Data analysis system and method based on multi-source data fusion

Abstract

The invention discloses a data analysis system and method based on multi-source data fusion, and relates to the technical field of data analysis, and the method comprises the steps: taking data of each source as target source data, and obtaining a timestamp of the target source data during data updating to obtain an updating instability coefficient and an updating delay coefficient; obtaining the number of updated data types when the target source data is updated each time to obtain an incomplete updating coefficient; obtaining a quality disqualification coefficient of the target source data according to the updated unstable coefficient, the updated delay coefficient and the updated incomplete coefficient; according to the quality disqualification coefficient of the target source data and a preset quality disqualification coefficient threshold value, whether the target source data can be added into data analysis of multi-source data fusion or not is judged; therefore, when data analysis is carried out based on multi-source data fusion, it can be determined that the data quality of the data of each source is qualified, the actual situation can not be accurately reflected, and the accuracy of final data analysis is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to a data analysis system and method based on multi-source data fusion. Background Art

[0002] Data analysis based on multi-source data fusion refers to the integration of data from different sources and forms, such as sensor data, images, text and structured data, through data fusion technology to form a comprehensive and comprehensive data set; this fusion can provide deeper insights and more accurate analysis than a single data source, supporting smarter decision-making. For example, in a smart city, by fusing multi-source data such as traffic monitoring, meteorological data, and social media information, real-time monitoring and dynamic scheduling of urban traffic can be achieved, thereby improving the efficiency and accuracy of urban management; this multi-source data fusion analysis is very extensive in application scenarios, especially in the fields of smart transportation, medical health, financial risk control, precision marketing, etc., through multi-level and multi-dimensional fusion analysis of data, the behavior and status of complex systems can be more comprehensively perceived and predicted, thereby achieving more efficient and scientific management and decision-making;

[0003] Existing data analysis methods based on multi-source data fusion often assume that the data from each source is of qualified quality and perform fusion analysis. However, if the data from a certain source is of unqualified quality and cannot accurately reflect the actual situation, it is still included in the overall data analysis, which may cause problems with the accuracy of the final data analysis. Summary of the invention

[0004] The purpose of the present invention is to solve the above-mentioned problems and provide a data analysis system and method based on multi-source data fusion.

[0005] In a first aspect of the present invention, a data analysis method based on multi-source data fusion is first proposed, the method comprising:

[0006] Taking the data from each source as the target source data, obtaining the timestamp of the target source data when it is updated to obtain the update instability coefficient and the update delay coefficient;

[0007] Obtain the number of data types updated each time the target source data is updated, and obtain an update incompleteness coefficient according to the number of updated data types;

[0008] Obtaining a quality failure coefficient of the target source data according to an update instability coefficient, an update delay coefficient, and an update incomplete coefficient;

[0009] The quality failure coefficient of the target source data is compared with the preset quality failure coefficient threshold, and based on the comparison result, it is determined whether the target source data can be included in the data analysis of multi-source data fusion.

[0010] Optionally, obtaining the timestamp of the updated data of the target source data to obtain the updated instability coefficient includes:

[0011] Obtain the actual timestamps of the target source data when the data is updated within a preset time period to obtain an actual timestamp set; and calculate the absolute differences between all adjacent timestamps in the actual timestamp set to obtain an actual update time interval set;

[0012] Calculate the average update time of the actual update time interval set. The calculation formula is: In the formula, is the mean update time of the actual update time interval set, D z represents the zth update time interval in the actual update time interval set, and w represents the total number of update time intervals in the actual update time interval set;

[0013] Calculate the update instability coefficient of the target source data. The calculation formula is:

[0014]

[0015] Where Dcx is the update instability coefficient of the target source data.

[0016] Optionally, obtaining the update delay coefficient by the timestamp of the update data of the target source data includes:

[0017] Obtain the actual timestamps of the target source data when the data is updated within a preset time period to obtain an actual timestamp set; and calculate the absolute differences between all adjacent timestamps in the actual timestamp set to obtain an actual update time interval set;

[0018] Compare each update time interval in the actual update time interval set with the preset maximum update time interval, and if the update time interval is not less than the preset maximum update time interval, record the corresponding update time interval as an abnormal update time interval;

[0019] The abnormal interval quantity coefficient is obtained by dividing the total number of abnormal update time intervals by the total number of update time intervals in the actual update time interval set. The calculation formula is: Wac=r / w, where Wac is the abnormal interval quantity coefficient, r is the total number of abnormal update time intervals, and w is the total number of update time intervals in the actual update time interval set.

[0020] Optionally, obtaining the timestamp of the update data of the target source data to obtain the update delay coefficient also includes:

[0021] Obtain the absolute difference between each abnormal update time interval and the preset longest update time interval to obtain the abnormal interval deviation coefficient; the calculation formula is: Where Rsd is the abnormal interval deviation coefficient, v represents the order number of the abnormal update time interval, Q is the longest update time interval, and r represents the total number of abnormal update time intervals;

[0022] The update delay coefficient is obtained according to the abnormal interval quantity coefficient and the abnormal interval deviation coefficient. The calculation formula is: PU = a1×Wac+a2×Rsd, where PU is the update delay coefficient of the target source data, Wac and Rsd are the abnormal interval quantity coefficient and the abnormal interval deviation coefficient, respectively, a1 and a2 are the preset proportional coefficients of Wac and Rsd, respectively, and a1 and a2 are both greater than 0.

[0023] Optionally, obtaining the updated incomplete coefficient according to the number of updated data types includes:

[0024] The number of data types updated each time the target source data is updated within a preset time period is obtained as the actual number of updated data types, and the preset number of standard updated data types is obtained each time the data is updated, and the actual number of updated data types and the preset number of standard updated data types are marked as F respectively. m and F, where m represents the order number of the target source data update times within the preset time period, m=1, 2, 3, 4, ..., g, g represents the total number of target source data update times, and g is a positive integer;

[0025] Calculate the update incompleteness coefficient of the target source data. The calculation formula is: Where Ulk is the update incompleteness coefficient of the target source data.

[0026] Optionally, obtaining a quality failure coefficient of the target source data according to the update instability coefficient, the update delay coefficient, and the update incomplete coefficient includes:

[0027]

[0028] Where Ekv is the quality unqualified coefficient of the target source data, Dcx, PU and Ulk are the update instability coefficient, update delay coefficient and update incomplete coefficient respectively, b1, b2 and b3 are the preset proportional coefficients of Dcx, PU and Ulk respectively, and b1, b2 and b3 are all greater than 0.

[0029] Optionally, the data analysis for determining whether the target source data can be added to the multi-source data fusion according to the comparison result includes:

[0030] Compare the quality failure coefficient of the target source data with the preset quality failure coefficient threshold. If the quality failure coefficient of the target source data is not less than the preset quality failure coefficient threshold, it means that the quality of the target source data is unqualified and the target source data cannot be included in the data analysis of multi-source data fusion.

[0031] If the quality unqualified coefficient of the target source data is less than the preset quality unqualified coefficient threshold, it means that the quality of the target source data is qualified, and the target source data can be added to the data analysis of multi-source data fusion.

[0032] In a second aspect of the present invention, a data analysis system based on multi-source data fusion is proposed, the system comprising:

[0033] Update time module: takes the data from each source as the target source data, obtains the timestamp of the update data of the target source data, and obtains the update instability coefficient and update delay coefficient;

[0034] Update incomplete module: obtain the number of data types updated each time the target source data is updated, and obtain the update incomplete coefficient according to the number of updated data types;

[0035] Unqualified quality module: obtains the unqualified quality coefficient of the target source data according to the update instability coefficient, update delay coefficient and update incomplete coefficient;

[0036] Analysis and judgment module: compares the quality failure coefficient of the target source data with the preset quality failure coefficient threshold, and judges whether the target source data can be included in the data analysis of multi-source data fusion based on the comparison results.

[0037] Beneficial effects of the present invention:

[0038] The present invention proposes a data analysis system and method based on multi-source data fusion, which takes the data from each source as the target source data, obtains the timestamp when the target source data is updated, and obtains the update incomplete coefficient; obtains the number of data types updated each time the target source data is updated, and obtains the update incomplete coefficient; obtains the quality unqualified coefficient of the target source data according to the update instability coefficient, the update delay coefficient and the update incomplete coefficient; and judges whether the target source data can be added to the data analysis of multi-source data fusion according to the quality unqualified coefficient of the target source data and a preset quality unqualified coefficient threshold; in this way, when performing data analysis based on multi-source data fusion, it is possible to determine whether the data quality of the data from each source is qualified and whether it can accurately reflect the actual situation, thereby ensuring the accuracy of the final data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The present invention will be further described below in conjunction with the accompanying drawings.

