Whole asset wealth protection and inheritance risk assessment method and system

Abstract

The invention belongs to the technical field of information matching and data assessment, and provides a full-asset wealth protection and inheritance risk assessment method and system, and the method specifically comprises the steps: firstly, recognizing a reference server and a redistribution server from a full-asset network, respectively recording the reference server and the redistribution server as a reference object and an evaluation object, and carrying out the assessment of the full-asset wealth protection and inheritance risk through an asset database of each evaluation object; and constructing a reference word adaptive portrait, then performing inheritance execution quality time sequence analysis by using the reference word adaptive portrait to obtain an inheritance execution score of each evaluation object, and finally performing inheritance risk evaluation in combination with the inheritance execution score of each evaluation object. According to the method, the coordination of the redistribution server between the reference server and the related database needing to execute inheritance processing in the whole asset wealth can be effectively quantified, and the risk that data of a server cluster is tampered and lost or individual asset data is failed to read and write in an inheritance task is reduced.

Description

Technical Field

[0001] The present invention belongs to the field of information matching and data evaluation technology, and specifically relates to a method and system for assessing the risk of total asset wealth protection and inheritance. Background Art

[0002] The inheritance of all-asset wealth plays a vital role in scenarios such as corporate mergers and acquisitions and family inheritance. The process of all-asset wealth inheritance mainly involves the accurate and correct transfer of various types of assets from the inheriting party to the inheritee. However, in the process of all-asset inheritance, the server that manages assets often does not simply complete the inheritance work through database reading and writing. This is because all assets are multi-dimensional data, and the different eras of the records and servers lead to different verification, registration or interaction methods, resulting in a long database reading and writing cycle for service providers, insufficient inheritance quality and efficiency of asset data, and a large number of adaptive adjustments are required. In addition, the replacement of asset service providers is also a common phenomenon in the inheritance of all-asset assets, which further amplifies the problem of insufficient inheritance quality and efficiency of asset data. In the process of the asset service provider performing the inheritance operation, in addition to selecting a server with sufficient data volume of the inheritee as the main storage server, it is also necessary to select a server to verify the quality of the storage server data, including whether the original data has been tampered with or whether there is a contradiction in the storage logic. Therefore, the asset data obtained historically required by the storage server for verification needs to be fully adaptable to the benchmark server in terms of storage type, interaction form, etc., to avoid the risk of declining verification quality during inheritance execution. Summary of the invention

[0003] The purpose of the present invention is to propose a method and system for assessing the risk of wealth protection and inheritance of all assets, so as to solve one or more technical problems existing in the prior art and at least provide a beneficial choice or create conditions.

[0004] In order to achieve the above object, according to one aspect of the present invention, a method for risk assessment of total asset wealth protection and inheritance is provided, the method comprising the following steps:

[0005] S100, identifying a reference server and a reallocation server from the entire asset network, and recording them as reference objects and evaluation objects respectively; S200, constructing a reference word adaptation portrait through the asset database of each evaluation object;

[0006] S300, using the benchmark word adaptation portrait to perform inheritance execution quality time series analysis to obtain inheritance execution scores for each evaluation object;

[0007] S400, conduct inheritance risk assessment based on the inheritance execution scores of each evaluation object.

[0008] Further, in step S100, the method of identifying the reference server and the reallocation server from the full asset network and recording them as the reference object and the evaluation object respectively is: the full asset network is composed of a plurality of servers, and the server includes a reference server and a plurality of reallocation servers, which are recorded as the reference object and the evaluation object respectively.

[0009] Since the process of full-asset wealth inheritance involves a large amount of data changes and editing, the number of data entries and categories is huge, and the number of servers involved is also large. Therefore, it is easy for developers to have vulnerabilities that cannot be predicted during network transmission, and it is easy to be illegally invaded. Therefore, in addition to the normal working reallocation server and benchmark server, it is usually necessary to select a server from the reallocation server as the deployment server. The benchmark server is used to verify the correctness of the data. Usually, the server with the most data deployed to the inheritor is selected. The deployment server directly exchanges data with the verification server, similar to the backup role. Therefore, the deployment server requires a relatively high level of interoperability in the data processing logic. The full-asset network refers to a server cluster composed of various servers serving asset management.

