CN116776274A

CN116776274A - Electronic seal data management system based on data analysis

Info

Publication number: CN116776274A
Application number: CN202311075425.4A
Authority: CN
Inventors: 庄玉龙; 李超; 赵云; 范红达
Original assignee: Beijing Dianju Information Technology Co ltd
Current assignee: Beijing Dianju Information Technology Co ltd
Priority date: 2023-08-25
Filing date: 2023-08-25
Publication date: 2023-09-19
Anticipated expiration: 2043-08-25
Also published as: CN116776274B

Abstract

The embodiment of the specification discloses an electronic seal data management system based on data analysis, which belongs to the field of electronic data processing and comprises the following components: the data acquisition module is used for acquiring the use data of the electronic seal; the anomaly detection module is used for detecting an anomaly value and acquiring an optimal value of a parameter K in an LOF algorithm; the data management module is used for managing the electronic seal use data by using an LOF algorithm; the abnormality detection module includes: an initial parameter sub-module for setting an initial value; a transformation parameter submodule for obtaining different parameters K; the correlation analysis submodule acquires correlations among different departments; the abnormality analysis submodule detects abnormality of the electronic seal use data; the difference evaluation sub-module is used for obtaining the difference of LOF values under different parameter K values; and the parameter sub-module is preferably used for acquiring a parameter K value when the difference is larger than a preset threshold value. The application improves the accuracy of anomaly detection on the use frequency of the electronic seal by using the LOF algorithm under different scenes.

Description

Electronic seal data management system based on data analysis

Technical Field

The application relates to the field of electronic data processing, in particular to an electronic seal data management system based on data analysis.

Background

The seal is an authoritative symbol, and in the paper office mode, the physical seal is an important mark for the effectiveness of documents and files. The electronic seal is a representation form of the electronic signature, which converts the electronic signature operation into a visual pattern with the same efficacy as paper file signature and stamping operation by using image processing and digital signature technology, and has a management and use mode close to the physical seal, thereby being convenient for use and ensuring the security of the printing process. The electronic seal technology not only can solve the security problem of electronic document transmission, but also can keep the consistency with the traditional office habit, so that the technology is more and more widely applied.

The use of the electronic seal is related to the running state of the company, so that the current running state of the company can be known by analyzing the use of the electronic seal. Based on the above, it is necessary to study an electronic seal data management system based on data analysis to analyze the use condition of the electronic seal, so as to help companies fully mine the value of the electronic seal and improve the management efficiency.

Disclosure of Invention

The embodiment of the specification provides an electronic seal data management system based on data analysis, which comprises: the data acquisition module is used for acquiring electronic seal use data, wherein the electronic seal use data comprises frequency data and time data; the anomaly detection module is used for carrying out anomaly detection on the electronic seal use data and obtaining an optimal value of the LOF algorithm parameter K; the data management module is used for managing the electronic seal use data based on the LOF algorithm and the optimal value of the parameter K; wherein, the anomaly detection module includes: an initial parameter sub-module for setting initial value of parameter K in LOF algorithm The method comprises the steps of carrying out a first treatment on the surface of the A transformation parameter submodule for transforming the parameter K value to obtain different parametersA number K; the correlation analysis submodule is used for analyzing the correlation of frequency of using the electronic seal among departments; the abnormality analysis submodule is used for detecting abnormality based on the parameter K and the frequency of using the electronic seal; the difference evaluation submodule is used for acquiring difference of LOF values of frequency of using the electronic seal by each department with different parameter K values>The method comprises the steps of carrying out a first treatment on the surface of the A preferred parameter sub-module for obtaining the difference of LOF values +.>And the value of the parameter K is an optimal value when the value is larger than a preset threshold value.

In some embodiments, the correlation analysis submodule includes: a history data analysis unit and a correlation acquisition unit; the historical data analysis unit is used for acquiring first frequency information of the electronic seal used by the department a on the same day and second frequency information of the electronic seal used by the department b on the same day in the ith historical data; the historical data analysis unit is also used for acquiring first average frequency information of the electronic seal used by the department a and second average frequency information of the electronic seal used by the department b in the n pieces of historical data; a correlation acquisition unit for acquiring the correlation of the electronic seal used by the department a and the department b according to the first frequency information, the second frequency information, the first average frequency information and the second average frequency information ；

Correlation of department a and department b using electronic sealCalculated by the following formula:

；

wherein ,representing the number of history data +.>Representing the frequency of using the electronic seal by the department a on the same day in the ith historical data; />Representing the frequency of using the electronic seal by the department b on the same day in the ith historical data; />Representing the frequency of using the electronic seal by the department a in the n historical data; />The n historical data are represented by the average frequency of using the electronic seal by the department b.

In some embodiments, the anomaly analysis sub-module further comprises: the system comprises a frequency analysis unit, a time analysis unit, a related department reference unit, a department duty analysis unit and a comprehensive judgment unit; the frequency analysis unit is used for analyzing whether the frequency of using the electronic seal by the department is abnormal or not based on the frequency data of using the electronic seal by the department; the time analysis unit is used for judging whether the time of using the electronic seal by each department is abnormal or not by analyzing the time of using the electronic seal by each department every day; the related department reference unit is used for referring to the related department use time to judge whether the department use time is abnormal or not; the department duty analysis unit is used for analyzing whether the frequency of the department use electronic seal is abnormal according to the proportion of the department use electronic seal to the total electronic seal usage; and the comprehensive judging unit is used for acquiring the abnormal results of the electronic seal used by each department according to the results of the frequency analysis unit, the time analysis unit, the related department reference unit and the department duty ratio analysis unit.

In some embodiments, the frequency analysis unit further comprises: a difference calculating subunit, a change rate calculating subunit and a frequency judging subunit; a difference calculating subunit for obtaining the difference between the frequency data of the department a using the electronic seal and the historical frequency dataThe method comprises the steps of carrying out a first treatment on the surface of the A change rate calculating subunit for acquiring the change rate difference between the frequency data of the department a using the electronic seal and the historical frequency data>The method comprises the steps of carrying out a first treatment on the surface of the A frequency judging subunit for acquiring the frequency abnormality possibility of the electronic seal used by the department a according to the difference and the change rate difference>；

Probability of frequency abnormalityCalculated by the following formula:

；

wherein ,representing the difference between the current frequency data and the stamp frequency used at the ith historical moment of the history; />The first change rate of the current frequency data and the adjacent data using the seal frequency is compared with the difference of the second change rate of the i-th historical moment and the adjacent moment using the seal frequency.

