Note: Descriptions are shown in the official language in which they were submitted.
CA 03168300 2022-07-18
METHOD FOR SELECTING MEMBER DATA FOR E-COMMERCE PLATFORMS
AND SYSTEM THEREOF
BACKGROUND OF THE INVENTION
Technical Field
[0001] The present invention relates to the technical field of data
processing, and more
particularly to a method and a system for selecting member data for e-commerce
platforms.
Description of Related Art
[0002] To e-commerce platforms, findings from analysis of member consumption
data are
important references for designing sales promotion. During analysis of member
consumption data, multi-latitudinal selection of data forms a key step for
acquiring
member consumption data. In the prior art, data selection is usually achieved
by pre-
aggregation selection (OLAP-Druid) and distributed memory computing selection
(SPARK). Therein, OLAP-Druid selection mainly adopts an HLL algorithm to
compute
and analyze consumption data of platform members. This algorithm tends to lose
precision in business scenarios related to deduplication, making the resulting
member
consumption data selection imprecise. As to SPARK selection, it involves
acquiring
original consumption data of platform members, and placing the original
consumption
data in memory of each distributed time node, in an extracting-detail manner,
for logical
computing, so as to generate the result of member consumption data selection
eventually.
It has been found in practical use that since the amount of original
consumption data of
members is huge, while the latter approach ensures precise selection of member
consumption data, it consumes more computing resources and memory resources,
leading
1
Date Regue/Date Received 2022-07-18
CA 03168300 2022-07-18
to low selection efficiency and poor user experience.
SUMMARY OF THE INVENTION
[0003] The objective of the present invention is to provide a method for
selecting member data
for e-commerce platforms and a system thereof, which reduce consumption of
memory
and computing resources while ensuring precise selection of member consumption
data,
thereby significantly improving selection efficiency of member consumption
data.
[0004] In order to achieve the foregoing objective, in one aspect, the present
invention provides
a method for selecting member data for e-commerce platforms, which comprises:
[0005] synchronizing member consumption data from a data warehouse so as to
create plural
data models;
[0006] based on member codes in the data models, generating plural integer
type identifiers that
are different from each other, and storing mapping relationship between the
member
codes and the integer type identifiers into a dictionary table;
[0007] associating the integer type identifiers with plural latitudinal
consumption fields of the
member consumption data in a one-to-one manner, so as to generate a bitmap
table; and
[0008] according to a query instruction from a user, performing bitwise
operation on the plural
latitudinal consumption fields in the bitmap table using the integer type
identifiers, and
outputting a selection result.
[0009] Preferably, after the step of associating the integer type identifiers
with plural latitudinal
consumption fields of the member consumption data in a one-to-one manner, so
as to
generate bitmap table, the method further comprises:
[0010] regularly updating field data in the bitmap table by means of
interpolation, so as to
generate a bitmap table corresponding to the present time node.
[0011] Specifically, the step of regularly updating field data in the bitmap
table by means of
interpolation, so as to generate a bitmap table corresponding to the present
time node
comprises:
[0012] based on the present time node, acquiring new member consumption data
from the data
2
Date Regue/Date Received 2022-07-18
CA 03168300 2022-07-18
warehouse, and synchronizing the data to the data models;
[0013] according to the mapping relationship of the member codes in the
dictionary table,
regularly interpolating the field data corresponding to the new member
consumption data
into the bitmap table, thereby updating the bitmap table by means of
interpolation.
[0014] More preferably, after the step of regularly updating field data in the
bitmap table by
means of interpolation, so as to generate a bitmap table corresponding to the
present time
node, the method further comprises:
[0015] cleaning the bitmap table, and eliminating irrelevant latitudinal field
data.
[0016] Further, before the step of according to a query instruction from a
user, performing
bitwise operation on the plural latitudinal consumption fields in the bitmap
table using
the integer type identifiers, outputting selection result, the method further
comprises:
[0017] presetting plural query instructions, performing a bitwise operation on
the cleaned bitmap
table in advance so as to obtain pre-selection results matching the plural
query
instructions; and
[0018] storing the pre-selection results into a temporary result list for user
query.
