Multi-dimensional account checking method and system based on group meal platform

Abstract

The invention relates to the technical field of data processing, in particular to a multi-dimensional account checking method and system based on a group meal platform, and the method comprises the steps: determining a child node state according to a stability magnitude and an association index, and determining a reference data selection mode according to the child node state; in single node direct selection, taking the to-be-reconciliated data corresponding to the child node with the maximum comprehensive evaluation value as reference data; in multi-node compensation selection, a compensation mode is determined according to the steady-state cross correlation degree and the steady-state equilibrium coefficient to select reference data; determining a reconciliation sub-node selection mode according to the reference comparison coefficient of the to-be-reconciliated data corresponding to the sub-node and the reference data and the cross correlation coefficient; the account checking processing efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a multi-dimensional reconciliation method and system based on a group dining platform. Background Art

[0002] As group dining platforms expand, transaction data volumes surge. Traditional manual reconciliation methods are no longer able to efficiently process this massive volume of data. Furthermore, abnormal data can easily be lost amidst normal transactions, resulting in poor reconciliation efficiency. Therefore, selecting key data for reconciliation to improve reconciliation efficiency is a pressing technical challenge facing those skilled in the art.

[0003] Chinese Patent Publication No. CN117149797B discloses a reconciliation method and system based on multi-dimensional data monitoring, including: determining the account update status values of different servers based on account update processing failure data and account update interruption data of different servers, and determining the probability of reconciliation problems based on the account update status values; determining the account update data of different servers based on the server's operation log data, and determining the account update busyness through the account update data of different servers; monitoring the operation data of different servers to obtain monitoring operation data, and determining the overall problem probability based on the reconciliation problem probability and account update busyness; and when the overall problem probability does not meet the requirements or reaches the preset reconciliation period, reconciling the account data of different servers. It can be seen that the above technical solution has the following problems: it mainly relies on the overall problem probability and the preset reconciliation period for reconciliation processing, lacks in-depth mining of the intrinsic correlation between different account data, and does not select reference data for dynamic optimization of the comparison benchmark, resulting in poor reconciliation processing efficiency and accuracy. Summary of the Invention

[0004] To this end, the present invention provides a multi-dimensional reconciliation method and system based on a group dining platform, which is used to overcome the problems in the prior art of lacking in-depth mining of the intrinsic correlations between different account data and failing to select reference data for dynamic optimization of the comparison benchmark, resulting in poor reconciliation processing efficiency and accuracy.

[0005] To achieve the above objectives, the present invention provides a multi-dimensional reconciliation method based on a group dining platform, comprising:

[0006] Determine the subnode status according to the steady-state measurement value and the correlation index, and determine the reference data selection method according to the subnode status. The reference data selection method is single-node direct selection or multi-node compensation selection;

[0007] In direct selection of a single node, the to-be-reconciled data corresponding to the sub-node with the largest comprehensive evaluation value is used as the reference data;

[0008] In multi-node compensation selection, the compensation method is determined based on the steady-state mutual correlation and the steady-state balance coefficient to select reference data. The compensation method is partial compensation of multiple stable sub-nodes or associated compensation of reference stable sub-nodes.

[0009] The reconciliation sub-node selection method is determined based on the reference comparison coefficient and mutual correlation coefficient between the data to be reconciled and the reference data corresponding to the sub-node. The reconciliation sub-node selection method is selected based on the interaction influence coefficient or comparison threshold;

[0010] The characteristic data is determined based on a combined bias influence and a balance coefficient, and the combined bias influence is determined based on a bias reference value of a cluster corresponding to the data to be reconciled.

[0011] Furthermore, if the state of the child node is that the steady-state quantity value is greater than or equal to the preset steady-state quantity value and the correlation index is greater than or equal to the preset correlation index, the reference data selection method is single-node direct selection.

[0012] Furthermore, if the state of the child node is that the steady-state quantity value is less than the preset steady-state quantity value or the correlation index is less than the preset correlation index, the reference data selection method is multi-node compensation selection.

[0013] Furthermore, if the steady-state mutual correlation is greater than or equal to the preset steady-state mutual correlation and the steady-state equalization coefficient is greater than or equal to the preset steady-state equalization coefficient, the compensation method is multi-stable sub-node partial compensation;

[0014] In the partial compensation of multi-stable sub-nodes, the sub-nodes are selected according to the correlation deviation and the multi-dimensional covariance weight, and the feature data in each selected sub-node are selected as reference data based on the difference coefficient and the category balance.

[0015] Furthermore, if the steady-state mutual correlation is less than the preset steady-state mutual correlation or the steady-state equalization coefficient is less than the preset steady-state equalization coefficient, the compensation method is reference steady-state sub-node correlation compensation;

[0016] Reference steady-state subnode associated compensation includes:

[0017] Determine the reference sub-node according to the evaluation threshold, and select the compensation sub-node according to the sub-influence coefficient and the combination of associated nodes;

[0018] Determine the data compensation range corresponding to each compensation sub-node based on the cross-influence threshold and the iterative correlation degree;

[0019] Determine the supplementary selection method based on the deviation reference coefficient of the data compensation range corresponding to each compensation sub-node;

[0020] All feature data in the reference sub-node and feature data in the data compensation range corresponding to each compensation sub-node are used as reference data.

[0021] Furthermore, the supplementary selection method is determined according to the deviation reference coefficient of the data compensation range corresponding to each compensation sub-node, including:

[0022] If the deviation reference value is greater than or equal to the preset deviation reference value, the supplementary selection method is to increase the number of compensation sub-nodes;

[0023] If the deviation reference value is less than the preset deviation reference value, the supplementary selection method is to increase the adjustment for the data compensation range.

