vessel and whether it undergoes regular maintenance; Fuel type label: corresponds to the type of fuel used on the ship; Destination label: The destination of the vessel's voyage; Safety labels: These correspond to the ship's safety level and whether it complies with certain safety regulations. Personalized personnel tags are generated through the generation of personnel's basic attributes or by analyzing personnel source data. Personalized personnel tags are as follows: Age group tags: Relevant personnel including youth, middle-aged, and elderly; Work experience tags: financial professionals, IT professionals, maintenance personnel, and related personnel; Regional tags: Relevant personnel including residents of East China and people from rural areas in southern China; Education level tags: relevant personnel including doctoral students and junior college graduates; Navigation habit tags: Personnel including those who sail year-round and those who sail occasionally; Violation tags: Users who violate marine environmental protection regulations, and related personnel suspected of illegal fishing; Crew information tags: professional crew members, hired temporary workers, and related personnel; Risk level labels: Individuals including those active in high-risk areas and those active in low-risk areas; Personalized vehicle tags are generated through the vehicle's basic attributes or by analyzing source vehicle data. Personalized vehicle tags are as follows: Port transportation tags: related vehicles including port trucks and port employee shuttle buses; Vehicle type tags: Related vehicles including large trucks and light vans; Violation tags: Vehicles including those speeding and overloaded; Transport volume tags: Related vehicles including high transport volume vehicles and low transport volume vehicles; Goods type label: Related vehicles including liquid chemical transport vehicles and grain transport vehicles; Loading time tags: related vehicles including vehicles loading during peak hours and vehicles loading at night; Vehicle speed tags: Related vehicles including high-speed vehicles and low-speed vehicles; Route tags: Related vehicles including those traveling along the river and those traveling on the cross-sea bridge; Personalized tags are generated using technologies including behavioral analysis and association rule mining. The specific processing flow is as follows: Data collection: By collecting data on the behavior of ships, vehicles, and people, we can obtain data on the behavioral preferences of ships, people, and vehicles. Data preprocessing: Performing relevant preprocessing on behavioral data, including cleaning and transformation, to meet the needs of subsequent analysis; Pattern mining: Using association rule mining and clustering techniques, behavioral data is analyzed to obtain association rules and information on ship, vehicle, and person categories, in order to generate personalized tags.

7. A system for creating virtual portraits of people, vehicles, and ships, characterized in that: include: Module M1: Acquires ship, personnel, and vehicle data as source data, performs preprocessing using relevant methods including data cleaning and data transformation, calculates and mines tags based on the preprocessed source data, and generates ship, personnel, and vehicle tags; Module M2: Defines hierarchical labels to divide ships, personnel, and vehicles into multiple non-overlapping parts according to manually defined features, with each part being a layer; Module M3: Defines cluster labels to group ships, personnel, and vehicles at each level according to more granular, manually defined features; Module M4: Extracts basic attributes of ships, personnel, and vehicles; Module M5: Define personalized tags. Add personalized tags to ships, personnel, and vehicles based on their basic attributes and respective source data. Generate personalized tags using relevant technologies, including behavioral analysis and association rule mining. Module M6: Defines the relationship labels between ships, personnel, and vehicles, and uses relevant methods, including association analysis and time series analysis, to mine the relationships; Module M7: Creates a label table in the database, adds the calculated labels to the label table, and then performs correlation analysis between the source data of ships, personnel, and vehicles and the label table according to business needs to obtain indicator data; Module M8: Provides the indicator data to the virtual portrait system for ships, vehicles, and people to generate charts; In module M6, the association tags between ships, personnel, and vehicles are defined manually based on actual business needs, as follows: Personnel and vehicle relationship tags: related tags including passenger tags and ownership tags; Personnel-Vessel Relationship Labels: Related labels including passenger label, ownership label, and employment label; Ship and vehicle labels: related labels including transport labels and illegal docking labels; For highly correlated source data on ships, vehicles, and people, association analysis and time series analysis methods are used to mine the relationships. The specific processing flow is as follows: Data collection: Collect interaction data between ships, personnel, and vehicles to obtain a dataset; Association analysis: By using association rule mining and frequent itemset mining techniques, we analyze datasets to identify frequent association patterns between ships, people, and vehicles. Time series analysis: Through time series analysis and sequence pattern mining techniques, analyze datasets to identify time series patterns among ships, personnel, and vehicles; In module M7, the tag table is designed with five fields: subject, object, feature type, feature value, and detailed description. An empty object field indicates that only the subject field is recorded. Subsequent tags added as the business expands can be directly added as row data. When the data volume is too large, the table is partitioned using hierarchical or grouped tags. Finally, the source data of ships, personnel, and vehicles are associated with the tag data, and the indicator data is analyzed according to business needs.