[0010] Furthermore, in step S200, a method for constructing a benchmark word adaptation portrait through the asset database of each evaluation object is as follows: each server includes an asset database, the asset database is a database composed of asset documents in JSON format or BSON format, and the stored data is unstructured or semi-structured data; the asset document is a collection of texts; every 5-10 natural days is taken as a parsing point, and a parsing point and 2-6 parsing points in the reverse time direction are taken as the mapping period of the parsing point; for any parsing point, the benchmark object merges each asset document of the corresponding mapping period to form a mapping text set, and each asset document of the evaluation object at the parsing point constitutes a parsing text set; the redistribution reference score of the evaluation object is calculated based on the mapping text set and the parsing text set of the parsing point, and the sequence composed of the redistribution reference scores of all parsing points is recorded as the benchmark word adaptation portrait.

[0011] Asset documents are usually used to store physical assets, including but not limited to equipment, buildings, and vehicles. Asset documents are also used to store digital assets, including but not limited to software, data sets, patents, and copyrights. The content of each asset is presented in text in the asset document.

[0012] Furthermore, the method for calculating the redistributed reference score of the evaluation object according to the mapping text set and the parsed text set of the parsing point is: for any parsing point, the text set is preprocessed, including text cleaning, word segmentation and stop word removal; the text set is identified with keywords through the LDA latent Dirichlet distribution model to obtain keywords and corresponding weights, and the keywords are classified into nouns and adjectives by using the CRF conditional random field algorithm and are respectively classified into related words and matching words; the weights of related words and matching words are normalized respectively;

[0013] Word embedding analysis is performed on each keyword in the text set and its corresponding weights to obtain word vectors. The word vector obtained based on the analysis of the mapped text set is the benchmark portrait, and the word vector obtained based on the analysis of the parsed text set is the evaluation portrait. The cosine similarity between the evaluation portrait and the benchmark portrait is calculated as the reference score for the redistribution of the parsing point.

[0014] The text set includes a mapping text set and a parsing text set; the word embedding method includes one of GloVe, Word2Vec or FastText; the LDA latent Dirichlet distribution model is referenced from the Gensim library in the python analysis software.

[0015] Further, in step S300, the method of using the benchmark word adaptation portrait to perform inheritance execution quality time series analysis to obtain the inheritance execution score of each evaluation object is: set a time period as the evaluation period EVtm, EVtm∈[1,10] years, and within the evaluation period, record the time scale of obtaining the redistributed reference score as a resolution point; wherein all evaluation objects are required to have an equal number of resolution points within the evaluation period;

[0016] For any evaluation object, obtain the redistributed reference scores of each analysis point and form a benchmark segmentation set; that is, the benchmark segmentation set is a set of redistributed reference scores;

[0017] Use K-means clustering algorithm to divide all elements into several clusters for the benchmark segmentation set; the default number of clusters is EVtm / 10;

[0018] Calculate the distance from any element to its corresponding cluster center and record it as the element's cluster center distance. If the cluster center distance of any element is greater than the 70th percentile of all cluster center distances in its corresponding cluster, the resolution point corresponding to the element is recorded as the long-distance adaptation point.

[0019] The 70th percentile of all distances in the corresponding cluster refers to the value of the 70th position of the distances between the cluster centers of all elements in the cluster after sorting from small to large.

[0020] The redistributed reference scores of each parsing point are normalized by the sigmoid activation function to obtain the adaptation scale;

[0021] For any evaluation object, the sequence of fitness scales of each remote fitness point is recorded as the remote evaluation sequence LS.PT; the fitness scales of all resolution points of the evaluation object are written into the sequence and recorded as the full-distance evaluation sequence LS.AP; any fitness scale is selected from the full-distance evaluation sequence, and the percentile of all fitness scales under the corresponding resolution point is recorded as Hta, and the corresponding percentile when the fitness scale is placed in the sequence LS.PT is recorded as Htb, then the fitness scale standard value IHT of the fitness scale is the root mean square value of Hta and Htb;

[0022] In the calculation of the percentiles of all adaptation scales at the corresponding resolution point, different evaluation objects have adaptation scales at the same resolution point, so any adaptation scale at the same resolution point can obtain percentiles.

[0023] The principle of obtaining the adaptation scale standard value here is actually based on the comparison of the adaptation scale, focusing on how to compare the adaptation scale of the remote adaptation point with the adaptation scale of the resolution point. The adaptation scale standard value is designed to measure the coordination between the reallocation server and the benchmark server in the storage logic of the inherited related database. The stronger the coordination, the higher the probability of overlap of asset-related types or attributes, indicating that the reallocation server can perform wealth information backup and verification work more safely and efficiently as a deployment server. When the adaptation scale standard value is low, it means that the reallocation server is prone to inefficiency in the backup and verification steps as a deployment server.

