CN113779044B - Data drilling method and system - Google Patents
Data drilling method and system Download PDFInfo
- Publication number
- CN113779044B CN113779044B CN202111310616.5A CN202111310616A CN113779044B CN 113779044 B CN113779044 B CN 113779044B CN 202111310616 A CN202111310616 A CN 202111310616A CN 113779044 B CN113779044 B CN 113779044B
- Authority
- CN
- China
- Prior art keywords
- layer
- data
- dimension
- drill
- drilling
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/2228—Indexing structures
- G06F16/2246—Trees, e.g. B+trees
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/2282—Tablespace storage structures; Management thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2474—Sequence data queries, e.g. querying versioned data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/248—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
- G06F16/258—Data format conversion from or to a database
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/80—Information retrieval; Database structures therefor; File system structures therefor of semi-structured data, e.g. markup language structured data such as SGML, XML or HTML
- G06F16/84—Mapping; Conversion
Abstract
The invention discloses a data drilling method, which comprises the following steps: s1, receiving and analyzing a user request, and generating a drill-down list with N layers, wherein N is more than or equal to 1 and is a positive integer; s2, acquiring and analyzing the drill-down data of each layer by layer from the first layer according to the drill-down list until the Nth layer is finished; and S3, processing the acquired data, generating format data meeting the direct display requirement of the front end, and returning the format data to the user. The invention also provides a data drilling system, which is used for drilling data layer by utilizing the layered drilling-down list, so that the data drilling efficiency is effectively improved.
Description
Technical Field
The invention belongs to the technical field of data processing, and particularly relates to a data drilling method and a data drilling system.
Background
Drilling is one of the characteristic functions in the report forms, and through the drilling function, a user can click a link in one report form to directly open other related report forms. As in the current intelligent system of online data analysis, there is such a result: the product category with the highest purchase rate of the customer group from the age of 30 to the age of 40. The user now also needs to know more detailed results, such as which products they have primarily purchased. In professional terms, it is necessary to know from which data this conclusion is based. This can be done by drilling out using the report.
When large data analysis is performed, the most used operation is drilling, which is to change the dimension level and change the analysis granularity. It includes drill-up and drill-down. The upward drilling is to summarize the low-level detail data to the high-level summarized data in a certain dimension or reduce the dimension; drilling down is the opposite, in that it looks deep into the detail data from the summary data or adds new dimensions. For example, when the user analyzes "sales conditions of regions and cities", the sales of a certain city may be subdivided into sales of various years, and the sales of a certain year may be further subdivided into sales of various seasons. Through the drilling function, the user can more deeply understand the data, can more easily find problems and make correct decisions.
If the application publication date is 2019, 2, and 22, the application publication number is CN109376177A, and the patent name is a chinese patent of a data drilling analysis method, a technical scheme is disclosed, so as to solve the technical problem that detailed data query of the power service information system in the prior art cannot realize cross-library customized query, and the method comprises the following steps: s1: creating a logic database; s2: creating a tree classification node and filling in node basic information; s3: writing an SQL query statement and performing compiling test; s4: determining a query parameter according to a query requirement, and configuring a drilling parameter; s5: judging whether a drill-down data set needs to be set according to the query requirement, and continuing to step S6 when the drill-down node data set needs to be set; s6: configuring a drill-down node data set; s7: and exporting and displaying the query result.
The application publication date is 12/4/2020, application publication number is CN112039989A, and the chinese patent with patent name of data drilling method, proxy server, service calling system and medium discloses a technical scheme, which includes the following steps: the proxy server receives a json character string sent by a request end; acquiring a name and a keyword initial subscript of a reference corresponding to the json character string, and storing the name and the keyword initial subscript as metadata corresponding to the json character string; serializing the json character strings to obtain a data packet in a json format; and when the data drilling function is triggered, drilling target data according to the metadata and the data packet.
However, when using drill-down, there are several problems as follows:
firstly, when drilling down, the data of a layer to be drilled down can be acquired in less than 1 second, and the data can be acquired in dozens of seconds or even minutes after the multi-layer drilling is continued. As more levels are drilled down, more data needs to be acquired per layer, and the time required to acquire data increases in geometric multiples.
Secondly, when drilling downwards, each layer of the drilling can be provided with own topN and sequencing rules, and ascending and descending orders can be carried out according to the layer or any other dimensionalities. The complex rule makes it difficult to obtain the required data by using one SQL statement, so the data is usually obtained by a hierarchical obtaining method. For example, drill down three layers, each layer requires a different topN and ordering rules:
(1) the layer 1 acquires 20 pieces of data meeting the conditions through one-time query according to a rule that the A field top20 is in descending order;
(2) the layer 2 acquires 600 pieces of data through 20 times of inquiry according to a rule that the B field top30 is in an ascending order and the 20 pieces of data acquired by the layer 1 are taken as conditions;
(3) by analogy, the layer 3 needs 600 queries to obtain the complete data of the third layer;
in order to solve the problem that the more the number of drilling layers is, the time consumption for acquiring complete data is too much due to the exponential increase of the number of times of querying the database, and in order to solve the problem that the time consumption for acquiring complete data is too much in the prior art, the following method is usually adopted for solving the problem that: or the problem is solved by limiting the number of layers to be drilled, for example, only 3 layers can be drilled; but this approach affects the efficiency of the user experience and analysis of the data; or the problem is solved by reducing the query times, not all the drill-down data are displayed, only the first layer of data are displayed, and the user can only select a single piece of data and then trigger the user to continue to drill down the current data; the mode increases the operation frequency of the user and influences the efficiency of data analysis.
Disclosure of Invention
1. Problems to be solved
The invention provides a data drilling method and a data drilling system, aiming at the problem of excessive time consumption in data query in the data analysis process in the prior art.
2. Technical scheme
In order to solve the problems, the technical scheme adopted by the invention is as follows: a method of data drilling, comprising the steps of:
s1, receiving and analyzing a user request, and generating a drill-down list with N layers, wherein N is more than or equal to 1 and is a positive integer;
s2, acquiring and analyzing the drill-down data of each layer by layer from the first layer according to the drill-down list until the Nth layer is finished;
and S3, processing the acquired data, generating format data meeting the direct display requirement of the front end, and returning the format data to the user. According to the technical scheme, the layered drill-down list is utilized, data are drilled layer by layer, and the data drilling efficiency is effectively improved.
