CN111291936B - Product life cycle prediction model generation method and device and electronic equipment - Google Patents

Product life cycle prediction model generation method and device and electronic equipment Download PDF

Info

Publication number
CN111291936B
CN111291936B CN202010108556.8A CN202010108556A CN111291936B CN 111291936 B CN111291936 B CN 111291936B CN 202010108556 A CN202010108556 A CN 202010108556A CN 111291936 B CN111291936 B CN 111291936B
Authority
CN
China
Prior art keywords
product
target
target product
life cycle
reference data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010108556.8A
Other languages
Chinese (zh)
Other versions
CN111291936A (en
Inventor
张思睿
陈立琨
韩佳音
丁亚曼
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Kingsoft Internet Security Software Co Ltd
Original Assignee
Beijing Kingsoft Internet Security Software Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Kingsoft Internet Security Software Co Ltd filed Critical Beijing Kingsoft Internet Security Software Co Ltd
Priority to CN202010108556.8A priority Critical patent/CN111291936B/en
Publication of CN111291936A publication Critical patent/CN111291936A/en
Application granted granted Critical
Publication of CN111291936B publication Critical patent/CN111291936B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Educational Administration (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application provides a method and a device for generating a product life cycle prediction model and electronic equipment, and belongs to the technical field of computer application. Wherein the method comprises the following steps: determining a target feature vector corresponding to a target product according to the newly increased user quantity and the retention change rate of the target product in K days, wherein K is a positive integer; acquiring a reference data set from historical data of each existing product according to the similarity between the target feature vector and each reference feature vector of each existing product; training a preset life cycle pre-estimated model by using the reference data set to generate a target pre-estimated model corresponding to the target product. Therefore, by the method for generating the product life cycle estimation model, sample data of a target product for training the product life cycle estimation model can be automatically expanded, so that labor cost is saved, and the estimation efficiency of the product life cycle is improved.

Description

Product life cycle prediction model generation method and device and electronic equipment
Technical Field
The present application relates to the field of computer application technologies, and in particular, to a method and an apparatus for generating a product life cycle prediction model, and an electronic device.
Background
The product life cycle is directly related to the establishment of product strategies and marketing strategies by enterprises, so that reasonable prediction of the life cycle of the product is of great significance. Because the life cycle is calculated by adding 1 day of retention rate to 360 days of retention rate of new daily product loading, the estimated core of the life cycle is the estimated retention rate.
In the related art, because the data characteristic of the new product is that there is no historical data, available sample data is very few, when the retention rate of the new product is estimated, the available data is usually found from the historical data of the historical product according to experience by a manual mode to expand the sample data so as to complete the estimation of the retention rate of the new product. However, the efficiency of manually expanding sample data of a new product is low, thereby reducing the estimated efficiency of the product life cycle.
Disclosure of Invention
The method, the device and the electronic equipment for generating the product life cycle estimation model are used for solving the problems that in the related technology, when the life cycle of a new product is estimated, the sample data of the new product is expanded in a manual mode, the efficiency is low, and the estimation efficiency of the life cycle of the product is reduced.
The method for generating the product life cycle estimation model provided by the embodiment of the application comprises the following steps: determining a target feature vector corresponding to a target product according to the newly increased user quantity and the retention change rate of the target product in K days, wherein K is a positive integer; acquiring a reference data set from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector of each existing product; training a preset life cycle pre-estimated model by using the reference data set to generate a target pre-estimated model corresponding to the target product.
Optionally, in an implementation manner of the embodiment of the first aspect, before determining the target feature vector corresponding to the target product according to the newly increased user quantity and the retention change rate of the target product on K days, the method further includes:
obtaining the newly increased user quantity of the target product in K days and a plurality of daily retention rates;
and calculating the retention change rate of the target product according to the plurality of daily retention rates.
Optionally, in another possible implementation form of the embodiment of the first aspect, the calculating a retention change rate of the target product according to the plurality of daily retention rates includes:
And calculating the retention change rate of the target product between the ith and jth days according to the i-day retention and the j-day retention, wherein i and j are different integers respectively.
Optionally, in a further possible implementation form of the embodiment of the first aspect, the determining a target feature vector corresponding to the target product includes:
determining a weight corresponding to the newly added user quantity of the target product on K days according to a preset mapping relation between the newly added user quantity range on the days and the weight;
and assigning the weight to an element used for representing the daily newly added user quantity in the target feature vector.
Optionally, in a further possible implementation form of the embodiment of the first aspect, the determining a target feature vector corresponding to the target product further includes:
and determining element values used for representing date attributes in the target feature vector according to the holiday attributes of the K days.
Optionally, in a further possible implementation form of the embodiment of the first aspect, each reference data in the reference data set includes a daily retention rate of the existing product;
after the reference data set is acquired, the method further comprises:
carrying out fusion processing on each same daily retention rate in each reference data to generate a training sample set;
The training the preset life cycle estimation model by using the reference data set includes:
and training a preset life cycle estimation model by using the training sample set.
Optionally, in another possible implementation form of the embodiment of the first aspect, the fusing processing is performed on each same daily retention rate in each reference data, including:
adding and averaging the same daily retention rate of each reference data;
or alternatively, the process may be performed,
determining the weight of each reference data according to the similarity between the existing product corresponding to each reference data and the target product;
and carrying out weighted average processing on each same daily retention rate in each reference data according to the weight of each reference data.
In another aspect, a method for generating a product life cycle according to an embodiment of the present application includes: acquiring a life cycle estimation request, wherein the estimation request comprises the type and the identification of a target product; determining the duration to be estimated corresponding to the target product according to the type of the target product; calculating the daily retention rate of the target product in the duration to be estimated by using a target estimation model corresponding to the identification of the target product; and determining the life cycle of the target product according to the daily retention rate of the target product in the duration to be estimated.
Optionally, in a possible implementation form of the embodiment of the second aspect, before determining, according to the type of the target product, a duration to be estimated corresponding to the target product, the method further includes:
and counting the life cycle of each type of product, and determining the duration to be estimated corresponding to each product type.
In another aspect, the device for generating the product life cycle estimation model provided by the embodiment of the application comprises: the determining module is used for determining a target feature vector corresponding to a target product according to the newly added user quantity and the retention change rate of the target product on K days, wherein K is a positive integer; the first acquisition module is used for acquiring a reference data set from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector of each existing product; and the training module is used for training a preset life cycle estimation model by using the reference data set to generate a target estimation model corresponding to the target product.
Optionally, in a possible implementation form of the embodiment of the third aspect, the apparatus further includes:
the second acquisition module is used for acquiring the newly increased user quantity and a plurality of daily retention rates of the target product in K days;
And the calculating module is used for calculating the retention change rate of the target product according to the plurality of daily retention rates.
Optionally, in another possible implementation form of the embodiment of the third aspect, the computing module is specifically configured to:
and calculating the retention change rate of the target product between the ith and jth days according to the i-day retention and the j-day retention, wherein i and j are different integers respectively.
Optionally, in a further possible implementation form of the embodiment of the third aspect, the determining module is specifically configured to:
determining a weight corresponding to the newly added user quantity of the target product on K days according to a preset mapping relation between the newly added user quantity range on the days and the weight;
and assigning the weight to an element used for representing the daily newly added user quantity in the target feature vector.
Optionally, in a further possible implementation form of the embodiment of the third aspect, the determining module is further configured to:
and determining element values used for representing date attributes in the target feature vector according to the holiday attributes of the K days.
Optionally, in a further possible implementation form of the embodiment of the third aspect, each reference data in the reference data set includes a daily retention rate of an existing product;
The device further comprises:
the fusion processing module is used for carrying out fusion processing on each same daily retention rate in each reference data to generate a training sample set;
the training module is specifically used for:
and training a preset life cycle estimation model by using the training sample set.
Optionally, in another possible implementation form of the embodiment of the third aspect, the fusion processing module is specifically configured to:
adding and averaging the same daily retention rate of each reference data;
or alternatively, the process may be performed,
determining the weight of each reference data according to the similarity between the existing product corresponding to each reference data and the target product;
and carrying out weighted average processing on each same daily retention rate in each reference data according to the weight of each reference data.
In another aspect, a product life cycle estimating device provided by an embodiment of the present application includes: the acquisition module is used for acquiring a life cycle estimation request, wherein the estimation request comprises the type and the identification of a target product; the first determining module is used for determining the duration to be estimated corresponding to the target product according to the type of the target product; the calculation module is used for calculating the daily retention rate of the target product in the duration to be estimated by using a target estimated model corresponding to the identification of the target product; and the second determining module is used for determining the life cycle of the target product according to the daily retention rate of the target product in the duration to be estimated.
Optionally, in a possible implementation form of the fourth aspect embodiment, the apparatus further includes:
and the third determining module is used for counting the life cycle of each type of product and determining the duration to be estimated corresponding to each product type.
An embodiment of the present application provides an electronic device, including: the system comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, and is characterized in that the processor realizes the product life cycle estimation model generation method or the product life cycle estimation method when executing the program.
In another aspect, the embodiment of the present application provides a computer readable storage medium having a computer program stored thereon, where the program when executed by a processor implements a method for generating a product life cycle estimation model or a method for estimating a product life cycle as described above.
In a further aspect, the present application provides a computer program, which when executed by a processor, implements the method for generating a product life cycle estimation model or the method for product life cycle estimation according to the embodiments of the present application.
According to the method, the device, the electronic equipment, the computer readable storage medium and the computer program for generating the product life cycle estimation model, the target feature vector corresponding to the target product is determined according to the newly increased user quantity and the retention change rate of the target product in K days, the reference data set is obtained from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector of each existing product, and the preset life cycle estimation model is trained by using the reference data set, so that the target estimation model corresponding to the target product is generated. Therefore, the target feature vector corresponding to the target product is determined through the existing sample data of the target product, the reference data set similar to the sample data of the target product is obtained from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector, and the target estimated model is trained, so that the sample data of the target product for the estimated model training is automatically expanded, the labor cost is saved, and the estimated efficiency of the product life cycle is improved.