In some embodiments, the time analysis unit further comprises: a time analysis subunit, a related department reference subunit and a time judgment subunit; a time analysis subunit for acquiring the time and frequency of using the electronic seal per hour of each department and acquiring the frequency difference per hour of using the electronic seal The method comprises the steps of carrying out a first treatment on the surface of the A related department reference subunit for acquiring the correlation of department a and related department +.>'A'; a time judging subunit forAccording to the frequency difference per hour->Acquisition of daily time abnormality->And according to the daily time abnormality +.>And correlation->' possibility of time abnormality of use of electronic stamp by acquisition department a->；

Daily time abnormalityCalculated by the following formula:

；

wherein ,time of presentation->Indicate->When the electronic seal is used by the hour department a, the frequency of the electronic seal used by the department a is different from the frequency of the electronic seal used by the department a acquired according to historical data;

time anomaly possibility of department a using electronic sealCalculated by the following formula:

；

wherein m represents the number of relevant departments,representing the correlation between the department a and the b-th department after normalizing the correlation between the department a and other m departments; />Indicating the abnormality of department a in time of day, +.>Indicating the time of day anomaly for department b.

In some embodiments, the related department reference unit further comprises: a time curve subunit, an abnormality degree analysis subunit and a department judgment subunit; a time curve subunit for acquiring the abnormal frequency time curve of each department using the electronic seal every day The method comprises the steps of carrying out a first treatment on the surface of the An abnormality degree analysis subunit for ++according to the frequency abnormality time curve>Acquiring the degree of abnormality DTW of a related department; a department judgment subunit for acquiring department abnormality degree +.f of department a based on the related department abnormality degree DTW>；

Degree of department abnormalityCalculated by the following formula:

；

wherein m represents the number of departments; f represents a reference time interval;representing the reference time interval asf, DTW distance of department a and department b; />Indicating the relevance of department a and department b to use the electronic seal.

In some embodiments, the department ratio analysis unit further comprises: a proportional relation analysis subunit, a proportional difference analysis subunit and a proportional judgment subunit; the proportion relation analysis subunit is used for acquiring a first proportion of the frequency of using the electronic seal by the department a to the frequency of using the electronic seal by the whole company on the same day; the proportional relation analysis subunit is further used for acquiring a second proportion of the frequency of using the electronic seal by the department a to the frequency of using the electronic seal by the whole company on the same day in the ith historical data; a proportion difference analysis subunit for acquiring the difference between the first proportion and the second proportion of the electronic seal used by the department aThe method comprises the steps of carrying out a first treatment on the surface of the The proportion difference analysis subunit is also used for acquiring a first difference of the frequency of using the electronic seal between the current analysis data of the department a and the adjacent data; the proportion difference analysis subunit is also used for acquiring a second difference of the frequency of using the electronic seal between the ith historical data of the department a and the adjacent data; a proportion difference analysis subunit for acquiring the difference of the first difference and the second difference of the department a The method comprises the steps of carrying out a first treatment on the surface of the A proportion judging subunit for judging the proportion according to-> and />Acquiring possibility of abnormality of department frequency of department a using electronic seal +.>；

Possibility of abnormal department frequencyBy the following stepsAnd (3) calculating a formula:

；

where n represents the number of n history data selected.

In some embodiments, the comprehensive decision unit further comprises a result acquisition subunit and an anomaly decision subunit; a result acquisition subunit for acquiring the degree of abnormality of the department a when using the electronic sealFrequent abnormality possibility of using electronic seal by department a->The possibility of frequency abnormality of using electronic seal by department b>Correlation with the b-th department obtained after normalization of the correlation between department a and other m departments ∈>Possibility of abnormality in department frequency->The method comprises the steps of carrying out a first treatment on the surface of the An abnormality determination subunit operable to determine ++the frequency of use of the electronic stamp by the result acquisition department a based on the result acquired by the result acquisition subunit>；

Frequency abnormality of department a using electronic sealCalculated by the following formula:

；

wherein ,a preset threshold value; norm () represents a normalization function; m represents m departments.

In some embodiments, the variance-evaluating sub-module further includes a LOF calculation unit and a variance calculation unit; the LOF calculation unit is used for obtaining an LOF value according to the K value and the frequency of using the electronic seal by the department; a difference calculation unit for obtaining the difference of LOF values of the electronic seal frequency used by each department under different K values ；

LOF value difference of frequency of using electronic seal by departmentCalculated by the following formula:

；

wherein exp () represents a power function; n represents the number of using frequency of the electronic seal, and m represents the number of departments;indicating the LOF value corresponding to the b department data in the i-th historical data acquired by using the LOF algorithm after normalization,>the frequency abnormality of using the electronic seal by department b in the ith historical data is indicated.

In some embodiments, the data management module further comprises: the system comprises an LOF calculation sub-module, a judging sub-module, a marking sub-module and a reporting sub-module; the LOF calculation sub-module is used for acquiring LOF values of the electronic seal use data of each department by using an LOF algorithm according to the optimal value of the parameter K; the judging submodule is used for judging whether the LOF value is larger than a preset threshold value or not; the marking submodule is used for marking the electronic seal use data when the LOF value is larger than a preset threshold value; and the reporting sub-module is used for reporting the marked electronic seal use data.

The electronic seal data management system based on data analysis provided in the embodiments of the present disclosure may have at least the following beneficial effects: (1) The application greatly increases the reliability of acquiring the abnormality of the use frequency of the electronic seal between departments by combining the correlation of the electronic seal used by each department of the company and analyzing the time of the electronic seal used by each department; (2) Meanwhile, according to the anomaly analysis result and LOF values obtained by using different K values, the corresponding anomaly detection result under the preferable K value is obtained, so that the anomaly detection time is greatly saved, and the anomaly detection reliability is increased; (3) The possibility of the company to find out the abnormal use of the company electronic seal is increased, and the use safety of the company seal is greatly improved.

Additional features will be set forth in part in the description which follows. As will become apparent to those skilled in the art upon review of the following and drawings, or may be learned by the production or operation of the examples. The features of the present specification can be implemented and obtained by practicing or using the various aspects of the methods, tools, and combinations set forth in the detailed examples below.

Drawings

The present specification will be further described by way of exemplary embodiments, which will be described in detail by way of the accompanying drawings. The embodiments are not limiting, in which like numerals represent like structures, wherein:

FIG. 1 is an exemplary block diagram of an electronic stamp data management system based on data analysis according to some embodiments of the present description;

FIG. 2 is an exemplary block diagram of an anomaly detection module shown in accordance with some embodiments of the present specification;

FIG. 3 is an exemplary block diagram of a correlation analysis sub-block shown in accordance with some embodiments of the present description;

FIG. 4 is an exemplary block diagram of an anomaly analysis sub-module shown in accordance with some embodiments of the present description;

FIG. 5 is a schematic diagram of a frequency analysis unit according to some embodiments of the present disclosure;

FIG. 6 is a schematic diagram of a time analysis unit according to some embodiments of the present description;

FIG. 7 is a schematic diagram of a related department reference unit shown in accordance with some embodiments of the present disclosure;

FIG. 8 is a schematic diagram of a department fraction analysis unit according to some embodiments of the present disclosure;

FIG. 9 is a schematic diagram of an integrated judgment unit according to some embodiments of the present description;

FIG. 10 is an exemplary block diagram of a variance-evaluating sub-module shown in accordance with some embodiments of the present description;

FIG. 11 is an exemplary block diagram of a data management module shown in accordance with some embodiments of the present description.