8. The virtual portrait creation system for ships, vehicles, and people according to claim 7, characterized in that, In module M1, the ship data is detected by radar, including relevant data such as unique ID, MMSI number, ship name, heading, speed, longitude, latitude, time, status, and data source; the personnel data is obtained from the personnel information database, including relevant data such as unique ID, name, age, gender, location, occupation, and address; the vehicle data is detected by Beidou, including relevant data such as unique ID, VIN code, license plate number, brand, vehicle series, vehicle type, vehicle class, and approved number of seats. The source data preprocessing uses the big data processing frameworks Spark and Hadoop. During data cleaning, outlier detection and handling methods are employed to remove abnormal data. The specific data processing flow is as follows: Data cleaning: By removing null values ​​and garbled data, correcting negative numbers, maximum values, and minimum values, and filling in average values ​​and empty strings, we can fix data quality problems and ensure the integrity and correctness of the source data. Data conversion: Converting data of different formats and sources into a unified format to meet the needs of subsequent processing; The module M2 defines hierarchical tags including: Ships are categorized according to artificially defined characteristics into four groups: Key Concern, Concern, General Concern, and Ordinary Concern. 1) Key vessel behaviors to be monitored: abnormal navigation plans, excessive speed, lingering, and illegal border crossings. 2) Vessel behaviors of concern: including vessels exceeding port hours, vessels concentrating in one area, vessels being overloaded, vessels disabling AIS, and vessels frequently operating at night; 3) Vessel behaviors of general concern: including vessels staying longer than expected, vessels berthing in inappropriate locations, and vessels berthing in prohibited areas; 4) Ordinary ship behavior: related behaviors including normal ship navigation, normal berthing, departure and operation in port; Personnel are categorized into four groups based on manually defined characteristics: key focus, focus, general focus, and ordinary focus. 1) Key personnel to focus on: those involved in security threats, sensitive information, and important clients or senior executives; 2) People of concern: those who enter and exit frequently, frequently change their mobile phone numbers or other personal information, engage in abnormal interpersonal interactions, and have personal security personnel; 3) People of general concern: those who spend too much time in crowded places, and those who frequently change their place of residence or workplace; 4) Ordinary people: Individuals whose travel history is normal and poses no safety risks, who abide by laws and regulations, and who have no record of misconduct; Vehicles are categorized into four groups based on manually defined characteristics: Priority Concern, Concern, General Concern, and Ordinary Concern. 1) Vehicles under special attention: Vehicles involved in the transportation of dangerous goods or high-value goods, vehicles that are frequently transferred in ownership, and vehicles with people on board; 2) Vehicles under surveillance: Vehicles that frequently change license plates, speed, follow irregular routes, or repeatedly appear at locations involved in illegal activities; 3) Vehicles of general concern: Vehicles that stay for too long, appear around sensitive locations, or have frequent changes in ownership or driver; 4) Ordinary vehicles: Vehicles that comply with regulations, are parked in designated areas, and have no illegal activities; The clustering labels defined in module M3 include: Ships at each level are grouped according to more granular, artificially defined characteristics, with grouping based on activity frequency and region, including: 1) Vessels that frequently operate in specific areas: Vessels that frequently enter or leave a certain port or stay in a certain sea area, or are included in the area of ​​interest pre-marked by the platform; 2) Vessels that frequently stay in fixed areas: Vessels that frequently stay in a certain dock, waterway, or sea area, with a fixed range of activity; 3) Vessels that sail irregularly but operate in limited areas: including vessels that navigate in coastal waters or fixed sea areas; 4) Vessels that frequently sail but whose operating areas are not fixed: Vessels that sail in different sea areas or ports; 5) Vessels that remain in port or dock for extended periods: vessels that are unable to navigate normally; Individuals at each level are grouped according to activity frequency and region based on more granular, manually defined characteristics, including: 1) People who frequently operate in a specific area: Crew members who frequently operate in a certain port, dock or on a specific shipping route, as well as related personnel including fishermen or tourists; 2) People who frequently stay in a fixed area: People on ships that are moored at a certain dock or port for a long time; 3) People who go out irregularly but whose activity area is limited: including crew members or fishermen whose activity range is limited to a certain sea area or river; 4) People who frequently travel but whose activity areas are not fixed: crew members of commercial vessels that frequently sail between different waters