Fusion management and application method based on personnel spatio-temporal data

Abstract

The invention relates to the technical field of data processing, in particular to a fusion management and application method based on personnel spatio-temporal data, which realizes automatic fusion and standardized processing of spatio-temporal data of different sources by constructing an intelligent data governance mechanism. The mechanism is combined with an advanced machine learning technology, data quality can be dynamically evaluated, and a fusion strategy is automatically adjusted, so that a complex spatio-temporal data track is deeply analyzed. The method not only improves the efficiency and accuracy of data processing, but also provides stronger support for various applications, and ensures that a timely and effective decision is made in a rapidly changing environment.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method for fusion management and application based on personnel spatiotemporal data. Background Art

[0002] With the rapid development of information technology, especially the widespread application of audio and video technology and sensor technology, the collection and processing of spatiotemporal data has become more convenient. These technologies enable devices to record users' behavior trajectories in real time. For example, surveillance cameras can not only capture the movement paths of people, but also combine with sensor data to comprehensively monitor the environmental status.

[0003] The differences in timeliness, reliability and format of data from different sources increase the risk of information bias and misjudgment during the fusion process. More importantly, traditional data fusion methods rely on the combination of data extraction, cleaning and big data analysis, which makes the data fusion process lack flexibility and adaptability, and cannot be adjusted autonomously according to real-time data quality and context changes. This limitation is particularly prominent in the fields of security monitoring, traffic management, environmental monitoring, and especially in social governance, such as abnormal activities of minors and the loss of the elderly.

[0004] In the existing technology, it is common to use ETL (extraction, transformation, loading) tools to connect data to the big data platform. After the connection, the data is governed, including deduplication, format standardization, coordinate system, etc. After the governance is completed, data analysis usually relies on some fixed rules (for example, scoring mechanism: scoring data according to the quality and historical performance of the data source; personnel recognition rate: calculating the success rate of personnel recognition through an algorithm, and only including data with a recognition rate exceeding a certain threshold in the final analysis; timestamp consistency: matching data based on timestamps to ensure that data close in time are fused, thereby assuming that these data have a higher relevance in terms of timeliness.) to judge the accuracy of information. However, this rule-based judgment method often lacks flexibility and is difficult to cope with changing environments and complex data characteristics, resulting in limited accuracy and reliability of data fusion.

[0005] To this end, this application designs a method for integrated management and application based on personnel spatiotemporal data, as well as solving the above problems. Summary of the invention

[0006] In order to make up for the deficiencies of insufficient flexibility and lack of accuracy caused by fixed rules in the prior art, the present invention provides a method for fusion management and application based on personnel spatiotemporal data.

[0007] A method for integrated management and application based on personnel spatiotemporal data, characterized in that it includes the following steps: S1, obtain the type of spatiotemporal data, the application system to which it belongs, the data source unit, and the type of storage database; S2, standardize the acquired spatiotemporal data, including data format conversion, coordinate system unification, and data type consistency check to ensure the consistency of data from different sources; S3, establish a data accuracy judgment mechanism, use machine learning algorithms to evaluate the quality of various types of data, and dynamically adjust the data fusion weight; S4, implements adaptive fusion strategy, automatically selects the optimal fusion method according to the characteristics and quality of real-time data sources to achieve efficient and accurate data fusion; S5 stores and manages the fused data to support subsequent real-time analysis and decision-making applications, including report generation and visualization.

[0008] Further, in order to better implement the present invention, the spatiotemporal data in S1 includes: audio point information, video point information, sensor point information, activity venue location information, checkpoint vehicle passing records, motor vehicle information, video analysis data, Internet cafe Internet access data, sensor data, public transportation real-time information, geographic information system data, and meteorological data; The activity venue location information includes the location information of supermarkets, communities, shopping malls, parks, hotels, and Internet cafes; the sensor data includes temperature, noise, humidity, light, and air quality; the public transportation real-time information includes the location of buses and subways; the meteorological data includes temperature, wind speed, and precipitation.