[0024] The inheritance execution score of each evaluation object is calculated based on the remote evaluation sequence and the adaptation scale value.

[0025] Since the calculation of the inheritance execution score is obtained by processing the long-range evaluation sequence and the adaptive scale standard value, it can effectively quantify the coordination between the relevant databases of the redistribution server that need to execute inheritance processing in the whole asset wealth and the benchmark server. However, the acquisition of the adaptive scale standard value is too dependent on the adaptive scale, which will lead to the problem of local generalization of the quantitative results and cause decision-making deviation. This is because the amount of data of the inherited asset data obtained in the server is not always huge. When the amount of data is insufficient, the full-range evaluation sequence data will fluctuate greatly and cause the adaptive scale to be insufficiently accurate. However, the existing technology cannot effectively compensate for this generalization phenomenon. In order to eliminate this influence, the present invention proposes a more preferred solution as follows:

[0026] Further, in step S300, the method of using the benchmark word adaptation portrait to perform inheritance execution quality time series analysis to obtain the inheritance execution score of each evaluation object is: set a time period as the evaluation period EVtm, EVtm∈[2,6] years; wherein all evaluation objects are required to have an equal number of resolution points within the evaluation period;

[0027] For any evaluation object, obtain the redistribution reference score of each resolution point and form a redistribution reference score sequence; in the redistribution reference score sequence, calculate the weighted average of all elements and record it as Mvae, where the weight of each element is the percentile of all elements in the corresponding redistribution reference score sequence; when any element in the redistribution reference score sequence is greater than Mvae, the resolution point corresponding to the element is recorded as a high fitness point, and the number of high fitness points is recorded as RQm;

[0028] A floating-point variable Laor is preset for the analysis points corresponding to all elements. If the analysis point is a high-fitness point, the inheritance priority is 1, otherwise the inheritance priority is 0.5. The purpose of setting the inheritance priority is to make the probability of high-fitness points being selected as inheritance estimation points in the subsequent process higher. The value of non-high-fitness points can also be between 0.3 and 0.7. The larger the value, the lower the ability to distinguish from high-fitness points.

[0029] The inheritance priority rate Faor of any analysis point is the ratio of its inheritance priority to the sum of the inheritance priorities of all analysis points;

[0030] In the redistribution reference score sequence, the cumulative inheritance priority FFaor of any element is the cumulative value of each inheritance priority between the element and the first element of the sequence; the first element of the sequence corresponds to the analysis point closest to the current moment; all the cumulative inheritance priorities are written into the sequence and recorded as a cumulative sequence;

[0031] Randomly generate RTN random numbers and record them as selected values. The range of the selected values ​​is RU∈(0,1); where RTN is a preset integer, RTN∈[500,2000];

[0032] Take k2 as the serial number of the analysis point, and for any selected value RU, set its estimation point condition to FFaor(k2)<RU≤FFaor(k2+1); search for the analysis point that meets the estimation point condition and mark the occurrence of an estimation point event at the analysis point; after marking the analysis points where estimation point events occur for all selected values, record the total number of estimation point events occurring at the analysis points as RQm_TH, sort RQm_TH from large to small and select the analysis points corresponding to the first RQm elements as the inheritance estimation points; where FFaor(k2) and FFaor(k2+1) are the cumulative inheritance priority rates of the analysis points corresponding to the k2th and k2+1th elements in the redistributed reference score sequence; calculate the inheritance execution score of each evaluation object based on the inheritance estimation point.

[0033] The principle of screening inheritance evaluation points here is actually based on the concepts of weighted average and genetic algorithm. By setting the inheritance priority and inheritance priority rate, the influence of each resolution point is quantified, and selection is made based on these quantified values. Finally, by comparing random numbers with the cumulative inheritance priority rates, a process similar to preferential selection is simulated to determine which resolution points should be more suitable to become inheritance evaluation points, thereby better explaining the document content relevance distance between the redistribution server and the benchmark server.

[0034] Beneficial effects: Since the adaptive scale value is a dynamic quantitative tool calculated based on the text word vectors of each evaluation object and the benchmark object, it can effectively quantify the coordination between the relevant databases of the redistribution server that needs to perform inheritance processing on the entire asset wealth and the benchmark server, and then explain the synchronization and adaptability of each redistribution server participating in the inheritance with the benchmark server in terms of asset content, thereby reducing the risk of data tampering, loss, or failure to read and write individual asset data in the server group during the inheritance task.

[0035] Further, in step S400, the method of conducting inheritance risk assessment in combination with the inheritance execution score of each evaluation object is as follows: the length of the benchmark word adaptation portrait is recorded as the parsing amount, that is, the number of parsing points of the evaluation object. The evaluation object with the maximum value in the parsing amount among the evaluation objects is searched as the first evaluation object. At any parsing point, if an evaluation object has a larger redistribution reference score than the first evaluation object, it is defined that the evaluation object has a high adaptation event, and the ratio of the number of high adaptation events of the evaluation object to its parsing amount is the high adaptation efficiency; the lower quartile value in the set composed of all high adaptation efficiencies is the adaptation efficiency threshold; the median value and standard deviation of all inheritance execution scores are recorded as MER and STDER respectively, and the evaluation risk overflow rate ERSR is preset, ERSR∈[1.5,3.5]; if the high adaptation efficiency of an evaluation object is less than the adaptation efficiency threshold and the inheritance execution score of the evaluation object is less than MER-ERSR×STDER, the evaluation object is evaluated to have an inheritance abnormality risk and is not selected as a deployment server, and each evaluation object with an inheritance abnormality risk is sent to the client server as an evaluation report.

[0036] Preferably, all undefined variables in the present invention, if not clearly defined, can be manually set thresholds.

[0037] The present invention also provides a total asset wealth protection and inheritance risk assessment system, the total asset wealth protection and inheritance risk assessment system comprises: a processor, a memory, and a computer program stored in the memory and executable on the processor, the processor implements the steps in the total asset wealth protection and inheritance risk assessment method when executing the computer program, the total asset wealth protection and inheritance risk assessment system can be run in computing devices such as desktop computers, laptop computers, PDAs, and cloud data centers, the executable system may include, but is not limited to, a processor, a memory, and a server cluster, the processor executes the computer program and runs it in the following system units:

[0038] The full-asset network storage object identification unit is used to identify the reference server and the reallocation server from the full-asset network, which are recorded as the reference object and the evaluation object respectively;

[0039] A benchmark word adaptation portrait construction unit, used to construct a benchmark word adaptation portrait through an asset database of each evaluation object;

[0040] The inheritance execution quality time series analysis unit is used to perform inheritance execution quality time series analysis using the benchmark word adaptation portrait to obtain the inheritance execution score of each evaluation object;

[0041] The inheritance risk assessment unit is used to conduct inheritance risk assessment based on the inheritance execution scores of each evaluation object.

[0042] The beneficial effects of the present invention are as follows: the present invention provides a method and system for assessing the risk of protecting and inheriting total asset wealth, which can effectively quantify the coordination between the relevant databases of the redistribution server that needs to perform inheritance processing on the total asset wealth and the benchmark server, and then explain the synchronization and adaptability between the asset content of each redistribution server participating in the inheritance and the benchmark server, thereby reducing the risk of data tampering, loss, or failure to read and write individual asset data in the server group during the inheritance task. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The above and other features of the present invention will become more obvious by describing in detail the embodiments shown in the accompanying drawings. The same reference numerals in the accompanying drawings of the present invention represent the same or similar elements. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other accompanying drawings can be obtained based on these accompanying drawings without creative work. In the accompanying drawings:

[0044] Figure 1 Shown is a flow chart of the full-asset wealth protection and inheritance risk assessment method;

[0045] Figure 2 Shown is the structure diagram of the full-asset wealth protection and inheritance risk assessment system. DETAILED DESCRIPTION

[0046] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects of the present invention, so as to fully understand the purpose, scheme and effect of the present invention. It should be noted that the embodiments and features in the embodiments of this application can be combined with each other without conflict.

[0047] like Figure 1 The following is a flow chart of the risk assessment method for all-asset wealth protection and inheritance. Figure 1 The full-asset wealth protection and inheritance risk assessment method according to an embodiment of the present invention is described below. The method comprises the following steps:

[0048] Embodiment 1

[0049] S100, identifying a reference server and a reallocation server from the entire asset network, and recording them as reference objects and evaluation objects respectively; S200, constructing a reference word adaptation portrait through the asset database of each evaluation object;

[0050] S300, using the benchmark word adaptation portrait to perform inheritance execution quality time series analysis to obtain inheritance execution scores for each evaluation object;

[0051] S400, conduct inheritance risk assessment based on the inheritance execution scores of each evaluation object.

[0052] Further, in step S100, the method of identifying the reference server and the reallocation server from the full asset network and recording them as the reference object and the evaluation object respectively is: the full asset network is composed of a plurality of servers, and the server includes a reference server and a plurality of reallocation servers, which are recorded as the reference object and the evaluation object respectively.

