CN114219540A - Method and device for determining user behavior period, electronic equipment and storage medium - Google Patents
Method and device for determining user behavior period, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN114219540A CN114219540A CN202111631056.3A CN202111631056A CN114219540A CN 114219540 A CN114219540 A CN 114219540A CN 202111631056 A CN202111631056 A CN 202111631056A CN 114219540 A CN114219540 A CN 114219540A
- Authority
- CN
- China
- Prior art keywords
- time
- behavior
- period
- preset
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0207—Discounts or incentives, e.g. coupons or rebates
- G06Q30/0224—Discounts or incentives, e.g. coupons or rebates based on user history
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Technology Law (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a method and a device for determining a user behavior cycle, electronic equipment and a storage medium, and relates to the field of financial technology, wherein the method for determining the user behavior cycle comprises the following steps: the method comprises the steps of obtaining behavior data of a target user in a preset time period and a preset period sequence, inputting the time sequence into a preset period model to obtain a time regular distance value, a time sequence length and a behavior period, calculating confidence of the time sequence about each behavior period based on the time regular distance value, the time sequence length and the behavior period, and determining the target behavior period as the behavior period of the time sequence under the condition that the confidence of the target behavior period is larger than a preset parameter threshold. The invention solves the technical problem that the recommended application function does not meet the actual use requirement of the user because the behavior period of the user cannot be accurately excavated in the related technology.
Description
Technical Field
The invention relates to the field of financial science and technology, in particular to a method and a device for determining a user behavior cycle, electronic equipment and a storage medium.
Background
At present, the application operation function in the user terminal is gradually enriched, and for most financial services, the user can handle the services only through the terminal. In order to meet the personalized demands of users, various functions, such as a "reserved transfer" function, have been introduced by financial institutions, which may provide great convenience to users who have the demand for periodic transfers. However, some functions (for example, functions for performing certain operations regularly) have a big problem in the promotion process, and if the functions are promoted comprehensively, the functions will cause the user who does not have the needs of the functions to feel the objections, and even cause the user to lose. How to excavate customers with periodic operational needs from the user's history is the key to solving this problem.
In the related art, there are more sophisticated algorithms in the aspect of timing prediction, such as: recurrent neural networks RNN, long-short term memory recurrent neural networks LSTM, linear regression, ARMA models, xgboost, Fourier transform, and the like. However, these existing algorithms cannot accurately find out the behavior cycle of the user, so that the recommended application function does not meet the actual use requirement of the user, mainly for the following two reasons: (1) the operation of the user through the financial platform is very diversified (for example, hundreds of transaction categories such as electricity payment, water payment and cash withdrawal), and due to different habits of each user, the period of user behavior is required to be accurately predicted, each user needs to be modeled, but a neural network model is separately trained for each operation of each user, which is impractical, and the requirement of timeliness cannot be met by adopting an excessively complex mode for prediction; (2) from the data level analysis, the operation of the user through the financial platform is not continuous, the generated time sequence is sparse, the electricity charge is taken as an example, the period is about 20-40 days, and the problem is not applicable to the common machine learning method.
In view of the above problems, no effective solution has been proposed.
Disclosure of Invention
The embodiment of the invention provides a method and a device for determining a user behavior cycle, electronic equipment and a storage medium, which are used for at least solving the technical problem that the recommended application function does not meet the actual use requirement of a user because the behavior cycle of the user cannot be accurately found in the related technology.
According to an aspect of the embodiments of the present invention, there is provided a method for determining a user behavior period, including: acquiring behavior data of a target user in a preset time period and a preset periodic sequence, wherein the behavior data at least comprises: the method comprises the steps of obtaining a time sequence and an operation behavior corresponding to each time point in the time sequence, wherein the cycle sequence comprises a plurality of behavior cycles; inputting the time sequence into a preset period model to obtain a time regular distance value, a time sequence length and the behavior period; calculating a confidence level of the time series with respect to each of the behavior periods based on the time warping distance value, the time series length, and the behavior period; and under the condition that the confidence of the target behavior period is greater than a preset parameter threshold, determining the target behavior period as the behavior period of the time sequence.
Optionally, after acquiring the behavior data of the target user within the preset time period, the determining method further includes: performing deduplication processing on the behavior data; and sequencing the behavior data subjected to the duplicate removal processing according to time points to obtain the behavior data sequenced according to a time sequence.
Optionally, after acquiring the behavior data of the target user within the preset time period, the determining method further includes: acquiring a start time node and an end time node in the time sequence; constructing an initial sequence of time nodes and node values associated based on the starting time node and the ending time node, wherein all the node values are assigned as preset numerical values in the initial sequence; and under the condition that the operation behavior occurs at a target time node, calculating the ratio between the time sequence length of the initial sequence and the total times of occurrence of the operation behavior, and replacing the node value corresponding to the target time node with the ratio.
Optionally, before inputting the time series into the preset period model, the determining method includes: acquiring user data of a plurality of users in a historical time period, wherein the user data at least comprises: the transaction part account is used for counting the third times that each user adopts the same transaction part account to generate the operation behavior; dividing the historical time period according to preset segmentation parameters to obtain a plurality of historical sub-time periods, and counting the fourth times of the operation behaviors in each historical sub-time period; determining user statistical data based on the first number of times, the second number of times, the third number of times, and the fourth number of times; and training the preset periodic model based on the user statistical data.
Optionally, the step of inputting the time series into a preset period model to obtain a time regular distance value, a time series length, and the behavior period includes: dividing the time sequence according to the interval duration of a preset period by adopting the preset period model to obtain a plurality of subsequences; expanding each subsequence according to a time schedule so that the time length of the subsequence is the same as that of the time sequence; calculating a dynamic time warping value between each subsequence and the time sequence; characterizing a minimum of the dynamic time-warping value as the time-warping distance value.