Further, the step S2 includes:
s21, acquiring dimension value data of the specified number, which are grouped by the dimension fields of the first layer of drill-down according to the list information of the first layer of drill-down, and sorted by the dimension fields of the first layer of drill-down by taking the time period requested by the user as a filtering condition;
s22, filtering the data to obtain a unique value in the first-layer drilling dimension field;
s23, according to the information of the first and second layers of drilling, the drilling dimension fields of the first and second layers are grouped, the time period requested by the user and the unique value of the drilling dimension field of the first layer in the step S22 are used as filtering conditions, the index dimension fields of the drilling of the second layer are sorted, and the specified number of dimension value data are distributed to the second layer of each first layer dimension field value;
s24, carrying out filtering repetition on the data in the step S23 to obtain a unique value of the second-layer drilling dimension field;
s25, according to the list information of the first layer, the second layer and the third layer of the drill-down, obtaining that the drill-down dimension fields of the first layer, the second layer and the third layer are grouped, taking the time period requested by a user, the unique value of the first layer dimension field in the step S22 and the unique value of the second layer of the drill-down dimension field in the step S24 as filtering conditions, sorting the drill-down index dimension fields of the third layer, and distributing the dimension value data of the appointed number to the third layer of each second layer dimension field value;
s26, carrying out filtering and repeating on the data in the step S25 to obtain a unique value of a third-layer drilling dimension field;
s27, and so on, until the entire drill-down list is traversed.
According to the technical scheme, the data are grouped, filtered, repeated and sequenced layer by layer, and drilling of each layer of data after the first layer obtains data from a limited data volume, so that the data drilling efficiency is greatly improved, and the user experience is improved.
Further, the step S3 includes:
s31, grouping the second layer data by the dimension field of the first layer drill-down to generate second layer grouped data;
s32, grouping the third layer data by the dimension fields of the first and second layer drill-down to generate third layer grouped data;
s33, by analogy, grouping the data of the Nth layer by the dimension fields drilled from the first layer to the Nth-1 layer to generate the data of the Nth layer;
and S34, starting from the first layer of the drill-down dimension value as the key field, finding corresponding data from the next layer of grouped data in a recursive mode, and splicing the data to form tree-structured data with the upper layer of the key field as a parent and the lower layer of the key field as a child, and feeding back the tree-structured data to the user. After the data are grouped, filtered and sequenced, a limited amount of data are obtained, and then data related to the dimension field of each layer are acquired layer by layer from the limited amount of data, so that the drilling efficiency is improved.
Further, the drill-to-annulus data includes: the method comprises the steps of obtaining current date data and previous date data, then performing grouping, sorting and filtering operation on the current date data and the previous date data respectively, wherein the obtained N-layer data of the previous date data needs to take a time period requested by a user and a unique value of a drill-down dimension field of the first N-1 layers of the current date data as a filtering condition.
Further, the step of generating the ring ratio data specifically includes:
s51, merging the previous-stage data of the first layer into the current-stage data of the first layer by taking the dimension field of the first layer as a connecting key, and calculating the difference value and the ring ratio of each index dimension field;
s52, merging the previous-period data of the second layer into the current-period data of the second layer by taking the dimension fields of the first layer and the second layer as a connecting key, and calculating the difference and the ring ratio of each index dimension field;
and S53, and so on, calculating the difference and the ring ratio of the index dimension field of each layer in the drill-down list. According to the technical scheme, the ring ratio value can be obtained, and the ring ratio value of the last layer is the required ring ratio data.
Further, the data format in step S3 is a JSON format.
The present invention also includes a data drilling system comprising:
the user request analysis module: the system is used for analyzing the user request and generating a drill-down list;
a data drilling module: and the data acquisition module is used for drilling data layer by layer according to the drill-down list generated by the user request analysis module and returning the acquired data to the user.
Furthermore, the data drilling module acquires a specified number of dimensional value data which are grouped by the dimensional fields of the first layer of drilling and sorted by the index dimensional fields of the first layer of drilling according to the first layer of information of the drilling and by taking the time period requested by a user as a filtering condition in the process of drilling the data; then, filtering the data to obtain a unique value in a first-layer drilling dimension field; according to the information of the first layer and the second layer of the drill-down, obtaining dimension fields of the drill-down of the first layer and the second layer to be grouped, taking the time period requested by a user and the unique value of the dimension field of the first layer as filtering conditions, sequencing the dimension fields of the index of the drill-down of the second layer, and distributing the specified number of dimension value data to the second layer of each dimension field value of the first layer; then, carrying out filtering and repeating on the data to obtain a unique value of a second-layer drilling dimension field; then according to the list information of the first layer, the second layer and the third layer of the drill-down, obtaining the drill-down dimension fields of the first layer, the second layer and the third layer to be grouped, taking the time period requested by a user, the unique value of the dimension field of the first layer of the drill-down and the unique value of the dimension field of the second layer of the drill-down as filtering conditions, sequencing the index dimension field of the drill-down of the third layer, and distributing the specified number of dimension value data to the third layer of each dimension field value of the second layer; filtering the dimension value data to obtain a unique value of a third layer of drilling dimension field; and so on until the entire drill-down list is traversed.
Further, the data drilling module obtains grouped data from each layer of data, specifically: grouping the data of the second layer by the dimension field of the first layer drill-down to generate the grouped data of the second layer; grouping the third layer data by using the dimension fields drilled by the first layer and the second layer to generate third layer grouped data; by analogy, grouping the data of the Nth layer by using the dimension fields drilled from the first layer to the Nth-1 layer to generate the data of the Nth layer; and then, starting from the first layer of the data with the dimension value as a key field, finding corresponding data from the next layer of grouped data in a recursive mode, and splicing the data to form tree-shaped structure data with the dimension field of the upper layer as a parent and the dimension field of the lower layer as a child, and feeding back the tree-shaped structure data to the user.
Further, the data drilling module acquires current date data and previous date data by taking user time as a filtering condition, generates ring ratio data according to the current date data and the previous date data of each layer, and feeds the ring ratio data back to the user.