The reference numerals in the figures illustrate: 100. a data acquisition module; 200. an anomaly detection module; 300. a data management module; 400. an electronic seal data management system based on data analysis; 210. an initial parameter sub-module; 220. a transformation parameter sub-module; 230. a correlation analysis sub-module; 240. an anomaly analysis sub-module; 250. a difference evaluation sub-module; 260. a preferred parameter sub-module; 231. a history data analysis unit; 232. a correlation acquisition unit; 241. a frequency analysis unit; 242. a time analysis unit; 243. a related department reference unit; 244. department ratio analysis unit; 245. a comprehensive judgment unit; 241A, a variance calculation subunit; 241B, a change rate calculation subunit; 241C, frequency judgment subunit; 242A, a time analysis subunit; 242B, related department reference subunits; 242C, a time judgment subunit; 243A, time curve subunit; 243B, an abnormality degree analysis subunit; 243C, department judging subunit; 244A, a proportional relationship analysis subunit; 244B, a proportion difference analysis subunit; 244C, a proportion judging subunit; 245A, a result acquisition subunit; 245B, an abnormality determination subunit; 251. a LOF calculation unit; 252. a difference calculation unit; 310. a LOF calculation sub-module; 320. judging a sub-module; 330. marking a sub-module; 340. and reporting the sub-module.

Detailed Description

In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the drawings that are required to be used in the description of the embodiments will be briefly described below. It is apparent that the drawings in the following description are only some examples or embodiments of the present specification, and it is possible for those of ordinary skill in the art to apply the present specification to other similar situations according to the drawings without inventive effort. Unless otherwise apparent from the context of the language or otherwise specified, like reference numerals in the figures refer to like structures or operations.

It should be appreciated that as used in this specification, a "system," "apparatus," "unit" and/or "module" is one method for distinguishing between different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

As used in this specification and the claims, the terms "a," "an," "the," and/or "the" are not specific to a singular, but may include a plurality, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.

A flowchart is used in this specification to describe the operations performed by the system according to embodiments of the present specification. It should be appreciated that the preceding or following operations are not necessarily performed in order precisely. Rather, the steps may be processed in reverse order or simultaneously. Also, other operations may be added to or removed from these processes.

Based on the technical problems, the specification provides an electronic seal data management system based on data analysis. The system provided in the embodiments of the present specification will be described in detail with reference to the accompanying drawings.

According to the application, through combining the correlation of using the electronic seal by each department of the company and analyzing the time of using the electronic seal by each department, the reliability of acquiring the frequency abnormality of using the electronic seal among departments is greatly increased, meanwhile, the LOF value acquired by using different K values is analyzed according to the abnormality analysis result, so that the corresponding abnormality detection result under the optimal K value is acquired, the time of abnormality detection is greatly saved, the reliability of abnormality detection is increased, the possibility of finding the electronic seal of the company for using the abnormality by the company is increased, and the use safety of the company seal is greatly increased.

And the use condition of the electronic seal is analyzed, so that whether the use of the seal is abnormal or not is judged. The specific scene aimed by the application is as follows: and analyzing and counting the use frequency of each electronic seal, and further managing the use data of the electronic seal.

Fig. 1 is an exemplary block diagram of an electronic stamp data management system based on data analysis according to some embodiments of the present description.

Referring to fig. 1, in some embodiments, an electronic stamp data management system 400 based on data analysis includes: the data acquisition module 100 is used for acquiring electronic seal use data, wherein the electronic seal use data comprises frequency data and time data; the anomaly detection module 200 is used for acquiring an optimal value of LOF algorithm parameter K according to the electronic seal usage data combined with the LOF algorithm; the data management module 300 is configured to tag the electronic seal usage data with an LOF algorithm that is an optimal value of the parameter K.

Specifically, the data acquisition module 100 acquires all electronic seal data to be analyzed using existing equipment. For companies, the electronic seal system can accurately record the seal time based on the time stamp technology in the operation process, so that the corresponding seal information is collected, wherein the seal information comprises the use department of the seal and the seal use time. And drawing a seal use frequency change curve of each department according to the seal use frequency of the system acquisition company per day.

The electronic seal is a technology which utilizes digital certificates and electronic signature technology to simulate the functions of the traditional seal. Electronic seals rely on digital certificates and electronic signatures to ensure the security of the electronic seals, and the electronic seals are difficult to forge. The electronic seal is similar to a physical seal in use, and is visual in operation. The electronic seal can be applied to electronic files such as electronic mails, electronic contracts, electronic reports and the like.

The electronic seal system has the function of recording seal information, and the data acquisition module is directly connected with the electronic seal system to acquire seal data. The acquired electronic seal data comprises: the seal belongs to a using department for determining which department uses the seal. Seal service time: the electronic seal system is based on a time stamp technology and can accurately record the seal time. Number of seal uses: the number of times of using each seal in a certain time is calculated. And then according to the acquired corresponding seal information: and calculating the total number of times of using the seal by each department every day, and drawing a seal using time change curve of each department on a time sequence plane.

Specifically, the anomaly detection module 200 is configured to detect an anomaly in use of the electronic seal, analyze frequencies of use of the electronic seal by each department of the company, obtain a relationship between frequencies of use of the electronic seal by each department, and analyze whether the use of the electronic seal is anomaly by combining with an LOF anomaly detection algorithm. And the management system performs abnormality detection on the seal data in the system at regular intervals for audit management of the electronic seal data, so as to analyze whether the electronic seal is used abnormally or not. The abnormality detection algorithm is an existing LOF abnormality detection algorithm, and detection is carried out based on the daily use times of the electronic seal and the use frequency of the electronic seal corresponding to each department. However, in the anomaly detection process, different detection results are obtained based on different parameters K, so that different parameters K need to be converted to obtain a preferable detection effect. In the application, the initial value of K is 2, and the parameter K is continuously converted every time K+1 is converted, so that the optimal detection effect is obtained.

Specifically, the data management module 300 performs exception analysis on the frequency of using the LOF algorithm by using the electronic seal for each department of the obtained company, and the data management system extracts and marks the data with the abnormal value LOF greater than 1, so that the reasons for the frequency of using the electronic seal, etc. need to be reported and analyzed. Wherein, select the data that LOF is greater than 1 as the unusual data and mark, have following advantage: most normal points (lof≡0) can be filtered out, focusing on capturing points of obvious abnormalities (LOF > > 1), confidence interval settings: it is desirable to detect outliers with 95% confidence intervals. Selecting LOF >1 may be implemented. And the electronic seal data with LOF more than 1 is selected for marking, so that focusing analysis and real abnormal problem solving are facilitated.