[0053] Further, in step S200, a method for constructing a benchmark word adaptation portrait through the asset database of each evaluation object is as follows: each server includes an asset database, the asset database is a database composed of asset documents in JSON format or BSON format, and the asset documents are a collection of texts; every 5 natural days is taken as a parsing point, and a parsing point and its 3 parsing points in the reverse time direction are taken as the mapping period of the parsing point; for any parsing point, the benchmark object merges each asset document of the corresponding mapping period to form a mapping text set, and each asset document of the evaluation object at the parsing point constitutes a parsing text set; the redistribution reference score of the evaluation object is calculated based on the mapping text set and the parsing text set of the parsing point, and the sequence composed of the redistribution reference scores of all parsing points is recorded as the benchmark word adaptation portrait.

[0054] Furthermore, the method for calculating the redistributed reference score of the evaluation object according to the mapping text set and the parsed text set of the parsing point is: for any parsing point, the text set is preprocessed, including text cleaning, word segmentation and stop word removal; the text set is identified with keywords through the LDA latent Dirichlet distribution model to obtain keywords and corresponding weights, and the keywords are classified into nouns and adjectives by using the CRF conditional random field algorithm and are respectively classified into related words and matching words; the weights of related words and matching words are normalized respectively;

[0055] Word embedding analysis is performed on each keyword in the text set and its corresponding weights to obtain word vectors. The word vector obtained based on the analysis of the mapped text set is the benchmark portrait, and the word vector obtained based on the analysis of the parsed text set is the evaluation portrait. The cosine similarity between the evaluation portrait and the benchmark portrait is calculated as the reference score for the redistribution of the parsing point.

[0056] Further, in step S300, the method of using the benchmark word adaptation portrait to perform inheritance execution quality time series analysis to obtain the inheritance execution score of each evaluation object is: set a time period as the evaluation period, taking a value of 2 years, and within the evaluation period, record the time scale of obtaining the redistribution reference score as the analysis point;

[0057] For any evaluation object, obtain the redistributed reference scores of each parsing point and form a benchmark segmentation set;

[0058] The K-means clustering algorithm is used on the benchmark segmentation set to divide all elements into 10 clusters;

[0059] Calculate the distance from any element to its corresponding cluster center and record it as the element's cluster center distance. If the cluster center distance of any element is greater than the 70th percentile of all cluster center distances in its corresponding cluster, the resolution point corresponding to the element is recorded as the long-distance adaptation point.

[0060] The redistributed reference scores of each parsing point are normalized by the sigmoid activation function to obtain the adaptation scale;

[0061] For any evaluation object, the sequence of fitness scales of each remote fitness point is recorded as the remote evaluation sequence LS.PT; the fitness scales of all analytical points of the evaluation object are written into the sequence and recorded as the full-distance evaluation sequence LS.AP; any fitness scale is selected from the full-distance evaluation sequence, and the percentile of all fitness scales under the corresponding analytical point is recorded as Hta, and the corresponding percentile when the fitness scale is placed in the sequence LS.PT is recorded as Htb, then the fitness scale standard value IHT of the fitness scale is the root mean square value of Hta and Htb, specifically: IHT = sqrt(Hta×Htb), where sqrt() is the square root function,

[0062] The inheritance execution score IURN of each evaluation object is calculated according to the remote evaluation sequence and the adaptive scale value: Where i1 is the cumulative variable, SnL is the number of elements in the remote evaluation sequence, LS.PT(i1) is the fitness scale corresponding to the i1th element in the remote evaluation sequence, min.PT is the minimum value in the fitness scale corresponding to each element in the remote evaluation sequence, exp() is the exponential function with the natural constant e as the base, hs<> is the harmonic mean function; lg() is the logarithmic function with 10 as the base, IHT i1 is the adaptation scale value corresponding to the i1th element in the remote evaluation sequence.

[0063] Further, in step S400, the method of conducting inheritance risk assessment in combination with the inheritance execution score of each evaluation object is as follows: the length of the benchmark word adaptation portrait is recorded as the parsing amount, the evaluation object with the maximum value in the parsing amount among each evaluation object is searched as the first evaluation object, and at any parsing point, if an evaluation object has a larger redistribution reference score than the first evaluation object, then the evaluation object is defined as having an adaptation event, and the ratio of the number of high adaptation events occurring in the evaluation object to its parsing amount is the high adaptation efficiency; the lower quartile value in the set consisting of all high adaptation efficiencies is the adaptation efficiency threshold; the median value and standard deviation of all inheritance execution scores are recorded as MER and STDER respectively, and the evaluation risk overflow rate ERSR is preset, ERSR∈[1.5,3.5]; if the high adaptation efficiency of an evaluation object is less than the adaptation efficiency threshold and the inheritance execution score of the evaluation object is less than MER-ERSR×STDER, then the evaluation object is evaluated to have an inheritance abnormality risk and is not selected as a deployment server, and each evaluation object with an inheritance abnormality risk is sent to the client's server as an evaluation report.