[0009] Further, in order to better implement the present invention, the standardization process of S2 is as follows: Coordinate system 1: The coordinate system of the longitude and latitude in the unified data. For example, Google Maps uses the WGS84 coordinate system, Amap and Tencent Maps use the GCJ02 coordinate system, and Baidu Maps uses the BD09 coordinate system. Consistency check: Ensure that the types of all data fields are consistent, such as unifying time information into UTC format; Missing value completion: For missing data, use basic data to complete it. For example, if the Internet cafe data only contains the name of an Internet cafe but no longitude and latitude, use the Internet cafe longitude and latitude information table to complete the longitude and latitude.

[0010] Further, in order to better implement the present invention, S3 is specifically: S31, Data Quality Feature Extraction: Define key features of data quality, such as data integrity, consistency, accuracy, and timeliness; extract corresponding quality features for different types of data to establish a comprehensive evaluation model; S32, training using deep learning models: Use deep learning models to build data quality assessment models; use labeled training data sets, including known high-quality and low-quality data, to train the model so that the model can learn the features that affect data quality; S33, dynamic quality assessment: During the data access process, the quality of each piece of data is assessed in real time to generate a quality score; based on the quality score output by the model, the data is dynamically classified into high-quality, low-quality or suspicious data; for low-quality data, data cleaning and reprocessing are performed to ensure the reliability of subsequent fusion; S34, fusion weight adjustment: dynamically adjust the weights of different data sources in the fusion process according to the quality score of the data; for example, data sources with high quality scores will receive greater fusion weights, while the weights of low-quality data sources will be reduced; S35, Establish a feedback mechanism: Regularly evaluate and optimize the data quality assessment model, and continuously improve the accuracy and adaptability of the model with the introduction of new data and feedback of analysis results; Monitor the quality of the fused data output, verify the effectiveness of the model through subsequent analysis and comparison with the actual results, and iteratively optimize it.

[0011] Further, in order to better implement the present invention, the S4 is specifically: S41, for each data source connected, evaluates its characteristics in real time, including data type, format, update frequency, and data quality score, to help the system understand the advantages and limitations of each data source; S42, assigning a weight to each data source according to its quality score; calculating a weighted average, giving a higher weight to a data source with better quality; S43, establish a feedback mechanism to monitor the deviation between the accuracy of the fusion results and the actual situation, and adjust the weight of the data source and the fusion strategy according to the deviation to achieve continuous optimization; for example, if a data source performs poorly in multiple recent fusion tasks, its weight in subsequent fusions can be reduced.

[0012] The beneficial effects of the present invention are: The present invention realizes the automatic fusion and standardized processing of spatiotemporal data from different sources by building an intelligent data governance mechanism. This mechanism, combined with advanced machine learning technology, can dynamically evaluate data quality and automatically adjust fusion strategies, thereby deeply analyzing complex spatiotemporal data trajectories. This method not only improves the efficiency and accuracy of data processing, but also provides stronger support for various applications, ensuring timely and effective decision-making in a rapidly changing environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0014] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Generally, the components of the embodiments of the present invention described and shown in the accompanying drawings here can be arranged and designed in various different configurations.

[0015] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0016] Figure 1 A specific embodiment of the present invention is a method for integrated management and application based on personnel spatiotemporal data, characterized in that it includes the following steps: Step 1: Obtain the type of spatiotemporal data, the application system to which it belongs, the data source unit, the type of storage database, etc. Spatiotemporal data includes: audio point information, video point information, sensor point information, location information of activity places (such as supermarkets, communities, shopping malls, parks, hotels, Internet cafes, etc.), checkpoint vehicle passing records, motor vehicle information, video analysis data, Internet cafe Internet access data, sensor data (such as temperature, noise, humidity, light, air quality, etc.), real-time public transportation information (such as bus and subway locations), geographic information system (GIS) data, meteorological data (such as temperature, wind speed, precipitation, etc.), etc.

[0017] Step 2: Standardize the acquired spatiotemporal data, including data format conversion, coordinate system unification, data type consistency check, etc., to ensure the consistency of data from different sources. Specifically include: Coordinate system 1: The coordinate system of longitude and latitude in unified data. For example, Google Maps uses the WGS84 coordinate system, Amap and Tencent maps use the GCJ02 coordinate system, and Baidu Maps uses the BD09 coordinate system.

[0018] Consistency check: Ensure that the types of all data fields are consistent, such as unifying time information into the UTC format (YYYY-MM-DDThh:mm:ssZ).

[0019] Missing value completion: For missing data, use basic data and other methods to complete it. For example, if the Internet cafe data only contains the name of an Internet cafe but no longitude and latitude, use the Internet cafe longitude and latitude information table to complete the longitude and latitude.

[0020] Step 3: Establish a data accuracy judgment mechanism, use machine learning algorithms to evaluate the quality of various types of data, and dynamically adjust the data fusion weight. Specific details include: Step 3.1: Data quality feature extraction: Define key features of data quality, such as data completeness (proportion of missing values), consistency (degree of standardization of formats and units), accuracy (difference from known true values), timeliness (data update time), etc. Extract corresponding quality features for different types of data (video, sensor, GPS, etc.) to establish a comprehensive evaluation model.

[0021] Step 3.2: Use deep learning model training: Use deep learning models to build data quality assessment models. Use labeled training data sets (including known high-quality and low-quality data) to train the model so that the model can learn the features that affect data quality.

[0022] Step 3.3: Dynamic quality assessment: During the data access process, the quality of each piece of data is assessed in real time to generate a quality score. Based on the quality score output by the model, the data is dynamically classified as high-quality, low-quality, or suspicious data. For low-quality data, data cleaning and reprocessing are performed to ensure the reliability of subsequent fusion.

[0023] Step 3.4: Fusion weight adjustment: According to the quality score of the data, the weights of different data sources in the fusion process are dynamically adjusted. For example, data sources with high quality scores will receive greater fusion weights, while the weights of low-quality data sources will be reduced.

[0024] Step 3.5: Establish a feedback mechanism: Regularly evaluate and optimize the data quality assessment model, and continuously improve the accuracy and adaptability of the model with the introduction of new data and feedback on analysis results.

[0025] Monitor the quality of the fused data output, verify the effectiveness of the model through subsequent analysis and comparison with the actual results, and iteratively optimize it.

[0026] Step 4: Implement an adaptive fusion strategy to automatically select the optimal fusion method based on the characteristics and quality of the real-time data source to achieve efficient and accurate data fusion. Specific details include: Step 4.1: For each data source connected, evaluate its characteristics in real time, including data type, format, update frequency, data quality score, etc., to help the system understand the advantages and limitations of each data source.

[0027] Step 4.2: Assign weights to each data source based on its quality score. Calculate the weighted average, giving higher weights to data sources with better quality.

[0028] Step 4.3: Establish a feedback mechanism to monitor the deviation between the accuracy of the fusion results and the actual situation, and adjust the weight of the data source and the fusion strategy according to the deviation to achieve continuous optimization. For example, if a data source performs poorly in multiple recent fusion tasks, its weight in subsequent fusions can be reduced.

[0029] Step 5: Store and manage the fused data to support subsequent real-time analysis and decision-making applications, including report generation, visualization, etc.

[0030] This embodiment aims to achieve technical innovation through the following aspects: Data standardization: Build an automated data standardization mechanism to unify the format and standardize the spatiotemporal data from different sources, ensuring that the data meets the analysis requirements at the access stage.

[0031] Data accuracy judgment: Introduce advanced machine learning algorithms to dynamically evaluate the accuracy of various types of data and make autonomous adjustments based on data quality and contextual information to achieve higher fusion accuracy.

[0032] Adaptive fusion mechanism: Design a flexible adaptive fusion strategy to automatically adjust the fusion method and weight according to the quality and characteristics of the data source, ensuring that multi-source spatiotemporal data can still be effectively fused in complex scenarios.