In summary, the application acquires the abnormal detection result under different parameters by continuously changing the LOF algorithm parameter K value by combining the electronic seal use data. And obtaining the optimal K value by comparing LOF value differences under different parameters. The corresponding abnormal detection result is more accurate. This greatly improves the accuracy of judging the abnormality of the frequency of use of the seal by the department. The application directly selects the optimal K value by comparing LOF value differences under different parameter K values. And the abnormal detection is carried out without repeatedly using different K values, so that a lot of repeated detection time is saved. By using the optimal K value parameter, the LOF algorithm can more accurately detect the condition of using the seal abnormally. The possibility of finding out the abnormal use seal is effectively improved, and the safe use of the seal is ensured.

The LOF algorithm, which is known as Local Outlier Factor, namely a local anomaly factor, is an algorithm for finding a density based on local anomalies, and detects anomaly points by calculating local density differences between a data object and its K neighbors. The "degree of isolation" of the detected object, i.e. the local density difference of its K-neighbors, is calculated, and the "degree of isolation" of the object is compared with the average "degree of isolation" of its K-neighbors, if the "degree of isolation" of an object is much greater than the average "degree of isolation" of its K-neighbors, it is likely to be an outlier.

FIG. 2 is an exemplary block diagram of an anomaly detection module shown in accordance with some embodiments of the present specification.

Referring to fig. 2, in some embodiments, the anomaly detection module 200 includes: wherein, the obtaining the optimal value of the parameter K by the anomaly detection module 200 includes: an initial parameter sub-module 210, configured to set an initial value of a parameter K in the LOF algorithm; a transformation parameter sub-module 220, configured to transform a parameter K value to obtain different K parameters; a correlation analysis sub-module 230 for analyzing the correlation between departments using the frequency of electronic seal; an anomaly analysis sub-module 240 for anomaly detection based on the parameter K and the frequency of using the electronic stamp; the difference evaluation sub-module 250 is used for obtaining the difference of LOF values of the frequency of using the electronic seal by each department under different K values; the preferred parameter sub-module 260 obtains the K value as the optimal value when the difference in the LOF values is greater than the preset threshold.

The accuracy of the frequency abnormality of the electronic seal used by the acquisition department is improved; the initial parameter sub-module 210 and the transformation parameter sub-module 220 continuously change K values to obtain abnormal detection results under different K parameters, the correlation analysis sub-module 230 analyzes the correlation among departments, the abnormal analysis sub-module 240 carries out abnormal detection based on the different K values, the difference evaluation sub-module 250 compares LOF value differences under the different K values, the optimal value is selected, and the corresponding abnormal detection results are more accurate. 2. The time for detecting the abnormality by repeatedly using different K values is saved; the LOF value difference under different K values is directly compared, so that the optimal K value is quickly selected, and the repeated detection time is saved. 3. Abnormal electronic seal data can be found more easily; by using the optimal K value parameter, the LOF algorithm can more accurately detect the abnormality, effectively discover the abnormality, and ensure the safe use of the electronic seal.

The method comprises the steps of calculating the number of neighbors of an object to be referred to in the isolation degree by using an LOF algorithm, wherein the parameter K value in the LOF algorithm is an important parameter in the LOF algorithm, and the number of the neighbors to be referred to is the number of the neighbors to be referred to. The larger the K value is, the more neighbor objects are referenced, and the more stable the LOF algorithm is; the smaller the K value, the fewer the reference neighbor objects, the more selective the LOF algorithm will be. For unorganized, centrally distributed data, a larger value of K is required. For locally aggregated, unevenly distributed data, smaller K values may be used. Therefore, the effect of the LOF algorithm on different data sets can be made different by properly adjusting the K value. Too large a value of K may ignore some outliers; too small a value of K may create some false outliers. The K value is adjusted according to the specific data set, so that a better LOF algorithm effect is obtained.

In summary, the K value parameter is continuously changed through the cooperation of all the sub-modules, and finally the optimal K value is selected.

The relationship between the frequency of using the seal between the departments is obtained, and because different departments may have coordination work in the working process, the relationship may correspond to the frequency of using the seal, so the relationship between the two departments (such as signing a customer by a business department, using a seal, and correspondingly requiring a production department, signing an order with a raw material manufacturer, using a seal, and producing a product required by the customer) of using the electronic seal can be obtained through analyzing the historical data.

Specifically, the relevant departments in the present application may refer to all other departments in the company except the department a; more specifically, the whole company has m departments other than the department a,indicating the relativity of department a with the j-th department,/-th department>Can be calculated by formula (1),>the value of (2) is 0 to 1, representing the degree of correlation between the two departments, will +.>After normalization treatment, the ∈10 is obtained>I.e. all->The sum of the correlations of all departments of the company (except department a) and department a is 1.

FIG. 3 is an exemplary block diagram of a correlation analysis sub-block shown in accordance with some embodiments of the present description.

Referring to fig. 3, in some embodiments, the correlation analysis sub-module 230 includes: a history data analysis unit 231 and a correlation acquisition unit 232; a history data analysis unit 231 for acquiringIn the ith historical data, the first frequency information of the electronic seal used by the department a on the same day and the second frequency information of the electronic seal used by the department b on the same day; the historical data analysis unit 231 is further configured to obtain, from the n pieces of historical data, first average frequency information of using the electronic seal by the department a and second average frequency information of using the electronic seal by the department b; a correlation acquisition unit 232 for acquiring the correlation of the electronic seal used by the departments a and b according to the first frequency information, the second frequency information, the first average frequency information and the second average frequency informationThe method comprises the steps of carrying out a first treatment on the surface of the Correlation of department a and department b using electronic seal ∈>Calculated by the following formula:

；

wherein ,representing the number of history data +.>Representing the frequency of using the electronic seal by the department a on the same day in the ith historical data; />Representing the frequency of using the electronic seal by the department b on the same day in the ith historical data; />Representing the frequency of using the electronic seal by department a in average in n historical data, ++>Representing the frequency of using the electronic seal by the department b in the n historical data; specifically, in the present application, n=90, i.e. one quarter detection; namely, when the official seal is used between two departments in the historical data The closer the ratio of numbers is, i.e. the +.>The smaller the correlation, the higher the frequency of using official seals by the two departments.

FIG. 4 is an exemplary block diagram of an anomaly analysis sub-module shown in accordance with some embodiments of the present specification.

Referring to fig. 4, in some embodiments, the anomaly analysis sub-module 240 includes: a frequency analysis unit 241, a time analysis unit 242, a related department reference unit 243, a department duty analysis unit 244, and a comprehensive judgment unit 245; a frequency analysis unit 241 for analyzing whether the frequency of using the electronic seal by the department is abnormal based on the frequency data of using the electronic seal by the department; a time analysis unit 242 for obtaining whether the time of using the electronic seal by each department is abnormal by analyzing the time of using the electronic seal by each department every day; a related department reference unit 243 for referring to the related department use time to determine whether the department use time is abnormal; a department's duty analysis unit 244 for analyzing whether the frequency of the department's use of the electronic seal is abnormal according to the proportion of the department's use of the electronic seal to the total electronic seal usage; and the comprehensive judging unit 245 is used for acquiring the result of whether the electronic seal is abnormal or not used by each department according to the result of each unit.

And analyzing the frequency abnormality of the used seal by the department according to the time frequency curve change, and further obtaining the abnormality degree of each item of data by analyzing the frequency of the used seal according to the acquired frequency of the used seal by the department.

Fig. 5 is a schematic diagram of a frequency analysis unit according to some embodiments of the present disclosure.

Referring to fig. 5, in some embodiments, the frequency analysis unit 241 includes: a difference calculation subunit 241A, a change rate calculation subunit 241B, and a frequency judgment subunit 241C; a difference calculating subunit 241A, configured to obtain a difference between frequency data of the electronic seal used by the department a and historical frequency data; a change rate calculating subunit 241B, configured to obtain a change rate difference between the frequency data of the electronic seal used by the department a and the historical frequency data; the frequency judging subunit 241C is configured to obtain, according to the difference and the change rate difference, a frequency anomaly possibility of using the electronic seal by the department a, where the frequency anomaly possibility is calculated according to the following formula:

；

wherein ,representing the difference between the current frequency data and the stamp frequency used at the ith historical moment of the history; />The first change rate of the current frequency data and the adjacent data using the seal frequency is compared with the difference of the second change rate of the i-th historical moment and the adjacent moment using the seal frequency. That is, the larger the difference between the current analysis data and the frequency of using the seal corresponding to other historical moments is, and the larger the difference between the current analysis data and the frequency of using the seal corresponding to the historical moments is, the more likely the current analysis data is abnormal.

Specifically, adjacent data in the present application refers to a time/data closest to the current analysis time/data in time sequence, and specifically refers to a previous or a next data record relative to the current frequency data in the formula; adjacent time refers to the data record of the previous or the next time relative to the i-th historical time; exemplary: if the current analysis is 8 months and 10 days data, the adjacent data refer to 8 months and 9 days data; similarly, if the i-th historical time (8 months 9 days) is analyzed: the adjacent time may refer to time data of 8 months, 8 days or 8 months, 10 days.

The analysis of the abnormality of the using time is performed, so that the using time of the seal is obtained by analyzing the historical data, and the abnormality of the corresponding using time is obtained by analyzing the daily using seal time of each department, namely, the abnormality of the using time of the seal corresponding to the current analysis data is obtained (namely, whether the analysis exists or not and the condition of using the seal in the non-working time is analyzed).

Fig. 6 is a schematic diagram of a time analysis unit according to some embodiments of the present description.

Referring to fig. 6, in some embodiments, the time analysis unit 242 includes: a time analysis subunit 242A, a related department reference subunit 242B, and a time determination subunit 242C; a time analysis subunit 242A, configured to obtain time and frequency of using the electronic seal per hour by each department, and obtain a difference of frequency of using the electronic seal per hour; a related department reference subunit 242B, configured to obtain a correlation between the department a and the related department; a time judging subunit 242C, configured to obtain a daily time abnormal condition according to the hourly frequency difference, and obtain a time abnormal possibility of using the electronic seal by the department a according to the daily time abnormal condition and the correlation;

Daily time abnormalityCalculated by the following formula:

；

meanwhile, as factors such as overtime exist possibly, the time difference between the using time of the electronic seal and the usual time is large, when the abnormality of the using time of the seal of the current department is analyzed, the using time of the seal of other departments is analyzed, and then whether the using time of the current department is abnormal or not is judged:

；

wherein m represents the number of relevant departments,representing the correlation between the department a and the b-th department obtained after normalization of the correlation between the department a and other m departments, namely, the sum of the correlations between all departments and the department a is 1; />Indicating the abnormality of department a in time of day, +.>The daily time abnormality of the department b is represented, namely, when the abnormality degree of the seal using times of each time of the current corresponding department a on the same day is solved to be larger, and the abnormality degree of the seal using times of the corresponding department is correspondingly smaller, the abnormality degree of the seal using times of the current department is explained to be smaller.

According to the analysis, the assistance work exists among the departments, but the assistance work may have time delay, so that the abnormal degree of the current department seal using frequency can be analyzed according to the acquired abnormal Ts-t time curve of the using frequency among the departments.

FIG. 7 is a schematic diagram of a related department reference unit according to some embodiments of the present description.

Referring to fig. 7, in some embodiments, the relevant department reference unit 243 includes: a time curve subunit 243A, an abnormality degree analysis subunit 243B, and a department determination subunit 243C; a time curve subunit 243A, configured to obtain a frequency abnormal time curve of each department using the electronic seal every day; an anomaly degree analysis subunit 243B, configured to obtain a related department anomaly degree DTW according to the frequency anomaly time curve; a department determination subunit 243C, configured to obtain a department abnormality degree of the department a according to the related department abnormality degree DTW;

degree of department abnormalityCalculated by the following formula:

；

wherein m represents the number of departments;frequency anomaly time curve representing daily usage electronic seal acquired based on department a and department b +.>Acquiring DTW distances of a department a and a department b; f represents a reference time interval, namely selecting a department a and a department b to calculate DTW distances of the department a and the department b by using frequency data of the electronic seal in f days; / >Indicating the relevance of department a and department b to use the electronic seal. Specifically, in the present application, f=15, i.e. the coordination period between departments is 7, so that the data of 7 days before and after the coordination period is selected for analysis, i.e. the required +.>The smaller the description department a uses the seal the lower the degree of abnormality.

In particular, according to the time curves of the a and b departmentsCalculating the DTW distance of the two time slots, selecting a short-time reference time interval f (15 days in this example), analyzing the data of 7 days before and after the short-time reference time interval f, and selecting a time sequence of the short-time reference time interval f (15 days in this example, namely 7 days before and after the short-time reference time interval f) from a TS-t time curve; calculating average use frequency of departments a and b in f days according to the selected time sequence; according to the average frequency of use of departments a, b over the f days, the formula ∈ ->Calculating their correlationThe DTW distance of the time series selection part of the departments a, b is calculated.

Wherein DTW, collectively, is Dynamic Time Warping, i.e., dynamic time warping. It is an algorithm for measuring the degree of similarity between two heterogeneous time series. DTW measures the similarity between time sequences of two different lengths and different multiplying powers by allowing "vertical movement" and "horizontal movement" when the time sequences match. By pairing two time series requires a shift across a nonlinear transformation, including compression, stretching and position in the time domain, by finding the best match between the time series segments and calculating the cumulative distance between them, the smaller the final DTW distance, the higher the similarity between the two time series.

Specifically, the frequency abnormal time curve is mainly used for analyzing whether the frequency of using the electronic seal by a certain department is abnormal or not, drawing a frequency curve of using the electronic seal by the department every day based on historical data, recording the frequency corresponding to the current analyzed data on the curve, comparing the current frequency with the historical frequency curve, calculating the difference of the current frequency and the historical frequency curve, and judging the possibility of abnormality of the current analyzed data according to the frequency curve and the difference.

According to the analysis of the proportion relation between the electronic seal used by the department and the total, the current analysis department may not have a related relation with other departments in the company, but the electronic seal used by the department has a certain relation with the electronic seal data used by the whole company, so the analysis is performed by the analysis department by using the proportion of the electronic seal used by the department to the frequency of all the electronic seals used by the company, so as to obtain a corresponding frequency-time curve, and further obtain the abnormal possibility of the frequency of the electronic seal used by the current department.

FIG. 8 is a schematic diagram of a department ratio analysis unit according to some embodiments of the present description.

Referring to fig. 8, in some embodiments, the department ratio analysis unit 244 further includes: a proportional relationship analysis subunit 244A, a proportional difference analysis subunit 244B, and a proportional determination subunit 244C; a proportional relationship analysis subunit 244A for acquiring the frequency of the electronic seal used by department a Using a first proportion of the frequency of the electronic seal as a whole; the proportional relation analysis subunit 244A is further configured to obtain a second proportion of the frequency of using the electronic seal by the department a to the frequency of using the electronic seal by the company on the same day in the ith historical data; a proportion difference analysis subunit 244B for acquiring the difference between the first proportion and the second proportion of the electronic seal used by the department aThe method comprises the steps of carrying out a first treatment on the surface of the The proportion difference analysis subunit 244B is further configured to obtain a first difference between the frequency of using the electronic seal by the current analysis data of the department a and the adjacent data thereof; the proportion difference analysis subunit 244B is further configured to obtain a second difference between the ith historical data of the department a and the frequency of using the electronic seal by the adjacent data; a proportion difference analysis subunit 244B for also obtaining the difference of the first difference and the second difference of department a +.>The method comprises the steps of carrying out a first treatment on the surface of the The proportion judging subunit 244C is configured to calculate, according to and obtain a department frequency abnormality probability of the department a using the electronic seal, the department frequency abnormality probability according to the following formula:

；

where n represents the number of n history data selected. I.e. the more regular the ratio of the number of seal uses by the current department to the total number of seal uses is, i.e. the The smaller the number of seals used by the current department, the smaller the likelihood of abnormality is. Specifically, the->, wherein ,/>Representing the proportion of the frequency of using the electronic seal by the department a to the frequency of using the electronic seal by the whole company on the same day; />In the ith historical data, the frequency of using the electronic seal by the department a accounts for the frequency of using the electronic seal by the whole company on the same day; />；/>；/>, wherein ,/>Indicating the frequency of use of the seal corresponding to the current analysis data of department a, +.>Representing the frequency of using seal corresponding to the data adjacent to the current analysis data of the department a; />Representing the ith historical data, and the proportion of department a to all stamps used on the same day of the company; />The i-th history data is shown, and the proportion of the time adjacent to the i-th history data, namely the adjacent day department a, to all the stamps used on the same day of the company is shown.

Fig. 9 is a schematic diagram of an integrated judgment unit according to some embodiments of the present description.

Referring to fig. 9, in some embodiments, the comprehensive judgment unit 245 includes: a result acquisition subunit 245A and an abnormality determination subunit 245B; the result obtaining subunit 245A is configured to obtain a degree of abnormality of the department a when the electronic seal is used, a frequency abnormality probability of the department b when the electronic seal is used, correlations between the department a and the m other departments after normalization, and a frequency abnormality probability of the department; an anomaly determination subunit 245B, configured to obtain, according to the result obtained by the result obtaining subunit 245A, frequency anomalies of using the electronic seal by the department a;

Department a uses electronic seal frequency abnormalityCalculated by the following formula:

；

wherein ,a preset threshold value; norm () represents a normalization function; m represents m departments. Specifically, from the above analysis, when the existence exists in the department related to the current department, i.e., the existence correlation is calculated to be larger than the set threshold, exemplary retrieval is performedWhen the current department is abnormal, whether the current department is abnormal or not can be analyzed according to the abnormality of the related department, otherwise, the current department is analyzed according to the proportion of all used seals of the department and the company, so that the related department is avoided, the reference effect is poor, inaccurate abnormal results are obtained, namely, when the current department is analyzed, the difference between the frequency of using the electronic seal on the corresponding day and the corresponding frequency data curve is large, namely, the current department is analyzed, namely, the current department is analyzed, and the current department is analyzed>The larger the abnormality obtained by analyzing the daily seal use time, the larger the abnormality is, i.e., the +.>The bigger the abnormal degree is obtained according to the proportion analysis of the frequency of the seal used by the related departments or the departments and the company;

；

that is, the larger the result of (8), the greater the possibility that the current department will use the stamp frequency data as abnormal use frequency corresponding to the same day.

The abnormal detection value is preferably obtained by parameters, the abnormal degree LOF of the daily seal frequency analyzed by the department is obtained according to the analysis of the corresponding department use frequency-time curve by the abnormal detection algorithm LOF, but different abnormal detection values are obtained by different parameters K, so the application evaluates the obtained LOF value by combining the analysis, and further analyzes and obtains the preferred abnormal detection result.

FIG. 10 is an exemplary block diagram of a variance-evaluating sub-block shown in accordance with some embodiments of the present description.

Referring to fig. 10, in some embodiments, the variance-evaluating sub-module 250 includes: a LOF calculation unit 251 and a variance calculation unit 252; an LOF calculation unit 251, configured to obtain an LOF value according to the K value and the frequency of using the electronic seal by the department; a difference calculating unit 252, configured to obtain differences of LOF values of frequencies of electronic seals used by various departments under different K values; the difference in LOF values for the frequency of departments using electronic seals is calculated by the following formula:

；

wherein exp () represents a power function; n represents the number of using frequency of the electronic seal, m represents the type of department,indicating the LOF value corresponding to the b department data in the i-th historical data acquired by using the LOF algorithm after normalization,>indicating the abnormality of the frequency of using the electronic seal by department b in the ith historical data +.>. I.e. the smaller the difference between the two obtained outliers is, the more likely the corresponding parameter K is the preferred parameter in the current LOF algorithm. Exemplary threshold μ=0.5, i.e., when the required Xi is greater than the threshold, the corresponding anomaly detection algorithm parameter K is considered to be the preferred parameter, rootAnd according to the preferred parameter K, using an LOF algorithm to detect abnormality of the follow-up data of the company.

The data management system can extract and mark the data with the abnormal value LOF more than 1 by using the LOF algorithm to carry out the abnormal analysis on the frequency of using the electronic seal by each department of the company, and report and analyze the reasons and the like of the abnormal using frequency of the electronic seal.

Referring to fig. 11, in some embodiments, the data management module 300 may include: the LOF calculation submodule 310, the judgment submodule 320, the marking submodule 330 and the reporting submodule 340; the LOF calculation submodule 310 is configured to obtain, according to the optimal value of the parameter K, an LOF value of the electronic seal usage data of each department by using an LOF algorithm; a judging sub-module 320, configured to judge whether the LOF value is greater than a preset threshold; a marking sub-module 330, configured to mark the electronic seal usage data when the LOF value is greater than a preset threshold value; and the reporting submodule 340 is used for reporting the marked electronic seal use data.

The exemplary threshold value in this embodiment is set to 1 and the marking submodule is exemplary as follows.

A flag bit is recorded for each electronic seal usage data, defaulting to 0. When the LOF value is greater than the threshold, the flag bit of the data is set to 1, indicating that it has been marked. Screening the data according to the flag bit of 0 or 1: data with flag bit 0: normal data is represented and no reporting is required. Data with flag bit 1: and the abnormal data is represented and needs to be reported, and the reporting sub-module reports the abnormal data with the marking bit of 1.

In summary, the possible benefits of the embodiments of the present disclosure include, but are not limited to: (1) By combining the correlation among departments and the time of using the electronic seal by each department, the accuracy of judging the frequency abnormality of using the electronic seal by the departments is greatly improved; (2) The optimal K value is selected by comparing LOF values under different K values, the corresponding abnormal detection result is more reliable, and the process saves the time of repeatedly using different K values for detection; (3) The condition of using the electronic seal abnormally is found more easily by optimizing the parameter K value of the LOF algorithm, and the safe use of the electronic seal is effectively ensured.

It should be noted that, the advantages that may be generated by different embodiments may be different, and in different embodiments, the advantages that may be generated may be any one or a combination of several of the above, or any other possible advantages that may be obtained.

While the basic concepts have been described above, it will be apparent to those skilled in the art that the foregoing detailed disclosure is by way of example only and is not intended to be limiting. Although not explicitly described herein, various modifications, improvements, and adaptations to the present disclosure may occur to one skilled in the art. Such modifications, improvements, and modifications are intended to be suggested within this specification, and therefore, such modifications, improvements, and modifications are intended to be included within the spirit and scope of the exemplary embodiments of the present invention.

Meanwhile, the specification uses specific words to describe the embodiments of the specification. Reference to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic is associated with at least one embodiment of the present description. Thus, it should be emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various positions in this specification are not necessarily referring to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of the present description may be combined as suitable.

Furthermore, those skilled in the art will appreciate that the various aspects of the specification can be illustrated and described in terms of several patentable categories or circumstances, including any novel and useful procedures, machines, products, or materials, or any novel and useful modifications thereof. Accordingly, aspects of the present description may be performed entirely by hardware, entirely by software (including firmware, resident software, micro-code, etc.) or by a combination of hardware and software. The above hardware or software may be referred to as a "data block," "module," "engine," "unit," "component," or "system. Furthermore, aspects of the specification may take the form of a computer product, comprising computer-readable program code, embodied in one or more computer-readable media.

Furthermore, the order in which the elements and sequences are presented in this specification, the use of numerical letters, or other designations, unless specifically indicated in the claims, is not intended to limit the order in which the processes and methods of this specification are performed. While certain presently useful inventive embodiments have been discussed in the foregoing disclosure, by way of various examples, it is to be understood that such details are merely illustrative and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements included within the spirit and scope of the embodiments of the present disclosure. For example, while the system components described above may be implemented by hardware devices, they may also be implemented solely by software solutions, such as installing the described system on an existing processing device or mobile device.

Likewise, it should be noted that in order to simplify the presentation disclosed in this specification and thereby aid in understanding one or more inventive embodiments, various features are sometimes grouped together in a single embodiment, figure, or description thereof. This method of disclosure, however, is not intended to imply that more features than are presented in the claims are required for the present description. Indeed, less than all of the features of a single embodiment disclosed above.

In some embodiments, numbers describing the components, number of attributes are used, it being understood that such numbers being used in the description of embodiments are modified in some examples by the modifier "about," "approximately" or "substantially. Unless otherwise indicated, "about," "approximately" or "substantially" indicate that the number allows for a 20% variation. Accordingly, in some embodiments, numerical parameters set forth in the specification and claims are approximations that may vary depending upon the desired properties sought to be obtained by the individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and employ a method for preserving the general number of digits. Although the numerical ranges and parameters set forth herein are approximations that may be employed in some embodiments to confirm the breadth of the range, in particular embodiments, the setting of such numerical values is as precise as possible.

Each patent, patent application publication, and other material, such as articles, books, specifications, publications, documents, etc., referred to in this specification is incorporated herein by reference in its entirety. Except for application history documents that are inconsistent or conflicting with the content of this specification, documents that are currently or later attached to this specification in which the broadest scope of the claims to this specification is limited are also. It is noted that, if the description, definition and/or use of a term in an attached material in this specification does not conform to or conflict with what is described in this specification, the description, definition and/or use of the term in this specification controls.

Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments of this specification. Other variations are possible within the scope of this description. Thus, by way of example, and not limitation, alternative configurations of embodiments of the present specification may be considered as consistent with the teachings of the present specification. Accordingly, the embodiments of the present specification are not limited to only the embodiments explicitly described and depicted in the present specification.

Claims

1. An electronic seal data management system based on data analysis, comprising:

the data acquisition module is used for acquiring electronic seal use data, wherein the electronic seal use data comprises frequency data and time data;

the anomaly detection module is used for carrying out anomaly detection on the electronic seal use data and obtaining an optimal value of an LOF algorithm parameter K;

the data management module is used for managing the electronic seal use data based on an LOF algorithm and the optimal value of the parameter K;

wherein, the abnormality detection module includes:

an initial parameter submodule for settingInitial value of parameter K in LOF algorithm；

The transformation parameter submodule is used for transforming the parameter K value and acquiring different parameters K;

the correlation analysis submodule is used for analyzing the correlation of frequency of using the electronic seal among departments;

The abnormality analysis submodule is used for detecting abnormality based on the parameter K and the frequency of using the electronic seal;

a difference evaluation sub-module for obtaining the difference of LOF values of the electronic seal frequency used by each department of different parameter K values；

A preferred parameter sub-module for obtaining the difference of LOF valuesAnd the value of the parameter K is an optimal value when the value is larger than a preset threshold value.

2. The electronic seal data management system based on data analysis of claim 1, wherein:

the correlation analysis submodule includes: a history data analysis unit and a correlation acquisition unit;

the historical data analysis unit is used for acquiring first frequency information of the electronic seal used by the department a on the same day and second frequency information of the electronic seal used by the department b on the same day in the ith historical data;

the historical data analysis unit is further used for acquiring first average frequency information of the electronic seal used by the department a and second average frequency information of the electronic seal used by the department b in the n pieces of historical data;

the correlation obtaining unit is configured to obtain correlation of using electronic seals by the department a and the department b according to the first frequency information, the second frequency information, the first average frequency information and the second average frequency information ；

Correlation of the department a and the department b using the electronic sealCalculated by the following formula:

3. The electronic seal data management system based on data analysis of claim 1, wherein:

the anomaly analysis sub-module further includes: the system comprises a frequency analysis unit, a time analysis unit, a related department reference unit, a department duty analysis unit and a comprehensive judgment unit;

the frequency analysis unit is used for analyzing whether the frequency of using the electronic seal by the department is abnormal or not based on the frequency data of using the electronic seal by the department;

the time analysis unit is used for judging whether the time of using the electronic seal by each department is abnormal or not by analyzing the time of using the electronic seal by each department every day;

the related department reference unit is used for referring to the related department use time to judge whether the department use time is abnormal or not;

The department duty analysis unit is used for analyzing whether the frequency of the department use electronic seal is abnormal according to the proportion of the department use electronic seal to the total electronic seal usage;

and the comprehensive judging unit is used for acquiring the abnormal results of the electronic seal used by each department according to the results of the frequency analysis unit, the time analysis unit, the related department reference unit and the department duty ratio analysis unit.

4. A data analysis-based electronic seal data management system according to claim 3, wherein:

the frequency analysis unit further includes: a difference calculating subunit, a change rate calculating subunit and a frequency judging subunit;

the difference calculating subunit is used for obtaining the difference between the frequency data of the department a using the electronic seal and the historical frequency data；

The change rate calculating subunit is used for obtaining the change rate difference between the frequency data of the department a using the electronic seal and the historical frequency data；

A frequency judging subunit for acquiring the frequency abnormality possibility of the electronic seal used by the department a according to the difference and the change rate differenceThe frequency ofPossibility of abnormality->Calculated by the following formula:

wherein ,representing the difference between the current frequency data and the stamp frequency used at the ith historical moment of the history; / >The first change rate of the current frequency data and the adjacent data using the seal frequency is compared with the difference of the second change rate of the i-th historical moment and the adjacent moment using the seal frequency.

5. A data analysis-based electronic seal data management system according to claim 3, wherein:

the time analysis unit further includes: a time analysis subunit, a related department reference subunit and a time judgment subunit;

the time analysis subunit is used for acquiring the time and the frequency of using the electronic seal per hour by each department and acquiring the frequency difference per hour of using the electronic seal；

The related department reference subunit is used for acquiring the relevance between the department a and the related department'；

The time judging subunit is configured to, according to the hourly frequency differenceAcquisition of daily time abnormality->And according to said daily time abnormality +.>And the correlation->' possibility of time abnormality of use of electronic stamp by acquisition department a->；

The daily time abnormal conditionCalculated by the following formula:

6. A data analysis-based electronic seal data management system according to claim 3, wherein:

the related department reference unit further includes: a time curve subunit, an abnormality degree analysis subunit and a department judgment subunit;

the time curve subunit is used for acquiring the frequency abnormal time curve of each department using the electronic seal every day；

The abnormality degree analysis subunit is configured to analyze the abnormal time curve according to the frequencyAcquiring the degree of abnormality DTW of a related department;

the department judging subunit is configured to obtain a department abnormality degree of the department a according to the related department abnormality degree DTW；

Degree of abnormality of the departmentCalculated by the following formula:

wherein m represents the number of departments; f represents a reference time interval;the DTW distance of the department a and the department b when the reference time interval is f; / >Indicating the relevance of department a and department b to use the electronic seal.

7. A data analysis-based electronic seal data management system according to claim 3, wherein:

the department ratio analysis unit further includes: a proportional relation analysis subunit, a proportional difference analysis subunit and a proportional judgment subunit;

the proportion relation analysis subunit is used for obtaining a first proportion of the frequency of using the electronic seal by the department a to the frequency of using the electronic seal by the whole company on the same day;

the proportion relation analysis subunit is further used for acquiring a second proportion of the frequency of using the electronic seal by the department a to the frequency of using the electronic seal by the whole company on the same day in the ith historical data;

the proportion difference analysis subunit is used for acquiring the difference between the first proportion and the second proportion of the electronic seal used by the department a；

The proportion difference analysis subunit is further used for acquiring a first difference of the frequency of using the electronic seal between the current analysis data of the department a and the adjacent data;

the proportion difference analysis subunit is further used for acquiring a second difference of the frequency of using the electronic seal between the ith historical data of the department a and the adjacent data;

the proportion difference analysis subunit is further used for acquiring the difference of the first difference and the second difference of the department a ；

The proportion judging subunit is used for judging the proportion of the current according to the currentAnd said->Acquiring possibility of abnormality of department frequency of department a using electronic seal +.>Possibility of abnormality of the department frequency +.>Calculated by the following formula:

where n represents the number of n history data selected.

8. A data analysis-based electronic seal data management system according to claim 3, wherein:

the comprehensive judging unit further comprises a result acquisition subunit and an abnormality judging subunit;

the result acquisition subunit is used for acquiring the degree of abnormality of the department a when the electronic seal is usedFrequent abnormality possibility of using electronic seal by department a->The possibility of frequency abnormality of using electronic seal by department b>Correlation with the b-th department obtained after normalization of the correlation between department a and other m departments ∈>Possibility of abnormality in department frequency->；

The abnormality judgment subunit is used for obtaining the frequency abnormality of the electronic seal used by the department a according to the result obtained by the result obtaining subunit；

Frequency abnormality of the department a using electronic sealCalculated by the following formula:

9. The electronic seal data management system based on data analysis of claim 1, wherein:

the difference evaluation sub-module further comprises an LOF calculation unit and a difference calculation unit;

the LOF calculation unit is used for obtaining an LOF value according to the K value and the frequency of using the electronic seal by departments;

the difference calculation unit is used for obtaining the difference of LOF values of the electronic seal frequency used by each department under different K values；

The departments use the difference of LOF values of the frequency of the electronic sealCalculated by the following formula:

10. The electronic seal data management system based on data analysis of claim 1, wherein:

the data management module further comprises: the system comprises an LOF calculation sub-module, a judging sub-module, a marking sub-module and a reporting sub-module;

the LOF calculation submodule is used for acquiring LOF values of the electronic seal use data of each department by using an LOF algorithm according to the optimal value of the parameter K;

The judging submodule is used for judging whether the LOF value is larger than a preset threshold value or not;

the marking submodule is used for marking the electronic seal use data when the LOF value is larger than a preset threshold value;

and the reporting sub-module is used for reporting the marked electronic seal use data.