3. Advantageous effects
Compared with the prior art, the invention has the beneficial effects that:
according to the invention, through carrying out filtering, grouping and sequencing on the data in the database and simultaneously limiting the number of groups, firstly, the number of times of query can be greatly reduced, the number of times of acquiring the data from the database is reduced, and the drilling efficiency is improved; in addition, the data of each layer is only acquired once when drilling is carried out, multiple acquisition is not needed, and the drilling efficiency is improved; and the obtained data can be directly returned to the front end for use after being converted into the format, other processing is not needed, and the data query efficiency is improved.
Drawings
FIG. 1 is a flow chart of example 1 of the present invention;
FIG. 2 is a first partial flowchart of example 2 of the present invention;
fig. 3 is a second partial flow chart of embodiment 2 of the present invention, and fig. 2 and fig. 3 constitute a complete flow chart of embodiment 2.
Detailed Description
The invention is further described with reference to specific examples.
First, some terms involved in the present invention are explained.
Indexes are as follows: units or methods for measuring the degree of development of a thing, such as: population, GDP, revenue, number of users, profit margin, retention, coverage, etc. The index needs to be obtained through a summary calculation manner such as summation, averaging and the like, and needs to be summarized and calculated under certain precondition such as time, place and range.
Dimension: the index is used for measuring the development degree of things, and if the development degree is good or bad, the development degree needs to be compared through different dimensions, and the dimension is a certain characteristic of things or phenomena, such as gender, region, time and the like. The time is a common and special dimension, and the development of things can be known to be good or bad through comparison before and after the time, for example, the number of users increases by 10% compared with the previous month and increases by 20% compared with the same year in the same year, which is the comparison in time and is also called vertical ratio; another comparison is the horizontal ratio, such as the comparison of population numbers and GDPs in different countries, the comparison of income of different provinces and number of users, the comparison between different companies and different departments, which are the comparison between units of the same level, which is called horizontal ratio for short.
Drill-down dimension field or dimension field: dimension fields that support drill-down operations, such as "gender", "region", "time", etc.;
index dimension field: units or methods for measuring drilling dimensions, such as "revenue", "number", etc.;
sorting dimension field: a field that can be used as a basis for drill-down dimension ordering; for example, the dimension field "gender" is arranged from large to small according to the amount of the dimension field "income", wherein the "income" is the dimension field of sorting; for example, the dimension field "gender" is arranged according to the Chinese pinyin of the "gender" value (male, female), wherein "gender" is also the dimension field of the rank.
Recursion: the programming technique of the procedure call itself is called recursion (recursion). Recursion is widely used in programming languages as an algorithm. A process or function has a method for directly or indirectly calling itself in its definition or description, which converts a large and complex problem layer by layer into a problem with a small scale similar to the original problem to solve, and the recursive strategy can describe the repeated calculation needed by the problem solving process only by a small amount of programs, thus greatly reducing the code amount of the programs.
The invention comprises a data drilling method and two data analysis methods which are commonly used in data analysis, wherein one method is to only see the current period data, and the other method is to see the ring ratio data, wherein the ring ratio = (the current period data-the previous period data)/the previous period data is multiplied by 100%.
Example 1
First, the following description will be given by taking the query of the current date as an example, as shown in fig. 1:
1. the server receives a request of the front end for drilling down display data;
2. the server analyzes the request, and the analyzing process specifically comprises the following steps: and generating N layers of drilling list information according to the drilling parameters, wherein N is more than or equal to 1 and is a positive integer, and the N comprises information such as a drilling dimension field of each layer, the number of dimension values, whether the sequencing dimension field and the sequencing dimension are sequenced in an ascending order or a descending order, the sequencing dimension field and a calculation formula thereof (the calculation formula can be preset by a user and stored at a server), and the like. For example, two user data, a and b, are currently analyzed, if in the original list, there are two groups of a, a and b are dimension fields, the number is a dimension field, the value of the dimension field is 5 and 2, respectively, there are two groups of b, 10 and 7, respectively, likewise, 10 and 7 are also dimension field values, the calculation formula is sum, that is, the sum is carried out, the number of a is 7, the number of b is 17, then the dimension field "number" is sorted, if it is descending order, b is arranged in front of a, otherwise, a is arranged in front of b.
3. According to the drill-down list information, starting from the first layer, drilling down layer by layer until the Nth layer is finished, and sequentially acquiring and analyzing the drill-down data of each layer at the current period:
3.1, according to the first layer information of the drill-down, acquiring the dimension value data of the specified number, which is sorted by the sorting dimension field of the drill-down of the first layer, by taking the time period requested by the user as the filtering condition (such as the requirement of the user for inquiring the data of one week or one month and the like);
3.2, filtering the data to obtain a unique value in the first-layer drilling dimension field, wherein the filtering is repeated data in the data, so that the operation amount in data drilling is reduced; for example, as shown in table 1, company and gender are drill-down dimension fields, people number is a mark dimension field, company is first-layer data, gender is second-layer data, and in the current table, there are four dimension fields: male, female, then the second layer after filtering is heavy has only two data: male and female; note that the filtering is for the drill-down dimension field, not the index dimension field.
TABLE 1
3.3 according to the information of the first and second layers of the drill-down, obtaining the dimension fields of the first and second layers of the drill-down to group, taking the time period requested by the user and the unique value of the dimension field of the first layer calculated in 3.2 as the filtering condition, sorting the sorted dimension fields of the drill-down of the second layer, and distributing the dimension value data of the appointed number to the second layer of each dimension field value of the first layer to obtain the limited number of data meeting the requirement, thereby reducing the number of times of inquiry.
The number of the query data can be limited by taking the unique value of the first-layer dimension field as a transition condition, so that the query range is limited in a small area, and the data query time is shortened; the function of allocating the designated number of dimensional value data to the second layer of each first layer dimensional field value is to group through fields and acquire the first X pieces of data of each group after grouping and sorting, wherein X is a positive integer, that is, the first X pieces of data of each group after grouping and sorting according to the dimensional fields drilled down by the first and second layers are acquired, more specifically, only the first X pieces of data of each group after sorting are displayed, and the data volume of the query result is limited in a small area, so that the efficiency of subsequent data processing is effectively improved.
3.4, carrying out filtering repetition on the data in the step 3.3 to obtain a unique value of the second-layer drilling dimension field;
3.5 according to the information of the first layer, the second layer and the third layer of the drill-down, obtaining the dimension fields of the first layer, the second layer and the third layer of the drill-down, grouping, taking the time period requested by a user and the unique values of the dimension fields of the first layer and the second layer calculated in the steps 3.2 and 3.4 as filtering conditions, sorting by using the sorting dimension fields of the third layer of the drill-down, and distributing the dimension value data with the appointed number to the third layer of each dimension field value of the second layer;
3.6, carrying out filtering and repeating on the data in the step 3.5 to obtain a unique value of a third-layer drilling dimension field;
3.7, by analogy, according to the information of the first to the Nth layers of the drill-down, obtaining the dimension fields of the first to the Nth layers of the drill-down, grouping, taking the time point requested by the user and the unique values of the dimension fields of the first to the Nth layers calculated in the 3.2 and 3.4.. as the filtering conditions, sequencing by using the sequencing dimension fields of the drill-down of the Nth layer, and distributing dimension value data of the appointed number of the dimension field values to the Nth layer of each dimension field value of the Nth-1 layer;
and 3.8, filtering and repeating the data in the previous step to obtain the unique value of the N-th layer down-drilling dimension field.
And 4, formatting data to generate format data meeting the direct display requirement of the front end:
4.1 grouping the second layer data according to the dimension field of the first layer drill-down to generate second layer grouped data;
4.2 grouping the third layer data by the dimension fields of the first and second layer drill-down to generate third layer grouped data;
4.3, by analogy, grouping the Nth layer data by the dimension fields drilled from the first layer to the N-1 layer to generate Nth layer grouped data;
and 4.4, starting from the first layer of the data with the drilling dimension value as the key field, finding corresponding data from the next layer of grouped data in a recursive mode, and splicing the data to form tree-shaped structure data with the upper layer of the data as a parent and the lower layer of the data corresponding to the key field as a child, and feeding back the tree-shaped structure data to the user.
It should be noted that, although both the packets in the data formatting and the packets in the data drill-down are packets, and the specific grouping method is the same, the purpose of the packets in the data drill-down is different, the packets in the data drill-down are to obtain data aggregated according to the packets, and after the packets in the data formatting are aggregated according to the packets, the corresponding data of the next layer is found in the next layer more quickly by using the dimension field value of the packet as an index, and the data is spliced, so that the tree-structured data in which the key field of the upper layer is used as a parent and the data corresponding to the key field of the lower layer is used as a child is found and fed back to the user.
In addition, in the invention, the information list of the drill-down is divided into N layers, each layer has its dimension field, dimension value number, sort dimension field and sort dimension whether ascending or descending sort, sort dimension field and its calculation formula, the data obtained is also layered, that is, the drill-down is performed according to the dimension field of each layer to obtain the data corresponding to each layer, and the tree structure data related according to the dimension field by the layer of data is finally fed back to the user.
5. And returning the formatted data to the front-end for presentation.
Example 2
In this embodiment, the query loop ratio data is taken as an example, and the specific steps are shown in fig. 2 and fig. 3:
1. the server receives a request of the front end for drilling down display data;
2. the server analyzes the request, and the analyzing process specifically comprises the following steps: and generating N layers of drilling list information according to the drilling parameters, wherein N is more than or equal to 1 and is a positive integer, and the N comprises information such as a drilling dimension field of each layer, the number of dimension values, whether the sequencing dimension field and the sequencing dimension are sequenced in an ascending order or a descending order, the sequencing dimension field and a calculation formula thereof (the calculation formula can be preset by a user and stored at a server), and the like. For example, two user data, a and b, are currently analyzed, if in the original list, there are two groups of a, a and b are dimension fields, the number is a dimension field, the value of the dimension field is 5 and 2, respectively, there are two groups of b, 10 and 7, respectively, likewise, 10 and 7 are also dimension field values, the calculation formula is sum, that is, the sum is carried out, the number of a is 7, the number of b is 17, then the dimension field "number" is sorted, if it is descending order, b is arranged in front of a, otherwise, a is arranged in front of b.
In specific implementation, the user request analysis module can receive the user request and analyze the user request to obtain the N layers of drill-down lists.
3. According to the drill-down list information, starting from the first layer, drilling down layer by layer until the Nth layer is finished, and sequentially acquiring and analyzing the drill-down data of each layer at the current period:
3.1, according to the first layer information of the drill-down, acquiring dimension fields of the drill-down of the first layer for grouping, and taking a time interval of the current period (namely the current period time in a time period requested by a user, if the user requests to inquire the cycle ratio data, the current period time is the current period, and the previous period time is the previous period) as a filtering condition, and sequencing the dimension value data of the specified number by sequencing dimension fields of the drill-down of the first layer;
3.2, filtering the data to obtain a unique value in the first-layer drilling dimension field, wherein the filtering is repeated data in the data, so that the operation amount in data drilling is reduced;
and 3.3, according to the information of the first and second layers of the drill-down, obtaining dimension fields of the first and second layer drill-down for grouping, taking the time interval of the current period and the unique value of the first layer dimension field calculated in the step 3.2 as a filtering condition, sequencing the sorted dimension fields of the second layer drill-down, and distributing the dimension value data with the appointed number to the second layer of each first layer dimension field value.
3.4, carrying out filtering repetition on the data in the step 3.3 to obtain a unique value of the second-layer drilling dimension field;
3.5 according to the information of the first layer, the second layer and the third layer of the drill-down, obtaining the dimension fields of the first layer, the second layer and the third layer of the drill-down, grouping, taking the time interval of the current period and the unique values of the dimension fields of the first layer and the second layer calculated in the steps 3.2 and 3.4 as filtering conditions, sequencing by using the sequencing dimension fields of the third layer of the drill-down, and distributing the dimension value data of the appointed number to the third layer of each dimension field value of the second layer;
3.6, carrying out filtering and repeating on the data in the step 3.5 to obtain a unique value of a third-layer drilling dimension field;
3.7, by analogy, according to the information of the first to the Nth layers of the drill-down, obtaining the dimension fields of the first to the Nth layers of the drill-down to group, taking the time interval of the current period and the unique values of the dimension fields of the first layer to the N-1 th layer calculated in the 3.2 and 3.4.. as the filtering condition, sequencing the dimension fields of the drill-down of the Nth layer, and distributing dimension value data of the appointed number of the dimension field values to the Nth layer of each dimension field value of the N-1 th layer;
and 3.8, filtering and repeating the data in the previous step to obtain the unique value of the N-th layer down-drilling dimension field.
4. Because the embodiment requires the ring ratio data, the time interval of the previous period needs to be calculated according to the time interval of the current period, and the previous-period drilling-down data of each layer is sequentially acquired and analyzed from the first layer to the nth layer:
4.1, according to the first layer information of the drill-down, obtaining the dimension value data of the appointed number which is obtained by grouping the dimension fields of the drill-down of the first layer, and sorting the dimension fields of the drill-down of the first layer by taking the time interval of the previous period and the unique value of the dimension field of the first layer calculated in the step 3.2 as the filtering condition;
4.2 according to the information of the first and second layers of the drill-down, obtaining the dimension fields of the first and second layer drill-down for grouping, taking the time interval of the upper period and the unique values of the dimension fields of the first and second layers calculated in 3.2 and 3.4 as the filtering condition, sorting the dimension fields of the second layer drill-down by using the sorting dimension fields of the second layer drill-down, and distributing the dimension value data of the appointed number to the second layer of each dimension field value of the first layer;
4.3 according to the information of the first, second and third layers of the drill-down, obtaining the dimension fields of the first, second and third layers of the drill-down to be grouped, taking the time interval of the previous period and the unique values of the dimension fields of the first, second and third layers calculated in 3.2, 3.4 and 3.6 as the filtering condition, sorting the dimension fields of the drill-down of the third layer, and distributing the dimension value data of the appointed number to the third layer of each dimension field value of the second layer;
and 4.4, by analogy, according to the information of the first to the Nth layers of the drill-down, obtaining that the dimension fields of the first to the Nth layers of the drill-down are grouped, taking the time interval in the previous period and the unique values of the dimension fields of the first to the Nth layers calculated in the 3.2 and the 3.4.
5. Merging the current data and the previous data, and generating ring ratio data:
5.1 from the first layer to the Nth layer, merging the current data and the previous data of each layer in sequence:
5.2, combining the upper-stage data of the first layer into the current-stage data of the first layer by taking the dimension field drilled in the first layer as a connecting key, and calculating the difference value and the ring ratio value of each index dimension field; the join key is a name of a dimension field used for joining, and is required to be present in the present period data and the previous period data, and the present period data and the previous period data are related to each other, and the ring ratio data is obtained.
5.3 the up period data of the second layer is merged into the current period data of the second layer by taking the dimension fields drilled down in the first and second layers as a connecting key, and the difference value and the ring ratio value of each index dimension field are calculated;
and 5.4, by analogy, combining the last-stage data of the Nth layer into the current-stage data of the Nth layer by taking the dimension fields drilled from the first layer to the Nth layer as connecting keys, and calculating the difference value and the ring ratio value of each index dimension field.
6. Formatting data, and generating format data meeting the direct display requirement of a front end:
6.1 grouping the second layer ring ratio data by using the dimension field of the first layer drill-down to generate second layer grouping ring ratio data;
6.2 grouping the third layer ring ratio data by using the dimension fields drilled by the first layer and the second layer to generate third layer grouping ring ratio data;
6.3, by analogy, grouping the N-th layer ring ratio data by using the dimension fields drilled from the first layer to the N-1 layer to generate N-th layer grouping ring ratio data;
and 6.4, starting from the first layer of the data with the drill-down dimension value as the key field, finding corresponding data from the next layer of the packet ring ratio data in a recursive mode, and splicing the data to form tree-structured data with the upper layer of the key field as a father and the lower layer of the key field corresponding data as a son, and feeding back the tree-structured data to the user.
7. And returning the formatted data to the front-end for presentation. In particular, the formatted data may be in JSON format.
In specific implementation, the data drilling module can perform the operations, group, filter and sort the data of each layer of the drill-down list layer by layer, drill the data, and then return the data meeting the requirements of the user to the user.
The following examples are given for the detailed description:
the first layer drill-down dimension field app _ id (application id), the number of dimension values is 3, if three apps are: app1, app2 and app3, wherein the dimension field drilled down in the first layer is id of the three apps, the dimension field is sorted into request, and the dimension fields are sorted in descending order;
a dimension field ad _ id (ad slot id) is drilled at the second layer, the number of dimension values is 5, the dimension field is sorted into imp (display), and the imp is sorted in a descending order, for example, the id of the ad slot of the top five of the three apps is obtained;
a third layer of dimension field network _ id (advertiser id), the number of dimension values is 5, dimension field (click) clicking is sorted, and the dimension fields are sorted in an ascending order;
the calculation formulas of the three index fields request, imp and click are sum, i.e. summation. Thus, the dimensions of each app in the database are summed and then sorted in descending or ascending order as desired.
Step 2, from the first layer to the end of the third layer, sequentially acquiring and analyzing the drilling data of each layer at the current period:
step 2.1, acquiring maximum 3 app _ id value data which are grouped by app _ id according to the first layer information of drilling, and are arranged in descending order by a sequencing dimension field request by taking 8 months and 1 day in 2021 as a filtering condition;
step 2.2, filtering the data in the step 2.1 to obtain unique values of app _ id, namely "app1", "app2" and "app 3";
step 2.3, according to the first and second layer information of drill-down, acquiring app _ id and ad _ id to be grouped, taking unique values of "app1", "app2" and "app3" of 1/8/2021 and app _ id as filtering conditions, arranging in an imp descending order, and distributing at most 5 ad _ id value data to each app _ id, wherein when the method is specifically implemented, a specific algorithm is as follows:
SQL
SELECT app_id,
ad_id,
sum(request) AS request,
sum(imp) AS imp,
sum(click) AS click
FROM report_data
WHERE date_hour >= '2021-08-01 10:00:00'
AND date_hour <= '2021-08-01 23:00:00'
AND app_id in [' app1', ' app2', ' app3']
GROUP BY app_id, ad_id
ORDER BY imp desc
LIMIT 5 BY app_id；
step 2.4, carrying out filtering on the data in the step 2.3 to obtain unique values of ad _ id, namely "ad1", "ad2", "ad3", "ad4", "ad5", "ad6", "ad7", "ad8", "ad9", "ad10", "ad11", "ad12", "ad13", "ad14", "ad 15";
step 2.5, according to the first, second and third layers of information of the drill-down, obtaining unique values "app1", "app2", "app3" and "ad1", "ad2", "ad3", "ad4", "ad5", "ad6", "ad7", "ad8", "ad9", "ad10", "ad11", "ad12", "ad13", "ad14" of ad _ id grouped by app _ id, unique values "ad1", "app _ id", "ad2", "ad12", "ad13", "ad14", "ad15" as filtering conditions, sorting and arranging in ascending order of sorting fields click, and distributing at most 5 values of data of network _ id values under each ad _ id, wherein the specific algorithm is as follows:
SQL
SELECT app_id,
ad_id,
network_id,
sum(request) AS request,
sum(imp) AS imp,
sum(click) AS click
FROM report_data
WHERE date_hour >= '2021-08-01 10:00:00'
AND date_hour <= '2021-08-01 23:00:00'
AND app_id in ['app1', 'app2', 'app3']
AND ad_id in ["ad1"、"ad2"、"ad3"、"ad4"、"ad5"、"ad6"、"ad7"、"ad8"、"ad9"、"ad10"、"ad11"、"ad12"、"ad13"、"ad14"、"ad15"]
GROUP BY app_id, ad_id,network_id
ORDER BY click asc
LIMIT 5 BY app_id, ad_id
step 2.6, carrying out filtering repetition on the data in the step 2.5 to obtain unique values 48,18,37 and 53 of the network _ id;
step 3, if the user requests to inquire the ring ratio data, calculating the time of the previous layer 2021, 7 months and 31 days according to the time of the current layer 2021, 8 months and 1 days, starting from the first layer to the end of the third layer, and sequentially acquiring and analyzing the previous drill-down data of each layer;
step 3.1, acquiring app _ id grouping according to the first layer information of drill-down, and taking unique values 'app 1', 'app 2' and 'app 3' of the app _ id and 31 days 7 and 7 in 2021 as filtering conditions, and sequencing at most 3 app _ id value data in descending order of field requests;
step 3.2, according to the first and second layers of information drilled down, acquiring the data grouped by app _ id, ad _ id, and unique values "app1", "app2", "app3" and "ad1", "ad2", "ad3", "ad4", "ad5", "ad6", "ad7", "ad8", "ad9", "ad10", "ad11", "ad12", "ad13", "ad14", and "ad15" as filtering conditions, and distributing 5 ad _ id data under each app _ id in descending order of sorting field imp; the specific algorithm refers to the above algorithm;
step 3.3, according to the first, second and third layers of information drilled down, acquiring unique values "app1", "app2", "app3" and "ad1", "ad2", "ad3", "ad4", "ad5", "ad6", "ad7", "ad8", "ad9", "ad10", "ad11", "ad12", "ad13", "ad14", "ad15" and unique values 48,18,37 and 53 of the network _ id as filtering conditions, sorting and arranging in ascending order of sorting fields click, and distributing maximum 5 word _ id value data under each ad _ id;
step 4, merging the current date and the previous date to generate ring ratio data, and merging the current date and the previous date of each layer in sequence from the first layer to the third layer;
step 4.1, merging the previous-period data of the first layer into the current-period data of the first layer by taking app _ id as a connecting key, and calculating the difference value and the ring ratio value of each index dimension field request, imp and click;
step 4.2, merging the previous data of the second layer into the current data of the second layer by taking app _ id and ad _ id as connecting keys, and calculating the difference and the ring ratio of each index dimension field request, imp and click;
step 4.3, merging the last-period data of the third layer into the current-period data of the third layer by taking app _ id, ad _ id and network _ id as connection keys, and calculating the difference value and the ring ratio value of each index dimension field request, imp and click;
step 5, formatting the data to generate format data meeting the direct display requirement of the front end;
step 5.1, grouping the second layer ring ratio data by app _ id to generate second layer grouping ring ratio data;
step 5.2, grouping the third layer ring ratio data by app _ id and ad _ id to generate third layer grouping ring ratio data;
step 5.3, in a recursive manner, starting with the first layer drill-down dimension app _ id values "app1", "app2", "app3" as key fields:
step 5.3.1, finding ad _ id data "ad1", "ad2", "ad3", "ad4", "ad5" grouped in "app1" from the second layer grouping ring ratio data with "app1" as a key field;
step 5.3.1.1, finding network _ id data 48, 37 grouped as "app1" and "ad1" from the third layer grouping ring ratio data with "app1" and "ad1" as key fields;
step 5.3.1.2, finding network _ id data 18,37 grouped in "app1" and "ad2" from the third layer grouped ring ratio data with "app1" and "ad2" as key fields;
step 5.3.1.3, finding network _ id data 18,37 grouped as "app1" and "ad3" from the third layer grouping ring ratio data with "app1" and "ad3" as key fields;
step 5.3.1.4, finding network _ id data 53, 37 grouped as "app1" and "ad4" from the third layer grouping ring ratio data with "app1" and "ad4" as key fields;
step 5.3.1.5, finding network _ id data 18,37 grouped in "app1" and "ad5" from the third layer grouped ring ratio data with "app1" and "ad5" as key fields;
step 5.3.2, finding the ad _ id data "ad6", "ad7", "ad8", "ad9", "ad10" grouped in "app2" from the second layer grouping ring ratio data with "app2" as the key field;
step 5.3.2.1, finding network _ id data 48, 37 grouped as "app2" and "ad6" from the third layer grouped ring ratio data with "app2" and "ad6" as key fields;
step 5.3.2.2: and so on;
step 5.4, the finally generated data format is a JSON (JavaScript Object Notation) format, and the specific structure is as follows:
[
{
"id": "app1",
"name": "app1",
"request": 3000,
"request_pre": 1000,
"request_detla": 2000,
"request_change": 2,
"imp": 3000,
"imp_pre": 1000,
"imp_detla": 2000,
"imp_change": 2,
"click": 3000,
"click_pre": 1000,
"click_detla": 2000,
"click_change": 2,
"children": [
{
"id": "ad1",
"name": "ad1",
"request": 1000,
"request_pre": 500,
"request_detla": 500,
"request_change": 1,
"imp": 2000,
"imp_pre": 1000,
"imp_detla": 1000,
"imp_change": 1,
"click": 100,
"click_pre": 100,
"click_detla": 0,
"click_change": 0,
"children": [
{
"id": "48",
"name": "48",
"request": 500,
"request_pre": 100,
"request_detla": 400,
"request_change": 4,
"imp": 1000,
"imp_pre": 500,
"imp_detla": 500,
"imp_change": 1,
"click": 100,
"click_pre": 100,
"click_detla": 0,
"click_change": 0,
"children": []
}
]
}
]
}
]
and 6, returning the formatted data to the front end for display.
By adopting the method, the data in the database are filtered, grouped and sequenced, and the grouping number is limited, so that the query times can be greatly reduced, the data acquisition times from the database are reduced, and the drilling efficiency is improved; in addition, the data of each layer is only acquired once when drilling is carried out, multiple acquisition is not needed, and the drilling efficiency is improved; and the obtained data can be directly returned to the front end for use after being formatted, other processing is not needed, and the data query efficiency is improved.
Claims (8)
1. A method of data drilling, comprising: the method comprises the following steps:
s1, receiving and analyzing a user request, and generating a drill-down list with N layers, wherein N is more than or equal to 1 and is a positive integer;
s2, acquiring and analyzing the drill-down data of each layer by layer from the first layer according to the drill-down list until the Nth layer is finished; the step S2 includes:
s21, acquiring a specified number of dimension value data which are grouped by the dimension fields of the first layer drill down according to the first layer list information of the drill down, and are sorted by the index dimension fields of the first layer drill down by taking the time period requested by a user as a filtering condition;
s22, filtering the data to obtain a unique value in the first-layer drilling dimension field;
s23, according to the information of the first and second layers of drilling, obtaining the dimension fields of the first and second layers of drilling to be grouped, taking the time period requested by the user and the unique value of the dimension field of the first layer in the step S22 as the filtering condition, sequencing the dimension fields of the indexes of the second layer of drilling, and distributing the dimension value data with the appointed number to the second layer of each dimension field value of the first layer;
s24, carrying out filtering repetition on the data in the step S23 to obtain a unique value of the second-layer drilling dimension field;
s25, according to the list information of the first layer, the second layer and the third layer of the drill-down, obtaining that the drill-down dimension fields of the first layer, the second layer and the third layer are grouped, sorting the drill-down index dimension fields of the third layer by taking the time period requested by a user, the unique value of the first layer dimension field in the step S22 and the unique value of the second layer drill-down dimension field in the step S24 as filtering conditions, and distributing the specified number of dimension value data to the third layer of each second layer dimension field value;
s26, carrying out filtering and repeating on the data in the step S25 to obtain a unique value of a third-layer drilling dimension field;
s27, repeating the steps until the whole drill-down list is traversed;
and S3, processing the acquired data and returning the data to the user.
2. The data drilling method of claim 1, wherein: the step S3 includes:
s31, grouping the data of the second layer by the drill-down dimension field of the first layer to generate the grouped data of the second layer;
s32, grouping the third layer data by the dimension fields drilled by the first layer and the second layer to generate third layer grouped data;
s33, by analogy, grouping the data of the Nth layer by the dimension fields drilled from the first layer to the Nth-1 layer to generate the data of the Nth layer;
and S34, starting from the first layer of the drill-down dimension value as the key field, finding corresponding data from the next layer of grouped data in a recursive mode, and splicing the data to form tree-structured data with the upper layer of dimension fields as parents and the lower layer of dimension fields corresponding to data as children, and feeding back the tree-structured data to the user.
3. A method of data drilling according to claim 1 or 2, characterized by: the drill-to-collar ratio data includes: acquiring the current date data and the previous date data, and then performing grouping, sorting and filtering operations on the current date data and the previous date data respectively, wherein the N-layer data for acquiring the previous date data needs to use the time period requested by the user and the unique value of the drill-down dimension field of the first N-1 layers of the current date data as the filtering condition in step S21.
4. A method of data drilling according to claim 3, characterized by: the step of drilling the ring ratio data specifically comprises the following steps:
s51, merging the previous-stage data of the first layer into the current-stage data of the first layer by taking the dimension field of the first layer as a connecting key, and calculating the difference value and the ring ratio of each index dimension field;
s52, merging the previous-period data of the second layer into the current-period data of the second layer by taking the dimension fields of the first layer and the second layer as a connecting key, and calculating the difference and the ring ratio of each index dimension field;
and S53, by analogy, calculating the difference value and the ring ratio value of the index dimension field of each layer in the drill-down list, wherein the ring ratio value of the last layer is the required ring ratio data.
5. A method of data drilling according to claim 1 or 2, characterized by: the data format in step S3 is a JSON format.
6. A data drilling system, characterized by: the method comprises the following steps:
the user request analysis module: the system is used for analyzing the user request and generating a drill-down list;
a data drilling module: the data drilling module is used for drilling data layer by layer according to a drill-down list generated by the user request analysis module and returning the acquired data to a user, and the data drilling module firstly acquires a specified number of dimensional value data which are grouped by using the dimensional fields of the drill-down of the first layer and sorted by using the index dimensional fields of the drill-down of the first layer according to the first layer information of the drill-down in the process of drilling the data and by using the time period requested by the user as a filtering condition; then, filtering the data to obtain a unique value in a first-layer drilling dimension field; according to the information of the first layer and the second layer of the drill-down, obtaining dimension fields of the drill-down of the first layer and the second layer to be grouped, taking the time period requested by a user and the unique value of the dimension field of the first layer as filtering conditions, sequencing the dimension fields of the index of the drill-down of the second layer, and distributing the specified number of dimension value data to the second layer of each dimension field value of the first layer; then, carrying out filtering and repeating on the data to obtain a unique value of a second-layer drilling dimension field; then according to the list information of the first layer, the second layer and the third layer of the drill-down, obtaining the drill-down dimension fields of the first layer, the second layer and the third layer to be grouped, taking the time period requested by a user, the unique value of the dimension field of the first layer of the drill-down and the unique value of the dimension field of the second layer of the drill-down as filtering conditions, sequencing the index dimension field of the drill-down of the third layer, and distributing the specified number of dimension value data to the third layer of each dimension field value of the second layer; filtering the dimension value data to obtain a unique value of a third layer of drilling dimension field; and so on until the entire drill-down list is traversed.
7. The data drilling system of claim 6, wherein: the data drilling module acquires grouped data from each layer of data, and specifically comprises the following steps: grouping the data of the second layer by the dimension field of the first layer drill-down to generate the grouped data of the second layer; grouping the third layer data by using the dimension fields drilled by the first layer and the second layer to generate third layer grouped data; by analogy, grouping the data of the Nth layer by using the dimension fields drilled from the first layer to the Nth-1 layer to generate the data of the Nth layer; and then, starting from the first layer of the data with the dimension value as a key field, finding corresponding data from the next layer of grouped data in a recursive mode, and splicing the data to form tree-shaped structure data with the dimension field of the upper layer as a parent and the dimension field of the lower layer as a child, and feeding back the tree-shaped structure data to the user.
8. The data drilling system of claim 6 or 7, wherein: and the data drilling module acquires the current date data and the previous date data, generates ring ratio data according to the current date data and the previous date data of each layer, and feeds the ring ratio data back to the user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111310616.5A CN113779044B (en) | 2021-11-08 | 2021-11-08 | Data drilling method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111310616.5A CN113779044B (en) | 2021-11-08 | 2021-11-08 | Data drilling method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113779044A CN113779044A (en) | 2021-12-10 |
| CN113779044B true CN113779044B (en) | 2022-02-11 |
Family
ID=78956622
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111310616.5A Active CN113779044B (en) | 2021-11-08 | 2021-11-08 | Data drilling method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113779044B (en) |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7139766B2 (en) * | 2001-12-17 | 2006-11-21 | Business Objects, S.A. | Universal drill-down system for coordinated presentation of items in different databases |
| CN101446964A (en) * | 2008-12-31 | 2009-06-03 | 中国建设银行股份有限公司 | Method of data mining and computer device |
| CN102135994A (en) * | 2011-03-17 | 2011-07-27 | 新太科技股份有限公司 | Online analytical processing (OLAP)-based intelligent analysis method |
| CN102521414B (en) * | 2011-12-28 | 2016-05-18 | 畅捷通信息技术股份有限公司 | Data mining device and data mining method |
| CN108446305B (en) * | 2018-01-30 | 2020-12-01 | 上海天旦网络科技发展有限公司 | System and method for multi-dimensional statistics of business data |
| CN108399251A (en) * | 2018-03-05 | 2018-08-14 | 贵州工程应用技术学院 | A kind of Intelligence Generator and data exhibiting process that multi-layer data shows |
| CN110069519A (en) * | 2018-08-23 | 2019-07-30 | 平安科技（深圳）有限公司 | Data information management method, apparatus, computer equipment and storage medium |
| CN110175184B (en) * | 2019-04-30 | 2023-01-20 | 创新先进技术有限公司 | Data dimension drilling-down method and system and electronic equipment |
| CN110263065A (en) * | 2019-05-22 | 2019-09-20 | 阿里巴巴集团控股有限公司 | Data mining method and apparatus |
| CN112039989B (en) * | 2020-08-31 | 2023-02-21 | 广州品唯软件有限公司 | Data drilling method, proxy server, service calling system and medium |
| CN113268491A (en) * | 2021-03-19 | 2021-08-17 | 广州天越电子科技有限公司 | Universal method for realizing data up-rolling and down-drilling |
-
2021
- 2021-11-08 CN CN202111310616.5A patent/CN113779044B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113779044A (en) | 2021-12-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11681694B2 (en) | Systems and methods for grouping and enriching data items accessed from one or more databases for presentation in a user interface | |
| Stolte et al. | Polaris: A system for query, analysis, and visualization of multidimensional relational databases | |
| US7007020B1 (en) | Distributed OLAP-based association rule generation method and system | |
| CN102968436B (en) | Chart is recommended | |
| US9075859B2 (en) | Parameterized database drill-through | |
| JP2005525642A (en) | Expert database forwarded back to link weighted association rules | |
| US20040015481A1 (en) | Patent data mining | |
| CN105740333A (en) | Visual advertisement management platform, and implementation method thereof | |
| CN107220365A (en) | Accurate commending system and method based on collaborative filtering and correlation rule parallel processing | |
| WO2013036688A2 (en) | Identifying product variants | |
| CN108241677A (en) | A kind of method and system for the tax revenue sorting code number for obtaining commodity | |
| Özdağoğlu et al. | Topic modelling-based decision framework for analysing digital voice of the customer | |
| Tan et al. | Time series clustering: A superior alternative for market basket analysis | |
| CN110297853A (en) | Frequent Set method for digging and device | |
| EP2939196A1 (en) | Techniques for measuring video profit | |
| CN106326318A (en) | Search method and device | |
| CN113487117B (en) | Method and system for simulating behavior data of electric business based on multi-dimensional user portrait | |
| CN113779044B (en) | Data drilling method and system | |
| WO2009006028A2 (en) | Explaining changes in measures thru data mining | |
| CN114429384B (en) | Intelligent product recommendation method and system based on e-commerce platform | |
| CN109697203A (en) | Index unusual fluctuation analysis method and equipment, computer storage medium, computer equipment | |
| CN110765100B (en) | Label generation method and device, computer readable storage medium and server | |
| Ying et al. | Research on E-commerce Data Mining and Managing Model in The Process of Farmers' Welfare Growth | |
| Diop et al. | TTProfiler: Computing types and terms profiles of assertional knowledge graphs | |
| Rahman | Development of self-service business intelligence for decision support system using microsoft business intelligence tools |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |