CN107463660B - Method for measuring and calculating data of active users of products and computer equipment - Google Patents
Method for measuring and calculating data of active users of products and computer equipment Download PDFInfo
- Publication number
- CN107463660B CN107463660B CN201710640388.5A CN201710640388A CN107463660B CN 107463660 B CN107463660 B CN 107463660B CN 201710640388 A CN201710640388 A CN 201710640388A CN 107463660 B CN107463660 B CN 107463660B
- Authority
- CN
- China
- Prior art keywords
- active user
- product
- amplification
- user data
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 31
- 230000003321 amplification Effects 0.000 claims abstract description 143
- 238000003199 nucleic acid amplification method Methods 0.000 claims abstract description 143
- 238000013178 mathematical model Methods 0.000 claims abstract description 68
- 238000012549 training Methods 0.000 claims abstract description 46
- 230000001419 dependent effect Effects 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000005259 measurement Methods 0.000 claims description 2
- 239000000047 product Substances 0.000 description 174
- 238000012544 monitoring process Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000002860 competitive effect Effects 0.000 description 2
- 238000005034 decoration Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003442 weekly effect Effects 0.000 description 2
- 241000282376 Panthera tigris Species 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 230000037211 monthly cycles Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a method for measuring and calculating product active user data. The product active user data measuring and calculating method comprises the steps of obtaining a first training sample set; training the first mathematical model by using a first training sample set to obtain parameters representing daily active user data and ranking sequence relation of the first mathematical model; acquiring a second training sample set; training a second mathematical model by using a second training sample set to obtain parameters of the relationship between the data amplification of the representative daily active user, the data amplification of the product keyword search index and the data amplification of the website access flow of the second mathematical model; and inputting the daily active user data amplification of the target product, the keyword search index data amplification of the target product and the website access flow data amplification as parameters into a second mathematical model, and measuring the daily active user data of the target product. By the method, the unknown daily active user data of the product can be estimated. The invention also provides computer equipment.
Description
Technical Field
The invention relates to the technical field of internet, in particular to a method for measuring and calculating product active user data and computer equipment.
Background
With the development of the internet, various internet-based application program products are gradually increased, and the competitive strength of similar products is also increased. If the daily active user data of the same type of products can be obtained, the development status, the trend and the law of the competitive products in the whole market can be known, the development opportunity of the own products can be mastered sharply, and the competitiveness of the own products is improved in intense competition.
However, the daily active user status of the same kind of product belongs to the internal secrets of the home company and changes constantly, and the detailed daily active user data status cannot be acquired by the external personnel.
Disclosure of Invention
The invention aims to provide a method for measuring and calculating product active user data and computer equipment, which are used for acquiring daily active user data of a product.
In order to achieve the purpose, the invention provides the following technical scheme:
a product active user data measurement method comprises the following steps:
acquiring a first training sample set; the first training sample set comprises daily active user data and ranking sequences of preset products; establishing a first mathematical model related to the correlation between the daily active user data and the ranking sequence, and training the first mathematical model by using a first training sample set to obtain parameters representing the relationship between the daily active user data and the ranking sequence of the first mathematical model;
acquiring a second training sample set; the second training sample set comprises daily active user data amplification of a preset product, product keyword search index data amplification and website access flow data amplification; establishing a second mathematical model related to the data amplification of daily active users, the data amplification of product keyword search indexes and the data amplification correlation of website access flow, and training the second mathematical model by using a second training sample set to obtain parameters of the second mathematical model, which represent the data amplification of daily active users, the data amplification of product keyword search indexes and the data amplification of website access flow;
obtaining a ranking sequence of a target product, an amplification of keyword search index data of the target product and an amplification of website access flow data; inputting the ranking sequence of the target product as a parameter into a first mathematical model to obtain daily active user data of the target product, and obtaining daily active user data amplification of the target product according to the daily active user data; and inputting the daily active user data increment of the target product, the keyword search index data amplification of the target product and the website access flow data amplification as parameters into a second mathematical model, and measuring the daily active user data of the target product.
Wherein the ranking sequence is a ranking sequence in a market leaderboard for the IOS application.
Wherein the daily active user data and ranking sequence of the preset products comprises: the daily active user data of the preset product on the same day, and the ranking of the preset product on the same day and two days later;
the ranking sequence of the target product comprises the ranking of the target product on the day and two days after the day.
The preset product and the target product are respectively subjected to daily active user data amplification, product keyword search index data amplification and website access flow data amplification, namely daily active user data month amplification, product keyword search index data month amplification and website access flow data month amplification.
Wherein the first mathematical model is a multilayer perceptron model.
Wherein the step of training the first mathematical model using the first set of training samples comprises: and taking the ranking sequence of the preset products as independent variables, and taking daily active user data as dependent variables to train the multilayer perceptron model.
Wherein the amplifying of the product keyword search index data of the preset product and the target product comprises: overall cycle ratio amplification of the hundredth index data of the product keyword, mobile cycle ratio amplification of the hundredth index data of the product keyword, or PC cycle ratio amplification of the hundredth index of the product keyword.
Wherein the amplification of the website access flow data of the preset product and the target product comprises: the method comprises the following steps of increasing the ring ratio of the Alexa website access flow ranking of a product, increasing the ring ratio of the number of people accessing the website per million netizens of the Alexa website, increasing the ring ratio of the number of pages of the website per million visited webpages of the Alexa website, or increasing the ring ratio of the number of IPs of the Alexa website.
Wherein the second mathematical model is a generalized linear model.
A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the product active user data estimation method of any of the above when executing the program.
Compared with the prior art, the scheme of the invention has the following advantages:
the method for measuring and calculating the data of the active users of the products obtains parameters representing the relation between the data of the active users and the ranking sequence in a first mathematical model according to the data of the active users of preset products and the ranking sequence, and obtains parameters representing the relation between the data amplification of the active users of the days, the data amplification of the keyword search index of the products and the data amplification of the website access flow in a second mathematical model according to the data amplification of the active users of the preset products, the data amplification of the keyword search index of the products and the data amplification of the website access flow. The invention aims to obtain the parameters for representing the rank sequence and the data correlation of daily active users of a product, and obtain the parameters for representing the data amplification of the daily active users, the data amplification of keyword search indexes of the product and the data amplification of website access flow. When the ranking sequence of the target product is input into the corresponding related parameters, the daily active user data of the target product can be obtained. And then the daily active user data amplification, the target product keyword search index data amplification and the website access flow data amplification of the target product are input into corresponding related parameters, so that the daily active user data of the target product can be measured.
Drawings
The foregoing and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a flowchart of a method for measuring and calculating product active user data according to an embodiment of the present invention;
FIG. 2 is a tabular diagram of the correlation of the IOS application product DAU with its ranking sequences, according to one embodiment of the invention;
FIG. 3 is a block diagram of a speculative DAU based on a preset product ranking sequence and an internal DAU, in accordance with one embodiment of the present invention;
FIG. 4 is a tabular diagram of the correlation of product DAU amplification with Baidu index, Alexa website traffic, in accordance with one embodiment of the present invention;
FIG. 5 is a schematic block diagram of the final measurement and calculation product DAU according to one embodiment of the present invention;
fig. 6 is a schematic view of a terminal portion structure of one embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative only and should not be construed as limiting the invention.
Active user scale: the number of users using the application product is counted during the statistical time.
And (3) DAU: a daily active user, referred to as daily life, is a user who uses an application product within a natural day.
Hundredth degree index: the scheme is also called a hundred-degree search heat degree, and shows the search attention degree and continuous change condition of the netizens on the keywords on the basis of hundred-degree mass netizen behavior data, so that the development trend of things is reflected to a certain degree.
Alexa website monitoring data: the website traffic data is also called in the scheme, and comprises traffic indexes such as website ranking, visitor quantity, page browsing quantity and the like, and the traffic and the popularity of the website can be reflected to a certain extent.
Ranking sequence: in the scheme, a ranking index set consisting of IOS free list ranking within a period of date is specified.
The invention provides a method for measuring and calculating product active user data. In this scenario, the product represents an application product. Specifically, the Application program includes a live game APP (Application). As shown in fig. 1, a method for measuring and calculating product active user data according to the present invention includes the steps of:
s10, acquiring a first training sample set; the first training sample set comprises daily active user data and ranking sequences of preset products; establishing a first mathematical model related to the correlation between the daily active user data and the ranking sequence, and training the first mathematical model by using a first training sample set to obtain parameters of the first mathematical model representing the relationship between the daily active user data and the ranking sequence.