Optionally, the step of calculating a dynamic time warping value between each of the subsequences and the time sequence includes: acquiring a node value corresponding to each time node in the subsequence; accumulating the node value and a preset alignment penalty value under the condition that the node value is a preset value to obtain an accumulated node value; under the condition that the node value is not the preset value, calculating a difference value between the alignment penalty value and the node value, and accumulating the node value and the difference value to obtain an accumulated node value; and calculating a dynamic time warping value between each subsequence and the time sequence based on the accumulated node value.
Optionally, after calculating the confidence of the time series with respect to each of the behavior periods, the determining method further includes: under the condition that the confidence degrees of all the behavior periods are less than or equal to a preset parameter threshold, sequencing all the confidence degrees to obtain a confidence degree sequencing result; determining a behavior cycle corresponding to the highest confidence in the confidence ranking results; based on a preset expansion strategy, expanding the behavior period corresponding to the highest confidence coefficient into a reference behavior period; and searching the behavior data of the target user in a preset time period according to the reference behavior period to calculate the confidence of the time series about the reference behavior period.
Optionally, the behavioral data includes at least one of: user identification, transaction type, transaction identification, transaction amount, transaction time, and transaction party account.
According to another aspect of the embodiments of the present invention, there is also provided an apparatus for determining a user behavior period, including: the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring behavior data of a target user in a preset time period and a preset cycle sequence, and the behavior data at least comprises: the method comprises the steps of obtaining a time sequence and an operation behavior corresponding to each time point in the time sequence, wherein the cycle sequence comprises a plurality of behavior cycles; the input unit is used for inputting the time sequence into a preset period model to obtain a time regular distance value, a time sequence length and the behavior period; a calculating unit configured to calculate a confidence of the time series with respect to each of the behavior periods based on the time warping distance value, the time series length, and the behavior period; the determining unit is used for determining the target behavior period as the behavior period of the time sequence under the condition that the confidence degree of the target behavior period is greater than a preset parameter threshold value.
Optionally, the determining means further comprises: the first duplicate removal module is used for performing duplicate removal processing on behavior data of a target user in a preset time period after the behavior data are acquired; and the first sequencing module is used for sequencing the behavior data subjected to the deduplication processing according to time points to obtain the behavior data sequenced according to a time sequence.
Optionally, the determining means further comprises: the first acquisition module is used for acquiring a starting time node and an ending time node in the time sequence after acquiring behavior data of a target user within a preset time period; a first construction module, configured to construct an initial sequence in which time nodes and node values are associated based on the start time node and the end time node, where in the initial sequence, all the node values are assigned to preset values; a first calculating module, configured to calculate, when the operation behavior occurs at a target time node, a ratio between a time series length of the initial sequence and a total number of times of occurrence of the operation behavior, and replace the node value corresponding to the target time node with the ratio.
Optionally, the determining means comprises: a second obtaining module, configured to obtain user data of a plurality of users in a historical time period before the time series is input to a preset periodic model, where the user data at least includes: the transaction part account is used for counting the third times that each user adopts the same transaction part account to generate the operation behavior; the first statistical module is used for dividing the historical time periods according to preset segmentation parameters to obtain a plurality of historical sub-time periods and counting the fourth times of the operation behaviors in each historical sub-time period; a first determining module, configured to determine user statistical data based on the first number, the second number, the third number, and the fourth number; and the first training module is used for training the preset period model based on the user statistical data.
Optionally, the input unit includes: the first dividing module is used for dividing the time sequence according to the interval duration of a preset period by adopting the preset period model to obtain a plurality of subsequences; the first expansion module is used for expanding each subsequence according to a time schedule so that the time length of the subsequence is the same as that of the time sequence; the second calculation module is used for calculating a dynamic time warping value between each subsequence and the time sequence; a first characterization module, configured to characterize a minimum dynamic time-warping value as the time-warping distance value.
Optionally, the second computing module comprises: the first obtaining sub-module is used for obtaining a node value corresponding to each time node in the sub-sequence; the first accumulation submodule is used for accumulating the node value and a preset alignment penalty value under the condition that the node value is a preset value, so as to obtain an accumulated node value; a second accumulation submodule, configured to calculate a difference value between the alignment penalty value and the node value and accumulate the node value and the difference value to obtain an accumulated node value when the node value is not the preset value; and the first calculating sub-module is used for calculating a dynamic time warping value between each subsequence and the time sequence based on the accumulated node value.
Optionally, the determining means further comprises: the second sequencing module is used for sequencing all the confidence degrees to obtain a confidence degree sequencing result under the condition that the confidence degrees of all the behavior periods are less than or equal to a preset parameter threshold after the confidence degree of the time sequence about each behavior period is calculated; the second determining module is used for determining the behavior cycle corresponding to the highest confidence in the confidence ranking results; the second expansion module is used for expanding the behavior cycle corresponding to the highest confidence coefficient into a reference behavior cycle based on a preset expansion strategy; and the third calculation module is used for searching the behavior data of the target user in a preset time period according to the reference behavior period so as to calculate the confidence of the time sequence about the reference behavior period.
Optionally, the behavioral data includes at least one of: user identification, transaction type, transaction identification, transaction amount, transaction time, and transaction party account.
According to another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program, and when the computer program runs, the apparatus where the computer-readable storage medium is located is controlled to execute any one of the above methods for determining a user behavior cycle.
According to another aspect of the embodiments of the present invention, there is also provided an electronic device, including one or more processors and a memory, where the memory is used for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement any one of the above-mentioned methods for determining a user behavior cycle.
According to the method, behavior data of a target user in a preset time period and a preset period sequence are obtained, the time sequence is input into a preset period model, a time regular distance value, a time sequence length and a behavior period are obtained, confidence of the time sequence about each behavior period is calculated based on the time regular distance value, the time sequence length and the behavior period, and the target behavior period is determined to be the behavior period of the time sequence under the condition that the confidence of the target behavior period is larger than a preset parameter threshold. In the application, the confidence of the time sequence corresponding to the operation behavior of the user can be calculated through a preset period model, the behavior period of the time sequence can be determined under the condition that the confidence is greater than a preset threshold, for the operation behavior with the behavior period, a function meeting the actual operation requirement of the user can be recommended through terminal application, better service is provided for the user, the satisfaction degree of the user is improved, and the technical problem that the recommended application function does not meet the actual use requirement of the user due to the fact that the behavior period of the user cannot be accurately excavated in the related technology is solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
FIG. 1 is a flow chart of an alternative method of determining a period of user behavior in accordance with an embodiment of the present invention;
FIG. 2 is a schematic diagram of an alternative method of determining periodic behavior of a user, according to an embodiment of the invention;
FIG. 3 is a schematic diagram of an alternative user behavior period determination apparatus according to an embodiment of the present invention;
fig. 4 is a block diagram of a hardware structure of an electronic device (or mobile device) for a method of determining a credit score value according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
To facilitate understanding of the invention by those skilled in the art, some terms or nouns referred to in the embodiments of the invention are explained below:
a Dynamic Time Warping (DTW) is a method for measuring the similarity of two Time sequences with different lengths, solves the matching problem of different lengths of the two Time sequences based on the idea of Dynamic programming, is widely applied to the field of isolated word identification, can automatically align the two Time sequences with similar trends, and has strong noise resistance.
It should be noted that the method and the apparatus for determining a user behavior period in the present disclosure may be used in the field of financial technology for determining a user behavior period, and may also be used in any field other than the field of financial technology for determining a user behavior period.
It should be noted that, the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for presentation, analyzed data, etc.) referred to in the present disclosure are information and data authorized by the user or sufficiently authorized by each party.
The embodiments of the invention described below may be applied in various systems/applications/devices that determine a periodic behavior of a user. The invention is used for excavating the behavior period of a user, the key problem of the periodic detection of the user behavior is the similarity measurement of time sequences, the character matching problem after the time sequences are symbolized is generally measured by using the Hamming distance, but the Hamming distance requires that two symbol sequences are required to be equal in length, the length is unequal, and the sequences with incompletely consistent time intervals are not suitable for being calculated by using the method. Since in the problem of time series similarity measure, it is almost impossible to have two sequences that are identical.
Therefore, the invention adopts the improved dynamic time warping DTW method to measure, adds the constraint to the periodic mining of the user, not only can accurately mine the behavior period (such as the periodicity of the transfer behavior) of the user in the financial services, but also can obtain the periodic pattern of the behavior with a small number of abnormal values and the behavior with multiple periodic intervals, can well solve the popularization problem of the financial function and has high service value.
The present invention will be described in detail with reference to examples.
Example one
In accordance with an embodiment of the present invention, there is provided an embodiment of a method for determining a user behavior period, it being noted that the steps illustrated in the flowchart of the drawings may be performed in a computer system such as a set of computer-executable instructions and that, although a logical order is illustrated in the flowchart, in some cases the steps illustrated or described may be performed in an order different than presented herein.
Fig. 1 is a flowchart of an alternative user behavior period determination method according to an embodiment of the present invention, and as shown in fig. 1, the method includes the following steps:
step S101, behavior data of a target user in a preset time period and a preset periodic sequence are obtained, wherein the behavior data at least comprise: the time sequence and the operation behavior corresponding to each time point in the time sequence, wherein the cycle sequence comprises a plurality of behavior cycles.
Step S102, inputting the time sequence into a preset period model to obtain a time regular distance value, a time sequence length and a behavior period.
Step S103, calculating the confidence of the time series about each behavior period based on the time warping distance value, the time series length and the behavior period.
And step S104, determining the target behavior period as a time-series behavior period under the condition that the confidence coefficient of the target behavior period is greater than a preset parameter threshold value.
Through the steps, behavior data of the target user in a preset time period and a preset period sequence can be obtained, the time sequence is input into a preset period model, a time regular distance value, a time sequence length and a behavior period are obtained, the confidence coefficient of the time sequence about each behavior period is calculated based on the time regular distance value, the time sequence length and the behavior period, and the target behavior period is determined to be the behavior period of the time sequence under the condition that the confidence coefficient of the target behavior period is larger than a preset parameter threshold value. In the embodiment of the invention, the confidence of the time sequence corresponding to the operation behavior of the user can be calculated through a preset period model, the behavior period of the time sequence can be determined under the condition that the confidence is greater than a preset threshold, and for the operation behavior with the behavior period, the function of meeting the actual operation requirement of the user can be recommended through terminal application, so that better service is provided for the user, the satisfaction degree of the user is improved, and the technical problem that the recommended application function does not meet the actual use requirement of the user because the behavior period of the user cannot be accurately excavated in the related technology is solved.
The following will explain the embodiments of the present invention in detail with reference to the above steps.
Step S101, behavior data of a target user in a preset time period and a preset periodic sequence are obtained, wherein the behavior data at least comprise: the time sequence and the operation behavior corresponding to each time point in the time sequence, wherein the cycle sequence comprises a plurality of behavior cycles.
In the embodiment of the present invention, behavior data of the target user in a preset time period (for example, the last year or last half year) may be acquired, where the behavior data includes a time series of occurrence of user operation behaviors (for example, a time series of occurrence dates of transfer behaviors), and an operation behavior corresponds to each time point in the time series. In addition, in this embodiment, a cycle sequence (for example, Cycles [5,10, 15., m/2], where m represents a length of the time sequence) to be found may be preset, the cycle sequence includes a plurality of behavior Cycles, in order to reduce the amount of calculation, an interval of the cycle sequence may be set to an initial cycle value (for example, the initial cycle value is set to 5), and if no suitable cycle is found, cycle adjustment is performed, and behavior cycle scrutiny is continued.
Optionally, the behavioural data comprises at least one of: user identification, transaction type, transaction identification, transaction amount, transaction time, and transaction party account.
Optionally, after acquiring the behavior data of the target user within the preset time period, the determining method further includes: performing duplicate removal processing on the behavior data; and sequencing the behavior data subjected to the duplicate removal processing according to the time points to obtain the behavior data sequenced according to the time sequence.
In the embodiment of the present invention, after behavior data of a target user in a preset time period is obtained, deduplication processing may be performed on the obtained behavior data (for example, duplicate transaction identifiers are removed), then transaction time of a corresponding transaction of the user may be obtained according to the transaction identifiers, and after the transaction time is sorted (that is, after deduplication processing is performed on the behavior data according to time points), behavior data sorted according to a time sequence may be obtained.
Optionally, after acquiring the behavior data of the target user within the preset time period, the determining method further includes: acquiring a start time node and an end time node in a time sequence; constructing an initial sequence of time nodes and node values associated based on a start time node and an end time node, wherein all the node values are assigned to preset values in the initial sequence; in the case where an operation action occurs at the target time node, a ratio between a time series length of the initial sequence (which may be 1 minute (or 2 minutes, 10 minutes) for a time series length of a certain sequence) and a total number of times of occurrence of the operation action is determined, and a node value corresponding to the target time node is replaced with the ratio.
In the embodiment of the present invention, after sorting a time sequence, the beginning and ending elements (i.e., the start time node and the end time node) in the time sequence may be obtained and may be denoted as start _ time and end _ time, an initial sequence (i.e., an initial sequence in which time nodes and node values are associated is constructed, wherein in the initial sequence, all node values are assigned as preset values) in which all the preset values from start _ time to end _ time are preset values (e.g., the preset values are set to 0/1) is generated based on the start time node and the end time node, if an operation behavior (e.g., a transfer behavior) occurs in a node in the initial sequence (i.e., in the case of an operation behavior occurring in a target time node), a ratio between a time sequence length of the initial sequence and a total number of times of the occurrence of the operation behavior is calculated, and a value corresponding to the target time node is replaced with the ratio of the node, a time series is obtained that can be input into a preset periodic model.
Optionally, before inputting the time series into the preset period model, the determining method includes: acquiring user data of a plurality of users in a historical time period, wherein the user data at least comprises the following steps: the transaction part account is used for counting the third times that each user adopts the same transaction part account to generate the operation behavior; dividing the historical time periods according to preset segmentation parameters to obtain a plurality of historical sub-time periods, and counting the fourth times of operation behaviors in each historical sub-time period; determining user statistical data based on the first times, the second times, the third times and the fourth times; and training the preset periodic model based on the user statistical data.
In an embodiment of the present invention, user data of a plurality of users in a historical period of time (e.g., in the past years) may be obtained, and the user data may include: user identification, transaction type, transaction identification, transaction amount, action occurrence date, transaction party account, and the like. For example, taking the transfer type as an example, a plurality of users are randomly selected according to the user identification, and user data of the transfer type in the past year is obtained.
In this embodiment, some rules of mining user operation behaviors may be tried to be analyzed by using a statistical method according to the acquired user data, and data for training the periodic model may be screened out. Taking the transfer behavior as an example, the data of the behavior is counted.
In this embodiment, the number of times that each user has a transfer behavior within the selected historical time period (that is, the user identifier is used to count the first number of times that the user has an operation behavior) may be analyzed according to the user identifier, and a user who has a very frequent transfer behavior may be obtained, and the user belongs to a very active user of the financial institution.
In this embodiment, the number of transfers occurring on each date may be counted according to the occurrence date of the action, and whether the transfer occurring on the user is related to the date may be analyzed (that is, the occurrence date of the action is used to count the second number of times of the operation occurring on each historical date).
In this embodiment, the distribution of the number of times of transfer from each user to the same person within the selected historical time period may be counted according to the transaction party account (that is, the transaction party account is used to count the third number of times of operation performed by each user using the same transaction party account).
In this embodiment, the history time period may be divided according to the preset segmentation parameters to obtain a plurality of history sub-time periods (for example, the history time period is divided into a beginning of the month, a middle of the month, and a last of the month, where if the transfer date is 1 to 10, the transaction is recorded as a transaction performed at the beginning of the month, similarly, 11 to 20 is recorded as a transaction in the month, and 21 to the end of the month is recorded as a transaction at the end of the month), and a fourth number of times of the operation performed in each history sub-time period is counted (for example, the number of times of the transfer performed by the user at the beginning of the month, in the middle of the month, and at the end of the month is counted).
In this embodiment, the user statistical data may be determined based on the obtained statistical times (including the first time, the second time, the third time, the fourth time, and the like), and then the preset period model may be trained based on the user statistical data.
In this embodiment, the total number of times of the operation behavior related to the fixed interval transaction duration may be counted, for example, for the transfer behavior occurring at an interval of 2 days, the behavior is recorded once, and the number of times of all the transfer behaviors occurring at an interval of 2 days is accumulated to obtain the total number of times of the transfer behaviors occurring at an interval of 2 days.
Optionally, after obtaining the user statistical data, performing deduplication processing on the obtained data (for example, removing repeated user identifiers), then obtaining operation behavior data of the user and corresponding transaction time according to the user identifiers, and sorting the transaction time to obtain a time sequence for training the period model.
Step S102, inputting the time sequence into a preset period model to obtain a time regular distance value, a time sequence length and a behavior period.
Optionally, the step of inputting the time sequence into a preset period model to obtain a time regular distance value, a time sequence length, and a behavior period includes: dividing the time sequence according to the interval duration of a preset period by adopting a preset period model to obtain a plurality of subsequences; expanding each subsequence according to the time schedule so that the time length of the subsequence is the same as that of the time sequence; calculating a dynamic time warping value between each subsequence and the time sequence; the minimum dynamic time-warping value is characterized as a time-warping distance value.
In the embodiment of the invention, the periodic sequence (Cycles [5,10, 15., m/2] to be searched is determined]Where m represents the time series length), the time series S (S) of the user is obtained1,s2,...,sm) And the number of times the operation behavior occurs is recorded as n. Optionally, in this embodiment, a preset cycle sequence may be traversed to obtain one value cyclei(i.e. a period interval duration), dividing the time sequence according to the period interval duration to obtain a plurality of subsequences '(S')1,...,scycle) Elongating the subsequence to have the same length as the time series S to obtain the sequence R (S)1,...,scycle,s1,...,scycle...) (i.e., each sub-sequence is spread according to a time schedule so that the time length of the sub-sequence is the same as the time length of the time sequence).
In this embodiment, an improved dynamic time warping algorithm DTW may be adopted to calculate a dynamic time warping value between each subsequence R and the time sequence S, characterize a minimum dynamic time warping value as a time warping distance value, and record a current period interval duration (i.e., a behavior period).
Optionally, the step of calculating a dynamic time warping value between each subsequence and the time series includes: acquiring a node value corresponding to each time node in the subsequence; accumulating the node value and a preset alignment penalty value under the condition that the node value is a preset value to obtain an accumulated node value; under the condition that the node value is not a preset value, calculating a difference value between the alignment penalty value and the node value, and accumulating the node value and the difference value to obtain an accumulated node value; and calculating a dynamic time warping value between each subsequence and the time sequence based on the accumulated node value.
In the embodiment of the present invention, the original DTW algorithm is calculated as follows:
given two time sequences r, s with the length of n, m respectively, an m × n matrix D is established, wherein the element cell (i, j) in D is riAnd sjThe original DTW algorithm has the following calculation formula (1):
wherein, cell (i-1, j), cell (i-1, j-1), cell (i, j-1) represent the previous element of cell (i, j) respectively, d (i, j) is riAnd sjThe distance of (c).
A path W from the element cell (1, 1) to the element cell (n, n) ═ W1,w2,...,wKAnd K is the number of paths, called the curved path. As can be seen from the DTW distance matrix D, there are many curved paths, and the goal of DTW calculation is to find the path with the smallest total length in the curved paths, as shown in the following equation (2):
the DTW algorithm can automatically align the sequences, and the sequences with similar trends can calculate smaller distances through a filling mode, but for an actual problem scene, the original automatic alignment mode of the DTW algorithm rather causes a larger error to occur during the calculation period, for example, there is no periodic user operation behavior, and the calculated distance is 0 in the automatic alignment characteristic of the original DTW algorithm, so that the operation behavior of the user is obtained to have periodicity, but obviously, the operation behavior is incorrect.
Therefore, in this embodiment, an improved DTW algorithm is adopted to calculate a dynamic time warping value between each subsequence and a time sequence, specifically:
the DTW algorithm is improved, when filling alignment is carried out, filling is not carried out by a previous time node, but a fixed constant L is filled, and from the calculation result, a penalty term is added to the distance when alignment is carried out. Specifically, after a node value corresponding to each time node in a subsequence is acquired, if the node value is a preset value (e.g., 0), accumulating the node value and a preset alignment penalty value L to obtain an accumulated node value; if the node value is not a preset value (assuming that the node value is Q), calculating a difference value (| L-Q |) between the alignment penalty value and the node value, accumulating the node value and the difference value to obtain an accumulated node value, and then calculating a dynamic time warping value between each subsequence and the time sequence based on the accumulated node value. The improved DTW algorithm can enable the calculated distance to be more consistent with the real situation, so that users with periodicity can be found more accurately, and the modified distance calculation formula (3) is as follows:
wherein, cell (i-1, j), cell (i-1, j-1), cell (i, j-1) represent the previous element of cell (i, j) respectively, d (i, L) represents riDistance from L, d (i),j) Is riAnd sjD (L, j) represents L and sjL is a fixed constant, riAnd sjRespectively, a value of an element in the two sequences.
Step S103, calculating the confidence of the time series about each behavior period based on the time warping distance value, the time series length and the behavior period.
In the embodiment of the present invention, after obtaining the time warping distance value DTW (R, S), the confidence of the time series with respect to each behavior period may be calculated by using the following formula (4) in combination with the length of the time series and the behavior period, where the confidence of the period indicates that a period pattern must continuously appear a certain number of times to consider that the time series has periodicity, when a period pattern frequently appears, the DTW (R, S) tends to 0, the confidence tends to 1, and indicates that the period is most reliable, when noise or other influences occur, the DTW (R, S) increases, the confidence tends to 0, and the confidence conf calculation formula (4) is as follows:
where n represents the time series length, p represents the current behavior period, and DTW (R, S) represents the time warping distance value.
And step S104, determining the target behavior period as a time-series behavior period under the condition that the confidence coefficient of the target behavior period is greater than a preset parameter threshold value.
In the embodiment of the present invention, when the confidence of the target behavior period is greater than a preset parameter threshold (for example, 0.8), it may be determined that the target behavior period is a time-series behavior period, and the traversal of the cycle sequence is ended.
Optionally, after calculating the confidence of the time series with respect to each behavior period, the determining method further includes: under the condition that the confidence degrees of all the behavior periods are less than or equal to a preset parameter threshold, sequencing all the confidence degrees to obtain a confidence degree sequencing result; determining a behavior period corresponding to the highest confidence in the confidence ranking results; based on a preset expansion strategy, expanding the behavior period corresponding to the highest confidence coefficient into a reference behavior period; and searching the behavior data of the target user in a preset time period according to the reference behavior period to calculate the confidence of the time series relative to the reference behavior period.
In the embodiment of the present invention, when the confidence degrees of all behavior periods are less than or equal to the preset parameter threshold, it indicates that no suitable period is included in the period sequence, and detailed examination needs to be performed, specifically: for allSequencing the confidence degrees to obtain a confidence degree sequencing result; determining the behavior cycle corresponding to the highest confidence in the confidence ranking result (i.e. determining the behavior cycle as the current optimal cycle)best) Based on a preset expansion strategy, the behavior cycle corresponding to the highest confidence coefficient is expanded to be a reference behavior cycle (for example, the current optimal cycle is cycled)bestExpanding the search range to obtain a reference behavior period (cycle)best-4,...,cyclebest-4,...,cyclebest+4) According to the reference behavior period, searching the behavior data of the target user in a preset time period to calculate the confidence of the time sequence with respect to the reference behavior period, and when the highest confidence is greater than a certain threshold (for example, 0.7, which may be lower than the previously set confidence threshold), determining that the time sequence has a period, otherwise, considering that the sequence does not have periodicity.
Fig. 2 is a schematic diagram of an optional determination of a user periodic behavior according to an embodiment of the present invention, as shown in fig. 2, for a certain transfer scenario, a user may transfer money through a transfer system on a terminal (e.g., a mobile phone), and while the user transfers money, it may be determined whether a historical behavior of the user has periodicity, and the requirement on timeliness is high, so that tags of the users may be generated before the transfer system is invoked, and stored in a server, which is equivalent to a process of only looking up a table when the transfer system is invoked, and the specific process is as follows:
acquiring user historical data (including user identification, transaction party account number, transaction time, consumption behavior and the like), calculating a historical behavior sequence of each user (the acquired user historical data can be input into an improved DTW algorithm for calculation), storing an obtained result into a user periodic behavior table (the table comprises the user identification and the transaction party account number and whether the periodic table has periodicity), and labeling the user (for example, 1: periodicity; 0: no periodicity); when a user transfers money by using a transfer system, a user identification and a transaction party account number are obtained, a user periodic behavior table is retrieved according to the two indexes to obtain a user label, whether periodicity exists or not is judged according to the label, if the periodicity exists in the obtained result, an 'appointment transfer' function is pushed for the client, the user requirement is met, and the user viscosity is improved.
The embodiment of the invention provides a user behavior period mining method based on an improved DTW algorithm, which is characterized in that a DTW algorithm distance formula is improved, and constraint is added to the periodic mining of user behaviors, so that a strong periodic mode of the user behaviors can be accurately found out, and a good effect is also achieved on a periodic sequence and a multi-periodic sequence containing noise, so that the user can be better served, and the service value is very high.
Example two
The device for determining the user behavior cycle provided in this embodiment includes a plurality of implementation units, and each implementation unit corresponds to each implementation step in the first embodiment.
Fig. 3 is a schematic diagram of an alternative user behavior period determination apparatus according to an embodiment of the present invention, and as shown in fig. 3, the determination apparatus may include: an acquisition unit 30, an input unit 31, a calculation unit 32, a determination unit 33, wherein,
an obtaining unit 30, configured to obtain behavior data of a target user in a preset time period and a preset cycle sequence, where the behavior data at least includes: the method comprises the steps that time sequences and operation behaviors corresponding to each time point in the time sequences are obtained, and a cycle sequence comprises a plurality of behavior cycles;
the input unit 31 is configured to input the time series to a preset period model, so as to obtain a time regular distance value, a time series length, and a behavior period;
a calculation unit 32 for calculating a confidence of the time series with respect to each behavior period based on the time warping distance value, the time series length, and the behavior period;
the determining unit 33 is configured to determine the target behavior period as a time-series behavior period when the confidence of the target behavior period is greater than a preset parameter threshold.
The determining device can obtain behavior data of a target user in a preset time period and a preset cycle sequence through the obtaining unit 30, input the time sequence into a preset cycle model through the input unit 31 to obtain a time warping distance value, a time sequence length and a behavior cycle, calculate the confidence of the time sequence about each behavior cycle through the calculating unit 32 based on the time warping distance value, the time sequence length and the behavior cycle, and determine the target behavior cycle as the behavior cycle of the time sequence through the determining unit 33 under the condition that the confidence of the target behavior cycle is greater than a preset parameter threshold. In the embodiment of the invention, the confidence of the time sequence corresponding to the operation behavior of the user can be calculated through a preset period model, the behavior period of the time sequence can be determined under the condition that the confidence is greater than a preset threshold, and for the operation behavior with the behavior period, the function of meeting the actual operation requirement of the user can be recommended through terminal application, so that better service is provided for the user, the satisfaction degree of the user is improved, and the technical problem that the recommended application function does not meet the actual use requirement of the user because the behavior period of the user cannot be accurately excavated in the related technology is solved.
Optionally, the determining device further includes: the first duplicate removal module is used for performing duplicate removal processing on the behavior data after the behavior data of the target user in a preset time period is acquired; and the first sequencing module is used for sequencing the behavior data subjected to the deduplication processing according to time points to obtain the behavior data sequenced according to the time sequence.
Optionally, the determining device further includes: the first acquisition module is used for acquiring a start time node and an end time node in a time sequence after acquiring behavior data of a target user within a preset time period; the device comprises a first construction module, a second construction module and a third construction module, wherein the first construction module is used for constructing an initial sequence of the association of time nodes and node values based on a starting time node and an ending time node, and all the node values are assigned to preset values in the initial sequence; and the first calculating module is used for calculating the ratio between the time sequence length of the initial sequence and the total times of the operation behaviors under the condition that the operation behaviors occur in the target time node, and replacing the node value corresponding to the target time node with the ratio.
Optionally, the determining means includes: a second obtaining module, configured to obtain user data of a plurality of users in a historical time period before inputting the time sequence into the preset periodic model, where the user data at least includes: the transaction part account is used for counting the third times that each user adopts the same transaction part account to generate the operation behavior; the first statistical module is used for dividing the historical time periods according to preset segmentation parameters to obtain a plurality of historical sub-time periods and counting the fourth times of operation behaviors occurring in each historical sub-time period; the first determining module is used for determining user statistical data based on the first times, the second times, the third times and the fourth times; and the first training module is used for training the preset periodic model based on the user statistical data.
Optionally, the input unit includes: the first dividing module is used for dividing the time sequence according to the interval duration of a preset period by adopting a preset period model to obtain a plurality of subsequences; the first expansion module is used for expanding each subsequence according to the time schedule so that the time length of the subsequence is the same as that of the time sequence; the second calculation module is used for calculating a dynamic time warping value between each subsequence and the time sequence; a first characterization module to characterize the minimum dynamic time regular integer value as a time regular distance value.