[0024] Furthermore, the clustering method is determined according to the data anomaly and increment coefficient corresponding to each child node, including:

[0025] For a single child node,

[0026] If the data variability is greater than or equal to the preset data variability or the increment coefficient is greater than or equal to the preset increment coefficient, the clustering method is to split the cluster according to the coefficient of variation;

[0027] If the data variability is less than the preset data variability and the increment coefficient is less than the preset increment coefficient, the clustering method is to perform overall clustering based on the correlation coefficient.

[0028] Furthermore, the characteristic data is the data to be reconciled whose combined bias influence is less than a preset combined bias influence and whose balance coefficient is greater than or equal to a preset balance coefficient.

[0029] Furthermore, the selection method of the reconciliation sub-node is determined based on the reference comparison coefficient and the mutual correlation coefficient between the to-be-reconciled data and the reference data corresponding to the sub-node, including:

[0030] If the reference comparison coefficient is greater than or equal to the preset reference comparison coefficient and the mutual correlation coefficient is greater than or equal to the preset mutual correlation coefficient, the reconciliation sub-node is selected based on the interaction influence coefficient;

[0031] If the reference comparison coefficient is less than the preset reference comparison coefficient or the mutual correlation coefficient is less than the preset mutual correlation coefficient, the reconciliation sub-node is selected based on the comparison threshold.

[0032] The present invention also provides a multi-dimensional reconciliation system based on a group dining platform, comprising:

[0033] A data collection module, which includes several sub-nodes for data collection;

[0034] A state analysis module, connected to the data acquisition module, is used to determine the state of the subnode based on the steady-state value and the correlation index, and to determine the reference data selection method based on the subnode state. The reference data selection method is single-node direct selection or multi-node compensation selection;

[0035] A first selection module, connected to the status analysis module, is used to use the to-be-reconciled data corresponding to the subnode with the largest comprehensive evaluation value as reference data in the direct selection of a single node;

[0036] A second selection module, connected to the state analysis module, is used to determine a compensation method to select reference data according to the steady-state mutual correlation and the steady-state balance coefficient in multi-node compensation selection, wherein the compensation method is partial compensation of multiple stable sub-nodes or associated compensation of reference stable sub-nodes;

[0037] a reconciliation selection module, connected to the first selection module and the second selection module, respectively, for determining a reconciliation sub-node selection method based on a reference comparison coefficient and a mutual correlation coefficient between the to-be-reconciled data and reference data corresponding to the sub-node, wherein the reconciliation sub-node selection method is selected based on an interaction influence coefficient or a comparison threshold;

[0038] The characteristic data is determined based on a combined bias influence and a balance coefficient, and the combined bias influence is determined based on a bias reference value of a cluster corresponding to the data to be reconciled.

[0039] Compared with the prior art, the beneficial effect of the present invention lies in that, in the technical solution of the present invention, the sub-node status is determined according to the steady-state measurement value and the correlation index, and the steady-state measurement value and the correlation index effectively reflect the stability and correlation of the data to be reconciled of each sub-node, and then different reference data selection methods are adaptively selected according to the sub-node status, so that the selection of the reference data selection method is more in line with the actual application scenario, avoiding the problem of poor reconciliation efficiency caused by inaccurate reference data selection.

[0040] Furthermore, the present invention effectively reflects the degree of interaction between stable sub-nodes and unstable sub-nodes and the proportion of stable sub-nodes through the steady-state mutual correlation degree and the steady-state balance coefficient, and then adaptively selects different compensation methods according to the steady-state mutual correlation degree and the steady-state balance coefficient. It can deeply explore the intrinsic correlation of data in different sub-nodes, and then dynamically adjust the combination of data in the sub-nodes, so as to achieve a balance between reducing computing overhead and improving fault tolerance, thereby improving the efficiency of reconciliation processing.

[0041] Furthermore, the present invention effectively reflects the degree of data anomaly and change status of sub-nodes through data anomaly and incremental coefficient, and then adaptively selects different clustering methods according to the data anomaly and incremental coefficient corresponding to each sub-node, so that the selected clustering method can automatically split the data to be reconciled in high-risk sub-nodes, balance the computing efficiency and accuracy requirements, and thus improve the efficiency of reconciliation processing.

[0042] Furthermore, the present invention effectively reflects the abnormal state of sub-data and the potential deviation influence of specific data clusters on the overall results by combining the bias influence and the balance coefficient, and then determines the characteristic data according to the combined bias influence and the balance coefficient, which can achieve a dynamic balance between accuracy and efficiency, improve the monitoring efficiency of hidden abnormal data, and make the selected reference data more representative. Then, the reconciliation sub-node selection method is determined according to the reference comparison coefficient and mutual correlation coefficient of the data to be reconciled corresponding to the sub-node and the reference data, so that the selection of the reconciliation sub-node is more accurate, thereby improving the efficiency and accuracy of the reconciliation processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 Schematic diagram of a multi-dimensional reconciliation method based on a group dining platform according to the present invention;

[0044] Figure 2 This is a flow chart of the present invention's method for determining reference data selection based on child node status;

[0045] Figure 3 This is a flow chart of the present invention for determining a compensation method based on a steady-state mutual correlation degree and a steady-state equalization coefficient;

[0046] Figure 4 This is a module connection diagram of the multi-dimensional reconciliation system based on the group dining platform of the present invention. DETAILED DESCRIPTION

[0047] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention.

[0048] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0049] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside", and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0050] Furthermore, it should be noted that, in the description of the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0051] See also Figures 1 to 3 As shown, the present invention provides a multi-dimensional reconciliation method based on a group dining platform, comprising:

[0052] Determine the subnode status according to the steady-state measurement value and the correlation index, and determine the reference data selection method according to the subnode status. The reference data selection method is single-node direct selection or multi-node compensation selection;

[0053] In direct selection of a single node, the to-be-reconciled data corresponding to the sub-node with the largest comprehensive evaluation value is used as the reference data;

[0054] In multi-node compensation selection, the compensation method is determined based on the steady-state mutual correlation and the steady-state balance coefficient to select reference data. The compensation method is partial compensation of multiple stable sub-nodes or associated compensation of reference stable sub-nodes.