[0064] Example 2

[0065] Embodiment 2 adopts the same evaluation method as Embodiment 1, with the difference that: in step S300, the method of using the benchmark word adaptation portrait to perform inheritance execution quality time series analysis to obtain the inheritance execution score of each evaluation object is: set a time period as the evaluation period EVtm, and take a value of 2 years;

[0066] For any evaluation object, obtain the redistribution reference score of each resolution point and form a redistribution reference score sequence; in the redistribution reference score sequence, calculate the weighted average of all elements and record it as Mvae, where the weight of each element is the percentile of all elements in the corresponding redistribution reference score sequence; when any element in the redistribution reference score sequence is greater than Mvae, the resolution point corresponding to the element is recorded as a high fitness point, and the number of high fitness points is recorded as RQm;

[0067] A floating-point variable is preset for the resolution points corresponding to all elements, recorded as the inheritance priority Laor. If the resolution point is a high-fitness point, the inheritance priority takes a value of 1, otherwise the inheritance priority takes a value of 0.5; the inheritance priority Faor of any resolution point is the ratio of its inheritance priority to the sum of the inheritance priorities of all resolution points, as follows;

[0068] Calculate the inheritance probability Faor of any analytical point: Where k1 is the cumulative variable, Tn is the number of analytical points;

[0069] In the redistribution reference score sequence, the cumulative inheritance priority FFaor of any element is the cumulative value of each inheritance priority between the element and the first element of the sequence; the first element of the sequence corresponds to the analysis point closest to the current moment; all the cumulative inheritance priorities are written into the sequence and recorded as a cumulative sequence;

[0070] Randomly generate RTN random numbers and record them as selected values. The range of the selected values ​​is RU∈(0,1); where RTN is a preset integer, RTN∈[500,2000];

[0071] Take k2 as the serial number of the parsing point, and for any selected value RU, set its estimation condition as FFaor(k2)<RU≤FFaor(k2+1); search for the parsing point that meets the estimation condition and mark the parsing point as having an estimation event; after marking the parsing points where estimation events occur for all selected values, record the total number of estimation events occurring at the parsing points as RQm_TH, sort RQm_TH from large to small and select the parsing points corresponding to the first RQm elements as the inherited estimation points; where FFaor(k2) and FFaor(k2+1) are the cumulative inheritance precedence rates of the parsing points corresponding to the k2th and k2+1th elements in the redistributed reference score sequence;

[0072] Calculate the inheritance execution score IURN of each evaluation object based on the inheritance evaluation points:

[0073]

[0074] Where k3 is the cumulative variable, BCMF k3 and Faor k3 are the redistribution reference score and inheritance priority rate of the k3th inheritance estimation point respectively, Avae is the average of the corresponding redistribution reference scores of all non-inheritance estimation points, ln() is the logarithmic function with the natural constant e as the base; non-inheritance estimation point refers to the analysis point that does not belong to the inheritance estimation point.

[0075] The embodiment of the present invention provides a full asset wealth protection and inheritance risk assessment system, such as Figure 2 Shown is a structural diagram of the total asset wealth protection and inheritance risk assessment system of the present invention. The total asset wealth protection and inheritance risk assessment system of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned total asset wealth protection and inheritance risk assessment method embodiment are implemented.

[0076] The system comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to run in the following units of the system:

[0077] The full-asset network storage object identification unit is used to identify the reference server and the reallocation server from the full-asset network, which are recorded as the reference object and the evaluation object respectively;

[0078] A benchmark word adaptation portrait construction unit, used to construct a benchmark word adaptation portrait through an asset database of each evaluation object;

[0079] The inheritance execution quality time series analysis unit is used to perform inheritance execution quality time series analysis using the benchmark word adaptation portrait to obtain the inheritance execution score of each evaluation object;

[0080] The inheritance risk assessment unit is used to conduct inheritance risk assessment based on the inheritance execution scores of each evaluation object.

[0081] The total asset wealth protection and inheritance risk assessment system can be run on computing devices such as desktop computers, laptops, PDAs, and cloud servers. The total asset wealth protection and inheritance risk assessment system can be operated on systems including, but not limited to, processors and memories. Those skilled in the art will appreciate that the examples are merely examples of the total asset wealth protection and inheritance risk assessment system and do not constitute a limitation on the total asset wealth protection and inheritance risk assessment system. The system can include more or fewer components than the examples, or a combination of certain components, or different components. For example, the total asset wealth protection and inheritance risk assessment system can also include input and output devices, network access devices, buses, and the like.

[0082] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the operating system of the total asset wealth protection and inheritance risk assessment system, and uses various interfaces and lines to connect the various parts of the entire total asset wealth protection and inheritance risk assessment system operating system.

[0083] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the total asset wealth protection and inheritance risk assessment system by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0084] Although the description of the present invention has been quite detailed and has been described in particular with respect to several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, so as to effectively cover the intended scope of the present invention. In addition, the present invention is described above with the embodiments foreseeable by the inventors, and its purpose is to provide a useful description, and those non-substantial changes to the present invention that are not currently foreseen may still represent equivalent changes of the present invention.

Claims

1. The full-asset wealth protection and inheritance risk assessment method is characterized by: The method comprises the following steps: S100, identifying a reference server and a reallocated server from the full asset network, which are recorded as a reference object and an evaluation object respectively; S200, constructing a benchmark word adaptation portrait through the asset database of each evaluation object; S300, using the benchmark word adaptation portrait to perform inheritance execution quality time series analysis to obtain inheritance execution scores for each evaluation object; S400, conduct inheritance risk assessment based on the inheritance execution scores of each evaluation object; The method in step S300 is specifically to calculate the inheritance execution score of each evaluation object according to the remote evaluation sequence and the adaptation scale value; or to calculate the inheritance execution score of each evaluation object according to the redistribution reference score and the inheritance priority rate.

2. The full-asset wealth protection and inheritance risk assessment method according to claim 1 is characterized in that: In step S100, a reference server and a reallocation server are identified from a full-asset network and recorded as a reference object and an evaluation object respectively. The full-asset network is composed of a plurality of servers, and the servers include a reference server and a plurality of reallocation servers, which are recorded as a reference object and an evaluation object respectively.

3. The full-asset wealth protection and inheritance risk assessment method according to claim 1 is characterized in that: In step S200, a method for constructing a benchmark word adaptation portrait through the asset database of each evaluation object is as follows: each server includes an asset database, the asset database is a database composed of asset documents in JSON format or BSON format, and the asset documents are a collection of texts; every 5-10 natural days is taken as a parsing point, and a parsing point and 2-6 parsing points in the reverse time direction are taken as the mapping period of the parsing point; for any parsing point, the benchmark object merges each asset document of the corresponding mapping period to form a mapping text set, and each asset document of the evaluation object at the parsing point constitutes a parsing text set; the redistribution reference score of the evaluation object is calculated based on the mapping text set and the parsing text set of the parsing point, and the sequence composed of the redistribution reference scores of all parsing points is recorded as the benchmark word adaptation portrait.

4. The full-asset wealth protection and inheritance risk assessment method according to claim 3 is characterized in that: The method of calculating the redistributed reference score of the evaluation object based on the mapping text set and the parsing text set of the parsing point is as follows: for any parsing point, the text set is preprocessed, including text cleaning, word segmentation, and stop word removal; The LDA latent Dirichlet distribution model is used to identify keywords in the text set, obtain keywords and corresponding weights, and the CRF conditional random field algorithm is used to classify keywords into nouns and adjectives and classify them into related words and matching words respectively. The weights of related words and matching words are normalized respectively; Word embedding analysis is performed on each keyword in the text set and its corresponding weights to obtain word vectors. The word vector obtained based on the analysis of the mapped text set is the benchmark portrait, and the word vector obtained based on the analysis of the parsed text set is the evaluation portrait. The cosine similarity between the evaluation portrait and the benchmark portrait is calculated as the reference score for the redistribution of the parsing point.

5. The full-asset wealth protection and inheritance risk assessment method according to claim 1 is characterized in that: In step S300, the method of using the benchmark word adaptation portrait to perform inheritance execution quality time series analysis to obtain inheritance execution scores of each evaluation object is: set a time period as an evaluation period, and within the evaluation period, record the time scale of obtaining the redistributed reference score as a parsing point; For any evaluation object, obtain the redistribution reference score of each resolution point and form a benchmark segmentation set; use the K-means clustering algorithm to divide all elements of the benchmark segmentation set into several clusters; calculate the distance from any element to its corresponding cluster center and record it as the element's cluster center distance. If the cluster center distance of any element is greater than the 70th percentile of all cluster center distances in its corresponding cluster, then the resolution point corresponding to the element is recorded as a long-distance adaptation point; normalize the redistribution reference score of each resolution point through the sigmoid activation function to obtain the adaptation scale; The sequence of fitness scales of each remote fitness point is recorded as the remote evaluation sequence LS.PT; the fitness scales of all resolution points of the evaluation object are written into a sequence recorded as the full-distance evaluation sequence LS.AP; any fitness scale is selected from the full-distance evaluation sequence, and the percentile of all fitness scales under the corresponding resolution point is recorded as Hta, and the corresponding percentile when the fitness scale is placed in the sequence LS.PT is recorded as Htb, then the fitness scale standard value IHT of the fitness scale is the root mean square value of Hta and Htb; the inheritance execution score of each evaluation object is calculated according to the remote evaluation sequence and the fitness scale standard value.

6. The full-asset wealth protection and inheritance risk assessment method according to claim 1 is characterized in that: In step S300, an alternative method for obtaining the inheritance execution score of each evaluation object by performing inheritance execution quality time series analysis using the benchmark word adaptation portrait is as follows: set a time period as the evaluation period EVtm; for any evaluation object, obtain the redistribution reference score of each parsing point, and form a redistribution reference score sequence; in the redistribution reference score sequence, calculate the weighted average of all elements and record it as Mvae, where the weight of each element is the percentile of all elements in the corresponding redistribution reference score sequence; when any element in the redistribution reference score sequence is greater than Mvae, the parsing point corresponding to the element is recorded as a high adaptation point, and the number of high adaptation points is recorded as RQm; A floating-point variable is preset for the resolution points corresponding to all elements, recorded as the inheritance priority Laor. If the resolution point is a high-fitness point, the inheritance priority is 1, otherwise the inheritance priority is 0.5; the inheritance priority Faor of any resolution point is the ratio of its inheritance priority to the sum of the inheritance priorities of all resolution points; in the redistribution reference score sequence, the cumulative inheritance priority FFaor of any element is the cumulative value of each inheritance priority between the element and the first element of the sequence; all cumulative inheritance priorities are written into the sequence and recorded as a cumulative sequence; Generate RTN random numbers randomly as selected values, and the range of selected values ​​is RU∈(0,1); take k2 as the serial number of the resolution point, and for any selected value RU, set its estimation point condition to FFaor(k2)<RU≤FFaor(k2+1); search for the resolution point that meets the estimation point condition and mark the occurrence of an estimation point event at the resolution point; The total number of estimation point events occurring at the analysis point is recorded as RQm_TH. RQm_TH is arranged from large to small and the analysis points corresponding to the first RQm elements are selected as inheritance estimation points. The inheritance execution score of each evaluation object is calculated based on the redistribution reference score and the inheritance priority rate.

7. The full-asset wealth protection and inheritance risk assessment method according to claim 1 is characterized in that: In step S400, the method of conducting inheritance risk assessment in combination with the inheritance execution scores of each evaluation object is as follows: the length of the benchmark word adaptation portrait is recorded as the parsing amount, and the evaluation object with the maximum value in the parsing amount among each evaluation object is searched as the first evaluation object. At any parsing point, if an evaluation object has a larger redistribution reference score than the first evaluation object, it is defined that the evaluation object has a high adaptation event, and the ratio of the number of high adaptation events of the evaluation object to its parsing amount is the high adaptation efficiency. The lower quartile value of the set of all high adaptation efficiencies is the adaptation efficiency threshold; the median and standard deviation of all inheritance execution scores are MER and STDER respectively, and the evaluation risk overflow rate ERSR is preset, ERSR∈[1.5,3.5]; if the high adaptation efficiency of an evaluation object is less than the adaptation efficiency threshold and the inheritance execution score of the evaluation object is less than MER-ERSR×STDER, then the evaluation object is evaluated to have an inheritance abnormality risk and is not selected as a deployment server. Each evaluation object with an inheritance abnormality risk is sent to the client's server as an evaluation report.

8. The full asset wealth protection and inheritance risk assessment system is characterized by: The total-asset wealth protection and inheritance risk assessment system includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the total-asset wealth protection and inheritance risk assessment method described in any one of claims 1 to 7 are implemented. The total-asset wealth protection and inheritance risk assessment system runs on desktop computers, laptop computers, PDAs, and computing devices in cloud data centers.

Images