In the first training sample set, the preset products are application program products with known daily active user data. In the present invention, the pre-set product is a self-owned product. The ranking sequence of the owned products is the ranking sequence in the IOS application market leaderboard. The ranked sequences of the owned products may be obtained at a third party monitoring data platform. The third-party monitoring data platform comprises App Annie, sensorTower and the like.
In the scheme, the daily active user data of the preset product comprises daily active user data of the current day. The ranking sequence of the preset products includes the ranking for the day and two days thereafter.
Specifically, the method for obtaining the ranking sequence of the preset products comprises the following steps:
the ranking of the daily free list of the preset products is collected by means of a third party monitoring platform (such as App Annie, SensorTower, and the like) of the IOS application market. Thus, a rank binary (date, rank) of the preset product may be obtained. Wherein, date is the date of the day, and rank is the ranking of the day. Since the day's DAU (day Active User) may be affected by the last week ranking, the ranking bigrams are expanded into a ranking sequence (date, rank [ order ]). Wherein, order takes a natural number less than 7, 0 represents the current day, 1 represents the following day, and so on.
In this embodiment, a first mathematical model is established regarding the correlation of daily active user data and ranking sequences. Wherein the first mathematical model may be a plurality of representations characterizing the correlation of the daily active user data and the ranking sequences. And will not be described in detail herein. By training the first mathematical model using the first training sample set, parameters of the first mathematical model that characterize the daily active user data and the ranking sequence relationships may be obtained.
Specifically, according to a ranking sequence obtained from a third-party monitoring data platform, the relevance of the daily active user data of the preset product and the ranking sequence is analyzed and obtained in combination with the daily active user data of the preset product. And establishing a first mathematical model according to the correlation of the first mathematical model and the second mathematical model, and training the first mathematical model by using a first training sample set so as to obtain parameters representing the daily active user data and ranking sequence relation in the first mathematical model. In the scheme, the daily active user data of the preset products and the ranking of the preset products on the day and two days later are adopted to train the first mathematical model. In particular, the first mathematical model is a multi-layer perceptron model. And training the multilayer perceptron model by taking the ranking sequence of the preset product as an independent variable and the daily active user data as a dependent variable, thereby obtaining parameters representing the relationship between the daily active user data and the ranking sequence in the multilayer perceptron model.
Specifically, a ranking sequence of a preset product (in the scheme, the ranking sequence comprises tiger live broadcast and YY live broadcast) for 3 months is collected, the relevance of the internal DAU of the preset product and the ranking sequence is analyzed, and the negative relevance of the DAU of the current day and the ranking sequences of the current day and two days later is found to be large. As shown in fig. 2. Wherein the correlation coefficients of the two are located in the range of [ -1,1], negative correlation is close to-1, positive correlation is close to 1, and irrelevant is close to 0. Therefore, the DAU of the preset product and the related data sets (date, DAU, rank [0], rank [1], rank [2]) in the ranking sequence can be used as a training sample set to train the multi-layer perceptron model related to the daily active user data and the ranking sequence relevance, so as to obtain the parameters representing the daily active user data and the ranking sequence relevance in the multi-layer perceptron model. Specifically, a multi-layer perceptron model is created by taking a relevant data set of a ranking sequence in a sample set as an independent variable and using the DAU as a dependent variable, and the multi-layer perceptron model is trained by using the ranking sequence and the DAU. As shown in fig. 3.
Wherein the first mathematical model (in this case, the multilayer perceptron model) may be an existing mathematical model. The method mainly aims to obtain the correlation between the daily active user data and the ranking sequence of the product, and obtain the parameters representing the relationship between the daily active user data and the ranking sequence in the existing first mathematical model according to the correlation. And by representing the parameters of the relationship between the two, the corresponding daily active user data can be obtained according to the ranking sequence.
S20, acquiring a second training sample set; the second training sample set comprises daily active user data amplification of a preset product, product keyword search index data amplification and website access flow data amplification; and establishing a second mathematical model related to the data amplification of the daily active users, the data amplification of the product keyword search index and the data amplification correlation of the website access flow, and training the second mathematical model by using a second training sample set to obtain parameters representing the data amplification of the daily active users, the data amplification of the product keyword search index and the data amplification of the website access flow of the second mathematical model.
In this embodiment, a second mathematical model is established regarding the correlation of daily active user data amplification, product keyword search index data amplification, and website access traffic data amplification. The second mathematical model may be a variety of representations that characterize the relevance of daily active user data amplification, product keyword search index data amplification, and website visitation traffic data amplification. And will not be described in detail herein. And training the second mathematical model by using the second training sample set to obtain parameters of the second mathematical model, representing daily active user data amplification, product keyword search index data amplification and website access flow data amplification correlation.
Specifically, the daily active user data of the preset products in the second training sample set is increased to a monthly active user data. The product keyword search index data of the preset product is increased to product keyword search index data monthly. And the amplification of the website access flow data of the preset product is the monthly amplification of the website access flow data. Of course, in other embodiments, the daily active user data amplification, the product keyword search index data amplification and the website access traffic data amplification of the preset product may also adopt other representation forms such as daily amplification or weekly amplification. The second mathematical model is a generalized linear model.
In the scheme, the daily active user data amplification of the preset product can be obtained according to the daily active user data of the preset product for several days. The product keyword search index data of the preset product adopts the hundredth index of the product. The amplification of the product keyword search index data of the preset product comprises the integral ring ratio amplification of the Baidu index data of the product keyword, the mobile ring ratio amplification of the Baidu index data of the product keyword or the PC ring ratio amplification of the Baidu index of the product keyword. The website access traffic data adopts Alexa website access traffic data. The amplification of the website access flow data of the preset products comprises the following steps: the method comprises the following steps of increasing the ring ratio of the Alexa website access flow ranking of a product, increasing the ring ratio of the number of people accessing the website per million netizens of the Alexa website, increasing the ring ratio of the number of pages of the website per million visited webpages of the Alexa website, or increasing the ring ratio of the number of IPs of the Alexa website.
In the scheme, in order to improve the accuracy of the measured and calculated DAU of the product, the Baidu index and Alexa website monitoring data of the product are further collected. And (3) combining the initially tested DAU (in the case, the daily active user data applied to training the first mathematical model), screening and creating an index system, and creating a multi-dimensional DAU presumption model spanning the data source to obtain the improved DAU of the final product.
Specifically, the Baidu index and the Alexa website traffic reflect changes of the monitoring APP (such as a live broadcast platform) DAU from different aspects to a certain extent, distribution and change relation among the initially measured DAU, the Baidu index, the multi-dimensional index value of the Alexa website traffic and the monthly cycle ratio increase of the index is deeply analyzed, and it is found that overall, the DAU increase of the monitoring APP, the increase of the Baidu index and the Alexa website traffic increase have change consistency, as shown in FIG. 4.
As shown in fig. 4, the performance of the IOS end monthly DAU is not consistent with the performance of the Baidu search heat and the website traffic, whereas the increase of the ring ratio of the IOS end monthly DAU is consistent with the ring ratio increase of the Baizhi (Baidu index) and the Alexa website access page number IP. The Personal Computer (PC) is in negative correlation and positive correlation with Alexa ranking and website traffic respectively, and shows that the websites with high search heat are ranked more forward and have higher traffic.
Therefore, the second mathematical model (namely, the generalized linear model) is trained by using the relevance of the ring ratio amplification data of the monthly active daily user of the preset product, the monthly average ring ratio amplification data of the hundredth index of the keyword and the monthly average ring ratio amplification data of the Alexa website flow, so as to obtain the parameters representing the relationship between the daily active user data amplification, the product keyword search index data amplification and the website access flow data amplification in the generalized linear model. See in particular fig. 5.
Wherein the second mathematical model may be an existing mathematical model. The method aims to obtain the correlation of daily active user data amplification, product keyword search index data amplification and website access flow data amplification, and train the second mathematical model according to the correlation to obtain the parameters of the second mathematical model representing the relationship between the daily active user data amplification, the product keyword search index data amplification and the website access flow data amplification. By characterizing the parameters of the relationship between the three, the final daily active user data can be obtained according to the initially measured daily active user data obtained in step S10.
S30, obtaining a ranking sequence of the target product, an amplification of keyword search index data of the target product and an amplification of website access flow data; inputting the ranking sequence of the target product as a parameter into a first mathematical model to obtain daily active user data of the target product, and obtaining daily active user data amplification of the target product according to the daily active user data; and inputting the daily active user data amplification of the target product, the keyword search index data amplification of the target product and the website access flow data amplification as parameters into a second mathematical model, and measuring the daily active user data of the target product.
In the scheme, the ranking sequence of the target products comprises the ranking of the current day and the two days later. The method for specifically acquiring the ranking sequence of the target product comprises the following steps: daily toll-free list rankings of target products are collected with the help of third party monitoring platforms (e.g., App anie, SensorTower, etc.) of the IOS application marketplace. Thus, a rank binary (date, rank) of the target product may be obtained. Wherein, date is the date of the day, and rank is the ranking of the day. Since the day's DAU (day Active User) may be affected by the last week ranking, the ranking bigrams are expanded into a ranking sequence (date, rank [ order ]). Wherein, order takes a natural number less than 7, 0 represents the current day, 1 represents the following day, and so on.
In the scheme, the daily active user data amplification of the target product can be achieved according to the daily active user data of the target product for several days, so that the amplification data of the target product can be obtained. The product keyword search index data of the target product adopts the hundredth index of the product. The product keyword search index data amplification of the target product comprises the overall ring ratio amplification of the Baidu index data of the product keyword, the mobile ring ratio amplification of the Baidu index data of the product keyword or the PC ring ratio amplification of the Baidu index of the product keyword. The website access traffic data adopts Alexa website access traffic data. The website access flow data amplification of the target product comprises the following steps: the method comprises the following steps of increasing the ring ratio of the Alexa website access flow ranking of a product, increasing the ring ratio of the number of people accessing the website per million netizens of the Alexa website, increasing the ring ratio of the number of pages of the website per million visited webpages of the Alexa website, or increasing the ring ratio of the number of IPs of the Alexa website.
In the scheme, the daily active user data of the target product in the second training sample set is increased to monthly active user data. The product keyword search index data of the target product is increased to product keyword search index data monthly. And the website access flow data of the target product is increased to monthly website access flow data. Of course, in other embodiments, the daily active user data amplification, the product keyword search index data amplification and the website access traffic data amplification of the target product may also adopt other representation forms such as daily amplification or weekly amplification.
In the scheme, the multilayer perceptron model is trained through the daily active user data and the ranking sequence of the preset product, so that parameters representing the relationship between the daily active user data and the ranking sequence in the first mathematical model are obtained. Thus, a ranked sequence of target products is input into the multi-tier perceptron to obtain daily active user data for the target products. By collecting the daily active user data of the target product for several days, the daily active user data amplification of the target product can be obtained. Specifically, the scheme acquires daily active user data of the target product within one month, and obtains monthly average daily active user data amplification of the target product.
And inputting the obtained daily active user data amplification of the target product, the target product keyword search index data amplification and the website access flow data amplification as parameters into a second mathematical model, so that the daily active user data of the target product can be measured.
The method for measuring and calculating the data of the active users of the products obtains parameters representing the relation between the data of the active users and the ranking sequence in a first mathematical model according to the data of the active users of preset products and the ranking sequence, and obtains parameters representing the relation between the data amplification of the active users of the days, the data amplification of the keyword search index of the products and the data amplification of the website access flow in a second mathematical model according to the data amplification of the active users of the preset products, the data amplification of the keyword search index of the products and the data amplification of the website access flow. The invention aims to obtain the parameters for representing the rank sequence and the data correlation of daily active users of a product, and obtain the parameters for representing the data amplification of the daily active users, the data amplification of keyword search indexes of the product and the data amplification of website access flow. When the ranking sequence of the target product is input into the corresponding related parameters, the daily active user data of the target product can be obtained. And then the daily active user data amplification, the target product keyword search index data amplification and the website access flow data amplification of the target product are input into corresponding related parameters, so that the daily active user data of the target product can be measured.
The invention also provides computer equipment. A computer device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. And when the processor executes the program, the method for measuring and calculating the active user data of any product in the scheme is realized.
See fig. 6. Fig. 6 is a block diagram of the structure of the relevant part of the computer device provided by the embodiment of the invention. In this embodiment, the computer device is specifically a mobile phone. The mobile phone comprises: a Radio Frequency (RF) circuit 610, a memory 620, an input unit 630, a display unit 640, a sensor 650, an audio circuit 660, a wireless fidelity (WiFi) module 670 (i.e., a WiFi chip module), a processor 680, and a power supply 690. The functions and interaction relationships of the relevant components of the mobile phone are not described in detail herein.
Those skilled in the art will appreciate that the handset configuration shown in fig. 6 is not intended to be limiting and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components.
The foregoing is only a partial 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 (9)
1. A method for measuring and calculating product active user data is characterized by comprising the following steps:
acquiring a first training sample set; the first training sample set comprises daily active user data and ranking sequences of preset products; establishing a first mathematical model related to the correlation between the daily active user data and the ranking sequence, and training the first mathematical model by using a first training sample set to obtain parameters representing the relationship between the daily active user data and the ranking sequence of the first mathematical model;
acquiring a second training sample set; the second training sample set comprises the monthly amplification of daily active user data, the monthly amplification of product keyword search index data and the monthly amplification of website access flow data of preset products; establishing a second mathematical model related to monthly amplification of daily active user data, monthly amplification of product keyword search index data and monthly amplification correlation of website access flow data, and training the second mathematical model by using a second training sample set to obtain parameters representing the monthly amplification of the daily active user data, the monthly amplification of the product keyword search index data and the monthly amplification of the website access flow data of the second mathematical model;
acquiring a ranking sequence of a target product, monthly amplification of keyword search index data of the target product and monthly amplification of website access flow data; inputting the ranking sequence of the target product as a parameter into a first mathematical model to obtain daily active user data of the target product, and obtaining monthly amplification of the daily active user data of the target product according to the daily active user data; inputting the monthly amplification of the keyword search index data of the target product and the monthly amplification of the website access flow data as parameters into a second mathematical model to obtain the monthly amplification of the daily active user data of the second mathematical model, and measuring and calculating the final daily active user data of the target product according to the monthly amplification of the daily active user data of the second mathematical model and the monthly amplification of the daily active user data of the target product.
2. The method for product active user data measurement and calculation of claim 1, wherein the ranking sequence is a ranking sequence in an IOS application market leaderboard.
3. The product active user data estimation method according to claim 2,
the daily active user data and ranking sequence of the preset products comprises: the daily active user data of the preset product on the same day, and the ranking of the preset product on the same day and two days later;
the ranking sequence of the target product comprises the ranking of the target product on the day and two days after the day.
4. The method for product active user data estimation according to claim 1, characterized in that the first mathematical model is a multi-layer perceptron model.
5. The method of claim 4, wherein the step of training the first mathematical model using the first set of training samples comprises: and taking the ranking sequence of the preset products as independent variables, and taking daily active user data as dependent variables to train the multilayer perceptron model.
6. The method for measuring and calculating product active user data according to claim 1, wherein the amplification of the product keyword search index data of the preset product and the target product comprises: overall cycle ratio amplification of the hundredth index data of the product keyword, mobile cycle ratio amplification of the hundredth index data of the product keyword, or PC cycle ratio amplification of the hundredth index of the product keyword.
7. The method for measuring and calculating product active user data according to claim 1, wherein the amplification of website access flow data of the preset product and the target product comprises: the method comprises the following steps of increasing the ring ratio of the Alexa website access flow ranking of a product, increasing the ring ratio of the number of people accessing the website per million netizens of the Alexa website, increasing the ring ratio of the number of pages of the website per million visited webpages of the Alexa website, or increasing the ring ratio of the number of IPs of the Alexa website.
8. The product active user data estimation method according to claim 1, characterized in that the second mathematical model is a generalized linear model.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the product active user data gauging method according to any one of claims 1-8 when executing said program.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710640388.5A CN107463660B (en) | 2017-07-31 | 2017-07-31 | Method for measuring and calculating data of active users of products and computer equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710640388.5A CN107463660B (en) | 2017-07-31 | 2017-07-31 | Method for measuring and calculating data of active users of products and computer equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107463660A CN107463660A (en) | 2017-12-12 |
CN107463660B true CN107463660B (en) | 2020-10-16 |
Family
ID=60547981
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710640388.5A Active CN107463660B (en) | 2017-07-31 | 2017-07-31 | Method for measuring and calculating data of active users of products and computer equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107463660B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109711652A (en) * | 2017-10-26 | 2019-05-03 | 厦门一品威客网络科技股份有限公司 | A kind of Chuan Ke team potential methods of marking |
CN108322783B (en) * | 2018-01-25 | 2021-08-17 | 广州虎牙信息科技有限公司 | Video website user scale presumption method, storage medium and terminal |
CN110458360B (en) * | 2019-08-13 | 2023-07-18 | 腾讯科技(深圳)有限公司 | Method, device, equipment and storage medium for predicting hot resources |
CN111475718A (en) * | 2020-03-30 | 2020-07-31 | 清华大学 | User activity model construction method and system based on preference diversity |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101266610A (en) * | 2008-04-25 | 2008-09-17 | 浙江大学 | Web active user website accessing mode on-line excavation method |
CN105787039A (en) * | 2016-02-25 | 2016-07-20 | 四川长虹电器股份有限公司 | Method for multi-period calculation and display of user retention rate |
CN105869022A (en) * | 2016-04-07 | 2016-08-17 | 腾讯科技(深圳)有限公司 | Application popularity prediction method and apparatus |
CN106126641A (en) * | 2016-06-24 | 2016-11-16 | 中国科学技术大学 | A kind of real-time recommendation system and method based on Spark |
CN106600344A (en) * | 2016-12-30 | 2017-04-26 | 广州虎牙信息科技有限公司 | Method and apparatus for obtaining active user data of target product |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9405838B2 (en) * | 2013-12-27 | 2016-08-02 | Quixey, Inc. | Determining an active persona of a user device |
-
2017
- 2017-07-31 CN CN201710640388.5A patent/CN107463660B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101266610A (en) * | 2008-04-25 | 2008-09-17 | 浙江大学 | Web active user website accessing mode on-line excavation method |
CN105787039A (en) * | 2016-02-25 | 2016-07-20 | 四川长虹电器股份有限公司 | Method for multi-period calculation and display of user retention rate |
CN105869022A (en) * | 2016-04-07 | 2016-08-17 | 腾讯科技(深圳)有限公司 | Application popularity prediction method and apparatus |
CN106126641A (en) * | 2016-06-24 | 2016-11-16 | 中国科学技术大学 | A kind of real-time recommendation system and method based on Spark |
CN106600344A (en) * | 2016-12-30 | 2017-04-26 | 广州虎牙信息科技有限公司 | Method and apparatus for obtaining active user data of target product |
Non-Patent Citations (1)
Title |
---|
基于时间线数据的微博活跃用户数估计;陈江宁 等;《计算机应用与软件》;20130815;第30卷(第8期);246-249 * |
Also Published As
Publication number | Publication date |
---|---|
CN107463660A (en) | 2017-12-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107463660B (en) | Method for measuring and calculating data of active users of products and computer equipment | |
WO2016107523A1 (en) | Access path analysis method and apparatus for website | |
Newman et al. | Mixture models and exploratory analysis in networks | |
RU2636133C2 (en) | Method and device for displaying application software | |
TWI567572B (en) | Data acquisition method, device and system | |
CN105183731B (en) | Recommendation information generation method, device and system | |
US20160125503A1 (en) | Method, apparatus and system for recommending product information | |
CN105868256A (en) | Method and system for processing user behavior data | |
Bellis et al. | Childhood happiness and violence: a retrospective study of their impacts on adult well-being | |
Liu et al. | A dengue fever predicting model based on Baidu search index data and climate data in South China | |
CN105183925A (en) | Content association recommending method and content association recommending device | |
CN106445971A (en) | Application recommendation method and system | |
KR20170076605A (en) | Method and device for providing adapted learning information to a user | |
CN113254472B (en) | Parameter configuration method, device, equipment and readable storage medium | |
Yang et al. | Outlier identification and robust parameter estimation in a zero-inflated Poisson model | |
Cortés et al. | The productivity of top researchers: a semi-nonparametric approach | |
CN103177086A (en) | Method and device for ranking order through using searching result correspondingly | |
Chmiel et al. | Scaling of human behavior during portal browsing | |
Li et al. | High-resolution time–frequency representation of EEG data using multi-scale wavelets | |
CN112735563A (en) | Recommendation information generation method and device and processor | |
WO2016131244A1 (en) | User health monitoring method, monitoring device, and monitoring terminal | |
CN111125507B (en) | Group activity recommendation method and device, server and computer storage medium | |
CN104657397B (en) | Information processing method and terminal | |
JP2011227720A (en) | Recommendation system, recommendation method and recommendation program | |
CN103425643A (en) | Recommending method and system for relevant search request strings |
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 |