[0033] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Other modifications or equivalent substitutions made to the technical solution of the present invention by ordinary technicians in the field should be included in the scope of the claims of the present invention as long as they do not depart from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for fusion management and application based on personnel spatiotemporal data, characterized in that: The following steps are involved: S1, obtain the type of spatiotemporal data, the application system to which it belongs, the data source unit, and the type of storage database; S2, standardize the acquired spatiotemporal data, including data format conversion, coordinate system unification, and data type consistency check to ensure the consistency of data from different sources; S3, establish a data accuracy judgment mechanism, use machine learning algorithms to evaluate the quality of various types of data, and dynamically adjust the data fusion weight; S4, implements adaptive fusion strategy, automatically selects the optimal fusion method according to the characteristics and quality of real-time data sources to achieve efficient and accurate data fusion; S5 stores and manages the fused data to support subsequent real-time analysis and decision-making applications, including report generation and visualization.

2. The method for integrated management and application based on personnel spatiotemporal data according to claim 1 is characterized in that: The spatiotemporal data in S1 include: audio point information, video point information, sensor point information, activity venue location information, checkpoint vehicle passing records, motor vehicle information, video analysis data, Internet cafe Internet access data, sensor data, public transportation real-time information, geographic information system data, and meteorological data; The activity venue location information includes the location information of supermarkets, communities, shopping malls, parks, hotels, and Internet cafes; the sensor data includes temperature, noise, humidity, light, and air quality; the public transportation real-time information includes the location of buses and subways; the meteorological data includes temperature, wind speed, and precipitation.

3. The method for integrated management and application based on personnel spatiotemporal data according to claim 1 is characterized in that: The standardization process of S2 is as follows: Coordinate system 1: The coordinate system of the longitude and latitude in the unified data. For example, Google Maps uses the WGS84 coordinate system, Amap and Tencent Maps use the GCJ02 coordinate system, and Baidu Maps uses the BD09 coordinate system. Consistency check: Ensure that the types of all data fields are consistent, such as unifying time information into UTC format; Missing value completion: For missing data, use basic data to complete it. For example, if the Internet cafe data only contains the name of an Internet cafe but no longitude and latitude, use the Internet cafe longitude and latitude information table to complete the longitude and latitude.

4. The method for integrated management and application based on personnel spatiotemporal data according to claim 1 is characterized in that: The S3 is specifically: S31, Data Quality Feature Extraction: Define key features of data quality, such as data integrity, consistency, accuracy, and timeliness; extract corresponding quality features for different types of data to establish a comprehensive evaluation model; S32, training using deep learning models: Use deep learning models to build data quality assessment models; use labeled training data sets, including known high-quality and low-quality data, to train the model so that the model can learn the features that affect data quality; S33, dynamic quality assessment: During the data access process, the quality of each piece of data is assessed in real time to generate a quality score; based on the quality score output by the model, the data is dynamically classified into high-quality, low-quality or suspicious data; for low-quality data, data cleaning and reprocessing are performed to ensure the reliability of subsequent fusion; S34, fusion weight adjustment: dynamically adjust the weights of different data sources in the fusion process according to the quality score of the data; for example, data sources with high quality scores will receive greater fusion weights, while the weights of low-quality data sources will be reduced; S35, Establish a feedback mechanism: Regularly evaluate and optimize the data quality assessment model, and continuously improve the accuracy and adaptability of the model with the introduction of new data and feedback of analysis results; Monitor the quality of the fused data output, verify the effectiveness of the model through subsequent analysis and comparison with the actual results, and iteratively optimize it.

5. The method for integrated management and application based on personnel spatiotemporal data according to claim 1 is characterized in that: The S4 is specifically: S41, for each data source connected, evaluates its characteristics in real time, including data type, format, update frequency, and data quality score, to help the system understand the advantages and limitations of each data source; S42, assigning weights to each data source according to its quality score; Calculate a weighted average, giving higher weight to data sources with better quality; S43, establish a feedback mechanism to monitor the deviation between the accuracy of the fusion results and the actual situation, and adjust the weight of the data source and the fusion strategy according to the deviation to achieve continuous optimization; for example, if a data source performs poorly in multiple recent fusion tasks, its weight in subsequent fusions can be reduced.

Images