or passengers on yachts; 5) People who stay on board for extended periods: Crew members or captains and other relevant personnel who work and live at sea or inland waters year-round; The vehicles at each layer are grouped according to activity frequency and region based on more granular, manually defined characteristics, including: 1) Vehicles that frequently operate in specific sea areas: Vehicles that are used for land transport, operations, or are permanently parked in a specific sea area; 2) Vehicles that are frequently parked in specific areas: Vehicles that are parked for extended periods at a specific dock, parking area, or berthing area; 3) Vehicles that travel irregularly but have limited activity areas: including private cars and official vehicles that operate in specific sea areas or rivers; 4) Vehicles that frequently travel but whose activity areas are not fixed: related vehicles including trucks and buses that frequently travel between different shores and docks; 5) Vehicles parked in fixed locations for extended periods: such as vehicles parked on the shore or in specific areas for extended periods; In module M4, the basic attributes of a ship include AIS number, ship name, ship type, ship size, ship tonnage, deadweight, year of construction, hull material, ship owner, and number of crew members. The basic attributes of an individual include name, gender, age, ID number, occupation, contact information, employer, work location, IP address, and criminal record. The basic attributes of a vehicle include its license plate number, vehicle type, vehicle color, brand and model, and vehicle owner. The personalized tags defined in module M5 include: Personalized ship tags are generated through the ship's basic attributes or by analyzing source ship data. Personalized ship tags are as follows: Speed ​​tag: corresponds to the ship's maximum speed, cruising speed, and other related attributes; Load capacity label: corresponds to the maximum load capacity of the vessel; Usage Environment Label: Relevant attributes of the vessel, including its usage environment, applicable routes, and sea conditions; Construction Year Tag: Corresponds to the ship's construction year and related attributes, including whether it is an old ship; Ship type tag: Relevant attributes including the ship type and whether it meets specific navigation conditions; Vessel category label: corresponds to the vessel type; Maintenance status label: Relevant attributes including the maintenance status of the vessel and whether it undergoes regular maintenance; Fuel type label: corresponds to the type of fuel used on the ship; Destination label: The destination of the vessel's voyage; Safety labels: These correspond to the ship's safety level and whether it complies with certain safety regulations. Personalized personnel tags are generated through the generation of personnel's basic attributes or by analyzing personnel source data. Personalized personnel tags are as follows: Age group tags: Relevant personnel including youth, middle-aged, and elderly; Work experience tags: financial professionals, IT professionals, maintenance personnel, and related personnel; Regional tags: Relevant personnel including residents of East China and people from rural areas in southern China; Education level tags: relevant personnel including doctoral students and junior college graduates; Navigation habit tags: Personnel including those who sail year-round and those who sail occasionally; Violation tags: Users who violate marine environmental protection regulations, and related personnel suspected of illegal fishing; Crew information tags: professional crew members, hired temporary workers, and related personnel; Risk level labels: Individuals including those active in high-risk areas and those active in low-risk areas; Personalized vehicle tags are generated through the vehicle's basic attributes or by analyzing source vehicle data. Personalized vehicle tags are as follows: Port transportation tags: related vehicles including port trucks and port employee shuttle buses; Vehicle type tags: Related vehicles including large trucks and light vans; Violation tags: Vehicles including those speeding and overloaded; Transport volume tags: Related vehicles including high transport volume vehicles and low transport volume vehicles; Goods type label: Related vehicles including liquid chemical transport vehicles and grain transport vehicles; Loading time tags: related vehicles including vehicles loading during peak hours and vehicles loading at night; Vehicle speed tags: Related vehicles including high-speed vehicles and low-speed vehicles; Route tags: Related vehicles including those traveling along the river and those traveling on the cross-sea bridge; Personalized tags are generated using technologies including behavioral analysis and association rule mining. The specific processing flow is as follows: Data collection: By collecting data on the behavior of ships, vehicles, and people, we can obtain data on the behavioral preferences of ships, people, and vehicles. Data preprocessing: Performing relevant preprocessing on behavioral data, including cleaning and transformation, to meet the needs of subsequent analysis; Pattern mining: Using association rule mining and clustering techniques, behavioral data is analyzed to obtain association rules and information on ship, vehicle, and person categories, in order to generate personalized tags.

Images