[0019] Preferably, the step of according to a query instruction from a user,
performing bitwise
operation on the plural latitudinal consumption fields in the bitmap table
using the integer
type identifiers, outputting selection result comprises:
[0020] receiving a query instruction from the user, and determining whether
the query instruction
is a said preset query instruction or not; and
[0021] if said determining has a positive result, directly outputting the
matching pre-selection
results from the temporary result list, and if said determining has a negative
result, based
on the bitmap table updated by interpolation, performing a logical operation
on plural
said latitudinal consumption fields through integer type identifiers and then
outputting a
selection result.
[0022] As compared to the prior art, the method for selecting member data for
e-commerce
platforms of the present invention provides the following beneficial effects.
[0023] The method for selecting member data for e-commerce platforms of the
present invention
first acquires the member consumption data from the data warehouse to create
the data
3
Date Regue/Date Received 2022-07-18
CA 03168300 2022-07-18
models. Therein, the data model comprises member codes, plural latitudinal
consumption
fields, and consumption dates. The method converts the member codes one by one
into
integer type identifiers, and stores the mapping relationship between the
integer type
identifiers and the member codes into the dictionary table. Then the integer
type
identifiers, the consumption fields, and the consumption date are used to
construct the
bitmap table. After the query instruction of the user is acquired, the integer
type identifiers
are called for logical bitwise operation on the plural latitudinal consumption
fields in the
bitmap table, thereby obtaining the selection result eventually.
[0024] It is thus clear that in the method for selecting member data for e-
commerce platforms of
the present invention, the integer type identifiers are used instead of member
codes and
the member consumption data are represented through a bitmap table, so that
selection of
member data can be easily accomplished by means of bitwise operation of sets
in the
bitmap table, thereby reducing use of computing and storing resources while
significantly
enhancing computing efficiency. Therefore, the disclosed method is
particularly suitable
for selection of massive member data.
[0025] In another aspect, the present invention provides a system for
selecting member data for
e-commerce platforms to be used in the method for selecting member data for e-
commerce platforms of the technical scheme described above. The system
comprises:
[0026] a data model creating unit, for synchronizing member consumption data
from a data
warehouse so as to create plural data models;
[0027] a dictionary table creating unit, for generating plural integer type
identifiers that are
different from each other based on member codes in the data models, and
storing mapping
relationship between the member codes and the integer type identifiers into a
dictionary
table;
[0028] a bitmap table generating unit, for associating the integers type
identifier with plural
latitudinal consumption fields of the member consumption data in a one-to-one
manner,
so as to generate a bitmap table; and
[0029] a query outputting unit, for according to a query instruction from a
user, performing
bitwise operation on the plural latitudinal consumption fields in the bitmap
table using
4
Date Regue/Date Received 2022-07-18
CA 03168300 2022-07-18
the integer type identifiers, and outputting a selection result.
[0030] Preferably, it further comprises a bitmap table updating unit connected
to the bitmap table
generating unit;
[0031] in which the bitmap table updating unit is for regularly updating field
data in the bitmap
table by means of interpolation, so as to generate a bitmap table
corresponding to the
present time node.
[0032] More preferably, the bitmap table updating unit comprises:
[0033] a data acquiring module, for acquiring new member consumption data from
the data
warehouse based on the present time node, and synchronizing the data to the
data models;
and
[0034] a bitmap table updating module, for according to the mapping
relationship of the member
codes into the dictionary table, regularly interpolating the field data
corresponding to the
new member consumption data into the bitmap table, thereby updating the bitmap
table
by means of interpolation.
[0035] More preferably, further comprises a data cleaning unit arranged
between the bitmap table
generating unit and the query outputting unit;
[0036] in which the data cleaning unit is for cleaning the bitmap table, and
eliminating irrelevant
latitudinal field data.
[0037] Preferably, it further comprises a pre-selection unit and a storing
unit, wherein the pre-
selection unit has an input end connected to an output end of the data
cleaning unit, and
the storing unit has an output end connected to an input end of the query
outputting unit;
[0038] the pre-selection unit, for presetting plural query instructions,
performing a bitwise
operation on the cleaned bitmap table in advance so as to obtain pre-selection
results
matching the plural query instructions; and
[0039] the storing unit, for storing the plural pre-selection results into a
temporary result list for
user query.
[0040] Preferably, the query outputting unit comprises:
[0041] a determining module, for receiving a query instruction from the user,
and determining
whether the query instruction is a said preset query instruction or not; and
Date Regue/Date Received 2022-07-18
CA 03168300 2022-07-18
[0042] an outputting module, for, when said determining has a positive result,
directly finding
and outputting a matching said pre-selection result from the temporary result
list, and if
said determining has a negative result, based on the bitmap table updated by
interpolation,
performing a logical operation on plural said latitudinal consumption fields
through
integer type identifiers and then outputting a selection result.
[0043] As compared to the prior art, the disclosed system for selecting member
data for e-
commerce platforms provides beneficial effects that are similar to those
provided by the
disclosed method for selecting member data for e-commerce platforms as
enumerated
above, and thus no repetitions are made herein.
BRIEF DESCRIPTION OF THE DRAWINGS
[0044] The accompanying drawings are provided herein for better understanding
of the present
invention and form a part of this disclosure. The illustrative embodiments and
their
descriptions are for explaining the present invention and by no means form any
improper
limitation to the present invention, wherein:
[0045] FIG. 1 is a flowchart of a method for selecting member data for e-
commerce platforms
according to Embodiment 1 of the present invention;
[0046] FIG. 2 shows exemplary selection results obtained using the method of
Embodiment 1 of
the present invention;
[0047] FIG. 3 is an example of a bitmap table A according to Embodiment 1 of
the present
invention;
[0048] FIG. 4 is an example of a bitmap table B according to Embodiment 1 of
the present
invention; and
[0049] FIG. 5 is a structural diagram of a system for selecting member data
for e-commerce
platforms according to Embodiment 2 of the present invention.
6
Date Regue/Date Received 2022-07-18
CA 03168300 2022-07-18
[0050] Reference numerals:
[0051] 1 ¨ data model creating unit 2¨ dictionary table creating
unit
[0052] 3 ¨ bitmap table generating unit 4¨ query outputting unit
[0053] 5 ¨ bitmap table updating module 6 ¨ data cleaning unit
[0054] 7¨ pre-selection unit 8 ¨ storing unit
[0055] 41 ¨ determining module 42¨ outputting module
[0056] 51 ¨ data acquiring module 52¨ bitmap table updating module
DETAILED DESCRIPTION OF THE INVENTION
[0057] To make the foregoing objectives, features, and advantages of the
present invention
clearer and more understandable, the following description will be directed to
some
embodiments as depicted in the accompanying drawings to detail the technical
schemes
disclosed in these embodiments. It is, however, to be understood that the
embodiments
referred herein are only a part of all possible embodiments and thus not
exhaustive. Based
on the embodiments of the present invention, all the other embodiments can be
conceived
without creative labor by people of ordinary skill in the art, and all these
and other
embodiments shall be encompassed in the scope of the present invention.
[0058] Embodiment 1
[0059] Referring to FIG. 1, the present embodiment provides a method for
selecting member
data for e-commerce platfouns, which comprises:
[0060] synchronizing member consumption data from a data warehouse so as to
create plural
data models; based on member codes in the data models, generating plural
integer type
identifiers that are different from each other, and storing mapping
relationship between
the member codes and the integer type identifiers into a dictionary table;
associating the
integers type identifier with plural latitudinal consumption fields of the
member
consumption data in a one-to-one manner, so as to generate a bitmap table; and
according
to a query instruction from a user, performing bitwise operation on the plural
latitudinal
consumption fields in the bitmap table using the integer type identifiers, and
outputting a
selection result.
[0061] The method for selecting member data for e-commerce platforms of the
present
7
Date Regue/Date Received 2022-07-18
CA 03168300 2022-07-18
embodiment first acquires the member consumption data from the data warehouse
to
create the data models. Therein, the data model comprises member codes, plural
latitudinal consumption fields, and consumption dates. The method converts the
member
codes one by one into integer type identifiers, and stores the mapping
relationship
between the integer type identifiers and the member codes into the dictionary
table. Then
the integer type identifiers, the consumption fields, and the consumption
dates are used
to construct the bitmap table. After the query instruction of the user is
acquired, the
integer type identifiers are called for logical bitwise operation on the
plural latitudinal
consumption fields in the bitmap table, thereby obtaining the selection result
eventually.
[0062] It is thus clear that in the method for selecting member data for e-
commerce platforms of
the present embodiment, the integer type identifiers are used instead of
member codes
and the member consumption data are represented through a bitmap table, so
that
selection of member data can be easily accomplished by means of bitwise
operation of
sets in the bitmap table, thereby reducing use of computing and storing
resources while
significantly enhancing computing efficiency. Therefore, the disclosed method
is
particularly suitable for selection of massive member data.
[0063] Still referring to FIG. 1, in view that the member consumption data
update every day, for
preventing hysteresis of the bitmap table data, in the present embodiment,
after the step
of associating the integers type identifier with plural latitudinal
consumption fields of the
member consumption data in a one-to-one manner, so as to generate a bitmap
table, the
method further comprises: regularly interpolating and updating the field data
in the
bitmap table, so as to generate a bitmap table corresponding to the present
time node.
[0064] Specifically, in the embodiment described previously, the step of
regularly interpolating
and updating the field data in the bitmap table, so as to generate a bitmap
table
corresponding to the time present node comprises:
[0065] based on the present time node, acquiring new member consumption data
from the data
warehouse, and synchronizing the data to the data models; according to the
mapping
relationship of the member codes in the dictionary table, regularly
interpolating the field
data corresponding to the new member consumption data into the bitmap table,
thereby
6
Date Regue/Date Received 2022-07-18
CA 03168300 2022-07-18
updating the bitmap table by means of interpolation.
[0066] In practical implementations, the member consumption data are regularly
acquired from
the data warehouse for updating. Then, based on the mapping relationship in
the
dictionary table, a designated complement function is used to perform
complement
operation, so as to achieve interpolation and updating of the field data in
the bitmap table.
An example is explained herein for facilitating understanding of the
interpolation and
updating process. First, the updated member consumption data are acquired from
the data
models and the member codes for which the consumption data have been updated
are
identified and then converted using the dictionary table, so as to get their
matching integer
type identifiers. Then the consumption field data updated on the day (flag=1)
corresponding to the integer type identifiers are acquired and stored into the
bitmap table.
The consumption field data summarized on the present day (flag=1) and the
consumption
field data summarized on the day before the present day (flag=2) are
integrated and used
as the consumption field data currently summarized (flag=2) and inserted into
the bitmap
table, thereby accomplishing interpolation and updating of the bitmap table.
It is to be
noted that flag=1 only represents the consumption field data updated on the
present day
and flag=2 represents all consumption field data existing currently. The
latter comprises
the consumption field data updated on the present day and all consumption
field data of
the present day and all previous days. It is known from the foregoing
implementation that
the present embodiment can continuously update the consumption field data in
the bitmap
table by accumulated integration instead of repeated computing for all
historical data, and
thus effectively reduce computing loads while ensuring a precise selection
result.
[0067] Optionally, referring to FIG. 1, in the embodiment described
previously, after the step of
regularly interpolating and updating the field data in the bitmap table, so as
to generate a
bitmap table corresponding to the present time node, the method further
comprises:
cleaning the bitmap table and eliminating irrelevant latitudinal field data.
[0068] In practical implementations, this step is equivalent to creating a
CUBE model. After the
field data of irrelevant dimensions are eliminated from the bitmap table, the
data
cardinality after grouping can be decreased, which means the query efficiency
can be
7
Date Regue/Date Received 2022-07-18
CA 03168300 2022-07-18
increased and the selection can be accelerated.
[0069] In order to further accelerate the process of selection, in the present
embodiment, before
the step according to a query instruction from a user, performing bitwise
operation on the
plural latitudinal consumption fields in the bitmap table using the integer
type identifiers,
outputting selection result, the method further comprises:
[0070] pre-setting plural query instruction, performing a bitwise operation on
the cleaned bitmap
table in advance so as to obtain pre-selection results matching the plural
query
instructions; and storing the pre-selection results into a temporary result
list for user query.
[0071] In practical implementations, since the amount of the field data in the
bitmap table is huge,
real-time computing can cause delayed output of the selection result. To
prevent this, in
the present embodiment, commonly used query instructions are stored in
advance, so that
the system can perform a bitwise operation on the cleaned bitmap table in
advance
according to these query instructions and get corresponding pre-selection
results. These
pre-selection results are stored into a temporary result list so as to be
directly called and
used in response to a user query.
[0072] Further, in the embodiment described previously, the step of according
to a query
instruction from a user, performing bitwise operation on the plural
latitudinal
consumption fields in the bitmap table using the integer type identifiers,
outputting
selection result comprises:
[0073] receiving a query instruction from a user, and determining whether the
query instruction
is a preset query instruction; if said determining has a positive result,
directly outputting
the matching pre-selection results from the temporary result list, and if said
determining
has a negative result, based on the bitmap table updated by interpolation,
performing a
logical operation on plural said latitudinal consumption fields through
integer type
identifiers and then outputting a selection result. It is thus clear that,
with the setting of
the foregoing two computing modes, a pre-selection result can be directly
called from the
temporary result list if the query instruction made by the user matches one of
the pre-
stored query instructions, thereby reducing waiting time required by
computing. If the
query instruction made by the user does not match any of the pre-stored query
instructions,
8
Date Regue/Date Received 2022-07-18
CA 03168300 2022-07-18
a selection result can be generated through a logical operation based on the
bitmap table
updated by interpolation according to the query instruction, which means that
the
selection result in this case is generated through real-time computing.
Thereby, the
present invention supports an expanded range of user queries and supports user
custom
multi-dimension selection.
[0074] For easy understanding, the present embodiment is now explained with an
example of
selection of consumption data of new and existing members. As shown in FIG. 2,
the
member consumption data of 0826 and 0827 are acquired from the data warehouse
and a
data model is created accordingly. The data model comprises consumption fields
of plural
latitudes, such as member codes, purchase channels, purchase categories,
purchase dates,
etc. By calling the mapping relationship in the dictionary table, the member
codes are
converted into integer type identifiers that facilitate operations of a bitmap
table. For easy
understanding and distinction, buyers are hereinafter represented letters A,
B, C, and D.
After the bitmap table is cleaned, the consumption data of new and existing
members
each day are summarized by purchase channels into a bitmap table A. The
statistical
criteria include purchase channels (online or offline) + flag (flag=1 or
flag=2) + purchase
date (0826 or 0827). The statistical data for each day include two entries.
One is the
bitmap table set of the members of the present day (flag=1), and the other is
the bitmap
table set of the currently existing (flag=2) members. Then the consumption
data of new
and existing members are summarized under the criteria including purchase
category +
purchase channel + flag (flag=1 or flag=2) + purchase date (0826 or 0827) into
bitmap table B. Similarly, the statistical data for each day include two
entries. The first
is the bitmap table set of the members of the present day (flag=1), and the
second is the
bitmap table set of the currently existing (flag=2) members. Afterward,
according to the
query instruction func( ), selection is made from the following three
scenarios:
[0075] Scenario I: the bitmap table A of online new buyers of the purchase
date 0827 is as
shown in FIG. 3. The selection process is essentially about a rb andnot
cardinality
bitwise operation of the bitmap set {A, DI and the bitmap set {A, C}. The
obtained
selection result is {D}, and the summed number of new buyers is 1.
9
Date Regue/Date Received 2022-07-18
CA 03168300 2022-07-18
[0076] Scenario II: the bitmap table B of online A/C (air conditioner) new
buyers of the
purchase date 0827 is as shown in FIG. 4. The process of identifying the
number of the
online A/C new buyers of the purchase date 0827 is essentially about a
rb andnot cardinality bitwise operation of the bitmap set {C, AI and the
bitmap set {C}.
The obtained selection result is {A}, and the summed number of new buyers is
1.
Similarly, the process of identifying the number of the online R&W
(refrigerators and
washing machines) new buyers of the purchase date 0827 is essentially about a
rb andnot cardinality bitwise operation of the bitmap set {D} and the bitmap
empty set.
The obtained selection result is {D}, and the summed number of new buyers is
1.
[0077] Scenario III: the process of identifying how many online air
conditioner new buyers of
the purchase date 0827 are online new buyers is essentially about a rb and
cardinality
operation of the bitmap table A and the bitmap table B. In other words, this
is about a
rb andnot cardinality bitwise operation of the bitmap set {A} and the bitmap
set {A, C}.
The obtained selection result is an empty set, and the summed number of new
buyers is
0. Similarly, the process of identifying how many of the online R&W new buyers
of the
purchase date 0827 are online new buyers is essentially about a rb andnot
cardinality
bitwise operation of the bitmap set {D} and the bitmap set {A, C}. The
obtained selection
result is {D}, and the summed number of new buyers is 1.
[0078] Embodiment 2
[0079] Referring to FIG. 1 and FIG. 5, the present embodiment provides a
system for selecting
member data for e-commerce platforms, which comprises:
[0080] a data model creating unit 1, for synchronizing member consumption data
from a data
warehouse so as to create plural data models;
[0081] a dictionary table creating unit 2, for generating plural integer type
identifiers that are
different from each other based on the member codes in the data model, and
storing
mapping relationship between the member codes and the integer type identifiers
into a
dictionary table;
[0082] a bitmap table generating unit 3, for associating the integers type
identifier with plural
Date Regue/Date Received 2022-07-18
CA 03168300 2022-07-18
latitudinal consumption fields of the member consumption data in a one-to-one
manner,
so as to generate a bitmap table; and
[0083] a query outputting unit 4, for according to a query instruction from a
user, performing
bitwise operation on the plural latitudinal consumption fields in the bitmap
table using
the integer type identifiers, and outputting a selection result.
[0084] Preferably, the system further comprises a bitmap table updating unit 6
connected to the
bitmap table generating unit 3. The bitmap table updating unit 6 is for
regularly
interpolating and updating the field data in the bitmap table, so as to
generate a bitmap
table corresponding to the present time node.
[0085] Preferably, the bitmap table updating unit 6 comprises:
[0086] a data acquiring module 51, for acquiring new member consumption data
from the data
warehouse based on the present time node, and synchronizing the data to the
data models;
and
[0087] a bitmap table updating module 52, for according to the mapping
relationship of the
member codes in the dictionary table, regularly interpolating the field data
corresponding
to the new member consumption data into the bitmap table, thereby updating the
bitmap
table by means of interpolation.
[0088] Preferably, the system further comprises a data cleaning unit 6
arranged between the
bitmap table generating unit 3 and the query outputting unit 4; and
[0089] a data cleaning unit 6, for cleaning the bitmap table, and eliminating
irrelevant latitudinal
field data.
[0090] Preferably, the system further comprises a pre-selection unit 7 and a
storing unit 8. The
pre-selection unit 7 has its input end connected to the output end of the data
cleaning unit
6, and the storing unit 8 has its output end connected to the input end of the
query
outputting unit 4;
[0091] a pre-selection unit 7, for presetting plural query instructions,
performing a bitwise
operation on the cleaned bitmap table in advance so as to obtain pre-selection
results
matching the plural query instructions; and
[0092] a storing unit 8, for storing the pre-selection results into a
temporary result list for user
11
Date Regue/Date Received 2022-07-18
CA 03168300 2022-07-18
query.
[0093] Preferably, the query outputting unit 4 comprises:
[0094] a determining module 41, for receiving a query instruction from a user,
and determining
whether the query instruction is a preset query instruction; and
[0095] an outputting module 42, for, if said determining has a positive
result, directly outputting
the matching pre-selection results from the temporary result list, and if said
determining
has a negative result, based on the bitmap table updated by interpolation,
performing a
logical operation on plural said latitudinal consumption fields through
integer type
identifiers and then outputting a selection result.
[0096] As compared to the prior art, the disclosed system for selecting member
data for e-
commerce platforms provides beneficial effects that are similar to those
provided by the
disclosed method for selecting member data for e-commerce platforms as
enumerated in
Embodiment 1, and thus no repetitions are made herein.
[0097] As will be appreciated by people of ordinary skill in the art,
implementation of all or a
part of the steps of the method of the present invention as described
previously may be
realized by having a program instruct related hardware components. The program
may
be stored in a computer-readable storage medium, and the program is about
performing
the individual steps of the methods described in the foregoing embodiments.
The storage
medium may be a ROM/RAM, a hard drive, an optical disk, a memory card or the
like.
[0098] The present invention has been described with reference to the
preferred embodiments
and it is understood that the embodiments are not intended to limit the scope
of the present
invention. Moreover, as the contents disclosed herein should be readily
understood and
can be implemented by a person skilled in the art, all equivalent changes or
modifications
which do not depart from the concept of the present invention should be
encompassed by
the appended claims. Hence, the scope of the present invention shall only be
defined by
the appended claims.
12
Date Regue/Date Received 2022-07-18