Optionally, the second calculating module includes: the first obtaining sub-module is used for obtaining a node value corresponding to each time node in the sub-sequence; the first accumulation submodule is used for accumulating the node value and a preset alignment penalty value under the condition that the node value is a preset value to obtain an accumulated node value; the second accumulation submodule is used for calculating a difference value between the alignment penalty value and the node value under the condition that the node value is not a preset value, and accumulating the node value and the difference value to obtain an accumulated node value; and the first calculating sub-module is used for calculating a dynamic time warping value between each sub-sequence and the time sequence based on the accumulated node value.
Optionally, the determining device further includes: the second sequencing module is used for sequencing all the confidence degrees to obtain a confidence degree sequencing result under the condition that the confidence degrees of all the behavior periods are less than or equal to a preset parameter threshold after the confidence degree of the time sequence about each behavior period is calculated; the second determining module is used for determining the behavior cycle corresponding to the highest confidence in the confidence ranking result; the second expansion module is used for expanding the behavior cycle corresponding to the highest confidence coefficient into a reference behavior cycle based on a preset expansion strategy; and the third calculation module is used for searching the behavior data of the target user in a preset time period according to the reference behavior period so as to calculate the confidence of the time sequence relative to the reference behavior period.
Optionally, the behavioural data comprises at least one of: user identification, transaction type, transaction identification, transaction amount, transaction time, and transaction party account.
The above-mentioned determining means may further comprise a processor and a memory, and the above-mentioned acquiring unit 30, the input unit 31, the calculating unit 32, the determining unit 33, and the like are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize the corresponding functions.
The processor comprises a kernel, and the kernel calls a corresponding program unit from the memory. The kernel can set one or more than one, and the target behavior period is determined to be the time sequence behavior period by adjusting the kernel parameters.
The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.
The present application further provides a computer program product adapted to perform a program for initializing the following method steps when executed on a data processing device: the method comprises the steps of obtaining behavior data of a target user in a preset time period and a preset period sequence, inputting the time sequence into a preset period model to obtain a time regular distance value, a time sequence length and a behavior period, calculating confidence of the time sequence about each behavior period based on the time regular distance value, the time sequence length and the behavior period, and determining the target behavior period as the behavior period of the time sequence under the condition that the confidence of the target behavior period is larger than a preset parameter threshold.
According to another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program, and when the computer program runs, the apparatus where the computer-readable storage medium is located is controlled to execute any one of the above methods for determining a user behavior cycle.
According to another aspect of embodiments of the present invention, there is also provided an electronic device, including one or more processors and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the method for determining a user behavior cycle of any one of the above.
Fig. 4 is a block diagram of a hardware structure of an electronic device (or a mobile device) for a method for determining a user behavior period according to an embodiment of the present invention. As shown in fig. 4, the electronic device may include one or more (shown as 102a, 102b, … …, 102 n) processors 102 (the processors 102 may include, but are not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA), and memory 104 for storing data. Besides, the method can also comprise the following steps: a display, an input/output interface (I/O interface), a Universal Serial Bus (USB) port (which may be included as one of the ports of the I/O interface), a network interface, a keyboard, a power supply, and/or a camera. It will be understood by those skilled in the art that the structure shown in fig. 4 is only an illustration and is not intended to limit the structure of the electronic device. For example, the electronic device may also include more or fewer components than shown in FIG. 4, or have a different configuration than shown in FIG. 4.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
In the above embodiments of the present invention, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
In the embodiments provided in the present application, it should be understood that the disclosed technology can be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units may be a logical division, and in actual implementation, there may be another division, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, units or modules, and may be in an electrical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (11)
1. A method for determining a user behavior period is characterized by comprising the following steps:
acquiring behavior data of a target user in a preset time period and a preset periodic sequence, wherein the behavior data at least comprises: the method comprises the steps of obtaining a time sequence and an operation behavior corresponding to each time point in the time sequence, wherein the cycle sequence comprises a plurality of behavior cycles;
inputting the time sequence into a preset period model to obtain a time regular distance value, a time sequence length and the behavior period;
calculating a confidence level of the time series with respect to each of the behavior periods based on the time warping distance value, the time series length, and the behavior period;
and under the condition that the confidence of the target behavior period is greater than a preset parameter threshold, determining the target behavior period as the behavior period of the time sequence.
2. The determination method according to claim 1, wherein after acquiring the behavior data of the target user within a preset time period, the determination method further comprises:
performing deduplication processing on the behavior data;
and sequencing the behavior data subjected to the duplicate removal processing according to time points to obtain the behavior data sequenced according to a time sequence.
3. The determination method according to claim 1, wherein after acquiring the behavior data of the target user within a preset time period, the determination method further comprises:
acquiring a start time node and an end time node in the time sequence;
constructing an initial sequence of time nodes and node values associated based on the starting time node and the ending time node, wherein all the node values are assigned as preset numerical values in the initial sequence;
and under the condition that the operation behavior occurs at a target time node, calculating the ratio between the time sequence length of the initial sequence and the total times of occurrence of the operation behavior, and replacing the node value corresponding to the target time node with the ratio.
4. The determination method according to claim 1, wherein before inputting the time series to a preset period model, the determination method comprises:
acquiring user data of a plurality of users in a historical time period, wherein the user data at least comprises: the transaction part account is used for counting the third times that each user adopts the same transaction part account to generate the operation behavior;
dividing the historical time period according to preset segmentation parameters to obtain a plurality of historical sub-time periods, and counting the fourth times of the operation behaviors in each historical sub-time period;
determining user statistical data based on the first number of times, the second number of times, the third number of times, and the fourth number of times;
and training the preset periodic model based on the user statistical data.