[0055] The reconciliation sub-node selection method is determined based on the reference comparison coefficient and mutual correlation coefficient between the data to be reconciled and the reference data corresponding to the sub-node. The reconciliation sub-node selection method is selected based on the interaction influence coefficient or comparison threshold;

[0056] The characteristic data is determined based on a combined bias influence and a balance coefficient, and the combined bias influence is determined based on a bias reference value of a cluster corresponding to the data to be reconciled.

[0057] The application scenario of the present invention is the selection of data to be reconciled when a group dining platform performs reconciliation processing. The present invention includes several sub-nodes, each of which contains several data to be reconciled. A single data to be reconciled includes several data names and values corresponding to each data name. The data names include but are not limited to order amount, delivery fee, settlement cycle, settlement amount, handling fee, and tax. This is easy for those skilled in the art to understand and will not be described in detail.

[0058] In the present invention, several historical records are correspondingly provided, and any of the historical records records the steady-state quantitative value, correlation index, comprehensive evaluation value, steady-state mutual correlation degree, steady-state equilibrium coefficient, difference coefficient, and category balance degree, etc., in the historical process of selecting the data to be reconciled at least once, and each historical record corresponds to a qualified mark, which records whether the selection of the data to be reconciled meets the user's requirements. The qualified mark can be recorded manually. It is understandable that the user can determine whether the selection process of the data to be reconciled meets the requirements based on self-set indicators. The self-set indicators can be, but are not limited to, the false positive rate, which will not be elaborated here. Among them, the false positive rate is the number of times the reconciliation sub-node is incorrectly selected;

[0059] The present invention is provided with a continuous cycle monitoring period. At the end of each monitoring period, the subnode status is determined once. The duration of the monitoring period can be set according to the user's needs. The greater the user's demand for monitoring accuracy, the shorter the duration of the monitoring period. A value of the monitoring period is provided, and the monitoring period is 1 hour.

[0060] The comprehensive evaluation value corresponding to a single sub-node = the steady-state reference value corresponding to the sub-node + the associated reference value corresponding to the sub-node. The associated reference value corresponding to a single sub-node = 1-(a1 / a). The average value of the name reference values corresponding to each reconciliation data collected by the sub-node during the target monitoring period is recorded as a1.

[0061] Specifically, if the state of the child node is that the steady-state quantity value is greater than or equal to the preset steady-state quantity value and the correlation index is greater than or equal to the preset correlation index, the reference data selection method is single-node direct selection.

[0062] The subnode state includes a first subnode state and a second subnode state. The first subnode state is that the steady-state value is greater than or equal to the preset steady-state value and the correlation index is greater than or equal to the preset correlation index. The second subnode state is that the steady-state value is less than the preset steady-state value or the correlation index is less than the preset correlation index.

[0063] The steady-state measurement value is the average of the steady-state reference values corresponding to each child node. The steady-state reference value is confirmed as follows: for a single child node, the child node is recorded as the target child node. The steady-state reference value corresponding to the target child node = 1 / (comparison reference value + fluctuation deviation value). The fluctuation deviation value is the maximum value of the fluctuation thresholds corresponding to each data name in the target child node. The comparison reference value = |comparison coefficient - preset comparison coefficient|. The comparison coefficient is the standard deviation of the fluctuation thresholds corresponding to each data name in the target child node. The preset comparison coefficient is the average of the comparison coefficients corresponding to the child nodes whose steady-state measurement values in the historical records that can meet user needs are greater than or equal to the preset steady-state measurement value. The fluctuation threshold corresponding to a single data name is the standard deviation of the values corresponding to the data name in each to-be-reconciled data received by the target child node in the target monitoring period. The previous monitoring period adjacent to the current monitoring period is recorded as the target monitoring period.

[0064] Correlation index = a2 / a, where a is the average of the name reference values corresponding to each piece of unreconciled data collected at each subnode during the target monitoring period. A2 is the number of data names present in each piece of unreconciled data collected at each subnode during the target monitoring period. The name reference value corresponding to a single piece of unreconciled data is the number of data names present in that piece of unreconciled data.

[0065] The values of the preset steady-state quantity value and the preset correlation index can be determined by the user according to the actual application scenario. The smaller the values of the preset steady-state quantity value and the preset correlation index, the greater the user's demand for direct selection of a single node. A preset steady-state quantity value and a preset correlation index are provided to detect the historical records of the user's direct selection of a single node, and the average value of the steady-state quantity value corresponding to the historical records that can meet the user's needs is recorded as the preset steady-state quantity value, and the average value of the correlation index corresponding to the historical records that can meet the user's needs is recorded as the preset correlation index.

[0066] Specifically, if the state of the child node is that the steady-state quantity value is less than the preset steady-state quantity value or the correlation index is less than the preset correlation index, the reference data selection method is multi-node compensation selection.

[0067] Specifically, if the steady-state mutual correlation is greater than or equal to the preset steady-state mutual correlation and the steady-state equalization coefficient is greater than or equal to the preset steady-state equalization coefficient, the compensation method is multi-stable sub-node partial compensation;

[0068] In the partial compensation of multi-stable sub-nodes, the sub-nodes are selected according to the correlation deviation and the multi-dimensional covariance weight, and the feature data in each selected sub-node are selected as reference data based on the difference coefficient and the category balance.

[0069] The subnodes whose comprehensive evaluation value is greater than or equal to the preset comprehensive evaluation value are recorded as stable subnodes, and the subnodes whose comprehensive evaluation value is less than the preset comprehensive evaluation value are recorded as unstable subnodes.

[0070] Steady-state correlation = number of data names that appear in both the target data set and the reference data set / number of unique data names that appear in the reference data set. The set of all to-be-reconciled data collected by all stable sub-nodes during the target monitoring period is recorded as the target data set, and the set of all to-be-reconciled data collected by all unstable sub-nodes during the target monitoring period is recorded as the reference data set.

[0071] Steady-state equilibrium coefficient = number of steady-state child nodes / number of all child nodes;

[0072] The values of the preset comprehensive evaluation value, the preset steady-state mutual correlation and the preset steady-state balance coefficient can be determined by the user according to the actual application scenario. The greater the user's demand for improving the accuracy of reconciliation processing, the larger the values of the preset comprehensive evaluation value, the preset steady-state mutual correlation and the preset steady-state balance coefficient are. A preset comprehensive evaluation value, a preset steady-state mutual correlation and a preset steady-state balance coefficient are provided. The minimum value of the comprehensive evaluation value corresponding to each stable-state sub-node in the historical records that can meet the user's needs is recorded as the preset comprehensive evaluation value. The historical records of partial compensation of multiple stable-state sub-nodes are detected, and the average value of the steady-state mutual correlation corresponding to the historical records that can meet the user's needs is recorded as the preset steady-state mutual correlation, and the average value of the steady-state balance coefficient corresponding to the historical records that can meet the user's needs is recorded as the preset steady-state balance coefficient.

[0073] Determine the selection of child nodes based on the correlation deviation degree and the multidimensional covariation weight, wherein the number of selected child nodes is determined based on the correlation deviation degree, and select steady-state child nodes in descending order of the multidimensional covariation weight until the number of selected child nodes is reached, and the number of selected child nodes is positively correlated with the correlation deviation degree;

[0074] Selecting feature data in each selected sub-node as reference data based on the difference coefficient and the category balance, wherein, for a single selected sub-node, feature data in the selected sub-node whose difference coefficient is greater than a preset difference coefficient or whose category balance is greater than a preset category balance is used as reference data;

[0075] Correlation deviation = standard deviation of similarity coefficient corresponding to each stable sub-node + standard deviation of fluctuation coefficient corresponding to each stable sub-node;

[0076] The similarity coefficient and the fluctuation coefficient are confirmed as follows: for a single steady-state subnode, the steady-state subnode is recorded as the target steady-state subnode, and the other steady-state subnodes other than the target steady-state subnode are recorded as reference steady-state subnodes. The similarity coefficient corresponding to the target steady-state subnode = 1-(the number of identical data names in the steady-state feature data corresponding to the target steady-state subnode and the steady-state feature data corresponding to each reference subnode / the number of different data names appearing in each feature data in the target steady-state subnode). The set of feature data in the steady-state subnode is recorded as steady-state feature data. Each steady-state subnode corresponds to steady-state feature data. The fluctuation coefficient corresponding to the target steady-state subnode is the average value of the sub-fluctuation coefficients corresponding to each target data name. The data name appearing in the steady-state feature data corresponding to the target steady-state subnode is recorded as the target data name. The sub-fluctuation coefficient corresponding to a single target data name = |the standard deviation of each value corresponding to the data name in the steady-state feature data corresponding to the target steady-state subnode - the standard deviation of each value corresponding to the data name in the steady-state feature data corresponding to each reference steady-state subnode|;

[0077] The multidimensional covariance weight corresponding to a single stable sub-node = the fluctuation coefficient corresponding to the stable sub-node / the similarity coefficient corresponding to the stable sub-node;

[0078] Difference coefficient = number of data names appearing in the feature data / number of different data names appearing in each feature data in the target steady-state sub-node; category balance = balance coefficient / combined bias influence;

[0079] The values of the preset difference coefficient and the preset category balance can be determined by the user according to the actual application scenario. The greater the user's demand for improving the accuracy of reconciliation processing, the larger the values of the preset difference coefficient and the preset category balance. A value of the preset difference coefficient and the preset category balance is provided, and the historical records of the user performing partial compensation of multi-stable sub-nodes are detected. The average value of the difference coefficients corresponding to the historical records that can meet the user's needs is recorded as the preset difference coefficient, and the average value of the category balance corresponding to the historical records that can meet the user's needs is recorded as the preset category balance.

[0080] Specifically, if the steady-state mutual correlation is less than the preset steady-state mutual correlation or the steady-state equalization coefficient is less than the preset steady-state equalization coefficient, the compensation method is reference steady-state sub-node correlation compensation;

[0081] Reference steady-state subnode associated compensation includes:

[0082] Determine the reference sub-node according to the evaluation threshold, and select the compensation sub-node according to the sub-influence coefficient and the combination of associated nodes;

[0083] Determine the data compensation range corresponding to each compensation sub-node based on the cross-influence threshold and the iterative correlation degree;

[0084] Determine the supplementary selection method based on the deviation reference coefficient of the data compensation range corresponding to each compensation sub-node;

[0085] All feature data in the reference sub-node and feature data in the data compensation range corresponding to each compensation sub-node are used as reference data.

[0086] wherein, a reference subnode is determined according to an evaluation threshold, wherein a steady-state subnode with the largest evaluation threshold is selected as the reference subnode;

[0087] Selecting a compensation sub-node based on the sub-influence coefficient and the associated node combination, wherein the associated node combination is determined based on the associated reference value, and the associated node combination whose sub-influence coefficient is greater than the preset sub-influence coefficient is recorded as the reference combination, and any sub-node in the reference combination is used as the compensation sub-node, and each reference combination corresponds to a compensation sub-node;

[0088] Determining an associated node combination according to an associated reference value includes: recording other child nodes other than a reference child node as child nodes to be selected, performing an associated analysis on each child node to be selected, when performing an associated analysis on a single child node to be selected, recording the child node to be selected as a target child node to be selected, recording other child nodes to be selected other than the target child node to be selected as reference child nodes to be selected, recording a set of reference child nodes to be selected and the target child node to be selected whose associated reference values with the target child node to be selected are greater than a preset associated reference value as an associated node combination, and continuing to perform an associated analysis on child nodes to be selected that are not recorded in the associated node combination until all child nodes to be selected are recorded in the associated node combination;

[0089] The method for confirming the correlation reference value is as follows: for any two child nodes to be selected, the correlation reference value = 1-(the absolute value of the difference between the comprehensive evaluation values corresponding to the two child nodes / the larger value of the comprehensive evaluation values corresponding to the two child nodes); the value of the preset correlation reference value can be determined by the user based on the actual application scenario. The greater the user's demand for improving the accuracy of reconciliation processing, the smaller the value of the preset correlation reference value. A preset correlation reference value is provided, and the preset correlation reference value is 70%;

[0090] The evaluation threshold corresponding to a single steady-state sub-node = the comprehensive evaluation value corresponding to the steady-state sub-node + the multidimensional covariation weight corresponding to the steady-state sub-node;

[0091] The sub-influence coefficient is confirmed in the following way: for a single associated node combination, each sub-node in the associated node combination is recorded as an influencing sub-node, and the sub-influence coefficient corresponding to the associated node combination is the average of the influence reference values corresponding to each influencing sub-node and the reference sub-node, the influence reference value corresponding to a single influencing sub-node and the reference sub-node = cross-influence threshold / deviation uniformity, cross-influence threshold = 1-(the number of identical data names in the feature data corresponding to the influencing sub-node and the feature data corresponding to the reference sub-node / the number of different data names in the feature data corresponding to the influencing sub-node), the different data names in the feature data corresponding to the influencing sub-node and the feature data corresponding to the reference sub-node are recorded as difference data names, the average of the sub-difference degrees corresponding to each difference data name is recorded as deviation uniformity, and the sub-difference degree corresponding to a single difference data name is the number of sub-nodes to be selected corresponding to the feature data with the difference name;

[0092] The value of the preset sub-influence coefficient can be determined by the user based on the actual application scenario. The greater the user's demand for improved reconciliation processing accuracy, the smaller the value of the preset sub-influence coefficient. A preset sub-influence coefficient value is provided, and the historical records of reference steady-state sub-node association compensation are detected. The average value of the sub-influence coefficients corresponding to the reference combination that can meet the user's needs is recorded as the preset sub-influence coefficient;

[0093] The data compensation range corresponding to each compensation sub-node is determined based on the cross-impact threshold and the iterative correlation degree, wherein the data compensation range corresponding to a single compensation sub-node is the to-be-reconciled data whose data evaluation index is greater than the preset data evaluation index corresponding to the compensation sub-node and is located in the compensation sub-node. The preset data evaluation index corresponding to a single compensation sub-node is positively correlated with the compensation coefficient corresponding to the compensation sub-node. The compensation coefficient corresponding to a single compensation sub-node = the cross-impact threshold corresponding to the compensation sub-node + the iterative correlation degree corresponding to the compensation sub-node;

[0094] The iterative correlation degree is confirmed in the following manner: for a single compensation subnode, the compensation subnode is recorded as the target compensation subnode, and the compensation subnodes other than the target compensation subnode are recorded as reference compensation subnodes. The iterative correlation degree is the average of the iterative reference values corresponding to the target compensation subnode and each reference compensation subnode. The iterative reference value corresponding to the target compensation subnode and the single reference compensation subnode = 1 / the absolute value of the difference between the cross-influence threshold corresponding to the target compensation subnode and the cross-influence threshold corresponding to the reference compensation subnode.

[0095] The data evaluation index corresponding to a single piece of data to be reconciled = the cross-impact threshold of the data to be reconciled + the category balance of the data to be reconciled.

[0096] Specifically, the supplement selection method is determined according to the deviation reference coefficient of the data compensation range corresponding to each compensation sub-node, including:

[0097] If the deviation reference value is greater than or equal to the preset deviation reference value, the supplementary selection method is to increase the number of compensation sub-nodes;

[0098] If the deviation reference value is less than the preset deviation reference value, the supplementary selection method is to increase the adjustment for the data compensation range.

[0099] Wherein, the deviation reference coefficient = the number of feature data in the data compensation range corresponding to each compensation sub-node / deviation similarity coefficient, the deviation similarity coefficient = 1 / |standard deviation of the data evaluation index corresponding to each feature data - preset standard deviation|, where the preset standard deviation is the standard deviation of the data evaluation index corresponding to each feature data in the data compensation range corresponding to each compensation sub-node in the historical records that perform the reference steady-state sub-node association compensation and can meet user needs;

[0100] The value of the preset deviation reference value can be determined by the user according to the actual application scenario. The larger the value of the preset deviation reference value, the greater the user's need for increasing the data compensation range. A value of the preset deviation reference value is provided, and the historical records of the user's increasing the data compensation range are detected. The average value of the deviation reference values corresponding to the historical records that can meet the user's needs is recorded as the preset deviation reference value;

[0101] When increasing the number of compensation sub-nodes, the increase in the number of compensation sub-nodes is positively correlated with the deviation reference value, and the adjustment sub-nodes are selected in descending order of the impact reference value until the number of compensation sub-nodes after the increase is reached. The adjustment sub-nodes are the other sub-nodes to be selected except the compensation sub-nodes.

[0102] When increasing the data compensation range, the data compensation range corresponding to each compensation sub-node is increased. When increasing the data compensation range of a single compensation sub-node, the preset data evaluation index corresponding to the compensation sub-node needs to be decreased. The decrease value of the preset data evaluation index corresponding to a single compensation sub-node is positively correlated with the deviation reference value.

[0103] Specifically, the clustering method is determined based on the data variance and increment coefficient corresponding to each child node, including:

[0104] For a single child node,

[0105] If the data variability is greater than or equal to the preset data variability or the increment coefficient is greater than or equal to the preset increment coefficient, the clustering method is to split the cluster according to the coefficient of variation;

[0106] If the data variability is less than the preset data variability and the increment coefficient is less than the preset increment coefficient, the clustering method is to perform overall clustering based on the correlation coefficient.

[0107] The data anomaly and incremental coefficient are determined as follows: for a single child node, the data anomaly is the maximum value of the fluctuation thresholds corresponding to each data name appearing in each pending reconciliation data collected by the child node during the target monitoring period; the incremental coefficient = |the average value of the fluctuation thresholds corresponding to each data name appearing in each pending reconciliation data collected by the child node - the average value of the adjacent fluctuation thresholds corresponding to each data name appearing in each pending reconciliation data collected by the child node|; the adjacent fluctuation threshold corresponding to a single data name is the standard deviation of the values corresponding to that data name in each pending reconciliation data collected by the child node during the previous monitoring period adjacent to the target monitoring period;

[0108] The values of the preset data variability and the preset incremental coefficient can be determined by the user according to the actual application scenario. The greater the user's demand for improving the accuracy of reconciliation processing, the smaller the values of the preset data variability and the preset incremental coefficient are. The values of the preset data variability and the preset incremental coefficient are provided. The historical records of the overall clustering based on the correlation coefficient are detected, and the average value of the data variability corresponding to the historical records that can meet the user's needs is recorded as the preset data variability, and the average value of the incremental coefficients corresponding to the historical records that can meet the user's needs is recorded as the preset incremental coefficient.

[0109] Splitting clusters based on the coefficient of variation includes: recording the number of data names contained in a single name combination as n, where n is negatively correlated with the coefficient of variation, where coefficient of variation = data variability + incremental coefficient, sorting each data name in descending order of abnormal reference values and recording it as a reference sequence, starting from the leftmost end of the reference sequence, selecting the first n data names in sequence to form a name combination, repeating the above selection process, and selecting n data names from the remaining sequence each time to form a name combination. When the number of remaining data names is less than n, all remaining data names are recorded as a name combination. Each data to be reconciled corresponds to several split groups. A single split combination includes each data name and the corresponding value corresponding to the single name combination. Split cluster analysis is performed on the split combination corresponding to each name combination. When split cluster analysis is performed on a single split combination corresponding to a single name combination, the split combination is recorded as a target split combination, and other split combinations other than the target split combination corresponding to the name combination are recorded as reference split combinations. The set of the reference split combination whose combination matching degree with the target split combination is greater than the preset combination matching degree and the target split combination is recorded as a split cluster, and the split cluster analysis is continued for the split combinations not recorded in the split cluster until all split combinations are recorded in the split cluster;

[0110] The abnormal reference value corresponding to a single data name = |the fluctuation threshold corresponding to the data name - the adjacent fluctuation threshold corresponding to the data name|;

[0111] Performing overall clustering based on the correlation coefficient includes: performing overall cluster analysis on each piece of to-be-reconciled data; when performing overall cluster analysis on a single piece of to-be-reconciled data, recording the to-be-reconciled data as target to-be-reconciled data; recording other to-be-reconciled data other than the target to-be-reconciled data as reference to-be-reconciled data; recording the set of the reference to-be-reconciled data and the target to-be-reconciled data having a correlation coefficient with the target to-be-reconciled data greater than a preset correlation coefficient as an overall cluster; and continuing to perform cluster analysis on the to-be-reconciled data not recorded in the overall cluster until all to-be-reconciled data are recorded in the overall cluster;

[0112] The combination matching degree is confirmed in the following way: for the two split combinations corresponding to a single name combination, the combination matching degree is the standard deviation of the difference coefficients corresponding to each data name in the name combination, and the difference coefficient corresponding to a single data name is the absolute value of the difference between the numerical values corresponding to the data name in the two split combinations;

[0113] For any two data to be reconciled, the correlation coefficient = 1-(the absolute value of the difference between the data evaluation indices corresponding to the two data to be reconciled / the larger value of the data evaluation indices corresponding to the two data to be reconciled);

[0114] The values of the preset combination matching degree and the preset correlation coefficient can be determined by the user according to the actual application scenario. The greater the user's demand for improving the accuracy of reconciliation processing, the larger the values of the preset combination matching degree and the preset correlation coefficient will be. The historical records of splitting clusters according to the coefficient of variation are detected, and the average value of the combination matching degree corresponding to each historical record that can meet the user's needs is recorded as the preset combination matching degree. The preset correlation coefficient is 80%.

[0115] Specifically, the characteristic data is the data to be reconciled whose combination bias influence is less than the preset combination bias influence and whose balance coefficient is greater than or equal to the preset balance coefficient.

[0116] The combined bias impact is confirmed as follows: for a single piece of data to be reconciled, the data to be reconciled is recorded as the target data to be reconciled. If the target data to be reconciled corresponds to a split cluster, the combined bias impact is the maximum value of the bias reference values corresponding to the split clusters corresponding to the target data to be reconciled. If the target data to be reconciled corresponds to an entire cluster, the combined bias impact is the bias reference value of the entire cluster corresponding to the target data to be reconciled. The bias reference value corresponding to a single cluster is confirmed as follows: the bias reference value corresponding to the cluster = |the sub-bias corresponding to the cluster - the preset sub-bias corresponding to the cluster|. The sub-bias corresponding to a single cluster is the maximum value of the abnormal reference values corresponding to each reference name. Each data name appearing in the cluster is recorded as a reference name. The abnormal reference value corresponding to a single reference name = |the average value of the values corresponding to the reference name in the cluster - the average value of the values corresponding to the reference name in the associated clusters|. Clusters corresponding to feature data in the historical records that have the same data names as those contained in the cluster are detected and recorded as associated clusters. The preset sub-bias corresponding to a single cluster is the average value of the sub-bias values corresponding to each associated cluster corresponding to the cluster.

[0117] The balance coefficient corresponding to a single data item to be reconciled is the average of the sub-balance coefficients corresponding to the data names appearing in the data item to be reconciled. The sub-balance coefficient corresponding to a single data name = 1 / |the value corresponding to the data name in the data item to be reconciled - the average of the values corresponding to the data name in the sub-node where the data item to be reconciled is located|;

[0118] The values of the preset combination bias influence and the preset balance coefficient can be determined by the user according to the actual application scenario. The greater the user's demand for improving the accuracy of reconciliation processing, the smaller the values of the preset combination bias influence and the preset balance coefficient. A value of a preset combination bias influence and a preset balance coefficient is provided, and the average value of the combination bias influence corresponding to each feature data in the historical records that can meet the user's needs is recorded as the preset combination bias influence, and the average value of the balance coefficient corresponding to each feature data in the historical records that can meet the user's needs is recorded as the preset balance coefficient.

[0119] Specifically, the selection method of the reconciliation sub-node is determined based on the reference comparison coefficient and mutual correlation coefficient between the to-be-reconciled data and the reference data corresponding to the sub-node, including:

[0120] If the reference comparison coefficient is greater than or equal to the preset reference comparison coefficient and the mutual correlation coefficient is greater than or equal to the preset mutual correlation coefficient, the reconciliation sub-node is selected based on the interaction influence coefficient;

[0121] If the reference comparison coefficient is less than the preset reference comparison coefficient or the mutual correlation coefficient is less than the preset mutual correlation coefficient, the reconciliation sub-node is selected based on the comparison threshold.

[0122] Wherein, reference comparison coefficient = (number of subnodes whose comparison threshold is greater than the preset comparison threshold) / number of all subnodes;

[0123] The comparison threshold is determined by taking the maximum value of the sub-comparison reference values corresponding to each data name in a single sub-node as the comparison threshold. The sub-comparison reference value corresponding to a single data name = |the fluctuation threshold corresponding to the data name in the sub-node - the standard deviation of the values corresponding to each reference data in which the data name appears|.

[0124] The mutual correlation coefficient is the standard deviation of the comparison threshold corresponding to each pre-reconciliation node. The child nodes whose comparison threshold is greater than the preset comparison threshold are recorded as pre-reconciliation nodes.

[0125] The values of the preset comparison threshold, preset reference comparison coefficient, and preset cross-correlation coefficient can be determined by the user based on the actual application scenario. The smaller the values of the preset comparison threshold, preset reference comparison coefficient, and preset cross-correlation coefficient, the greater the user's need for optimized detection and selection. The average value of the comparison threshold corresponding to each pre-reconciliation node in the historical records that can meet the user's needs is recorded as the preset comparison threshold. The historical records selected by the user for optimized detection are detected, and the average value of the reference comparison coefficients corresponding to the historical records that can meet the user's needs is recorded as the preset reference comparison coefficient. The average value of the cross-correlation coefficients corresponding to the historical records that can meet the user's needs is recorded as the preset cross-correlation coefficient.

[0126] When selecting based on the interaction influence coefficient, the pre-reconciliation node with an interaction influence coefficient greater than the preset interaction influence coefficient is selected as the reconciliation sub-node;

[0127] When selecting based on the comparison threshold, the child node whose comparison threshold is greater than the preset comparison threshold is used as the reconciliation child node;

[0128] For a single pre-reconciliation node, record it as the target pre-reconciliation node, and record all other pre-reconciliation nodes as reference pre-reconciliation nodes. The interaction influence coefficient corresponding to the target pre-reconciliation node = the average sub-comparison reference value corresponding to the target pre-reconciliation node for the data names that appear in both the target pre-reconciliation node and the reference pre-reconciliation node / the number of data names that appear in both the target pre-reconciliation node and the reference pre-reconciliation node.

[0129] The value of the preset interaction influence coefficient can be determined by the user based on the actual application scenario. The greater the user's demand for improving the accuracy of reconciliation processing, the smaller the value of the preset interaction influence coefficient. The historical records selected by the user based on the interaction influence coefficient are detected, and the average value of the interaction influence coefficient corresponding to each reconciliation sub-node selected from the historical records that can meet the user's needs is recorded as the preset interaction influence coefficient.

[0130] See also Figure 4 As shown, it is a module connection diagram of the multi-dimensional reconciliation system based on the group dining platform of the present invention. The present invention also provides a multi-dimensional reconciliation system based on the group dining platform, including:

[0131] A data collection module, which includes several sub-nodes for data collection;

[0132] A state analysis module, connected to the data acquisition module, is used to determine the state of the subnode based on the steady-state value and the correlation index, and to determine the reference data selection method based on the subnode state. The reference data selection method is single-node direct selection or multi-node compensation selection;

[0133] A first selection module, connected to the status analysis module, is used to use the to-be-reconciled data corresponding to the subnode with the largest comprehensive evaluation value as reference data in the direct selection of a single node;

[0134] A second selection module, connected to the state analysis module, is used to determine a compensation method to select reference data according to the steady-state mutual correlation and the steady-state balance coefficient in multi-node compensation selection, wherein the compensation method is partial compensation of multiple stable sub-nodes or associated compensation of reference stable sub-nodes;

[0135] a reconciliation selection module, connected to the first selection module and the second selection module, respectively, for determining a reconciliation sub-node selection method based on a reference comparison coefficient and a mutual correlation coefficient between the to-be-reconciled data and reference data corresponding to the sub-node, wherein the reconciliation sub-node selection method is selected based on an interaction influence coefficient or a comparison threshold;

[0136] The characteristic data is determined based on a combined bias influence and a balance coefficient, and the combined bias influence is determined based on a bias reference value of a cluster corresponding to the data to be reconciled.

[0137] Thus far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present invention.

[0138] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that the present invention is susceptible to various modifications and variations. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A multi-dimensional reconciliation method based on a group dining platform, characterized in that: include: Determine the subnode status according to the steady-state measurement value and the correlation index, and determine the reference data selection method according to the subnode status. The reference data selection method is single-node direct selection or multi-node compensation selection; In the case of direct selection of a single node, the to-be-reconciled data corresponding to the sub-node with the largest comprehensive evaluation value is used as the reference data; In multi-node compensation selection, the compensation method is determined based on the steady-state mutual correlation and the steady-state balance coefficient to select reference data. The compensation method is partial compensation of multiple stable sub-nodes or associated compensation of reference stable sub-nodes. The reconciliation sub-node selection method is determined based on the reference comparison coefficient and mutual correlation coefficient between the data to be reconciled and the reference data corresponding to the sub-node. The reconciliation sub-node selection method is selected based on the interaction influence coefficient or comparison threshold; The characteristic data is determined based on a combined bias influence and a balance coefficient, and the combined bias influence is determined based on a bias reference value of a cluster corresponding to the data to be reconciled.

2. The multi-dimensional reconciliation method based on the group dining platform according to claim 1 is characterized in that: If the subnode status is that the steady-state quantity value is greater than or equal to the preset steady-state quantity value and the correlation index is greater than or equal to the preset correlation index, the reference data selection method is single node direct selection.

3. The multi-dimensional reconciliation method based on the group dining platform according to claim 2 is characterized in that: If the sub-node state is that the steady-state quantity value is less than the preset steady-state quantity value or the correlation index is less than the preset correlation index, the reference data selection method is multi-node compensation selection.

4. The multi-dimensional reconciliation method based on the group dining platform according to claim 3 is characterized in that: If the steady-state mutual correlation is greater than or equal to the preset steady-state mutual correlation and the steady-state balance coefficient is greater than or equal to the preset steady-state balance coefficient, the compensation method is multi-stable sub-node partial compensation; In the partial compensation of multi-stable sub-nodes, the sub-nodes are selected according to the correlation deviation and the multi-dimensional covariance weight, and the feature data in each selected sub-node are selected as reference data based on the difference coefficient and the category balance.

5. The multi-dimensional reconciliation method based on the group dining platform according to claim 4 is characterized in that: If the steady-state mutual correlation is less than the preset steady-state mutual correlation or the steady-state balance coefficient is less than the preset steady-state balance coefficient, the compensation method is reference steady-state sub-node correlation compensation; Reference steady-state subnode associated compensation includes: Determine the reference sub-node according to the evaluation threshold, and select the compensation sub-node according to the sub-influence coefficient and the combination of associated nodes; Determine the data compensation range corresponding to each compensation sub-node based on the cross-influence threshold and the iterative correlation degree; Determine the supplementary selection method based on the deviation reference coefficient of the data compensation range corresponding to each compensation sub-node; All feature data in the reference sub-node and feature data in the data compensation range corresponding to each compensation sub-node are used as reference data.

6. The multi-dimensional reconciliation method based on the group dining platform according to claim 5 is characterized in that: The supplement selection method is determined based on the deviation reference coefficient of the data compensation range corresponding to each compensation sub-node, including: If the deviation reference value is greater than or equal to the preset deviation reference value, the supplementary selection method is to increase the number of compensation sub-nodes; If the deviation reference value is less than the preset deviation reference value, the supplementary selection method is to increase the adjustment for the data compensation range.

7. The multi-dimensional reconciliation method based on the group dining platform according to claim 1 is characterized in that: The clustering method is determined based on the data variance and increment coefficient corresponding to each child node, including: For a single child node, If the data variability is greater than or equal to the preset data variability or the increment coefficient is greater than or equal to the preset increment coefficient, the clustering method is to split the cluster according to the coefficient of variation; If the data variability is less than the preset data variability and the increment coefficient is less than the preset increment coefficient, the clustering method is to perform overall clustering based on the correlation coefficient.

8. The multi-dimensional reconciliation method based on the group dining platform according to claim 1 is characterized in that: The characteristic data is the data to be reconciled whose combination bias influence is less than the preset combination bias influence and whose balance coefficient is greater than or equal to the preset balance coefficient.

9. The multi-dimensional reconciliation method based on the group dining platform according to claim 1 is characterized in that: The reconciliation sub-node selection method is determined based on the reference comparison coefficient and mutual correlation coefficient between the data to be reconciled and the reference data corresponding to the sub-node, including: If the reference comparison coefficient is greater than or equal to the preset reference comparison coefficient and the mutual correlation coefficient is greater than or equal to the preset mutual correlation coefficient, the reconciliation sub-node is selected based on the interaction influence coefficient; If the reference comparison coefficient is less than the preset reference comparison coefficient or the mutual correlation coefficient is less than the preset mutual correlation coefficient, the reconciliation sub-node is selected based on the comparison threshold.

10. A reconciliation system using the multi-dimensional reconciliation method based on a group dining platform according to any one of claims 1 to 9, characterized in that: include: A data collection module, which includes several sub-nodes for data collection; A state analysis module, connected to the data acquisition module, is used to determine the state of the subnode based on the steady-state value and the correlation index, and to determine the reference data selection method based on the subnode state. The reference data selection method is single-node direct selection or multi-node compensation selection; A first selection module, connected to the status analysis module, is used to use the to-be-reconciled data corresponding to the subnode with the largest comprehensive evaluation value as reference data in the direct selection of a single node; A second selection module, connected to the state analysis module, is used to determine a compensation method to select reference data according to the steady-state mutual correlation and the steady-state balance coefficient in multi-node compensation selection, wherein the compensation method is partial compensation of multiple stable sub-nodes or associated compensation of reference stable sub-nodes; a reconciliation selection module, connected to the first selection module and the second selection module, respectively, for determining a reconciliation sub-node selection method based on a reference comparison coefficient and a mutual correlation coefficient between the to-be-reconciled data and reference data corresponding to the sub-node, wherein the reconciliation sub-node selection method is selected based on an interaction influence coefficient or a comparison threshold; The characteristic data is determined based on a combined bias influence and a balance coefficient, and the combined bias influence is determined based on a bias reference value of a cluster corresponding to the data to be reconciled.

Images