CN114022051A - Index fluctuation analysis method, storage medium and electronic equipment - Google Patents

Index fluctuation analysis method, storage medium and electronic equipment Download PDF

Info

Publication number
CN114022051A
CN114022051A CN202111657169.0A CN202111657169A CN114022051A CN 114022051 A CN114022051 A CN 114022051A CN 202111657169 A CN202111657169 A CN 202111657169A CN 114022051 A CN114022051 A CN 114022051A
Authority
CN
China
Prior art keywords
data
fluctuation
index
dimension
sample data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111657169.0A
Other languages
Chinese (zh)
Inventor
汪芳羽
傅文林
邓自立
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alibaba Cloud Computing Ltd
Original Assignee
Alibaba Cloud Computing Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alibaba Cloud Computing Ltd filed Critical Alibaba Cloud Computing Ltd
Priority to CN202111657169.0A priority Critical patent/CN114022051A/en
Publication of CN114022051A publication Critical patent/CN114022051A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/901—Indexing; Data structures therefor; Storage structures
    • G06F16/9027—Trees
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/903—Querying
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • Educational Administration (AREA)
  • Marketing (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • General Business, Economics & Management (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Computational Linguistics (AREA)
  • Software Systems (AREA)
  • Game Theory and Decision Science (AREA)
  • Technology Law (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The application discloses an index fluctuation analysis method, a storage medium and an electronic device, wherein the index fluctuation analysis method comprises the following steps: determining log data generated by a service application platform in an enterprise digital center station as sample data to be analyzed; the at least one piece of sample data to be analyzed comprises a dimension combination of at least two pieces of dimension data and index data corresponding to the dimension combination; analyzing the index data of the dimension combination in the sample data to be analyzed according to a set first correlation requirement to obtain abnormal sample data; according to a set second association requirement, mining a frequent item set of the dimension combination in the abnormal sample data; and determining fluctuation data corresponding to the dimension combination in the frequent item set according to a set third association requirement. Therefore, attribution results are constrained from different angles, accuracy of fluctuation data is guaranteed, and accuracy of attribution results is improved.