5. The method for determining according to claim 1, wherein the step of inputting the time series into a preset period model to obtain a time warping distance value, a time series length and the behavior period comprises:
dividing the time sequence according to the interval duration of a preset period by adopting the preset period model to obtain a plurality of subsequences;
expanding each subsequence according to a time schedule so that the time length of the subsequence is the same as that of the time sequence;
calculating a dynamic time warping value between each subsequence and the time sequence;
characterizing a minimum of the dynamic time-warping value as the time-warping distance value.
6. The method of claim 5, wherein the step of calculating a dynamic time warping value between each of the subsequences and the time sequence comprises:
acquiring a node value corresponding to each time node in the subsequence;
accumulating the node value and a preset alignment penalty value under the condition that the node value is a preset value to obtain an accumulated node value;
under the condition that the node value is not the preset value, calculating a difference value between the alignment penalty value and the node value, and accumulating the node value and the difference value to obtain an accumulated node value;
and calculating a dynamic time warping value between each subsequence and the time sequence based on the accumulated node value.
7. The determination method according to claim 1, characterized in that after calculating the confidence of the time series with respect to each of the behavior periods, the determination method further comprises:
under the condition that the confidence degrees of all the behavior periods are less than or equal to a preset parameter threshold, sequencing all the confidence degrees to obtain a confidence degree sequencing result;
determining a behavior cycle corresponding to the highest confidence in the confidence ranking results;
based on a preset expansion strategy, expanding the behavior period corresponding to the highest confidence coefficient into a reference behavior period;
and searching the behavior data of the target user in a preset time period according to the reference behavior period to calculate the confidence of the time series about the reference behavior period.
8. The method of any one of claims 1 to 7, wherein the behavior data includes at least one of: user identification, transaction type, transaction identification, transaction amount, transaction time, and transaction party account.
9. An apparatus for determining a period of user behavior, comprising:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring behavior data of a target user in a preset time period and a preset cycle sequence, and the behavior data at least comprises: the method comprises the steps of obtaining a time sequence and an operation behavior corresponding to each time point in the time sequence, wherein the cycle sequence comprises a plurality of behavior cycles;
the input unit is used for inputting the time sequence into a preset period model to obtain a time regular distance value, a time sequence length and the behavior period;
a calculating unit configured to calculate a confidence of the time series with respect to each of the behavior periods based on the time warping distance value, the time series length, and the behavior period;
the determining unit is used for determining the target behavior period as the behavior period of the time sequence under the condition that the confidence degree of the target behavior period is greater than a preset parameter threshold value.
10. A computer-readable storage medium, comprising a stored computer program, wherein the computer program, when executed, controls an apparatus in which the computer-readable storage medium is located to perform the method for determining a user behavior cycle according to any one of claims 1 to 8.
11. An electronic device comprising one or more processors and memory storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of determining a user behavior cycle of any of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111631056.3A CN114219540A (en) | 2021-12-28 | 2021-12-28 | Method and device for determining user behavior period, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111631056.3A CN114219540A (en) | 2021-12-28 | 2021-12-28 | Method and device for determining user behavior period, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114219540A true CN114219540A (en) | 2022-03-22 |
Family
ID=80706572
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111631056.3A Pending CN114219540A (en) | 2021-12-28 | 2021-12-28 | Method and device for determining user behavior period, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114219540A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116109394A (en) * | 2023-03-23 | 2023-05-12 | 北京芯盾时代科技有限公司 | Group mining method, device, electronic equipment and computer readable storage medium |
| CN116738033A (en) * | 2022-09-05 | 2023-09-12 | 荣耀终端有限公司 | Method and device for recommending service |
-
2021
- 2021-12-28 CN CN202111631056.3A patent/CN114219540A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116738033A (en) * | 2022-09-05 | 2023-09-12 | 荣耀终端有限公司 | Method and device for recommending service |
| CN116109394A (en) * | 2023-03-23 | 2023-05-12 | 北京芯盾时代科技有限公司 | Group mining method, device, electronic equipment and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107689008A (en) | A kind of user insures the method and device of behavior prediction | |
| CN109840782B (en) | Click rate prediction method, device, server and storage medium | |
| CN114219540A (en) | Method and device for determining user behavior period, electronic equipment and storage medium | |
| CN110880127B (en) | Consumption level prediction method and device, electronic equipment and storage medium | |
| CN110782291A (en) | Advertisement delivery user determination method and device, storage medium and electronic device | |
| CN110247974A (en) | Information-pushing method, device, computer and storage medium based on block chain | |
| CN111861605A (en) | Business object recommendation method | |
| CN115204985A (en) | Shopping behavior prediction method, device, equipment and storage medium | |
| CN113592605A (en) | Product recommendation method, device, equipment and storage medium based on similar products | |
| WO2021077226A1 (en) | Method and system for individual demand forecasting | |
| CN116127184A (en) | Product recommendation method and device, nonvolatile storage medium and electronic equipment | |
| CN115271886A (en) | Financial product recommendation method and device, storage medium and electronic equipment | |
| CN112990989B (en) | Value prediction model input data generation method, device, equipment and medium | |
| CN111325255B (en) | Specific crowd delineating method and device, electronic equipment and storage medium | |
| CN111325572B (en) | Data processing method and device | |
| CN116501979A (en) | Information recommendation method, information recommendation device, computer equipment and computer readable storage medium | |
| CN116610821A (en) | Knowledge graph-based enterprise risk analysis method, system and storage medium | |
| CN109462635B (en) | Information pushing method, computer readable storage medium and server | |
| CN110389963A (en) | The recognition methods of channel effect, device, equipment and storage medium based on big data | |
| CN116186119A (en) | User behavior analysis method, device, equipment and storage medium | |
| CN115345663A (en) | Marketing strategy evaluation method and device, electronic equipment and storage medium | |
| CN114677233A (en) | Information recommendation method and device, storage medium and electronic equipment | |
| CN113724044A (en) | User portrait based commodity recommendation, apparatus, computer device and storage medium | |
| CN113886697A (en) | Clustering algorithm based activity recommendation method, device, equipment and storage medium | |
| CN111709764B (en) | Method and device for determining related parameters of multimedia content and storage medium |
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 |