Additional aspects and advantages of the application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the application.
Drawings
The foregoing and/or additional aspects and advantages of the application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:
FIG. 1 is a flowchart of a method for generating a product life cycle estimation model according to an embodiment of the present application;
FIG. 2 is a flowchart of another method for generating a product life cycle estimation model according to an embodiment of the present application;
FIG. 3 is a flowchart of a product life cycle estimation method according to an embodiment of the present application;
FIG. 4 is a schematic structural diagram of a device for generating a product life cycle estimation model according to an embodiment of the present application;
FIG. 5 is a schematic structural diagram of a product life cycle estimation device according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
Reference will now be made in detail to embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the like or similar elements throughout. The embodiments described below by referring to the drawings are illustrative and intended to explain the present application and should not be construed as limiting the application.
Aiming at the problems that in the related art, when the life cycle of a new product is estimated, the sample data of the new product is expanded manually, the efficiency is low, and the estimated efficiency of the life cycle of the product is reduced, the embodiment of the application provides a method for generating an estimated model of the life cycle of the product.
According to the method for generating the product life cycle estimation model, the target feature vector corresponding to the target product is determined according to the newly-increased user quantity and the retention change rate of the target product in K days, the reference data set is obtained from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector of each existing product, and the preset life cycle estimation model is trained by the reference data set, so that the target estimation model corresponding to the target product is generated. Therefore, the target feature vector corresponding to the target product is determined through the existing sample data of the target product, the reference data set similar to the sample data of the target product is obtained from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector, and the target estimated model is trained, so that the sample data of the target product for the estimated model training is automatically expanded, the labor cost is saved, and the estimated efficiency of the product life cycle is improved.
The method for generating the product life cycle estimation model, the device, the electronic equipment, the storage medium and the computer program provided by the application are described in detail below with reference to the accompanying drawings.
The following describes in detail a method for generating a product life cycle estimation model according to an embodiment of the present application with reference to fig. 1.
Fig. 1 is a flowchart of a method for generating a product life cycle estimation model according to an embodiment of the present application.
As shown in FIG. 1, the method for generating the product life cycle estimation model comprises the following steps:
and step 101, determining a target feature vector corresponding to the target product according to the newly increased user quantity and the retention change rate of the target product on K days, wherein K is a positive integer.
The target product refers to a product needing life cycle estimation. For example, any internet product, especially a newly-on-line internet product, is possible.
Wherein K refers to any date from the current date to the current date of the target product on line. In practical use, K is treated as an integer, so K may be a positive integer. For example, the on-line time of the target product is 1 month and 1 day 2020, and the current date is 1 month and 5 days 2020, then K may be any one of 1 month and 1 day and 5 days 2020, for example, K is 4 months 2020, and then K is converted into a positive integer of 20200104.
The retention change rate refers to the slope of a straight line where corresponding points in the two-dimensional space exist in any two daily retention rates; wherein the X-axis of the two-dimensional space represents day, and the Y-axis of the two-dimensional space represents day retention. As a possible implementation manner, the retention change rate may be calculated according to the acquired newly increased user quantity of the target product on K days and a plurality of daily retention rates. That is, in one possible implementation manner of the embodiment of the present application, before the step 101, the method may further include:
obtaining the newly increased user quantity and a plurality of daily retention rates of a target product in K days;
and calculating the retention change rate of the target product according to the plurality of daily retention rates.
The Daily retention of the K day is the ratio of the user amount on line on any date after the K day to the newly added user amount (DNU) of the K day. For example, when the K day is 1 month and 1 day in 2020, DNU is 1000 in 1 month and 1 day in 2020, and 500 out of 1000 users newly added in 1 month and 1 day in 2020 are on line in 1 month and 2 days in 2020, the 1 day retention rate in 1 month and 1 day in 2020 is 0.5; correspondingly, the retention of 1 st month and 3 rd year in 2020 is 2 nd day retention of 1 st month and 1 st year in 2020, and so on.
In the embodiment of the application, DNU of all dates of the target product and a plurality of daily retention rates corresponding to DNU of each date can be obtained from the statistical data of the historical data of the target product. It should be noted that, since the target product is usually a new product, the history data may be relatively small, and only DNU and a plurality of daily retention rates on individual dates may be obtained. For example, as shown in table 1, the obtained sample data of the target product, i.e., DNU and 1 day retention and 3 days retention of the target product at 25 days of 1 month in 2019, are shown.
TABLE 1
Data item Date of day DNU 1 day retention 3 day retention
Numerical value 20190125 342 0.473684 0.368421
After obtaining the DNU of K days and a plurality of daily retention rates, the retention change rate of the target product can be calculated according to the plurality of daily retention rates. For example, as shown in a set of data in table 1, the slope of a straight line where a corresponding point in the two-dimensional space of the 1-day retention and the 3-day retention of the target product is: (0.368421-0.473684)/(3-1) = -0.052632, i.e. the retention rate of change of the target product is-0.052632.
As a possible implementation, the retention change rate during any two days, i.e. one retention change rate of the target product, may be calculated from any two of the plurality of daily retention rates. That is, in one possible implementation manner of the embodiment of the present application, calculating the retention change rate of the target product according to the plurality of daily retention rates may include:
and calculating the retention change rate of the target product between the ith and jth days according to the i-day retention and the j-day retention, wherein i and j are different integers respectively.
In the embodiment of the application, the retention change rate of the target product between the ith day and the jth day can be determined according to the i day retention rate, the j day retention rate and the difference between j and i. Namely, the retention change rate of the target product between the ith day and the jth day is as follows: (P) j -P i ) /(j-i), where P j For j days retention, P i The retention was i days.
The change rate between any two daily retention rates can be determined according to the plurality of daily retention rates corresponding to DNU of the target product on K days, so that when the obtained daily retention rates are more than 2, the plurality of retention change rates of the target product can be determined. That is, the number of retention change rates of the target product is the number of combinations of two of the plurality of daily retention rates arbitrarily taken out:where n is the number of daily survivors.
In the embodiment of the application, the target feature vector can be used for carrying out vector representation on DNU and retention change rate of the target product on K days. If a plurality of K days exist, DNU and target feature vectors corresponding to the retention change rate of each K day can be respectively determined.
In the embodiment of the application, three types of elements can be included in the target feature vector. Wherein, a class of elements is used for representing date attributes, a class of elements is used for representing DNU, and a class of elements is used for representing retention change rate.
As a possible implementation manner, the first element of the target feature vector may be used as an element for characterizing the date attribute, i.e., K may be used as the value of the first element of the target feature vector; the second element of the target feature vector can be adopted as an element for representing DNU, namely DNU on the K days can be adopted as the value of the second element of the target feature vector; and then taking the retention change rate of the target product as the value of other elements of the target feature vector, for example, taking the retention change rate of the target product as the value of the third element of the target feature vector if the retention change rate of the target product is 1, taking the retention change rate of the target product as the values of the third and fourth elements of the target feature vector if the retention change rate of the target product is 2, and so on.
For example, from the data shown in Table 1, it may be determined that the retention change rate of the target product is-0.052632, so that the target feature vector corresponding to the target product determined from the data shown in Table 1 is [20190125,342, -0.052632].
Further, because the user may have a large difference between the holiday and the working day on the use condition of the product, the element value of the date attribute in the target feature vector can be determined according to the holiday attribute of the K days. That is, in one possible implementation manner of the embodiment of the present application, the step 103 may include:
and determining the element value used for representing the date attribute in the target feature vector according to the holiday attribute of the K days.
As a possible implementation, the preset value may be used to represent the holiday attribute of K days, for example, 0 and 1 may be used to represent the holiday attribute of K days. Specifically, when the K day is "workday", a "0" may be determined as a holiday attribute of the K day, and a "0" may be determined as an element value for characterizing the date attribute in the target feature vector; when the K-day is "holiday", it is possible to determine "1" as the holiday attribute of the K-day and "1" as the element value for characterizing the date attribute in the target feature vector.
Further, the DNU may also be discretized to divide the DNU from a continuous value into a limited number of classification attributes. That is, in one possible implementation manner of the embodiment of the present application, the step 103 may include:
determining a weight corresponding to the newly added user quantity of the target product on K days according to a preset mapping relation between the newly added user quantity range on the days and the weight;
and assigning the weight value to an element used for representing the daily newly added user quantity in the target feature vector.