[0040] Figure 1 It is a flow chart of the data analysis method based on multi-source data fusion;

[0041] Figure 2 This is a framework diagram of a data analysis system based on multi-source data fusion. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of the present invention.

[0044] The embodiment of the present invention provides a data analysis method based on multi-source data fusion. Figure 1 , Figure 1 A flow chart of a data analysis method based on multi-source data fusion provided by an embodiment of the present invention. The method comprises the following steps:

[0045] Taking the data from each source as the target source data, obtaining the timestamp of the target source data when it is updated to obtain the update instability coefficient and the update delay coefficient;

[0046] Obtain the number of data types updated each time the target source data is updated, and obtain an update incompleteness coefficient according to the number of updated data types;

[0047] Obtaining a quality failure coefficient of the target source data according to an update instability coefficient, an update delay coefficient, and an update incomplete coefficient;

[0048] The quality failure coefficient of the target source data is compared with the preset quality failure coefficient threshold, and based on the comparison result, it is determined whether the target source data can be included in the data analysis of multi-source data fusion.

[0049] Based on the data analysis method based on multi-source data fusion provided by the embodiment of the present invention, through the above-mentioned method, when performing data analysis based on multi-source data fusion, it is possible to determine whether the data quality of the data from each source is qualified and whether it can accurately reflect the actual situation, thereby ensuring the accuracy of the final data analysis.

[0050] In one embodiment, obtaining the timestamp of the updated data of the target source data to obtain the updated instability coefficient includes:

[0051] Obtain the actual timestamps of the target source data when the data is updated within a preset time period to obtain an actual timestamp set; and calculate the absolute differences between all adjacent timestamps in the actual timestamp set to obtain an actual update time interval set;

[0052] Calculate the average update time of the actual update time interval set. The calculation formula is: In the formula, is the mean update time of the actual update time interval set, D z represents the first z update time intervals in the actual update time interval set, and w represents the total number of update time intervals in the actual update time interval set;

[0053] Calculate the update instability coefficient of the target source data. The calculation formula is:

[0054]

[0055] Where Dcx is the update instability coefficient of the target source data.

[0056] It should be noted that the preset time period is set by professionals based on actual circumstances and is not limited or elaborated on in detail.

[0057] It should be noted that the update process of the target source data can be obtained through the recording system of the monitoring system of the multi-source data fusion platform, and the timestamp of each data update can be recorded to form an actual timestamp set; these timestamps can be obtained through database query, log record or API acquisition; it can also be other acquisition methods, which are not limited or elaborated in detail.

[0058] It should be noted that the larger the update instability coefficient of the target source data, the greater the degree of unqualified quality of the target source data, and the data cannot accurately reflect the actual situation. If it is still included in the overall data analysis, it may cause problems with the accuracy of the final data analysis. The reason is that the larger the update instability coefficient of the target source data, the greater the fluctuation of the update interval of the data, reflecting the irregularity and instability of the data update. This instability will cause the data to fail to reflect the actual situation in a timely and accurate manner, especially in application scenarios that require real-time monitoring and rapid response. If such data is still included in the multi-source data fusion analysis, it will have a negative impact on the final analysis results, because unstable data may introduce errors or delays, destroy the consistency between data, and even lead to wrong conclusions or decisions. For example, in intelligent traffic analysis, if the update frequency of some traffic data is unstable, it may incorrectly reflect the traffic conditions, resulting in inaccurate traffic scheduling and optimization plans. Therefore, in order to ensure the accuracy of data analysis, those data sources with unstable updates must be excluded to avoid affecting the overall analysis accuracy and decision-making effect.

[0059] In one embodiment, obtaining the update delay coefficient by obtaining the timestamp of the update data of the target source data includes:

[0060] Obtain the actual timestamps of the target source data when the data is updated within a preset time period to obtain an actual timestamp set; and calculate the absolute differences between all adjacent timestamps in the actual timestamp set to obtain an actual update time interval set;

[0061] Compare each update time interval in the actual update time interval set with the preset maximum update time interval, and if the update time interval is not less than the preset maximum update time interval, record the corresponding update time interval as an abnormal update time interval;

[0062] The abnormal interval quantity coefficient is obtained by dividing the total number of abnormal update time intervals by the total number of update time intervals in the actual update time interval set. The calculation formula is: Wac=r / w, where Wac is the abnormal interval quantity coefficient, r is the total number of abnormal update time intervals, and w is the total number of update time intervals in the actual update time interval set.

[0063] In one embodiment, obtaining the timestamp of the update data of the target source data to obtain the update delay coefficient further includes:

[0064] Obtain the absolute difference between each abnormal update time interval and the preset longest update time interval to obtain the abnormal interval deviation coefficient; the calculation formula is: Where Rsd is the abnormal interval deviation coefficient, v represents the order number of the abnormal update time interval, Q is the longest update time interval, and r represents the total number of abnormal update time intervals;

[0065] The update delay coefficient is obtained according to the abnormal interval quantity coefficient and the abnormal interval deviation coefficient. The calculation formula is: PU = a1×Wac+a2×Rsd, where PU is the update delay coefficient of the target source data, Wac and Rsd are the abnormal interval quantity coefficient and the abnormal interval deviation coefficient, respectively, a1 and a2 are the preset proportional coefficients of Wac and Rsd, respectively, and a1 and a2 are both greater than 0.

[0066] It should be noted that a1 and a2 are set by professionals according to actual conditions. Generally, the sum of a1 and a2 is 1. For example, a1 and a2 can be 0.5 and 0.5 respectively, or other numbers, which are not specifically limited.

[0067] It should be noted that the preset maximum update time interval is set by professionals based on actual conditions and is not specifically limited or elaborated.

[0068] It should be noted that the larger the update delay coefficient of the target source data, the greater the degree of unqualified quality of the target source data, and the data cannot accurately reflect the actual situation. If it is still included in the overall data analysis, it may cause problems with the accuracy of the final data analysis. The reason is that the larger the update delay coefficient of the target source data, the longer the update delay or the more abnormal intervals of the data source, which means that the data is not updated in time or is updated inaccurately at certain time points. This delay will increase the deviation between the data and the actual situation. Especially in scenarios that require rapid response and real-time decision-making, the timeliness and accuracy of the data are crucial; if such data with large delays is still included in the multi-source data fusion analysis, it may cause the final analysis results to lag behind the actual situation, thereby producing misleading conclusions; for example, in an intelligent transportation system, if there is a large update delay in traffic flow data, the system may not be able to adjust the traffic signal in time, resulting in the failure to solve the traffic congestion problem in time; therefore, if the data source with large update delay continues to participate in the analysis, it will significantly reduce the accuracy of the data analysis and even affect the correctness and efficiency of the decision-making;

[0069] In one embodiment, obtaining the updated incomplete coefficient according to the updated number of data types includes:

[0070] The number of data types updated each time the target source data is updated within a preset time period is obtained as the actual number of updated data types, and the preset number of standard updated data types is obtained each time the data is updated, and the actual number of updated data types and the preset number of standard updated data types are marked as F respectively. mand F, where m represents the order number of the target source data update times within the preset time period, m=1, 2, 3, 4, ..., g, g represents the total number of target source data update times, and g is a positive integer;

[0071] Calculate the update incompleteness coefficient of the target source data. The calculation formula is: Where Ulk is the update incompleteness coefficient of the target source data.

[0072] It should be noted that the number of updated data types refers to the number of data of different categories or dimensions involved in each data update. For example, in an intelligent transportation system, the number of updated data types may include multiple different data types such as traffic flow, vehicle speed, road conditions, weather conditions, etc. If an update only contains traffic flow data and does not update vehicle speed or weather conditions, the number of updated data types will be small; another example is in a medical and health system, where updated data types may include a patient's body temperature, blood pressure, heart rate, and diagnostic information. If an update only involves body temperature data and other information is not updated, the number of updated data types will also be small; in addition, the preset number of standard updated data types for each data update is set by professionals based on actual conditions, and no specific restrictions or elaborations are made.

[0073] It should be noted that the larger the update incompleteness coefficient of the target source data, the greater the degree of unqualified quality of the target source data, which cannot accurately reflect the actual situation. If it is still added to the overall data analysis, it may cause problems with the accuracy of the final data analysis. The reason is that the larger the update incompleteness coefficient of the target source data, the less expected data type is expected each time the data is updated, resulting in incomplete dimensions and content of the data; such incomplete data cannot fully reflect the actual situation, especially in application scenarios involving interactive analysis of multi-source data types, the lack of some key information will lead to distorted analysis results; for example, in smart city management, if the update of traffic data lacks road conditions or weather information, then the traffic forecast based on these incomplete data will be inaccurate, which may lead to errors in traffic scheduling decisions. Therefore, if such incomplete data is still involved in multi-source data fusion analysis, it will seriously affect the final analysis accuracy and may lead to wrong conclusions and decisions.

[0074] In one embodiment, obtaining the quality failure coefficient of the target source data according to the update instability coefficient, the update delay coefficient and the update incomplete coefficient includes:

[0075]

[0076] Where Ekv is the quality unqualified coefficient of the target source data, Dcx, PU and Ulk are the update instability coefficient, update delay coefficient and update incomplete coefficient respectively, b1, b2 and b3 are the preset proportional coefficients of Dcx, PU and Ulk respectively, and b1, b2 and b3 are all greater than 0.

[0077] It should be noted that b1, b2, and b3 are set by professionals according to actual conditions. Generally, the sum of b1, b2, and b3 is 1. For example, b1, b2, and b3 can be 0.3, 0.4, and 0.3, respectively, or other numbers, without specific limitation.

[0078] In one embodiment, the data analysis for determining whether the target source data can be added to the multi-source data fusion according to the comparison result includes:

[0079] Compare the quality failure coefficient of the target source data with the preset quality failure coefficient threshold. If the quality failure coefficient of the target source data is not less than the preset quality failure coefficient threshold, it means that the quality of the target source data is unqualified and the target source data cannot be included in the data analysis of multi-source data fusion.

[0080] If the quality unqualified coefficient of the target source data is less than the preset quality unqualified coefficient threshold, it means that the quality of the target source data is qualified, and the target source data can be included in the data analysis of multi-source data fusion.

[0081] It should be noted that the preset unqualified quality coefficient threshold is set by professionals based on actual conditions, and no detailed description or limitation is given.

[0082] In one implementation, through the above method, when performing data analysis based on multi-source data fusion, it is possible to determine whether the data quality of the data from each source is qualified and whether it can accurately reflect the actual situation, thereby ensuring the accuracy of the final data analysis.

[0083] Based on the same inventive concept, the present invention also provides a data analysis system based on multi-source data fusion. Figure 2 , Figure 2 A framework diagram of a data analysis system based on multi-source data fusion provided by an embodiment of the present invention, the system comprising:

[0084] Update time module: takes the data from each source as the target source data, obtains the timestamp of the update data of the target source data, and obtains the update instability coefficient and update delay coefficient;

[0085] Update incomplete module: obtain the number of data types updated each time the target source data is updated, and obtain the update incomplete coefficient according to the number of updated data types;

[0086] Unqualified quality module: obtains the unqualified quality coefficient of the target source data according to the update instability coefficient, update delay coefficient and update incomplete coefficient;

[0087] Analysis and judgment module: compares the quality failure coefficient of the target source data with the preset quality failure coefficient threshold, and judges whether the target source data can be included in the data analysis of multi-source data fusion based on the comparison results.

[0088] Based on the data analysis system based on multi-source data fusion provided by the embodiment of the present invention, through the above-mentioned method, when performing data analysis based on multi-source data fusion, it is possible to determine whether the data quality of the data from each source is qualified and whether it can accurately reflect the actual situation, thereby ensuring the accuracy of the final data analysis.

[0089] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A data analysis method based on multi-source data fusion, characterized in that: The following steps are involved: Taking the data from each source as the target source data, obtaining the timestamp of the target source data when it is updated to obtain the update instability coefficient and the update delay coefficient; Obtain the number of data types updated each time the target source data is updated, and obtain an update incompleteness coefficient according to the number of updated data types; Obtaining a quality failure coefficient of the target source data according to an update instability coefficient, an update delay coefficient, and an update incomplete coefficient; The quality failure coefficient of the target source data is compared with the preset quality failure coefficient threshold, and based on the comparison result, it is determined whether the target source data can be included in the data analysis of multi-source data fusion.

2. The data analysis method based on multi-source data fusion according to claim 1 is characterized in that: The timestamp of the updated data of the target source data is obtained to obtain the update instability coefficient including: Obtain the actual timestamps of the target source data when the data is updated within a preset time period to obtain an actual timestamp set; and calculate the absolute differences between all adjacent timestamps in the actual timestamp set to obtain an actual update time interval set; Calculate the average update time of the actual update time interval set. The calculation formula is: In the formula, is the mean update time of the actual update time interval set, D z represents the zth update time interval in the actual update time interval set, and w represents the total number of update time intervals in the actual update time interval set; Calculate the update instability coefficient of the target source data. The calculation formula is: Where Dcx is the update instability coefficient of the target source data.

3. The data analysis method based on multi-source data fusion according to claim 1 is characterized in that: The update delay coefficient obtained by obtaining the timestamp of the update data of the target source data includes: Obtain the actual timestamps of the target source data when the data is updated within a preset time period to obtain an actual timestamp set; and calculate the absolute differences between all adjacent timestamps in the actual timestamp set to obtain an actual update time interval set; Compare each update time interval in the actual update time interval set with the preset maximum update time interval, and if the update time interval is not less than the preset maximum update time interval, record the corresponding update time interval as an abnormal update time interval; The abnormal interval quantity coefficient is obtained by dividing the total number of abnormal update time intervals by the total number of update time intervals in the actual update time interval set. The calculation formula is: Wac=r / w, where Wac is the abnormal interval quantity coefficient, r is the total number of abnormal update time intervals, and w is the total number of update time intervals in the actual update time interval set.

4. The data analysis method based on multi-source data fusion according to claim 3 is characterized in that: The timestamp of the updated data of the target source data is obtained to obtain the update delay coefficient, which also includes: Obtain the absolute difference between each abnormal update time interval and the preset longest update time interval to obtain the abnormal interval deviation coefficient; the calculation formula is: Where Rsd is the abnormal interval deviation coefficient, v represents the order number of the abnormal update time interval, Q is the longest update time interval, and r represents the total number of abnormal update time intervals; The update delay coefficient is obtained according to the abnormal interval quantity coefficient and the abnormal interval deviation coefficient. The calculation formula is: PU = a1×Wac+a2×Rsd, where PU is the update delay coefficient of the target source data, Wac and Rsd are the abnormal interval quantity coefficient and the abnormal interval deviation coefficient, respectively, a1 and a2 are the preset proportional coefficients of Wac and Rsd, respectively, and a1 and a2 are both greater than 0.

5. The data analysis method based on multi-source data fusion according to claim 1 is characterized in that: The update incomplete coefficients obtained based on the number of data types updated include: The number of data types updated each time the target source data is updated within a preset time period is obtained as the actual number of updated data types, and the preset number of standard updated data types is obtained each time the data is updated, and the actual number of updated data types and the preset number of standard updated data types are marked as F respectively. m and F, where m represents the order number of the target source data update times within the preset time period, g=1, 2, 3, 4, ..., g, g represents the total number of target source data update times, and g is a positive integer; Calculate the update incompleteness coefficient of the target source data. The calculation formula is: Where Ulk is the update incompleteness coefficient of the target source data.

6. The data analysis method based on multi-source data fusion according to claim 1 is characterized in that: The quality unqualified coefficients of the target source data obtained according to the update instability coefficient, update delay coefficient and update incomplete coefficient include: Where Ekv is the quality unqualified coefficient of the target source data, Dcx, PU and Ulk are the update instability coefficient, update delay coefficient and update incomplete coefficient respectively, b1, b2 and b3 are the preset proportional coefficients of Dcx, PU and Ulk respectively, and b1, b2 and b3 are all greater than 0.

7. The data analysis method based on multi-source data fusion according to claim 1 is characterized in that: Data analysis based on the comparison results to determine whether the target source data can be added to the multi-source data fusion includes: Compare the quality failure coefficient of the target source data with the preset quality failure coefficient threshold. If the quality failure coefficient of the target source data is not less than the preset quality failure coefficient threshold, it means that the quality of the target source data is unqualified and the target source data cannot be included in the data analysis of multi-source data fusion. If the quality unqualified coefficient of the target source data is less than the preset quality unqualified coefficient threshold, it means that the quality of the target source data is qualified, and the target source data can be included in the data analysis of multi-source data fusion.

8. A data analysis system based on multi-source data fusion, used to implement the data analysis method based on multi-source data fusion according to any one of claims 1 to 7, characterized in that: The system comprises: Update time module: takes the data from each source as the target source data, obtains the timestamp of the target source data when it is updated, and obtains the update instability coefficient and update delay coefficient; Update incomplete module: obtain the number of data types updated each time the target source data is updated, and obtain the update incomplete coefficient according to the number of updated data types; Unqualified quality module: obtains the unqualified quality coefficient of the target source data according to the update instability coefficient, update delay coefficient and update incomplete coefficient; Analysis and judgment module: compares the quality failure coefficient of the target source data with the preset quality failure coefficient threshold, and judges whether the target source data can be included in the data analysis of multi-source data fusion based on the comparison results.

Images