Description

Index fluctuation analysis method, storage medium and electronic equipment
Technical Field
The application relates to the technical field of computer application, in particular to an index fluctuation analysis method and device and an index fluctuation analysis interaction method and device of a resource service application platform. The application also relates to a computer storage medium and an electronic device.
Background
With the continuous development of science and technology, computer foundations such as informatization, intellectualization and the like are widely applied to various industry fields and daily life. Driven by industry and technology, digital transformation has become a consensus of enterprises. The Digital transformation (Digital transformation) is established on the basis of Digital transformation (Digitization) and Digital upgrading (Digitization), and can touch the core business of an enterprise and establish transformation with a business operation mode as a target.
With the continuous promotion and deepening of enterprise sharing capacity in a digital service mode, in order to deposit the core capacity of an enterprise to a platform in a digital form along with the continuous development of services, an effective system form and an organization method for carrying out digital transformation on the enterprise are formed by a digital middle station which takes the services as a center and constructs a data closed-loop operation system by the business middle station and the data middle station, so that the enterprise can more efficiently carry out service exploration and innovation, and the purpose of constructing the core differentiated competitiveness of the enterprise in a digital asset form is realized.
The enterprise digital middle station is used as a service platform for enterprise digital transformation, the operation monitoring of core operation business becomes an important task of the enterprise digital middle station, and the task can acquire relevant data of operation conditions aiming at the operation monitoring and make reference for subsequent adjustment and development.
Disclosure of Invention
The application provides an index fluctuation analysis method, which is used for solving the problem that a conventional index fluctuation analysis method in the prior art is not suitable for analyzing multi-dimensional data in an enterprise digital middlebox; on the other hand, even if multi-dimensional data analysis can be performed, the calculation cost is high, and the analysis result is inaccurate.
The application provides an index fluctuation analysis method, which comprises the following steps:
determining log data generated by a service application platform in an enterprise digital center station as sample data to be analyzed; the at least one piece of sample data to be analyzed comprises a dimension combination of at least two pieces of dimension data and index data corresponding to the dimension combination;
analyzing the index data of the dimension combination in the sample data to be analyzed according to a set first correlation requirement to obtain abnormal sample data;
according to a set second association requirement, mining a frequent item set of the dimension combination in the abnormal sample data;
and determining fluctuation data corresponding to the dimension combination in the frequent item set according to a set third association requirement.
In some embodiments, the analyzing, according to the set first association requirement, the index data of the dimension combination in the sample data to be analyzed to obtain abnormal sample data includes:
determining a set index variation threshold as the first association requirement;
determining the index variation of the sample data to be analyzed according to the index data of the standard date of the sample data to be analyzed and the set index data of the reference date of the sample data to be analyzed;
and determining the abnormal sample data according to the index variation and the index variation threshold.
In some embodiments, the indicator variation threshold is set according to a variation distribution state of the indicator data in the sample data to be analyzed.
In some embodiments, the mining a frequent item set of the dimension combinations in the abnormal sample data according to the set second association requirement includes:
determining the set support degree as a requirement condition of the second association requirement;
acquiring target abnormal sample data according to the support degree;
and determining the frequent item set according to a frequent pattern tree constructed by the target abnormal sample data.
In some embodiments, the obtaining target exception sample data according to the support degree includes:
counting the abnormal sample data according to the occurrence times of the same dimension value in the dimension data to obtain candidate abnormal sample data;
and determining the target abnormal sample data according to the dimension value which is greater than or equal to the support threshold value in the selected candidate abnormal sample data.
In some embodiments, the determining the frequent item set according to the frequent pattern tree constructed by the target abnormal sample data includes:
sequencing the target abnormal sample data according to the occurrence times of the dimension value;
constructing the frequent pattern tree according to the sorted target abnormal sample data;
and searching the frequent item set which meets the requirements of conditional frequent item sets and/or frequent items according to the frequent pattern tree.
In some embodiments, the determining, according to the set third association requirement, fluctuation data corresponding to a dimension combination in the frequent item set includes:
determining a set confidence threshold as the third association requirement;
and determining fluctuation data corresponding to the dimension combination in the frequent item set according to the comparison between the obtained fluctuation score of the dimension combination in the frequent item set and the confidence coefficient threshold value.
In some embodiments, the determining, according to the obtained fluctuation score of the dimension combination in the frequent item set and the comparison of the confidence threshold, fluctuation data corresponding to the dimension combination in the frequent item set includes:
determining the fluctuation fraction according to the contribution degree and the abnormality degree of the dimension combination;
and determining the dimensional combination with the fluctuation score larger than or equal to the confidence coefficient threshold value as the fluctuation data corresponding to the dimensional combination in the frequent item set.
In some embodiments, the determining the fluctuation score according to the contribution degree and the abnormality degree of the dimension combination includes:
determining the score of the contribution degree according to the sum of the comparison date index values of the dimension combinations in the frequent item set, the sum of the comparison date index values of all the dimension combinations, the sum of the reference date index values of the dimension combinations in the frequent item set and the sum of the reference date index values of all the dimension combinations;
determining the value of the abnormal degree according to the prior probability of the dimension combination in the frequent item set and the posterior probability of the dimension combination in the frequent item set;
and determining the fluctuation score according to the weighted sum of the contribution score and the abnormal degree score.
In some embodiments, the determining the score of the degree of abnormality according to the prior probability of the dimension combination in the set of frequent items and the posterior probability of the dimension combination in the set of frequent items includes:
determining the prior probability according to the sum of the reference date index values of the dimension combinations in the frequent item set and the sum of the reference date index values of all the dimension combinations;
determining the posterior probability according to the sum of the comparison date index values of the dimension combinations in the frequent item set and the sum of the comparison date index values of all the dimension combinations;
and determining the abnormal degree score according to the prior probability and the posterior probability.
In some embodiments, further comprising:
classifying the dimension combination corresponding to the fluctuation data into a cause set;
sorting the dimension combinations in the cause set according to fluctuation scores;
and outputting the sorted attribution sets.
In some embodiments, further comprising:
and carrying out data combination and/or data cleaning on the log data according to the dimension data.
The application also provides a transaction index fluctuation analysis interaction method of the resource service application platform, which comprises the following steps:
transmitting transaction sample data to be analyzed based on triggering of a transaction index fluctuation analysis request of a resource service application platform in an enterprise digital center;
responding to the transaction index fluctuation analysis request, and acquiring abnormal transaction sample data generated by the resource service application platform for the transaction sample data to be analyzed according to a first association requirement;
performing frequent item set mining on the abnormal transaction sample data according to a second association requirement;
determining fluctuation data corresponding to the dimensionality combination in the frequent item set according to a third correlation requirement;
and displaying the fluctuation data in a visual mode.
In some embodiments, the transmitting the transaction sample data to be analyzed based on the trigger of the transaction index fluctuation analysis request of the resource service application platform in the enterprise digital center comprises:
selecting the transaction sample data to be analyzed from a database based on triggering of a transaction index fluctuation analysis request of a resource service application platform in an enterprise digital center; the data table established according to the transaction log data generated by the resource service application platform is stored in the database, and at least one piece of transaction sample data to be analyzed in the data table comprises a dimension combination of at least two dimension data and index data corresponding to the dimension combination;
and transmitting the selected transaction sample data to be analyzed to an index fluctuation analysis side.
In some embodiments, the visually presenting the fluctuation data includes:
displaying the fluctuation data according to the selected display quantity of the fluctuation data; the surge data includes: generating at least one or more of a dimensional combination of a fluctuation, index data information corresponding to the dimensional combination, fluctuation analysis information, adjustment reference information provided according to the fluctuation analysis information, and history adjustment example information; wherein the fluctuation analysis information is determined from the fluctuation data.
The application also provides a display method of the fluctuation data, which comprises the following steps:
responding to display triggering operation of the fluctuation data, and displaying the acquired fluctuation data on a fluctuation data output interface of the electronic equipment; wherein the fluctuation data includes: generating at least one or more of a dimensional combination of a fluctuation, index data information corresponding to the dimensional combination, fluctuation analysis information, adjustment reference information provided according to the fluctuation analysis information, and history adjustment example information; wherein the fluctuation analysis information is determined from the fluctuation data.
The application also provides a computer storage medium for storing the data generated by the network platform and a program for processing the data generated by the network platform;
the program, when read and executed by a processor, executes the index fluctuation analysis method as described above; or, executing the transaction index fluctuation analysis interaction method of the resource service application platform; alternatively, the method of presenting the fluctuation data as described above is performed.
The present application further provides an electronic device, comprising:
a processor;
a memory for storing a program for processing data generated by a network platform, the program, when read and executed by the processor, executing the index fluctuation analysis method as described above; or, executing the transaction index fluctuation analysis interaction method of the resource service application platform; alternatively, the method of presenting the fluctuation data as described above is performed.
Compared with the prior art, the method has the following advantages:
according to the index fluctuation analysis method, the data information under different association requirements is acquired for the acquired sample data to be analyzed through at least three set association requirements, so that the attribution result is constrained from different angles, the accuracy of the fluctuation data is ensured, and the accuracy of the attribution result is improved. In addition, the relation between the dimension combinations is determined by constructing the fp trees, and the calculation speed of index fluctuation analysis in a multi-dimension data scene is improved.
Drawings
Fig. 1 is a flowchart of an embodiment of an index fluctuation analysis method provided in the present application.
Fig. 2 is a schematic structural diagram of an fp tree in an embodiment of an index fluctuation analysis method provided by the present application.
Fig. 3 is a schematic structural diagram of a conditional fp tree in an embodiment of an index fluctuation analysis method provided by the present application.
Fig. 4 is a schematic structural diagram of an embodiment of an index fluctuation analysis apparatus provided in the present application.
Fig. 5 is a flowchart of an embodiment of an index fluctuation analysis interaction method of a resource service application platform provided in the present application.
Fig. 6 is a schematic structural diagram of an embodiment of a transaction index fluctuation analysis interaction apparatus of a resource service application platform provided in the present application.
Fig. 7 is a schematic structural diagram of an embodiment of an electronic device provided in the present application.
Detailed Description
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. This application is capable of implementation in many different ways than those herein set forth and of similar import by those skilled in the art without departing from the spirit of this application and is therefore not limited to the specific implementations disclosed below.
The terminology used in the description herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The description used in this application and in the appended claims is for example: the terms "a," "an," "first," and "second," etc., are not intended to be limiting in number or order, but rather are used to distinguish one type of information from another.
Based on the above background art, it is known that the enterprise digital center station needs to perform real-time anomaly detection on various Key Performance Indicators (KPIs) of the core operation service, and then needs to perform anomaly positioning and Root Cause Analysis (RCA) on the detected anomaly indicators so as to further repair and stop loss operations. The invention concept of the index fluctuation analysis method provided by the application is derived from the problems that the conventional index fluctuation analysis method in the prior art cannot be applied to multi-dimensional data of enterprise digital middleboxes, and the accuracy of an analysis result is poor. Specifically, the conventional index fluctuation analysis method may include an attribute method or a method based on a rule tree of the contribution degree.
The adtributor method firstly needs to calculate the abnormal degree of all dimension values under each dimension, then arranges the dimension values of each dimension in a descending order according to the abnormal degree, then calculates the contribution degree of each dimension value, eliminates the dimension values with small contribution degree, adds the dimension combination meeting the threshold condition of the contribution degree into the root cause set, calculates the total abnormal degree value of the dimension combination, and finally arranges and outputs the dimension combination in the root cause set in a descending order according to the total abnormal degree value. The problems of the method are that: if the index fluctuation is attributed to the element combination of one dimension, the possible internal relation among multiple dimensions can be ignored, and the method is not suitable for index fluctuation analysis caused by the dimension combination of multiple dimensions.
The method for the rule tree based on the contribution degree converts the relation between the dimensions into a multi-branch tree structure or a decision tree structure, the dimension combination is all nodes passed by each tree node along the way, the contribution degree of each node is calculated, and the dimension combination is arranged and output according to the descending order of the contribution degree. The problems of the method are that: when the dimension is more or the dimension value is more, the calculation amount is inevitably increased; the dimension combinations with high contribution degree only represent more data and cannot provide value for analysis.
Therefore, the two methods are not suitable for the scenario of index fluctuation analysis of multidimensional data involved in the enterprise digital middleboxes, or the two methods cannot provide a more accurate analysis result for the index fluctuation analysis of multidimensional data involved in the enterprise digital middleboxes, and increase the calculation cost and the resource cost. Therefore, the application provides an index fluctuation analysis method, which comprises the following specific contents.
As shown in fig. 1, fig. 1 is a flowchart of an embodiment of an index fluctuation analysis method provided in the present application, where the embodiment of the method includes:
step S101: determining log data generated by a service application platform in an enterprise digital center station as sample data to be analyzed; the at least one piece of sample data to be analyzed comprises a dimension combination of at least two pieces of dimension data and index data corresponding to the dimension combination.
The enterprise digital center in step S101 may be understood as a center product that performs digital transformation for an enterprise, is service-oriented capability combination and multiplexing, provides an integrated solution, and aims to improve research and development efficiency and reduce innovation cost. The enterprise digital center station can comprise users, combinations, platforms, data, standards and specifications, and is an integral system of users and systems. Namely: the enterprise digital middle platform is an enterprise level capability multiplexing platform which abstracts the common requirements of enterprises, creates platform and component system capabilities and shares the system capabilities to each service unit in the forms of interfaces, components and the like.
The log data is log data that records procedural events generated by the computer architecture system. Under the requirement of enterprise digital transformation, various service application platforms or business operations related in the enterprise digital middleboxes can continuously generate a large amount of log data. Such as financial industry, shopping platforms, etc., log data may be generated daily at the TB level and even at the PB level.
The specific implementation process of step S101 may be to obtain log data generated by a service application platform in the enterprise digital center, where the log data may be stored in a form of a data table, or to construct a data table through the log data. The data table may be at least two data tables within a set time. Each row of data in the data table can be used as a piece of sample data to be analyzed, and the sample data to be analyzed can also be selected and determined from the data table. At least one sample data to be analyzed in the data table may include a dimension combination of at least two dimension data and index data corresponding to the dimension combination, which is exemplified as follows:
Figure 58174DEST_PATH_IMAGE001
Figure 208532DEST_PATH_IMAGE002
the province, the age, the occupation and the weather are dimension combinations of dimensions, and the transaction amount is index data. The data in each dimension is a dimension value, for example: the upper sea in the first sample data is the dimension value of provinces, the young is the dimension value of ages, college students are the dimension value of profession, and snow is the dimension value of weather.
For the above example, data tables in two different time periods may be constructed according to log data, and the dimension data in the data tables may be preprocessed. Therefore, it may further include:
and carrying out data combination and/or data cleaning on the log data according to the dimension data. The specific implementation process may be to perform data merging and/or data cleaning on the constructed at least two data tables. The data merging may be to merge two data tables together, that is, merge the same index data in the same dimension.
Figure DEST_PATH_IMAGE003
The data cleaning can be at least one cleaning mode of filling data in the data table, removing dimensions with only one dimension value and removing dimensions with dimension values exceeding the preset required number. The preset requirement number can be determined according to an actual application scene or according to the total data volume, the specific preset requirement number is not limited, and the requirement for removing dimensions with more dimension values can be met.
It should be noted that, because the two data tables are only exemplified by a small amount of data, the two data tables are only involved in splicing, and in an actual application scenario, a large amount of data needs to be processed in the manner of merging and/or cleaning.
In step S101, the merged data table may be used as sample data to be analyzed, or the sample data to be analyzed may be selected from the merged data table.
Step S102: and analyzing the index data of the dimension combination in the sample data to be analyzed according to a set first association requirement to obtain abnormal sample data.
The first association requirement in step S102 may be understood as a division requirement, that is, to divide sample data to be analyzed into abnormal data and normal data, and a specific division may be determined according to a set index variation threshold, that is, to use the index variation threshold as the first association requirement. Therefore, the specific implementation process of step S102 may include:
step S102-1: determining a set index variation threshold as the first association requirement; in this embodiment, the setting of the index variation threshold may be set according to a variation distribution state of index data in sample data; for example: selecting a median of index variation distribution of all sample data, namely the median of index variation; or, the mean value of the index variation distribution of all sample data, that is, the mean value of the index variation is selected.
Step S102-2: determining the index variation of the sample data to be analyzed according to the index data of the standard date of the sample data to be analyzed and the set index data of the reference date of the sample data to be analyzed; in this embodiment, the index variation may be calculated by using the following formula:
r =2×（m2-m1）/（m2+m1）；
where r represents the index change amount of the sample data to be analyzed, m1 represents the index value of the reference date of the sample data to be analyzed (i.e., the index data described above), and m2 represents the index value of the comparison date of the sample data to be analyzed.
Step S102-3: and determining the abnormal sample data according to the index variation and the index variation threshold. In this embodiment, when the index variation (r) is greater than or equal to the set index variation threshold, it is determined as abnormal sample data. For example: suppose the outlier sample data is:
Figure 534953DEST_PATH_IMAGE004
step S103: and mining a frequent item set of the dimension combination in the abnormal sample data according to a set second association requirement.
The second association requirement can be understood as a frequent item set filtering requirement, and filtering can be realized through the support degree. The specific implementation process may include:
step S103-1: determining the set support degree as a requirement condition of the second association requirement; the support degree may be set according to an empirical value. Wherein the support degree can be 20 to 40 percent from bottom to top according to the frequency distribution of occurrence of the statistical dimension value. Of course, the setting of the support degree may be determined according to an actual application scenario, and is only an example here.
Step S103-2: and acquiring target abnormal sample data according to the support degree.
Step S103-3: and determining the frequent item set according to a frequent pattern tree constructed by the target abnormal sample data.
In this embodiment, the specific implementation process of step S103-2 may include:
step S103-21: counting the abnormal sample data according to the occurrence times of the same dimension value in the dimension data to obtain candidate abnormal sample data; in this embodiment, the statistics may be understood as performing statistics on the occurrence frequency of the dimension value, for example: 3 times in young, 1 time in middle-aged, 1 time in old, 2 times in college students, 1 time in teaching staff, 1 time in free occupation, 1 time in white collar, 1 time in snow, 2 times in sunny day, 1 time in cloudy day, 1 time in rain, 1 time in Shanghai, 1 time in Chongqing, 2 times in Jiangsu, 1 time in Zhejiang, and the like.
Step S103-22: and determining the target abnormal sample data according to the dimension value which is greater than or equal to the support threshold value in the selected candidate abnormal sample data.
In this embodiment, the step S103-22 may be understood as determining, as the target abnormal sample data, the dimension value whose number of occurrences of the dimension value is greater than or equal to 2 when the support degree is 2, in other words, filtering and deleting the dimension value whose number of occurrences of the dimension value is less than the support degree 2. Continuing with the example in step S103-21 above, the filtering may be to determine young, college students, sunny, and juncea as target abnormal sample data, and to filter middle-aged, elderly, teaching workers, free occupation, white collar, snow, cloudy, rain, shanghai, chongqing, and zhejiang.
The specific implementation process of step S103-3 may include:
step S103-31: sequencing the target abnormal sample data according to the occurrence times of the dimension value; in this embodiment, the sorting may be descending sorting, that is, each sample data in the target abnormal sample data is descending sorted according to the frequency of occurrence of the dimension value.
Step S103-32: and constructing the frequent pattern tree according to the sorted target abnormal sample data. The frequent pattern tree may also be referred to as FP-tree (FP Growth algorithm), and following the above example, the target exception sample data may be:
1. young and college student
2. Young, sunny day
3. University student of Jiangsu
4. Jiangsu in sunny days
5. Young people
Constructing fp trees, hanging each dimension value of sample data under one tree, and counting the occurrence frequency, for example: the first sample data is Root-youth 1-college 1; the second sample data is Root-young 2-college 1, Root-young 2-sunny 1, and so on. The above is only a summary description of the general procedure for constructing the fp-tree in the present embodiment, and actually the construction of the fp-tree belongs to the prior art, and the data structure for constructing the fp-tree includes the original data, the item header table and the fp-tree. First, an entry header table needs to be established according to the original data, and this process may correspond to the above-mentioned determination and sorting of the target exception sample data. An fp-tree is then constructed based on the entry header table, i.e.: reading in ordered target abnormal sample data when the fp tree is established, and inserting the nodes into the fp tree according to the ordered sequence, wherein the node in the front of the order is an ancestor node, and the node in the back is a descendant node. When there is a common ancestor node, then 1 is added to the common ancestor node count. When a new node appears, the node corresponding to the item head table can be linked with the new node through the linked list. And completing the construction of the fp tree until all target abnormal sample data are inserted into the fp tree. Following the above example, the fp-tree constructed in this embodiment is shown in fig. 2.
Step S103-33: and searching the frequent item set which meets the requirements of conditional frequent item sets and/or frequent items according to the frequent pattern tree. After the fp tree is constructed for the target abnormal sample data, the corresponding conditional fp tree can be obtained for each dimension value. The conditional mode base is determined with the set of paths ending with the looked-up dimension value. Each path may be understood as a prefix path. The above example can be used for searching the frequent item set in the present embodiment, for example: the fp-tree constructed as shown in fig. 2 is shown in fig. 3 for a conditional fp-tree with dimension values of college students. When the support degree is 2, the frequent item sets are young and college students, jiangsu and college students, young and sunny days, jiangsu and college students, and so on, the frequent item sets of all the dimensional values are calculated, and the 4 frequent item sets are taken as an example in this embodiment for explanation.
Step S104: and determining fluctuation data corresponding to the dimension combination in the frequent item set according to a set third association requirement.
The third association requirement in step S104 may be understood as a screening requirement, and the screening may be implemented by confidence. The specific implementation process may include:
step S104-1: determining a set confidence threshold as the third association requirement;
step S104-2: and determining fluctuation data corresponding to the dimension combination in the frequent item set according to the comparison between the obtained fluctuation score of the dimension combination in the frequent item set and the confidence coefficient threshold value.
In this embodiment, the confidence threshold may be set based on an empirical value, and the confidence threshold may be set empirically through analysis of historical fluctuation data.
The specific implementation process of the step S104-2 may include:
step S104-21: and determining the fluctuation fraction according to the contribution degree and the abnormality degree of the dimension combination.
Step S104-22: and determining the dimensional combination with the fluctuation score larger than or equal to the confidence coefficient threshold value as the fluctuation data corresponding to the dimensional combination in the frequent item set.
In this embodiment, the contribution degree in step S104-21 may determine the score of the contribution degree according to the sum of the comparison date index values of the dimension combinations in the frequent item set, the sum of the comparison date index values of all the dimension combinations, the sum of the reference date index values of the dimension combinations in the frequent item set, and the sum of the reference date index values of all the dimension combinations. Specifically, the score of the contribution degree may be calculated by the following formula:
Figure 327328DEST_PATH_IMAGE005
；
wherein, A isi(m) represents the sum of the comparison date index values of the frequent item set dimension combination i, A (m) represents the sum of the comparison date index values of all dimension combinations, Fi(m) represents the sum of the reference date indicator values for the frequent itemset dimension combination i, and F (m) represents the sum of the reference date indicator values for all dimension combinations.
In this embodiment, the degree of abnormality in step S104-21 may determine the score of the degree of abnormality according to the prior probability of the dimension combination in the frequent item set and the posterior probability of the dimension combination in the frequent item set. Specifically, the computation of the score of the degree of abnormality may be performed by first computing a prior probability and a posterior probability, and then computing the score of the degree of abnormality.
In this embodiment, the prior probability may be calculated by using the following formula:
Figure 465048DEST_PATH_IMAGE006
；
wherein p isi(m) represents the prior probability, Fi(m) represents the sum of the reference date indicator values for the frequent itemset dimension combination i, and F (m) represents the sum of the reference date indicator values for all dimension combinations.
In this embodiment, the posterior probability may be calculated by the following formula:
Figure 786308DEST_PATH_IMAGE008
；
wherein q isi(m) represents the posterior probability, Ai(m) represents the sum of the control date index values for the frequent itemset dimension combination i, and A (m) represents the sum of the control date index values for all dimension combinations.
The degree of abnormality score may be given by the following formula:
order to
Figure 9479DEST_PATH_IMAGE009
，
Figure 339967DEST_PATH_IMAGE010
；
Figure 332193DEST_PATH_IMAGE011
；
Wherein, the Si(m) represents the score of the degree of abnormality.
The fluctuation score in step S104-21 may be determined by weighted summation of the score of the contribution degree and the score of the abnormality degree.
Based on the above, the specific implementation process of the step S104-21 may include:
step S104-211: determining the score of the contribution degree according to the sum of the comparison date index values of the dimension combinations in the frequent item set, the sum of the comparison date index values of all the dimension combinations, the sum of the reference date index values of the dimension combinations in the frequent item set and the sum of the reference date index values of all the dimension combinations;
step S104-212: determining the value of the abnormal degree according to the prior probability of the dimension combination in the frequent item set and the posterior probability of the dimension combination in the frequent item set;
step S104-213: and determining the fluctuation score according to the weighted sum of the contribution score and the abnormal degree score.
The specific implementation process of the steps S104-212 may include:
step S104-2121: determining the prior probability according to the sum of the reference date index values of the dimension combinations in the frequent item set and the sum of the reference date index values of all the dimension combinations;
step S104-2122: determining the posterior probability according to the sum of the comparison date index values of the dimension combinations in the frequent item set and the sum of the comparison date index values of all the dimension combinations;
step S104-2123: and determining the abnormal degree score according to the prior probability and the posterior probability.
Based on the above, the present embodiment may further include:
step S105: classifying the dimension combination corresponding to the fluctuation data into a cause set;
step S106: sorting the dimension combinations in the cause set according to fluctuation scores;
step S107: and outputting the sorted attribution sets.
The sorting of step S106 may be in descending order of the volatility score. In step S107, attribute sets collected in descending order are output, and the dimensionality related to the index fluctuation can be obtained by displaying the attribute sets on the terminal device.
The above is a description of an embodiment of the index fluctuation analysis method provided by the present application, and it can be known through this embodiment that the embodiment of the index fluctuation analysis method provided by the present application obtains data information under different association requirements through at least three association requirements set for sample data to be analyzed, so as to constrain attribution results from different angles, ensure accuracy of fluctuation data, and further improve accuracy of attribution results. In addition, the relation between the dimension combinations is determined by constructing the fp trees, and the calculation speed of index fluctuation analysis in a multi-dimension data scene is improved.
The above is a specific description of an embodiment of an index fluctuation analysis method provided in the present application, and corresponds to the foregoing provided embodiment of an index fluctuation analysis method, and the present application also discloses an embodiment of an index fluctuation analysis apparatus, please refer to fig. 4. The device embodiments described below are merely illustrative.
As shown in fig. 4, fig. 4 is a schematic structural diagram of an embodiment of an index fluctuation analysis apparatus provided in the present application, where the apparatus embodiment includes:
a first determining unit 401, configured to determine log data generated by a service application platform in an enterprise digital middlebox as sample data to be analyzed; the at least one piece of sample data to be analyzed comprises a dimension combination of at least two pieces of dimension data and index data corresponding to the dimension combination; for a specific implementation process of the first determining unit 401, reference may be made to the content of step S101, and details are not repeated here.
An obtaining unit 402, configured to analyze the index data of the dimension combination in the sample data to be analyzed according to a set first association requirement, and obtain abnormal sample data;
a mining unit 403, configured to mine a frequent item set of the dimension combination in the abnormal sample data according to a set second association requirement;
a second determining unit 404, configured to determine, according to a set third association requirement, fluctuation data corresponding to the dimension combination in the frequent item set.
The obtaining unit 402 may include: a first determining subunit, a second determining subunit and a third determining subunit;
the first determining subunit is configured to determine a set indicator variation threshold as the first association requirement; and setting the index variation threshold according to the variation distribution state of the index data in the sample data to be analyzed.
The second determining subunit is configured to determine an index variation of the sample data to be analyzed according to the index data of the standard date of the sample data to be analyzed and the set index data of the reference date of the sample data to be analyzed;
and the third determining subunit is configured to determine the abnormal sample data according to the index variation and the index variation threshold.
For a specific implementation process of the obtaining unit 402, reference may be made to the content of step S102, and details are not repeated here.
The excavation unit 403 may include: the first determining subunit is used for acquiring the sub-unit and the second determining subunit;
the first determining subunit is configured to determine a set support degree as a requirement condition of the second association requirement;
the obtaining subunit is configured to obtain target abnormal sample data according to the support degree;
and the second determining subunit is configured to determine the frequent item set according to a frequent pattern tree constructed by the target abnormal sample data.
The acquiring subunit may include: a statistics subunit and a target determination subunit; the statistics subunit is configured to perform statistics on the abnormal sample data according to the occurrence times of the same dimension value in the dimension data, and acquire candidate abnormal sample data; the target determining subunit is configured to determine the target abnormal sample data according to the dimension value greater than or equal to the support threshold in the selected candidate abnormal sample data.
The second determining subunit may include: the sequencing subunit is used for constructing the subunit and searching the subunit; the sorting subunit is configured to sort the target abnormal sample data according to the number of occurrences of the dimensional value; the constructing subunit is configured to construct the frequent pattern tree according to the sorted target abnormal sample data; and the searching subunit searches the frequent item sets meeting the requirements of conditional frequent item sets and/or frequent item sets according to the frequent pattern tree.
For the specific implementation process of the mining unit 403, reference may be made to the content of step S103, and details are not repeated here.
The second determining unit 404 may include: a requirement determination subunit and a data determination subunit; the requirement determining subunit is configured to determine a set confidence threshold as the third association requirement; and the comparison subunit is configured to determine fluctuation data corresponding to the dimension combination in the frequent item set according to the comparison between the obtained fluctuation score of the dimension combination in the frequent item set and the confidence threshold.
The comparison subunit may include: a score determination subunit and a data determination subunit; the score determining subunit is configured to determine the fluctuation score according to the contribution degree and the abnormality degree of the dimension combination; and the data determining subunit is configured to determine the dimension combination with the fluctuation score being greater than or equal to the confidence threshold as the fluctuation data corresponding to the dimension combination in the frequent item set.
The score determining subunit may include a contribution score determining subunit, an abnormality score determining subunit, and a fluctuation score determining subunit; and the contribution degree score determining subunit is used for determining the score of the contribution degree according to the sum of the comparison date index values of the dimension combinations in the frequent item set, the sum of the comparison date index values of all the dimension combinations, the sum of the reference date index values of the dimension combinations in the frequent item set and the sum of the reference date index values of all the dimension combinations. And the abnormal degree score determining subunit is used for determining the abnormal degree score according to the prior probability of the dimension combination in the frequent item set and the posterior probability of the dimension combination in the frequent item set. The fluctuation score determining subunit is configured to determine the fluctuation score according to a weighted sum of the contribution score and the abnormality score.
The abnormality degree score determining subunit may include: a first probability determination subunit and a second probability determination subunit, the first probability determination subunit being configured to determine the prior probability based on a sum of reference date index values of the dimension combinations in the frequent item set and a sum of reference date index values of all the dimension combinations; the second probability determination subunit is used for determining the posterior probability according to the sum of the comparison date index values of the dimension combinations in the frequent item set and the sum of the comparison date index values of all the dimension combinations; the abnormality degree score determination subunit is specifically configured to determine the abnormality degree score according to the prior probability and the posterior probability.
For a specific implementation process of the second determining unit 404, reference may be made to the content of step S102, and details are not repeated here.
The embodiment may further include: the device comprises a classification unit, a sorting unit and an output unit; the classification unit is used for classifying the dimension combination corresponding to the fluctuation data into a cause set; the sorting unit is used for sorting the dimension combinations in the attribution sets according to fluctuation scores; the output unit is used for outputting the sorted cause sets.
The embodiment may further include: and the data preprocessing unit is used for carrying out data combination and/or data cleaning on the log data according to the dimension data.
The above is a description of an embodiment of the index fluctuation analysis apparatus provided in the present application, and as to the specific content of the embodiment of the apparatus, reference may be made to the content of the above method embodiment, and therefore, a description will not be repeated.
Based on the above, the present application further provides a transaction index fluctuation analysis interaction method for a resource service application platform, as shown in fig. 5, fig. 5 is a flowchart of an embodiment of the index fluctuation analysis interaction method for the resource service application platform provided in the present application; the interaction method embodiment can include:
step S501: and transmitting the transaction sample data to be analyzed based on the trigger of the transaction index fluctuation analysis request of the resource service application platform in the enterprise digital center.
In this embodiment, the resource service application platform in step S501 may be understood as an application platform capable of providing a commodity transaction service, transaction log data may be generated on the resource service application platform, and the transaction log data may be stored in a database in a form of a data table. In this embodiment, the executing party of the trigger of the transaction index fluctuation analysis request may be a client, specifically, the trigger may be automatically triggered according to a set trigger period, or a trigger event may be generated according to touch; the former can be bound with a database, and when a trigger event is received, transaction sample data to be analyzed is automatically uploaded from the database; the latter may be that when a trigger event is generated according to touch, a user uploads transaction sample data to be analyzed that needs to be analyzed. Therefore, the specific implementation process of step S501 may include:
step S501-1: selecting the transaction sample data to be analyzed from a database based on triggering of a transaction index fluctuation analysis request of a resource service application platform in an enterprise digital center; the data table established according to the transaction log data generated by the resource service application platform is stored in the database, and at least one piece of transaction sample data to be analyzed in the data table comprises a dimension combination of at least two dimension data and index data corresponding to the dimension combination.
Step S501-2: and transmitting the selected transaction sample data to be analyzed to an index fluctuation analysis side.
Step S502: and responding to the transaction index fluctuation analysis request, and acquiring abnormal transaction sample data generated by the resource service application platform for the transaction sample data to be analyzed according to a first association requirement.
Step S503: and performing frequent item set mining on the abnormal transaction sample data according to a second association requirement.
Step S504: and determining fluctuation data corresponding to the dimension combination in the frequent item set according to a third correlation requirement.
The specific implementation process of step S502 to step S504 may refer to the content of step S101 to step S104, and will not be described repeatedly here.
Step S505: and displaying the fluctuation data in a visual mode.
In step S505, the fluctuation data may be displayed in a visual manner, and a merchant of the resource service application platform may visually obtain the index fluctuation condition, for example: information of a combination of dimensions related to the index fluctuation. Of course, the enterprise digital center management party can also intuitively know the index fluctuation condition. In this embodiment, the displayed fluctuation data may include information of a dimension combination related to the index fluctuation, and may also be related information such as an index fluctuation attribution result, output reference data, and/or a parameter adjustment suggestion; the method can also comprise comparison between the past historical index fluctuation condition and the current index fluctuation condition and the like. The present embodiment may include: generating at least one or more of a dimensional combination of a fluctuation, index data information corresponding to the dimensional combination, fluctuation analysis information, adjustment reference information provided according to the fluctuation analysis information, and history adjustment example information; wherein the fluctuation analysis information is determined from the fluctuation data; the history adjustment example information may be example information adjusted for the fluctuation data, and/or fluctuation change state information of the history fluctuation data adjusted by the example information.
The content of the fluctuation data is not limited, and information related to the fluctuation data can be displayed in a visual mode.
The specific implementation process of step S505 may include:
step S505-1: displaying the fluctuation data according to the selected display quantity of the fluctuation data; wherein the fluctuation data includes the dimension combination and/or index data corresponding to the dimension combination.
The above is a specific description of an embodiment of an index fluctuation analysis interaction method for a resource service application platform provided in the present application, and corresponds to the foregoing embodiment of an index fluctuation analysis interaction method for a resource service application platform, and the present application also discloses an embodiment of an index fluctuation analysis interaction apparatus for a resource service application platform, please refer to fig. 6. The device embodiments described below are merely illustrative.
As shown in fig. 6, fig. 6 is a schematic structural diagram of an embodiment of a transaction index fluctuation analysis interaction device of a resource service application platform provided in the present application; the interaction device embodiment comprises:
the transmission unit 601 is configured to transmit transaction sample data to be analyzed based on triggering of a transaction index fluctuation analysis request of a resource service application platform in the enterprise digital middlebox.
An obtaining unit 602, configured to, in response to the transaction index fluctuation analysis request, obtain, according to a first association requirement, abnormal transaction sample data generated by the resource service application platform for the transaction sample data to be analyzed.
A mining unit 603, configured to perform frequent item set mining on the abnormal transaction sample data according to a second association requirement;
a determining unit 604, configured to determine, according to a third association requirement, fluctuation data corresponding to a dimension combination in the frequent item set;
the display unit 605 is configured to display the fluctuation data in a visual manner.
In this embodiment, the transmission unit 601 may include: the selecting subunit is used for selecting the transaction sample data to be analyzed from the database based on the triggering of a transaction index fluctuation analysis request of a resource service application platform in the enterprise digital center station; the data table established according to the transaction log data generated by the resource service application platform is stored in the database, and at least one piece of transaction sample data to be analyzed in the data table comprises a dimension combination of at least two dimension data and index data corresponding to the dimension combination. The transmission unit 601 is specifically configured to transmit the transaction sample data to be analyzed selected by the selecting subunit to an index fluctuation analysis side.
In this embodiment, the display unit 605 is specifically configured to display the fluctuation data according to the selected display quantity of the fluctuation data; namely: the fluctuation data is displayed in a visual mode, and a merchant of the resource service application platform can visually acquire the index fluctuation condition, such as: information of a combination of dimensions related to the index fluctuation. Of course, the enterprise digital center management party can also intuitively know the index fluctuation condition. In this embodiment, the displayed fluctuation data may include information of a dimension combination related to the index fluctuation, and may also be related information such as an index fluctuation attribution result, output reference data, and/or a parameter adjustment suggestion; the method can also comprise comparison between the past historical index fluctuation condition and the current index fluctuation condition and the like. The present embodiment may include: generating at least one or more of a dimensional combination of a fluctuation, index data information corresponding to the dimensional combination, fluctuation analysis information, adjustment reference information provided according to the fluctuation analysis information, and history adjustment example information; wherein the fluctuation analysis information is determined from the fluctuation data; the history adjustment example information may be example information adjusted for the fluctuation data, and/or fluctuation change state information of the history fluctuation data adjusted by the example information.
Specific contents of the embodiment of the transaction index fluctuation analysis interaction device regarding the resource service application platform may refer to the description of step S501 to step S505 and the description of step S101 to step S104 described above.
Based on the above, the present application further provides a method for displaying fluctuation data, including:
responding to display triggering operation of the fluctuation data, and displaying the acquired fluctuation data on a fluctuation data output interface of the electronic equipment; wherein the fluctuation data includes: generating at least one or more of a dimensional combination of a fluctuation, index data information corresponding to the dimensional combination, fluctuation analysis information, adjustment reference information provided according to the fluctuation analysis information, and history adjustment example information; wherein the fluctuation analysis information is determined from the fluctuation data; the history adjustment example information may be example information adjusted for the fluctuation data, and/or fluctuation change state information of the history fluctuation data adjusted by the example information.
Continuing to use the example in the embodiment of the index fluctuation analysis method, as shown in fig. 3, assuming that log data generated by a service application platform in an enterprise digital middleboard is transaction data, fluctuation data corresponding to a determined frequent item set dimension combination is fluctuation data of frequent items of youth and Jiangsu, and an e-commerce application service party can analyze the content of the fluctuation data, such as potential reasons of Jiangsu & youth transaction fluctuation data, and information such as fluctuation occurrence caused by transaction time; the time of fluctuation of Jiangsu & youth in historical fluctuation data can be displayed, so that an e-commerce application service side or a third party can judge the rule of fluctuation data and establish a coping strategy, for example: relevant information of preferential activities is introduced at the peak of the morning and the evening for Jiangsu and youth; historical coping strategy information of historical fluctuation data similar to or identical to that of Jiangsu & youth fluctuation data can also be displayed, such as: the historical application strategy can be the version of the application platform adjusted (such as fluctuation caused by updating the version or fluctuation caused by updating the version), and can also be the comparison between the operation strategy of the e-commerce application platform before the historical fluctuation data and the operation strategy of the e-commerce application platform after adjustment. The presented information may also include: after the historical fluctuation data is adjusted according to the memorability of the coping strategy information, the fluctuation change state information of the fluctuation data is, for example: the fluctuation state information may be information on the time length for which the fluctuation occurs again, the amplitude of the fluctuation change, and the like.
Of course, the drilling can be continued according to the fluctuation data, and the specific information of the fluctuation caused by the analysis of the fluctuation data and the e-commerce transaction recorded data is combined, such as: the product purchase quantity is suddenly increased or decreased, so that the e-commerce side can adjust the product in a targeted manner.
Correspondingly, this application still provides a display device of fluctuation data, includes:
the display unit is used for responding to display triggering operation of the fluctuation data and displaying the acquired fluctuation data on a fluctuation data output interface of the electronic equipment; wherein the fluctuation data includes: generating at least one or more of a dimensional combination of a fluctuation, index data information corresponding to the dimensional combination, fluctuation analysis information, adjustment reference information provided according to the fluctuation analysis information, and history adjustment example information; wherein the fluctuation analysis information is determined from the fluctuation data; the history adjustment example information may be example information adjusted for the fluctuation data, and/or fluctuation change state information of the history fluctuation data adjusted by the example information.
Based on the above, the present application further provides a computer storage medium for storing data generated by a network platform and a program for processing the data generated by the network platform;
the program, when being read and executed by a processor, executes the steps in the index fluctuation analysis method embodiment as described above; or, executing the steps in the embodiment of the transaction index fluctuation analysis interaction method of the resource service application platform.
Based on the above, the present application further provides an electronic device, as shown in fig. 7, fig. 7 is a schematic structural diagram of an embodiment of the electronic device provided in the present application, where the embodiment includes:
a processor 701;
a memory 702 for storing a program for processing data generated by a network platform, the program, when being read and executed by the processor, performing the steps in the index fluctuation analysis method embodiment as described above; or, executing the steps in the embodiment of the transaction index fluctuation analysis interaction method of the resource service application platform.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
1. Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include non-transitory computer readable media (transient media), such as modulated data signals and carrier waves.
2. As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
Although the present application has been described with reference to the preferred embodiments, it is not intended to limit the present application, and those skilled in the art can make variations and modifications without departing from the spirit and scope of the present application, therefore, the scope of the present application should be determined by the claims that follow.

Claims (18)

1. An index fluctuation analysis method, characterized by comprising:
determining log data generated by a service application platform in an enterprise digital center station as sample data to be analyzed; the at least one piece of sample data to be analyzed comprises a dimension combination of at least two pieces of dimension data and index data corresponding to the dimension combination;
analyzing the index data of the dimension combination in the sample data to be analyzed according to a set first correlation requirement to obtain abnormal sample data;
according to a set second association requirement, mining a frequent item set of the dimension combination in the abnormal sample data;
and determining fluctuation data corresponding to the dimension combination in the frequent item set according to a set third association requirement.
2. The index fluctuation analysis method according to claim 1, wherein the analyzing, according to a set first association requirement, the index data of the dimension combination in the sample data to be analyzed to obtain abnormal sample data includes:
determining a set index variation threshold as the first association requirement;
determining the index variation of the sample data to be analyzed according to the index data of the standard date of the sample data to be analyzed and the set index data of the reference date of the sample data to be analyzed;
and determining the abnormal sample data according to the index variation and the index variation threshold.
3. The index fluctuation analysis method according to claim 2, wherein the index variation threshold is set according to a variation distribution state of index data in the sample data to be analyzed.
4. The index fluctuation analysis method according to claim 1, wherein the mining a frequent item set of the dimension combinations in the abnormal sample data according to the set second association requirement includes:
determining the set support degree as a requirement condition of the second association requirement;
acquiring target abnormal sample data according to the support degree;
and determining the frequent item set according to a frequent pattern tree constructed by the target abnormal sample data.
5. The method according to claim 4, wherein the obtaining target abnormal sample data according to the support degree includes:
counting the abnormal sample data according to the occurrence times of the same dimension value in the dimension data to obtain candidate abnormal sample data;
and determining the target abnormal sample data according to the dimension value which is greater than or equal to the support threshold value in the selected candidate abnormal sample data.
6. The method according to claim 5, wherein the determining the frequent item set according to the frequent pattern tree constructed from the target abnormal sample data includes:
sequencing the target abnormal sample data according to the occurrence times of the dimension value;
constructing the frequent pattern tree according to the sorted target abnormal sample data;
and searching the frequent item set which meets the requirements of conditional frequent item sets and/or frequent items according to the frequent pattern tree.
7. The index fluctuation analysis method according to claim 1, wherein the determining fluctuation data corresponding to the dimension combination in the frequent item set according to the set third association requirement includes:
determining a set confidence threshold as the third association requirement;
and determining fluctuation data corresponding to the dimension combination in the frequent item set according to the comparison between the obtained fluctuation score of the dimension combination in the frequent item set and the confidence coefficient threshold value.
8. The index fluctuation analysis method according to claim 7, wherein the determining fluctuation data corresponding to the frequent item set dimension combination according to the comparison between the obtained fluctuation score of the frequent item set dimension combination and the confidence threshold comprises:
determining the fluctuation fraction according to the contribution degree and the abnormality degree of the dimension combination;
and determining the dimensional combination with the fluctuation score larger than or equal to the confidence coefficient threshold value as the fluctuation data corresponding to the dimensional combination in the frequent item set.
9. The index fluctuation analysis method according to claim 8, wherein the determining the fluctuation score based on the degree of contribution and the degree of abnormality of the dimensional combination includes:
determining the score of the contribution degree according to the sum of the comparison date index values of the dimension combinations in the frequent item set, the sum of the comparison date index values of all the dimension combinations, the sum of the reference date index values of the dimension combinations in the frequent item set and the sum of the reference date index values of all the dimension combinations;
determining the value of the abnormal degree according to the prior probability of the dimension combination in the frequent item set and the posterior probability of the dimension combination in the frequent item set;
and determining the fluctuation score according to the weighted sum of the contribution score and the abnormal degree score.
10. The index fluctuation analysis method according to claim 9, wherein the determining the score of the degree of abnormality based on the prior probability of the combination of dimensions in the frequent item set and the posterior probability of the combination of dimensions in the frequent item set comprises:
determining the prior probability according to the sum of the reference date index values of the dimension combinations in the frequent item set and the sum of the reference date index values of all the dimension combinations;
determining the posterior probability according to the sum of the comparison date index values of the dimension combinations in the frequent item set and the sum of the comparison date index values of all the dimension combinations;
and determining the abnormal degree score according to the prior probability and the posterior probability.
11. The index fluctuation analysis method according to claim 1, characterized by further comprising:
classifying the dimension combination corresponding to the fluctuation data into a cause set;
sorting the dimension combinations in the cause set according to fluctuation scores;
and outputting the sorted attribution sets.
12. The index fluctuation analysis method according to claim 1, characterized by further comprising:
and carrying out data combination and/or data cleaning on the log data according to the dimension data.
13. A transaction index fluctuation analysis interaction method of a resource service application platform is characterized by comprising the following steps:
transmitting transaction sample data to be analyzed based on triggering of a transaction index fluctuation analysis request of a resource service application platform in an enterprise digital center;
responding to the transaction index fluctuation analysis request, and acquiring abnormal transaction sample data generated by the resource service application platform for the transaction sample data to be analyzed according to a first association requirement;
performing frequent item set mining on the abnormal transaction sample data according to a second association requirement;
determining fluctuation data corresponding to the dimensionality combination in the frequent item set according to a third correlation requirement;
and displaying the fluctuation data in a visual mode.
14. The method according to claim 13, wherein the transmitting the transaction sample data to be analyzed based on the trigger of the request for transaction index fluctuation analysis of the resource service application platform in the enterprise digital center comprises:
selecting the transaction sample data to be analyzed from a database based on triggering of a transaction index fluctuation analysis request of a resource service application platform in an enterprise digital center; the data table established according to the transaction log data generated by the resource service application platform is stored in the database, and at least one piece of transaction sample data to be analyzed in the data table comprises a dimension combination of at least two dimension data and index data corresponding to the dimension combination;
and transmitting the selected transaction sample data to be analyzed to an index fluctuation analysis side.
15. The method of claim 13, wherein the visually presenting the fluctuation data comprises:
displaying the fluctuation data according to the selected display quantity of the fluctuation data; the surge data includes: generating at least one or more of a dimensional combination of a fluctuation, index data information corresponding to the dimensional combination, fluctuation analysis information, adjustment reference information provided according to the fluctuation analysis information, and history adjustment example information; wherein the fluctuation analysis information is determined from the fluctuation data.
16. A method for displaying fluctuation data is characterized by comprising the following steps:
responding to display triggering operation of the fluctuation data, and displaying the acquired fluctuation data on a fluctuation data output interface of the electronic equipment; wherein the fluctuation data includes: generating at least one or more of a dimensional combination of a fluctuation, index data information corresponding to the dimensional combination, fluctuation analysis information, adjustment reference information provided according to the fluctuation analysis information, and history adjustment example information; wherein the fluctuation analysis information is determined from the fluctuation data.
17. A computer storage medium for storing network platform generated data and a program for processing the network platform generated data;
the program, when read and executed by a processor, performs the index fluctuation analysis method according to any one of claims 1 to 12; or, executing the transaction index fluctuation analysis interaction method of the resource service application platform according to any one of the claims 13 to 15; alternatively, a method of presenting fluctuating data according to claim 16 is performed.
18. An electronic device, comprising:
a processor;
a memory for storing a program for processing data generated by a network platform, the program, when read and executed by the processor, performing the index fluctuation analysis method according to any one of claims 1 to 12; or, executing the transaction index fluctuation analysis interaction method of the resource service application platform according to any one of the claims 13 to 15; alternatively, a method of presenting fluctuating data according to claim 16 is performed.
CN202111657169.0A 2021-12-31 2021-12-31 Index fluctuation analysis method, storage medium and electronic equipment Pending CN114022051A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111657169.0A CN114022051A (en) 2021-12-31 2021-12-31 Index fluctuation analysis method, storage medium and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111657169.0A CN114022051A (en) 2021-12-31 2021-12-31 Index fluctuation analysis method, storage medium and electronic equipment

Publications (1)

Publication Number Publication Date
CN114022051A true CN114022051A (en) 2022-02-08

Family

ID=80069450

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111657169.0A Pending CN114022051A (en) 2021-12-31 2021-12-31 Index fluctuation analysis method, storage medium and electronic equipment

Country Status (1)

Country Link
CN (1) CN114022051A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114726593A (en) * 2022-03-23 2022-07-08 阿里云计算有限公司 Data analysis method, data analysis device, abnormal information identification method, abnormal information identification device, and storage medium
CN117312296A (en) * 2023-11-29 2023-12-29 山东街景智能制造科技股份有限公司 Multi-platform collaboration data processing method

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080288430A1 (en) * 2007-02-26 2008-11-20 International Business Machines Corporation System and method to infer anomalous behavior of members of cohorts and inference of associate actors related to the anomalous behavior
CN108346011A (en) * 2018-05-15 2018-07-31 阿里巴巴集团控股有限公司 Index fluction analysis method and device
CN109961199A (en) * 2017-12-25 2019-07-02 北京京东尚科信息技术有限公司 A kind of method and apparatus for analyzing data fluctuations
CN112132485A (en) * 2020-09-30 2020-12-25 上海众源网络有限公司 Index data processing method and device, electronic equipment and storage medium
CN112365146A (en) * 2020-11-06 2021-02-12 腾讯科技（北京）有限公司 Method, device and equipment for acquiring dimension of index transaction and storage medium
CN112667827A (en) * 2020-12-23 2021-04-16 北京奇艺世纪科技有限公司 Data anomaly analysis method and device, electronic equipment and storage medium
CN113128875A (en) * 2021-04-22 2021-07-16 阳光保险集团股份有限公司 Root cause positioning method and device for index abnormity of multi-dimensional data set
CN113537685A (en) * 2020-10-10 2021-10-22 腾讯科技（深圳）有限公司 Data processing method and device
WO2021217855A1 (en) * 2020-04-30 2021-11-04 平安科技（深圳）有限公司 Abnormal root cause positioning method and apparatus, and electronic device and storage medium

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080288430A1 (en) * 2007-02-26 2008-11-20 International Business Machines Corporation System and method to infer anomalous behavior of members of cohorts and inference of associate actors related to the anomalous behavior
CN109961199A (en) * 2017-12-25 2019-07-02 北京京东尚科信息技术有限公司 A kind of method and apparatus for analyzing data fluctuations
CN108346011A (en) * 2018-05-15 2018-07-31 阿里巴巴集团控股有限公司 Index fluction analysis method and device
WO2021217855A1 (en) * 2020-04-30 2021-11-04 平安科技（深圳）有限公司 Abnormal root cause positioning method and apparatus, and electronic device and storage medium
CN112132485A (en) * 2020-09-30 2020-12-25 上海众源网络有限公司 Index data processing method and device, electronic equipment and storage medium
CN113537685A (en) * 2020-10-10 2021-10-22 腾讯科技（深圳）有限公司 Data processing method and device
CN112365146A (en) * 2020-11-06 2021-02-12 腾讯科技（北京）有限公司 Method, device and equipment for acquiring dimension of index transaction and storage medium
CN112667827A (en) * 2020-12-23 2021-04-16 北京奇艺世纪科技有限公司 Data anomaly analysis method and device, electronic equipment and storage medium
CN113128875A (en) * 2021-04-22 2021-07-16 阳光保险集团股份有限公司 Root cause positioning method and device for index abnormity of multi-dimensional data set

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114726593A (en) * 2022-03-23 2022-07-08 阿里云计算有限公司 Data analysis method, data analysis device, abnormal information identification method, abnormal information identification device, and storage medium
CN117312296A (en) * 2023-11-29 2023-12-29 山东街景智能制造科技股份有限公司 Multi-platform collaboration data processing method
CN117312296B (en) * 2023-11-29 2024-02-20 山东街景智能制造科技股份有限公司 Multi-platform collaboration data processing method

Similar Documents

Publication Publication Date Title
CN108537544B (en) Real-time monitoring method and monitoring system for transaction system
US11816120B2 (en) Extracting seasonal, level, and spike components from a time series of metrics data
US8676818B2 (en) Dynamic storage and retrieval of process graphs representative of business processes and extraction of formal process models therefrom
US20200192894A1 (en) System and method for using data incident based modeling and prediction
CN106462583B (en) System and method for rapid data analysis
US8442863B2 (en) Real-time-ready behavioral targeting in a large-scale advertisement system
US6763354B2 (en) Mining emergent weighted association rules utilizing backlinking reinforcement analysis
US20030187677A1 (en) Processing user interaction data in a collaborative commerce environment
CN114022051A (en) Index fluctuation analysis method, storage medium and electronic equipment
TW201327233A (en) Personalized information pushing method and device
Japec et al. AAPOR report on big data
CN113342976A (en) Method, device, storage medium and equipment for automatically acquiring and processing data
CN110637317A (en) Distributed node cluster for establishing digital contact points across multiple devices on a digital communications network
CN110795613A (en) Commodity searching method, device and system and electronic equipment
KR20200019741A (en) Data Analysis Support System and Data Analysis Support Method
CN115577991B (en) Business intelligent data analysis system and analysis method based on big data
US20210357401A1 (en) Automatic frequency recommendation for time series data
CN114495137A (en) Bill abnormity detection model generation method and bill abnormity detection method
CN113393169A (en) Financial industry transaction system performance index analysis method based on big data technology
Shabana et al. A study on Big data advancement and Big data analytics
Kumar et al. Jaccard similarity based mining for high utility webpage sets from weblog database.
KR20220069229A (en) The method of coupling with heterogeneous data using relation of fields in data
CN111191126A (en) Keyword-based scientific and technological achievement accurate pushing method and device
CN110992109A (en) Real estate customer analysis method, device and storage medium based on association rule
Andriichuk et al. Usage of expert decision-making support systems in information operations detection

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
REG Reference to a national code

Ref country code: HK

Ref legal event code: DE

Ref document number: 40067481

Country of ref document: HK

RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20220208