As one possible implementation, the DNU may be discretized using a discretization mechanism as shown in table 2. Table 2 shows a mapping relationship between a preset DNU range and a weight, so that the weight corresponding to the DNU of the target product on the K day can be determined according to the DNU of the target product on the K day in table 2, and then the weight corresponding to the DNU of the target product on the K day is determined as an element value for representing the DNU in the target feature vector.
TABLE 2
DNU <4 <15 <50 <100 <400 <1000 <2000 <3000 <4000 <5000 >＝5000
Weight value 10 9 8 7 6 5 4 3 2 1 0
For example, according to the data shown in table 1, it may be determined that the retention change rate of the target product is-0.052632, and 25 in 2019 is "working day", so that when the first element of the target feature vector is 0 and dnu is 342, the corresponding weight is 6, and thus the target feature vector corresponding to the target product determined according to the data shown in table 1 is [0.000000,6.000000, -0.052632].
Step 102, obtaining a reference data set from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector of each existing product.
The reference feature vector is a feature vector generated according to the DUN and the retention rate of the existing product on each day. The method for generating each reference feature vector is the same as the method for generating the target feature vector corresponding to the target product, that is, the DUN on each day and the plurality of day retention rates corresponding to the DUN on each day included in the history data can be obtained according to the history data of the existing product, and then each reference feature vector is determined according to the DUN on each day and the plurality of day retention rates corresponding to the DUN on each day.
The similarity between the target feature vector and the reference feature vector can be measured by using a vector distance between the target feature vector and the reference feature vector. Specifically, if the vector distance between the target feature vector and the reference feature vector is smaller, the higher the similarity between the target feature vector and the reference feature vector is indicated; if the vector distance between the target feature vector and the reference feature vector is larger, the similarity between the target feature vector and the reference feature vector is lower.
As a possible implementation manner, the euclidean distance between the target feature vector and the reference feature vector may be used, the similarity between the target feature vector and the reference feature vector is measured, and the reference data with the euclidean distance smaller than the distance threshold value between the reference feature vector and the target feature vector is obtained from the historical data of the existing product, so as to form the reference data set.
When similarity between the target feature vector and the reference feature vector is measured by using the euclidean distance, the number of elements in the reference feature vector needs to be the same as the number of elements in the target feature vector, so that the reference feature vector with the same number of elements as the number of elements of the target feature vector can be selected from the reference feature vector corresponding to each data in the existing product history data, and further the euclidean distance between the target feature vector and each selected reference feature vector is determined.
In actual use, the distance threshold may be preset according to actual needs, which is not limited in the embodiment of the present application. For example, the distance threshold may be 0.04.
And step 103, training a preset life cycle estimation model by using the reference data set to generate a target estimation model corresponding to the target product.
In the embodiment of the application, because each reference data in the reference data set is determined according to the similarity between the corresponding target feature vector of the target product and each reference feature vector of the existing product, each reference data in the reference data set has higher similarity with the historical data of the target product. Therefore, the life cycle of the target product can be effectively predicted by using the target estimated model trained by the reference data set.
As a possible implementation manner, the preset life cycle estimation model may be obtained by integrating a daily retention decay model shown in formula (1):
wherein y is x-day retention, x is an independent variable, the value range of x is related to the type of the target product, and in actual use, the value range of x can be determined according to the specific target product, for example, the value range of x can be an integer of [0,360 ], a, b 1 And b 2 For the parameters in the pre-set life cycle pre-estimated model, L is the life cycle, the process of training the target pre-estimated model is to determine a and b 1 And b 2 Is a specific value process.
In the embodiment of the application, a preset life cycle estimation model is trained by utilizing a reference data set, each day retention rate contained in one reference data is used as a y value, and the day corresponding to each day retention rate is used as an x value to be substituted into a formula (1) to determine a group of a and b 1 And b 2 Is a value of (2); alternatively, if the number of daily retention in one reference data is small, a and b cannot be determined from one reference data 1 And b 2 The specific value of (a) and (b) can be determined by determining a group of a and b by the daily retention rate included in the plurality of reference data 1 And b 2 Is a value of (a).
In the case of the determined a, b 1 And b 2 When the values of (a) are multiple, multiple groups of a,b 1 And b 2 The average value of (a) is determined as a, b in the target estimated model 1 And b 2 Specific values of (2). Determining a and b in a target pre-estimated model 1 And b 2 After the specific value of (3), a target pre-estimated model can be generated.
According to the method for generating the product life cycle estimation model, the target feature vector corresponding to the target product is determined according to the newly-increased user quantity and the retention change rate of the target product in K days, the reference data set is obtained from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector of each existing product, and the preset life cycle estimation model is trained by the reference data set, so that the target estimation model corresponding to the target product is generated. Therefore, the target feature vector corresponding to the target product is determined through the existing sample data of the target product, the reference data set similar to the sample data of the target product is obtained from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector, and the target estimated model is trained, so that the sample data of the target product for the estimated model training is automatically expanded, the labor cost is saved, and the estimated efficiency of the product life cycle is improved.
In one possible implementation form of the application, the reference data set obtained from the historical data of the existing product may include a plurality of reference data, so that the plurality of reference data may be subjected to fusion processing to reduce the number of times and complexity of model training.
The method for generating the product life cycle estimation model provided by the embodiment of the application is further described below with reference to fig. 2.
Fig. 2 is a flowchart of another method for generating a product life cycle estimation model according to an embodiment of the present application.
As shown in FIG. 2, the method for generating the product life cycle estimation model comprises the following steps:
step 201, determining a target feature vector corresponding to the target product according to the newly increased user quantity and the retention change rate of the target product on K days, wherein K is a positive integer.
Step 202, obtaining a reference data set from historical data of each existing product according to the similarity between the target feature vector and each reference feature vector of each existing product.
The specific implementation and principles of the steps 201 to 202 may refer to the detailed description of the embodiments, and are not repeated here.
And 203, carrying out fusion processing on each same daily retention rate in each reference data in the reference data set to generate a training sample set.
In the embodiment of the application, if more reference data are included in the reference data set, the preset life cycle estimated model is respectively trained by using the daily retention rate in each reference data, so that the model training times are more, and the calculation amount and the complexity of the model training are increased. Thus, in one possible implementation form of the embodiment of the present application, the same daily retention rate included in each reference data may be subjected to a fusion process to generate a training sample set including a plurality of fused daily retention rates.
For example, if the reference data set includes three reference data as shown in table 3, the 1 day retention a can be obtained by performing the fusion processing for each same day retention in each reference data 1 、b 1 、c 1 Fusing to generate 1 day retention rate after fusing, and obtaining 2 day retention rate a 2 、b 2 、c 2 Fusing to generate 2-day retention rate after fusing, and obtaining 3-day retention rate a 3 、b 3 、c 3 And fusing to generate 3-day retention after fusion, wherein the generated training sample set comprises three sample data of 1-day retention after fusion, 2-day retention after fusion and 3-day retention after fusion.
TABLE 3 Table 3
Date of day DNU 1 day retention Retention of 2 days 3 day retention
Reference data 1 Day A M 1 a 1 a 2 a 3
Reference data 2 Day B M 2 b 1 b 2 b 3
Reference data 3 Day C M 3 c 1 c 2 c 3
As a possible implementationIn this way, the average process may be performed to sum up and average each identical daily retention rate in each reference data to fuse each identical daily retention rate in each reference data. For example, when three reference data in Table 3 are fused by addition and averaging, the 1-day retention after fusion is N 1 ＝(a 1 +b 1 +c 1 ) 3, 2-day retention after fusion of N 2 ＝(a 2 +b 2 +c 2 ) 3, 3-day retention after fusion of N 3 ＝(a 3 +b 3 +c 3 ) 3, i.e. the training sample set generated is { (x=1, y=n) 1 ),(x＝2,y＝N 2 ),(x＝3,y＝N 3 )}。
As a possible implementation manner, the weight of each reference data may be determined according to the similarity between the existing product corresponding to each reference data and the target product, and the weighted average process may be performed on each same daily retention rate in each reference data according to the weight of each reference data.
It should be noted that, to ensure accuracy of model training, a reference data set for model training may be obtained from historical data of a plurality of existing products, where the type, application field, and the like of each existing product may be different from those of the target product, so that the weight of each reference data may be determined according to the similarity between the existing product and the target product.
Specifically, the higher the similarity between the existing product and the target product is, the higher the weight of the reference data corresponding to the existing product is; otherwise, the lower the weight of the reference data corresponding to the existing product is.
For example, the reference data 1 and the reference data 2 in table 3 are obtained from the historical data of a certain social software, the reference data 3 is obtained from the historical data of a certain online game, and the target product is a social software product, so that the reference data 1 and the reference data 2 can be determined to have the same weight and be greater than the weight of the reference data 3. For example, it can be determined that the weight of reference data 1 and reference data 2 is 0.7, the weight of reference data 3 is 0.3, and thus the 1-day retention after fusion is N 1 ＝(0.7a 1 +0.7b 1 +0.3c 1 ) 3, 2-day retention after fusion of N 2 ＝(0.7a 2 +0.7b 2 +0.3c 2 ) 3, 3-day retention after fusion of N 3 ＝(0.7a 3 +0.7b 3 +0.3c 3 )/3。
Step 204, training the preset life cycle estimation model by using the training sample set, and generating a target estimation model corresponding to the target product.
In the embodiment of the application, after fusion processing is performed on each same daily retention rate in each reference data to generate a training sample set, each training sample (i.e. each group of x and y values) in the training sample set may be substituted into a preset life cycle estimation model to determine parameters a and b included in the preset life cycle model 1 And b 2 And then generating a target pre-estimation model corresponding to the target product.
For example, the training sample set determined is { (x=1, y=n) 1 ),(x＝2,y＝N 2 ),(x＝3,y＝N 3 ) And (x=1, y=n) 1 )、(x＝2,y＝N 2 )、(x＝3,y＝N 3 ) The three groups of data are respectively substituted into the formula (1) to generate the parameters a and b 1 And b 2 Further, parameters a and b are calculated according to the equation formed by the three equations 1 And b 2 Is a value of (a).
According to the method for generating the product life cycle estimation model, the target feature vector corresponding to the target product is determined according to the newly-increased user quantity and the retention change rate of the target product in K days, the reference data set is obtained from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector of each existing product, and then fusion processing is carried out on each identical day retention rate in each reference data in the reference data set to generate a training sample set, and further the training sample set is utilized to train a preset life cycle estimation model to generate the target estimation model corresponding to the target product. Therefore, according to the similarity between the target feature vector and each reference feature vector, a reference data set similar to sample data of the target product is obtained from historical data of each existing product, and after fusion processing is carried out on each reference data in the reference data set, the target estimated model is trained by utilizing the fused reference data, so that the sample data of the target product is automatically expanded, the labor cost is saved, the calculation amount and the complexity of the estimated model training are reduced, and the estimated efficiency of the life cycle of the product is further improved.
The following describes in detail the product life cycle estimation method provided by the embodiment of the present application with reference to fig. 3.
Fig. 3 is a flowchart of a product life cycle estimation method according to an embodiment of the present application.
As shown in fig. 3, the product life cycle estimation method includes the following steps:
step 301, a life cycle estimation request is obtained, wherein the estimation request comprises a type and an identification of a target product.
It should be noted that, after training, the product life cycle estimation model in the embodiment of the present application may be configured in any electronic device, so as to estimate the life cycle of the target product when the life cycle estimation request is obtained.
The types of the target products can be divided according to the application fields of the target products. For example, the types of target products may include social categories, gaming categories, news categories, and so forth. In actual use, the types of the target products can be divided according to specific application scenes, and the embodiment of the application does not limit the types.
The identification of the target product may be information that the name, version number, etc. of the target product can uniquely determine the target product.
In the embodiment of the application, the life cycle estimation request can be obtained through the input equipment (such as a keyboard, a mouse, a touch screen and the like) of the electronic equipment where the product life cycle estimation model is located. As one possible implementation, the product lifecycle prediction model may be configured in an electronic device for use by a user in the form of an application. Specifically, when the user clicks the application program space of the product life cycle estimation model, a life cycle estimation request input control is provided in an application program interface, and the user is allowed to input information such as the type and the identification of the target product to be estimated currently, so that a life cycle estimation request is generated according to the obtained information such as the type and the identification of the target product, and the acquisition of the life cycle estimation request is completed.
Step 302, determining a duration to be estimated corresponding to the target product according to the type of the target product.
In the embodiment of the application, because of the life cycle of different types of target products, large differences can exist, such as longer life cycle of social products and news products, and shorter life cycle of game products; the life cycle of the product is obtained by integrating the daily retention rate of the product, so that the time range for estimating the life cycle of the target product can be determined first before estimating the life cycle of the target product.
As a possible implementation manner, the life cycle of the existing product of the same type as the target product can be determined according to the type of the target product, and then the duration to be estimated corresponding to the target product is determined according to the life cycle of the product of the same type as the target product. For example, the average life cycle of each product of the same type as the target product can be determined as the duration to be estimated corresponding to the target product.
Furthermore, the mapping relation between the product type and the estimated time length can be preset in advance, so that the time length to be estimated corresponding to the target product can be determined conveniently, the calculation complexity of product life cycle estimation is reduced, and the efficiency of product life cycle estimation is improved. That is, in one possible implementation manner of the embodiment of the present application, before the step 302, the method may further include:
And counting the life cycle of each type of product, and determining the duration to be estimated corresponding to each product type.
As a possible implementation manner, the life cycle of each existing product type can be counted, and the counted result is determined as the duration to be estimated corresponding to each product type. For example, the average life cycle of the existing various types of products can be determined, and the average life cycle is determined as the duration to be estimated corresponding to each product type.
Step 303, calculating the daily retention rate of the target product in the duration to be estimated by using the target estimation model corresponding to the identification of the target product.
In the embodiment of the application, the life cycle of different target products can be estimated by using the same estimated model, a plurality of estimated models can be trained in advance, and different estimated models can be used for estimating different target products. Therefore, in the embodiment of the application, the mapping relation between the product identifier and the pre-estimated model can be preset, so that the pre-estimated model corresponding to the identifier of the target product can be determined as the target pre-estimated model according to the identifier of the target product, and the daily retention rate of the target product in the duration to be pre-estimated is calculated by using the target pre-estimated model.
For example, if the duration to be estimated of the target product is 360 days, 1-360 may be substituted into the target estimation model to determine the 1-day retention rate to 360-day retention rate of the target product.
Step 304, determining the life cycle of the target product according to the daily retention rate of the target product in the duration to be estimated.
In the embodiment of the application, after determining the daily retention rate of the target product in the duration to be estimated, the sum of the daily retention rates of the target product in the duration to be estimated can be determined as the life cycle of the target product.
For example, if the duration to be estimated of the target product is 360 days, the sum of the 1-day retention rate and the 360-day retention rate of the target product may be determined as the life cycle of the target product.
According to the product life cycle estimating method provided by the embodiment of the application, the life cycle estimating request comprising the type and the identification of the target product is obtained, the duration to be estimated corresponding to the target product is determined according to the type of the target product, then the daily retention rate of the target product in the duration to be estimated is calculated by utilizing the target estimating model corresponding to the identification of the target product, and the life cycle of the target product is determined according to the daily retention rate of the target product in the duration to be estimated. Therefore, the life cycle of the target product is estimated through the pre-trained product life cycle estimation model, so that the labor cost is saved, and the product life cycle estimation efficiency is improved. In order to achieve the above embodiment, the present application further provides a device for generating a product life cycle estimation model.
Fig. 4 is a schematic structural diagram of a device for generating a product life cycle estimation model according to an embodiment of the present application.
As shown in fig. 4, the product life cycle estimation model generating device 40 includes:
the determining module 41 is configured to determine a target feature vector corresponding to the target product according to the newly increased user quantity and the retention change rate of the target product on K days, where K is a positive integer;
a first obtaining module 42, configured to obtain a reference data set from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector of each existing product;
the training module 43 is configured to train a preset life cycle estimation model by using the reference data set, and generate a target estimation model corresponding to the target product.
In practical use, the device for generating the product life cycle estimation model provided by the embodiment of the application can be configured in any electronic equipment to execute the method for generating the product life cycle estimation model.
According to the product life cycle estimation model generation device provided by the embodiment of the application, the target feature vector corresponding to the target product is determined according to the newly-increased user quantity and the retention change rate of the target product in K days, the reference data set is obtained from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector of each existing product, and the preset life cycle estimation model is trained by using the reference data set, so that the target estimation model corresponding to the target product is generated. Therefore, the target feature vector corresponding to the target product is determined through the existing sample data of the target product, the reference data set similar to the sample data of the target product is obtained from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector, and the target estimated model is trained, so that the sample data of the target product for the estimated model training is automatically expanded, the labor cost is saved, and the estimated efficiency of the product life cycle is improved.
In one possible implementation form of the present application, the product life cycle estimation model generating device 40 further includes:
the second acquisition module is used for acquiring the newly increased user quantity and the multiple daily retention rate of the target product in K days;
and the calculating module is used for calculating the retention change rate of the target product according to the plurality of daily retention rates.
Further, in another possible implementation form of the present application, the computing module is specifically configured to:
and calculating the retention change rate of the target product between the ith and jth days according to the i-day retention and the j-day retention, wherein i and j are different integers respectively.
Further, in still another possible implementation form of the present application, the determining module 41 is specifically configured to:
determining a weight corresponding to the newly added user quantity of the target product on K days according to a preset mapping relation between the newly added user quantity range on the days and the weight;
and assigning the weight value to an element used for representing the daily newly added user quantity in the target feature vector.
Further, in still another possible implementation form of the present application, the determining module 41 is further configured to:
and determining the element value used for representing the date attribute in the target feature vector according to the holiday attribute of the K days.
In one possible implementation form of the present application, each reference data in the reference data set includes a daily retention rate of the existing product; correspondingly, the product life cycle estimation model generating device 40 further includes:
the fusion processing module is used for carrying out fusion processing on each same daily retention rate in each reference data to generate a training sample set;
correspondingly, the training module 43 is specifically configured to:
and training the preset life cycle estimation model by using the training sample set.
Further, in another possible implementation form of the present application, the above fusion processing module is specifically configured to:
adding and averaging the same daily retention rate of each reference data;
or alternatively, the process may be performed,
determining the weight of each reference data according to the similarity between the existing product corresponding to each reference data and the target product;
and carrying out weighted average processing on each same daily retention rate in each reference data according to the weight of each reference data.
It should be noted that the foregoing explanation of the embodiment of the method for generating a product life cycle estimation model shown in fig. 1 and 2 is also applicable to the device 40 for generating a product life cycle estimation model of this embodiment, and will not be repeated here.
According to the product life cycle prediction model generation device provided by the embodiment of the application, the target feature vector corresponding to the target product is determined according to the newly increased user quantity and the retention change rate of the target product in K days, the reference data set is obtained from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector of each existing product, and then the retention rate of each same day in each reference data in the reference data set is fused to generate a training sample set, so that the training sample set is utilized to train the preset life cycle prediction model, and the target prediction model corresponding to the target product is generated. Therefore, according to the similarity between the target feature vector and each reference feature vector, a reference data set similar to sample data of the target product is obtained from historical data of each existing product, and after fusion processing is carried out on each reference data in the reference data set, the target estimated model is trained by utilizing the fused reference data, so that the sample data of the target product is automatically expanded, the labor cost is saved, the calculation amount and the complexity of the estimated model training are reduced, and the estimated efficiency of the life cycle of the product is further improved.
In order to implement the above embodiment, the present application further provides a product life cycle estimating device.
Fig. 5 is a schematic structural diagram of a product life cycle estimating device according to an embodiment of the present application.
As shown in fig. 5, the product life cycle estimating device 50 includes:
the obtaining module 51 is configured to obtain a lifecycle prediction request, where the prediction request includes a type and an identifier of a target product;
the first determining module 52 is configured to determine a duration to be estimated corresponding to the target product according to the type of the target product;
the calculating module 53 is configured to calculate a daily retention rate of the target product within a duration to be estimated using a target estimation model corresponding to the identifier of the target product;
the second determining module 54 is configured to determine a life cycle of the target product according to a daily retention rate of the target product in the duration to be estimated.
In practical use, the product life cycle estimating device provided by the embodiment of the application can be configured in any electronic equipment to execute the product life cycle estimating method.
In one possible implementation form of the present application, the product life cycle estimating device 50 further includes:
and the third determining module is used for counting the life cycle of each type of product and determining the duration to be estimated corresponding to each product type.
According to the product life cycle estimating device provided by the embodiment of the application, the life cycle estimating request comprising the type and the identifier of the target product is obtained, the duration to be estimated corresponding to the target product is determined according to the type of the target product, then the daily retention rate of the target product in the duration to be estimated is calculated by utilizing the target estimating model corresponding to the identifier of the target product, and the life cycle of the target product is determined according to the daily retention rate of the target product in the duration to be estimated. Therefore, the life cycle of the target product is estimated through the pre-trained product life cycle estimation model, so that the labor cost is saved, and the product life cycle estimation efficiency is improved.
In order to achieve the above embodiment, the present application further provides an electronic device.
Fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
As shown in fig. 6, the electronic device 200 includes:
the memory 210 and the processor 220, the bus 230 connecting different components (including the memory 210 and the processor 220), the memory 210 stores a computer program, and when the processor 220 executes the program, the product life cycle estimation model generating method or the product life cycle estimation method according to the embodiments of the present application is implemented.
Bus 230 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus, micro channel architecture (MAC) bus, enhanced ISA bus, video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.
Electronic device 200 typically includes a variety of electronic device readable media. Such media can be any available media that is accessible by electronic device 200 and includes both volatile and nonvolatile media, removable and non-removable media.
Memory 210 may also include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM) 240 and/or cache memory 250. The electronic device 200 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 260 may be used to read from or write to non-removable, nonvolatile magnetic media (not shown in FIG. 6, commonly referred to as a "hard disk drive"). Although not shown in fig. 6, a magnetic disk drive for reading from and writing to a removable non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading from or writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In such cases, each drive may be coupled to bus 230 via one or more data medium interfaces. Memory 210 may include at least one program product having a set (e.g., at least one) of program modules configured to carry out the functions of embodiments of the application.
Program/utility 280 having a set (at least one) of program modules 270 may be stored in, for example, memory 210, such program modules 270 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment. Program modules 270 generally perform the functions and/or methods of the embodiments described herein.
The electronic device 200 may also communicate with one or more external devices 290 (e.g., keyboard, pointing device, display 291, etc.), one or more devices that enable a user to interact with the electronic device 200, and/or any device (e.g., network card, modem, etc.) that enables the electronic device 200 to communicate with one or more other computing devices. Such communication may occur through an input/output (I/O) interface 292. Also, electronic device 200 may communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the Internet, through network adapter 293. As shown, network adapter 293 communicates with other modules of electronic device 200 over bus 230. It should be appreciated that although not shown, other hardware and/or software modules may be used in connection with electronic device 200, including, but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, data backup storage systems, and the like.
The processor 220 executes various functional applications and data processing by running programs stored in the memory 210.
It should be noted that, the implementation process and the technical principle of the electronic device in this embodiment refer to the foregoing explanation of the product life cycle estimation model generation method or the product life cycle estimation method in the embodiment of the present application, and are not repeated herein.
The electronic equipment provided by the embodiment of the application can execute the method for generating the product life cycle estimation model, the target feature vector corresponding to the target product is determined according to the newly added user quantity and the retention change rate of the target product in K days, the reference data set is obtained from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector of each existing product, and the preset life cycle estimation model is trained by using the reference data set, so that the target estimation model corresponding to the target product is generated. Therefore, the target feature vector corresponding to the target product is determined through the existing sample data of the target product, the reference data set similar to the sample data of the target product is obtained from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector, and the target estimated model is trained, so that the sample data of the target product for the estimated model training is automatically expanded, the labor cost is saved, and the estimated efficiency of the product life cycle is improved.
In order to implement the above embodiments, the present application also proposes a computer-readable storage medium.
The computer readable storage medium stores a computer program which, when executed by a processor, implements the method for generating a product life cycle estimation model or the method for estimating a product life cycle according to the embodiments of the present application.
In order to achieve the foregoing embodiments, an embodiment of the present application provides a computer program, which when executed by a processor, implements a product life cycle estimation method according to the embodiment of the present application.
In alternative implementations, the present embodiments may employ any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the consumer electronic device, partly on the consumer electronic device, as a stand-alone software package, partly on the consumer electronic device and partly on the remote electronic device or entirely on the remote electronic device or server. In the case of remote electronic devices, the remote electronic device may be connected to the consumer electronic device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external electronic device (e.g., connected through the internet using an internet service provider).
Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. This application is intended to cover any variations, uses, or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the application pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.
It is to be understood that the application is not limited to the precise arrangements and instrumentalities shown in the drawings, which have been described above, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims (18)

1. The method for generating the product life cycle estimation model is characterized by comprising the following steps of:
determining a target feature vector corresponding to a target product according to the newly increased user quantity and the retention change rate of the target product in K days, wherein K is a positive integer;
acquiring a reference data set from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector of each existing product;
Training a preset life cycle pre-estimated model by using the reference data set to generate a target pre-estimated model corresponding to the target product;
the determining the target feature vector corresponding to the target product comprises the following steps:
and determining element values used for representing date attributes in the target feature vector according to the holiday attributes of the K days.
2. The method of claim 1, wherein before determining the target feature vector corresponding to the target product according to the newly added user quantity and the retention change rate of the target product on K days, further comprises:
obtaining the newly increased user quantity of the target product in K days and a plurality of daily retention rates;
and calculating the retention change rate of the target product according to the plurality of daily retention rates.
3. The method of claim 2, wherein calculating a retention rate of change of the target product based on the plurality of daily retention rates comprises:
and calculating the retention change rate of the target product between the ith and jth days according to the i-day retention and the j-day retention, wherein i and j are different integers respectively.
4. The method of claim 1, wherein the determining the target feature vector for the target product comprises:
Determining a weight corresponding to the newly added user quantity of the target product on K days according to a preset mapping relation between the newly added user quantity range on the days and the weight;
and assigning the weight to an element used for representing the daily newly added user quantity in the target feature vector.
5. The method of any of claims 1-4, wherein each reference data in the reference data set includes a daily retention of an existing product;
after the reference data set is acquired, the method further comprises:
carrying out fusion processing on each same daily retention rate in each reference data to generate a training sample set;
the training the preset life cycle estimation model by using the reference data set includes:
and training a preset life cycle estimation model by using the training sample set.
6. The method of claim 5, wherein said fusing each of the same daily retention in the respective reference data comprises:
adding and averaging the same daily retention rate of each reference data;
or alternatively, the process may be performed,
determining the weight of each reference data according to the similarity between the existing product corresponding to each reference data and the target product;
And carrying out weighted average processing on each same daily retention rate in each reference data according to the weight of each reference data.
7. A method for product lifecycle estimation, comprising:
acquiring a life cycle estimation request, wherein the estimation request comprises the type and the identification of a target product;
determining the duration to be estimated corresponding to the target product according to the type of the target product;
calculating the daily retention rate of the target product in the duration to be estimated by using a target estimated model corresponding to the identification of the target product, wherein the target estimated model is generated by training according to the product life cycle estimated model generation method of any one of the claims 1-6;
and determining the life cycle of the target product according to the daily retention rate of the target product in the duration to be estimated.
8. The method of claim 7, wherein before determining the duration to be estimated corresponding to the target product according to the type of the target product, further comprises:
and counting the life cycle of each type of product, and determining the duration to be estimated corresponding to each product type.
9. A product lifecycle prediction model generating apparatus, comprising:
The determining module is used for determining a target feature vector corresponding to a target product according to the newly added user quantity and the retention change rate of the target product on K days, wherein K is a positive integer;
the first acquisition module is used for acquiring a reference data set from the historical data of each existing product according to the similarity between the target feature vector and each reference feature vector of each existing product;
the training module is used for training a preset life cycle estimation model by utilizing the reference data set to generate a target estimation model corresponding to the target product;
the determining module is further configured to:
and determining element values used for representing date attributes in the target feature vector according to the holiday attributes of the K days.
10. The apparatus as recited in claim 9, further comprising:
the second acquisition module is used for acquiring the newly increased user quantity and a plurality of daily retention rates of the target product in K days;
and the calculating module is used for calculating the retention change rate of the target product according to the plurality of daily retention rates.
11. The apparatus of claim 10, wherein the computing module is configured to:
and calculating the retention change rate of the target product between the ith and jth days according to the i-day retention and the j-day retention, wherein i and j are different integers respectively.
12. The apparatus of claim 9, wherein the determining module is specifically configured to:
determining a weight corresponding to the newly added user quantity of the target product on K days according to a preset mapping relation between the newly added user quantity range on the days and the weight;
and assigning the weight to an element used for representing the daily newly added user quantity in the target feature vector.
13. The apparatus of any of claims 9-12, wherein each reference data in the reference data set includes a daily retention of an existing product;
the device further comprises:
the fusion processing module is used for carrying out fusion processing on each same daily retention rate in each reference data to generate a training sample set;
the training module is specifically configured to:
and training a preset life cycle estimation model by using the training sample set.
14. The apparatus of claim 13, wherein the fusion processing module is specifically configured to:
adding and averaging the same daily retention rate of each reference data;
or alternatively, the process may be performed,
determining the weight of each reference data according to the similarity between the existing product corresponding to each reference data and the target product;
And carrying out weighted average processing on each same daily retention rate in each reference data according to the weight of each reference data.
15. A product life cycle estimation device, comprising:
the acquisition module is used for acquiring a life cycle estimation request, wherein the estimation request comprises the type and the identification of a target product;
the first determining module is used for determining the duration to be estimated corresponding to the target product according to the type of the target product;
the computing module is used for computing the daily retention rate of the target product in the duration to be estimated by using a target estimated model corresponding to the identification of the target product, and the target estimated model is generated by training according to the product life cycle estimated model generating method of any one of the claims 1-6;
and the second determining module is used for determining the life cycle of the target product according to the daily retention rate of the target product in the duration to be estimated.
16. The apparatus as recited in claim 15, further comprising:
and the third determining module is used for counting the life cycle of each type of product and determining the duration to be estimated corresponding to each product type.
17. An electronic device, comprising: a memory, a processor and a program stored in the memory and executable on the processor, wherein the processor implements the method for generating a product life cycle estimation model according to any one of claims 1 to 6 or the method for estimating a product life cycle according to any one of claims 7 to 8 when executing the program.
18. A computer readable storage medium having stored thereon a computer program, wherein the program when executed by a processor implements the method of generating a product lifecycle estimation model according to any one of claims 1-6 or the method of product lifecycle estimation according to any one of claims 7-8.
CN202010108556.8A 2020-02-21 2020-02-21 Product life cycle prediction model generation method and device and electronic equipment Active CN111291936B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010108556.8A CN111291936B (en) 2020-02-21 2020-02-21 Product life cycle prediction model generation method and device and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010108556.8A CN111291936B (en) 2020-02-21 2020-02-21 Product life cycle prediction model generation method and device and electronic equipment

Publications (2)

Publication Number Publication Date
CN111291936A CN111291936A (en) 2020-06-16
CN111291936B true CN111291936B (en) 2023-10-17

Family

ID=71020570

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010108556.8A Active CN111291936B (en) 2020-02-21 2020-02-21 Product life cycle prediction model generation method and device and electronic equipment

Country Status (1)

Country Link
CN (1) CN111291936B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111784008A (en) * 2020-06-30 2020-10-16 北京金山安全软件有限公司 Product life cycle estimation method and device, electronic equipment and storage medium
CN114742569A (en) * 2021-01-08 2022-07-12 广州视源电子科技股份有限公司 User life stage prediction method and device, computer equipment and storage medium
CN113269370B (en) * 2021-06-18 2023-12-12 腾讯科技(成都)有限公司 Active user prediction method and device, electronic equipment and readable storage medium

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8326673B1 (en) * 2006-12-28 2012-12-04 Sprint Communications Company L.P. Carrier data based product inventory management and marketing
CN104759100A (en) * 2015-03-23 2015-07-08 小米科技有限责任公司 Prediction method and device of game life cycle and game operator server
CN104767811A (en) * 2015-04-03 2015-07-08 北京奇虎科技有限公司 Information recommending method and device
WO2017019834A1 (en) * 2015-07-28 2017-02-02 Mastercard International Incorporated Enhanced smart refrigerator systems and methods
CN106897904A (en) * 2017-02-24 2017-06-27 北京金山安全软件有限公司 Product life cycle modeling method and device and electronic equipment
CN107464023A (en) * 2017-08-16 2017-12-12 新疆金风科技股份有限公司 The on-line study method and apparatus of life cycle forecast model
CN108596652A (en) * 2018-03-28 2018-09-28 麒麟合盛网络技术股份有限公司 Active users prediction technique and device
CN108805313A (en) * 2017-04-26 2018-11-13 株式会社日立制作所 Needing forecasting method, device towards global supply chain and system
CN110503290A (en) * 2019-04-12 2019-11-26 北京理工大学 The twin volume data management method of number towards product lifecycle

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030018952A1 (en) * 2001-07-13 2003-01-23 Roetzheim William H. System and method to estimate resource usage for a software development project
US20160283953A1 (en) * 2015-03-26 2016-09-29 International Business Machines Corporation Demand-supply matching with a time and virtual space network
US10712738B2 (en) * 2016-05-09 2020-07-14 Strong Force Iot Portfolio 2016, Llc Methods and systems for industrial internet of things data collection for vibration sensitive equipment

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8326673B1 (en) * 2006-12-28 2012-12-04 Sprint Communications Company L.P. Carrier data based product inventory management and marketing
CN104759100A (en) * 2015-03-23 2015-07-08 小米科技有限责任公司 Prediction method and device of game life cycle and game operator server
CN104767811A (en) * 2015-04-03 2015-07-08 北京奇虎科技有限公司 Information recommending method and device
WO2017019834A1 (en) * 2015-07-28 2017-02-02 Mastercard International Incorporated Enhanced smart refrigerator systems and methods
CN106897904A (en) * 2017-02-24 2017-06-27 北京金山安全软件有限公司 Product life cycle modeling method and device and electronic equipment
CN108805313A (en) * 2017-04-26 2018-11-13 株式会社日立制作所 Needing forecasting method, device towards global supply chain and system
CN107464023A (en) * 2017-08-16 2017-12-12 新疆金风科技股份有限公司 The on-line study method and apparatus of life cycle forecast model
CN108596652A (en) * 2018-03-28 2018-09-28 麒麟合盛网络技术股份有限公司 Active users prediction technique and device
CN110503290A (en) * 2019-04-12 2019-11-26 北京理工大学 The twin volume data management method of number towards product lifecycle

Also Published As

Publication number Publication date
CN111291936A (en) 2020-06-16

Similar Documents

Publication Publication Date Title
CN109670267B (en) Data processing method and device
CN111291936B (en) Product life cycle prediction model generation method and device and electronic equipment
US9195910B2 (en) System and method for classification with effective use of manual data input and crowdsourcing
CN108399564B (en) Credit scoring method and device
US20190340641A1 (en) Simulation-based evaluation of a marketing channel attribution model
CN108629608B (en) User data processing method and device
CN107103049A (en) A kind of recommendation method and the network equipment
CN110020773A (en) A kind of data processing method, equipment and computer-readable medium
CN110889725B (en) Online advertisement CTR estimation method, device, equipment and storage medium
CN112883257B (en) Behavior sequence data processing method and device, electronic equipment and storage medium
CN113538154A (en) Risk object identification method and device, storage medium and electronic equipment
CN109582906A (en) Determination method, apparatus, equipment and the storage medium of data reliability
CN111159241A (en) Click conversion estimation method and device
CN108647714A (en) Acquisition methods, terminal device and the medium of negative label weight
CN111160638A (en) Conversion estimation method and device
CN109711849B (en) Ether house address portrait generation method and device, electronic equipment and storage medium
CN114997414A (en) Data processing method and device, electronic equipment and storage medium
CN108021713B (en) Document clustering method and device
KR20220006580A (en) Visit Prediction
CN115345600B (en) RPA flow generation method and device
JP2023029604A (en) Apparatus and method for processing patent information, and program
CN111125272B (en) Regional characteristic acquisition method, regional characteristic acquisition device, computer equipment and medium
CN111191142B (en) Electronic resource recommendation method and device and readable medium
CN113742564A (en) Target resource pushing method and device
CN116383883B (en) Big data-based data management authority processing method and system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant