CN113590925A - User determination method, device, equipment and computer storage medium - Google Patents
User determination method, device, equipment and computer storage medium Download PDFInfo
- Publication number
- CN113590925A CN113590925A CN202010363873.4A CN202010363873A CN113590925A CN 113590925 A CN113590925 A CN 113590925A CN 202010363873 A CN202010363873 A CN 202010363873A CN 113590925 A CN113590925 A CN 113590925A
- Authority
- CN
- China
- Prior art keywords
- attention
- user
- behavior data
- determining
- target object
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 75
- 230000006399 behavior Effects 0.000 claims description 121
- 238000012545 processing Methods 0.000 claims description 28
- 238000010606 normalization Methods 0.000 claims description 21
- 238000004590 computer program Methods 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000009795 derivation Methods 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 12
- 230000006870 function Effects 0.000 description 9
- 230000008569 process Effects 0.000 description 8
- 238000007477 logistic regression Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 5
- 238000005457 optimization Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000001419 dependent effect Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000010365 information processing Effects 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000007429 general method Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000003014 reinforcing effect Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012546 transfer 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
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a user determination method, a user determination device, user determination equipment and a computer storage medium. The method comprises the following steps: acquiring online behavior data of a plurality of users within a preset time period; determining the attention of each user to the target object according to the online behavior data; determining attention trend curves of the multiple users to the target object according to the online behavior data of the multiple users and attention degrees corresponding to the online behavior data; determining the attention stage of each user to the target object according to the attention trend curve of each user to the target object; and determining a user corresponding to the target attention stage from the plurality of users according to the attention stage of each user to the target object. According to the embodiment of the invention, the target user meeting the preset condition can be quickly and accurately determined.
Description
Technical Field
The present invention relates to the field of information processing, and in particular, to a user determination method, apparatus, device, and computer storage medium.
Background
With the continuous expansion of the scale of the users of the operators, the services are continuously enriched. While providing services for users, the network service platform generally determines potential target users through online behavior data, and then recommends products that may be of interest to the target users.
At present, in a scheme related to target user identification, a preference threshold of online behavior data is usually set based on artificial subjectivity, and the method relies on artificial subjective judgment to a great extent, so that not only is manpower consumed greatly and efficiency low, but also the problem of inaccurate identification is caused by artificial subjectivity.
Therefore, how to quickly and accurately determine the target users meeting the preset conditions becomes a problem to be solved.
Disclosure of Invention
The embodiment of the invention provides a user determination method, a user determination device, user determination equipment and a computer storage medium, which can quickly and accurately determine a target user meeting preset conditions.
In a first aspect, the present application provides a user determination method, including: acquiring online behavior data of a plurality of users within a preset time period; determining the attention of each user to the target object according to the online behavior data; determining attention trend curves of the multiple users to the target object according to the online behavior data of the multiple users and attention degrees corresponding to the online behavior data; determining the attention stage of each user to the target object according to the attention trend curve of each user to the target object; and determining a user corresponding to the target attention stage from the plurality of users according to the attention stage of each user to the target object.
In one possible implementation, determining a focus stage of each user on the target object according to the focus trend curve includes: performing derivation calculation on the attention trend curve, and determining a first characteristic value and a second characteristic value of the attention trend curve, wherein the first characteristic value is smaller than the second characteristic value; determining the user corresponding to the online behavior data which is greater than the first characteristic value and smaller than the second characteristic value as the user corresponding to the first attention stage; determining the user corresponding to the on-line behavior data larger than the second characteristic value as the user corresponding to the second attention stage; correspondingly, the step of determining the user corresponding to the target attention stage from the plurality of users comprises the following steps: determining a user corresponding to the first attention stage as a user corresponding to the target attention stage; or determining the user corresponding to the first attention stage and the user corresponding to the second attention stage as the users corresponding to the target attention stage.
In one possible implementation, the online behavior data includes the number of days of interest of each user to the target object and/or the number of times of interest of each user to the target object.
In one possible implementation, after determining the user corresponding to the target attention stage from the plurality of users, the method further includes: determining a plurality of first attention degrees according to attention days of a user corresponding to the target attention stage to a plurality of target objects; determining a plurality of second attention degrees according to the attention times of the user corresponding to the target attention stage to the target objects; and determining the target object of which the sum of the first attention and the second attention is greater than or equal to a first threshold value and the absolute value of the difference between the first attention and the second attention is greater than or equal to a second threshold value as the first target object of the user corresponding to the target attention stage.
In one possible implementation, determining an attention trend curve of a plurality of users to a target object according to online behavior data of the plurality of users and attention degrees corresponding to the online behavior data includes: the following steps are performed for each user's online behavior data: preprocessing the online behavior data of the user to obtain preprocessed online behavior data; determining the attention of a user to a target object according to the preprocessed online behavior data; carrying out normalization processing on the attention degree to obtain the attention degree after the normalization processing; and determining an attention trend curve according to the preprocessed on-line behavior data and the attention degree after the normalization processing.
In one possible implementation, the preprocessing the online behavior data of the user to obtain the preprocessed online behavior data includes: calculating on-line behavior data corresponding to the target object based on a Lauda criterion, and determining a data distribution range of the on-line behavior data; determining a third threshold and a fourth threshold corresponding to the target object according to the data distribution range and the preset value-taking interval; replacing the on-line behavior data smaller than the third threshold value in the on-line behavior data corresponding to the target object with the third threshold value; and replacing the on-line behavior data which is larger than the fourth threshold value in the on-line behavior data corresponding to the target object with the fourth threshold value.
In a second aspect, an embodiment of the present invention provides a user determination apparatus, where the apparatus includes: the acquisition module is used for acquiring online behavior data of a plurality of users within a preset time period; the first determining module is used for determining the attention degree of each user to the target object according to the online behavior data; the second determination module is used for determining an attention trend curve of each user to the target object according to the online behavior data of the plurality of users and the attention degree corresponding to the online behavior data; the third determination module is used for determining the attention stage of each user to the target object according to the attention trend curve of each user to the target object; and the fourth determining module is used for determining the user corresponding to the target attention stage from the plurality of users according to the attention stage of each user to the target object.
In a third aspect, an embodiment of the present invention provides a computing device, where the device includes: a processor and a memory storing computer program instructions; the processor, when executing the computer program instructions, implements the user determination method as provided by embodiments of the present invention.
In a fourth aspect, an embodiment of the present invention provides a computer storage medium, where computer program instructions are stored, and when executed by a processor, implement the user determination method provided in the embodiment of the present invention.
According to the user determination method, the user determination device, the user determination equipment and the computer storage medium, the attention trend curve of the user to the target object is determined according to the online behavior data of the plurality of users, then the user corresponding to the target attention stage is determined according to the attention trend curve, and the accuracy of determining the user meeting the preset conditions can be improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments of the present invention will be briefly described below, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic diagram of a Laudea criterion according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a method for implementing multi-dimensional multi-angle recognition according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a reinforced regression curve provided in accordance with an embodiment of the present invention;
fig. 4 is a flowchart illustrating a user determination method according to an embodiment of the present invention;
FIG. 5 is a schematic view of a attention trend curve provided by an embodiment of the present invention;
FIG. 6 is a schematic view of another attention trend curve provided by the embodiment of the present invention;
FIG. 7 is a schematic view of another attention trend curve provided by the embodiment of the present invention;
FIG. 8 is a schematic view of a further attention trend curve provided by an embodiment of the present invention;
fig. 9 is a schematic diagram of an implementation method of extremum processing according to an embodiment of the present invention;
fig. 10 is a schematic structural diagram of a user determination device according to an embodiment of the present invention;
fig. 11 is a schematic diagram of an exemplary hardware architecture provided by an embodiment of the present invention.
Detailed Description
Features and exemplary embodiments of various aspects of the present invention will be described in detail below, and in order to make objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not to be construed as limiting the invention. It will be apparent to one skilled in the art that the present invention may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present invention by illustrating examples of the present invention.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
With the continuous development of internet technology, a network service platform can determine potential target users through online behavior data.
Currently, in the related art for identifying the preference of the client (i.e., the potential target user) and further subdividing the preference degree according to the identified user, there are mainly the following implementation manners.
First, in the related art of customer preference identification, customer preference identification is mainly achieved by cluster subdivision, customer preference identification is achieved by an RFM model method, and customer preference identification is achieved by a method using simple business rules. These several ways of identifying customer preferences are significantly affected by human subjectivity.
The method comprises the steps of identifying client preferences by clustering subdivision, preparing various data of client target preferences, such as financing times, days, reading times, days, gourmet search times, days and the like, setting clustering categories by means of algorithms such as K-means or K-means and the like, and summarizing different clustering center points into different preference categories according to clustering results.
The RFM model method is used for realizing client preference identification, and needs to prepare a target preference Recency (time of recent search), a Frequency (Frequency of search within a period of time) and a Monetary (time, duration or flow occupied within a period of time), then obtain an RFM score by weighting, and set a preference threshold value manually and subjectively.
The customer preference identification is realized by simple business rules, and the preferred rules are set artificially and subjectively through business understanding and data expression, such as the financing preference rules (the monthly use days of the financing app are more than 20 days, and the flow is more than 200M), so that the artificial subjective influence is large.
Then, in the related art relating to the customer preference degree subdivision, the preference degree subdivision methods are different mainly by corresponding to different customer preference identification methods, but are also largely influenced by subjectivity.
The cluster subdivision realizes preference identification, and the cluster subdivision is divided into strong preference and weak preference according to the level of a cluster central point, but each preference can not necessarily distinguish the strength; the RFM method distinguishes the preference of different degrees by subjectively setting a plurality of threshold values; simple business rules differentiate preference degrees by differentiating rule thresholds according to business understanding and data measurement. The several implementation modes of the customer preference degree subdivision all need manual participation and depend on artificial subjective influence.
Finally, the optimization of a client preference model is related, historical data is used as a training sample in the current related technology, whether a client prefers a certain characteristic in a certain period is analyzed, and the client preference model needs to be updated continuously along with social development and human thinking change. However, the optimization of the customer preference model also needs manual participation, depends on the influence of human subjectivity, greatly consumes manpower, and has low efficiency.
Therefore, in the prior art of identifying the preference of the client and subdividing the preference degree, the preference threshold of the online behavior data is set based on artificial subjectivity, more manual participation is needed, the identification method is low in efficiency, the subjective influence is large, and the identification is inaccurate.
In order to solve the problem that the target user meeting the preset condition cannot be determined quickly and accurately at present, the target user meeting the preset condition can be determined by analyzing online behavior data of the user. Based on this, the embodiment of the invention provides a user determination method.
In order to better explain the methods according to the embodiments of the present invention, the algorithms according to the present invention are described below.
First, an interquartile range (IQR), also called an interquartile range, is introduced.
The quartile range is a method in descriptive statistics to determine the difference between the third quartile and the first quartile. The variance and standard deviation are the same, and represent the dispersion of variables in the statistical data, but the quarter-difference is more a robust statistic (robust statistic). The quartering distance is typically used to construct a box plot, as well as a brief chart summary of the probability distribution. For a symmetric distribution (where the number of bits is necessarily equal to the arithmetic mean of the third quartile and the first quartile), one-half of the quartile is equal to the Median Absolute Difference (MAD). Median is a reflection of central tendency.
The expression related to the four-bit distance is specifically as follows: quartering pitch, IQR-Q3-Q1; upper bound of the interquartile range: upper _ limit ═ Q3+1.5 × IQR; lower bound of the interquartile range: lower _ limit ═ Q1-1.5 × IQR.
Next, the 3 σ criterion, also called the raleigh criterion, is introduced.
The 3 sigma criterion is that a group of detection data is assumed to only contain random errors, the detection data is calculated to obtain standard deviation, an interval is determined according to a certain probability, the error exceeding the interval is considered not to belong to the random errors but to be a coarse error, and the data containing the error is removed. And is suitable for use when there are more groups of data. The discrimination processing principle and method are only limited to processing sample data of normal or approximately normal distribution, and the method is based on the premise that the measurement times are sufficiently large, and the method is not reliable enough to remove gross errors by using a criterion when the measurement times are few. Therefore, in the case of a small number of measurements, it is preferable not to select this criterion but to use other criteria.
In a normal distribution, σ represents the standard deviation and μ represents the mean. And x is the symmetry axis of the image. As shown in fig. 1, the 3 σ principle is:
the probability of the numerical distribution in (μ - σ, μ + σ) is 0.6827;
the probability of the numerical distribution in (μ -2 σ, μ +2 σ) is 0.9545;
the probability of the numerical distribution in (μ -3 σ, μ +3 σ) is 0.9973;
it is considered that the values of Y are almost entirely concentrated in the (μ -3 σ, μ +3 σ) range, and the possibility of exceeding this range is only less than 0.3%.
Next, a Sigmoid function, also called Sigmoid growth curve, is introduced.
In the information science, due to the properties of single increment and single increment of an inverse function, a Sigmoid function is often used as a threshold function of a neural network, and variables are mapped to be between 0 and 1. The first derivative of the function can be expressed as s' (x) ═ s (x) (1-s (x)) by itself, and both functions are continuous, monotonically increasing numerical functions, which are often applied in neural networks based on the BP (back propagation of errors) algorithm. The sum, the target value, the error value and the like of a certain neuron in the BP neural network can be obtained by using the Sigmoid function or the derivative thereof.
then, Logistic regression, also known as Logistic regression analysis, is introduced.
Logistic regression is a generalized linear regression model, and thus has many similarities to multiple linear regression analysis, and is commonly used in the fields of data mining, disease automatic diagnosis, economic prediction, and the like. Dependent variables of logistic regression can be either two-class or multi-class, but two classes are more common and easier to interpret. So in practice the most common is the logistic regression of the two classes.
The applicable conditions of the Logistic regression model are as follows: (1) the dependent variable is a categorical variable of the two categories or the occurrence of an event, and is a numerical variable. However, it should be noted that the double counting phenomenon indicator is not suitable for Logistic regression. (2) Both residual and dependent variables are subject to binomial distribution. The binomial distribution corresponds to a classification variable, so that the binomial distribution is not normal distribution, and further, the equation estimation and test problems are solved by using a maximum likelihood method instead of a least square method. (3) The argument and Logistic probability are linear relationships. (4) The observation objects are independent of each other.
The invention provides an MPMD (Multi-reactive and Multi-dimensional) method in combination with online behavior data of a user. The MPMD algorithm refers to a method for evaluating objects from multiple angles and multiple dimensions so as to accurately position the objects. As shown in FIG. 2, the angles of discrimination 1, 2, … …, and X may be determined from a plurality of discrimination dimensions (w) for each of the discrimination angles1、w2… …, and wnEtc.) to evaluate the object. For example, the judgment angle may be the number of times a user uses a certain behavior within a period of time, or the number of days that the user uses the certain behavior within a period of time; the discriminative dimension may be a degree of attention of the user to something.
Constructing a client preference model from multiple dimensions and angles, such as using the times F (the times of a certain behavior of a client in a period of time) and the days T (the days of the certain behavior of the client in the period of time) of a certain thing preferred by the client, calculating an inflection coefficient of each client preferred thing by calculating an enhanced regression V, reserving parameters of the inflection coefficient, and objectively dividing the preference degree: in practical application, the inflection point of the initial full period is used as a threshold value for identifying the preference of the client, and whether the reinforced regression V is greater than zero can be used as a judgment standard for identifying the preference of the client.
And by reinforcing regression, judging inflection points and setting self-learning parameters, evaluating and subdividing, wherein the improved algorithm is suitable for the client preference subdivision of the client portrait, and the client preference degree is subdivided: beginning full period, peak period, and final full period.
V1 ═ logistic (t) formula (1)
V2 ═ logistic (f) formula (2)
V ═ logistic (t) + logistic (f) formula (3)
By calculating the inflection coefficients of the reinforced regression, the subdivision preference degree, as shown in fig. 3, is placed on a two-dimensional plane according to the reinforced regression formula (1) above, similar to the logistic curve, resulting in two inflection points.
according to the method, a more objective method is provided in customer preference identification and customer preference degree subdivision, optimizable parameters (b and c) are reserved, the two parameters are reserved, data change can be achieved, iterative learning parameters b and c are achieved, automatic optimization is achieved by using big data processing capacity and machine learning automatic optimization technology through continuous optimization of the parameters, timeliness and accuracy of customer preference identification are improved, and customer preference identification is enabled to be more intelligent.
According to the user determination method provided by the embodiment of the invention, the attention trend curve of the user to the target object is determined according to the online behavior data of the plurality of users, and then the user corresponding to the target attention stage is determined according to the attention trend curve, so that the accuracy of determining the user meeting the preset condition can be improved.
Fig. 4 is a flowchart illustrating a user determination method according to an embodiment of the present invention.
As shown in fig. 4, the user determination method may include S401-S405, and the method is applied to a server, and specifically as follows:
s401, acquiring online behavior data of a plurality of users in a preset time period.
S402, determining the attention of each user to the target object according to the on-line behavior data.
And S403, determining attention trend curves of the multiple users to the target object according to the online behavior data of the multiple users and attention degrees corresponding to the online behavior data.
S404, determining the attention stage of each user to the target object according to the attention trend curve of each user to the target object.
S405, determining a user corresponding to the target attention stage from the plurality of users according to the attention stage of each user to the target object.
According to the user determination method, the target user meeting the preset conditions is determined by analyzing the online behavior data of the user, and the accuracy of target user identification can be improved.
The contents of S401 to S405 are described below, respectively:
first, referring to S401, in particular, the above-mentioned online behavior data includes the number of days of attention of each user to the target object and/or the number of times of attention of each user to the target object.
Illustratively, acquiring online behavior data related to life of a plurality of users in a preset time period, wherein the online behavior data can comprise days of middle school attention, days of school district attention, days of villa attention, days of house change attention, days of immigration attention and days of visa attentions attention; and the times of paying attention to middle schools, the times of paying attention to school districts, the times of paying attention to villas, the times of paying attention to house changing, the times of paying attention to immigration and the times of paying attention to visa.
In a possible embodiment, online behavior data of a plurality of users within a preset time period is obtained, that is, data such as internet search keywords and app usage records of the users are obtained.
Taking the middle school of interest as an example, the search keywords may include: physics in junior middle schools, association of school names, mathematics in high schools, etc. And middle school related app usage records, such as usage records on apps of some tutoring jobs.
Illustratively, the online behavior data of a plurality of users within a preset time period is shown in table 1:
TABLE 1
Optionally, in one embodiment, the following steps are performed for each user's online behavior data: preprocessing the online behavior data of the user to obtain preprocessed online behavior data; determining the attention of a user to a target object according to the preprocessed online behavior data; carrying out normalization processing on the attention degree to obtain the attention degree after the normalization processing; and determining an attention trend curve according to the preprocessed on-line behavior data and the attention degree after the normalization processing.
The above step of normalizing the attention degree to obtain the normalized attention degree is described in the section of S403.
The step of preprocessing the online behavior data of the user to obtain the preprocessed online behavior data may specifically include:
calculating on-line behavior data corresponding to the target object based on a Lauda criterion, and determining a data distribution range of the on-line behavior data; determining a third threshold and a fourth threshold corresponding to the target object according to the data distribution range and the preset value-taking interval; replacing the on-line behavior data smaller than the third threshold value in the on-line behavior data corresponding to the target object with the third threshold value; and replacing the on-line behavior data which is larger than the fourth threshold value in the on-line behavior data corresponding to the target object with the fourth threshold value.
Optionally, in a possible embodiment, the data preprocessing mainly deals with null value problem and extreme value problem, such as the sample data shown in table 1 above, the concerns of the customer number 2907 are all 0, there is no preference, and there is no any correction effect on the model during model training, so the data is deleted; extremum processing uses a 3 σ criterion and an interquartile range (IQR) to identify extrema.
Calculating the on-line behavior data corresponding to the target object based on the Lauda criterion, and determining the data distribution range of the on-line behavior data; in the step of determining the third threshold and the fourth threshold corresponding to the target object according to the data distribution range and the preset value-taking interval, the method specifically includes:
firstly, calculating the online behavior data of the user in a preset range corresponding to a target object, and determining the data distribution range of the online behavior data, wherein the online behavior data can be the number of days when the user pays attention to the target object or the number of times when the user pays attention to the target object; and secondly, determining a minimum value (third threshold) and a maximum value (fourth threshold) corresponding to the target object according to the data distribution range and the preset value-taking interval.
After the minimum (third threshold) and maximum (fourth threshold) values are determined, data greater than the extreme value and data less than the minimum value may be processed. The invention relates to a method for processing an extreme value, which comprises two general methods, one is a deleting method, namely deleting data larger than a maximum value and data smaller than a minimum value, and the other is an assigning method, namely assigning the data larger than the extreme value as the maximum value and assigning the data smaller than the minimum value as the minimum value.
For example, as shown in table 1, the maximum value of the number of middle school concerned is 2129, the maximum value of the number of school district concerned is 188, and the maximum value of the number of house change concerned is 596, and data higher than the maximum value is assigned as the maximum value. The data after treatment are shown in table 2:
TABLE 2
The parenthesized data in table 2 is the processed data.
Second, referring to S402, in one possible embodiment, the customer preferences are divided for the number of days and the number of times that a certain point of interest of the customer is used, i.e., the customer preferences are divided by the degree of interest of each user with respect to the target object.
The number of usage days (f) is used to indicate the number of days a client uses an application for a period of time, and the number of usage times (t) is used to indicate the number of times a client uses an application for a period of time. And determining the initial attention of the user to the target object according to the using days and the using times and the following formula. (i.e., logistic (t) and logistic (f) determined from the initial coefficient b and the initial coefficient c). The initial coefficient b and the initial coefficient c may first adopt random values, then converge and correct the attention trend curve according to the actual value of the attention degree (such as days and times) of the user to the target object, and then determine the corrected coefficient b and coefficient c according to the inflection point of the attention trend curve.
The Logistic curve is in an S shape, and derivatives are taken to obtain inflection points, and three period distribution of the Logistic curve is obtained, so that a certain attention day score and/or attention frequency score (attention degree of a user to a target object) is calculated.
Then, referring to S403, optionally, in a possible embodiment, performing normalization processing on the attention degree to obtain the attention degree after the normalization processing; and determining an attention trend curve according to the preprocessed on-line behavior data and the attention degree after the normalization processing.
The step of performing normalization processing on the attention degree to obtain the attention degree after the normalization processing may specifically be: the original value (attention) is linearly transformed so that the result value (attention after normalization) is mapped between [0-1 ].
Fig. 5 and 6 show the following steps of determining the attention trend curve of each user to the target object according to the online behavior data of the plurality of users and the attention degree after the normalization processing.
The horizontal axis of fig. 5 represents the number of days of attention of all users to the target object, and the vertical axis of fig. 5 represents the degree of attention of users to the target object, i.e., logistic (t)). For example, when the attention object is a school zone, the curve corresponding to the school zone is a graph distribution of the days of attention of all users to the school zone, and as can be seen from fig. 5, the attention degree corresponding to a user with 7 days of attention is about 0.7. Besides the school district, attention trend curves corresponding to room changing and visa of the attention object are shown.
The horizontal axis of fig. 6 represents the number of times of attention of all users to the target object, and the vertical axis of fig. 6 represents the degree of attention of users to the target object, i.e., logistic (f). For example, when the attention object is a school zone, the curve corresponding to the school zone in fig. 6 is a graph distribution of the number of times that all users pay attention to the school zone, and as can be seen from the graph, the attention degree corresponding to a user who pays attention to the school zone for 15 days is about 0.7. Besides the school district, attention trend curves corresponding to room changing, visa, middle school and immigration of the attention object are shown.
In addition, in order to further analyze the attention degree of the user to the target object, the restoration processing may be performed on fig. 5 and fig. 6 (attention trend curve determined according to the attention degree after the normalization processing) to obtain fig. 7 and fig. 8. That is, in fig. 5 and fig. 6, the attention of the vertical axis is normalized, for example, the actual number of times of all users is 1 to 1000 times, and after normalization, the value becomes between 0 and 1, and it is necessary to restore to the actual 1 to 1000 times.
Illustratively, as shown in fig. 7, the plateau region in fig. 7 corresponds to the beginning of the top of the attention trend curve shown in fig. 3, the slow decrease region in fig. 7 corresponds to the peak of the attention trend curve, and the dip region in fig. 7 corresponds to the end of the top of the attention trend curve. As shown in fig. 8, the first falling region in fig. 8 corresponds to the top of the attention trend curve shown in fig. 3, the steady rising region in fig. 7 corresponds to the top of the attention trend curve, and the second falling region in fig. 7 corresponds to the bottom of the attention trend curve.
Then, referring to S404, optionally, in an embodiment, a derivation calculation is performed on the attention trend curve to determine a first characteristic value and a second characteristic value of the attention trend curve, where the first characteristic value is smaller than the second characteristic value; determining the user corresponding to the online behavior data which is greater than the first characteristic value and smaller than the second characteristic value as the user corresponding to the first attention stage; and determining the user corresponding to the on-line behavior data larger than the second characteristic value as the user corresponding to the second attention stage.
The above step of performing derivation calculation on the attention trend curve and determining the first characteristic value and the second characteristic value of the attention trend curve may specifically include: the attention trend curve is determined according to the attention degree of the user to the target object, the inflection point (the first characteristic value and the second characteristic value) of the attention trend curve can be calculated, then the inflection point coefficient of the attention trend curve is determined according to the inflection point, and the parameter of the inflection point coefficient is reserved. When new online behavior data of the user are acquired, the parameters (b and c) of the inflection point coefficient can be updated by the new online behavior data of the user, so that the parameters are continuously updated, and finally a first characteristic value and a second characteristic value of the updated attention trend curve are determined.
For example, the results of the first characteristic value (left inflection point) and the second characteristic value (right inflection point) of the attention trend curve determined according to table 2 may be as shown in table 3:
TABLE 3
and determining the user corresponding to the online behavior data smaller than the first characteristic value to pay attention to the user corresponding to the prosperous period.
and determining the user corresponding to the online behavior data which is greater than the first characteristic value and smaller than the second characteristic value as the user corresponding to the first attention stage, namely the user corresponding to the attention peak period.
and determining the user corresponding to the on-line behavior data larger than the second characteristic value as the user corresponding to the second attention stage, namely the user corresponding to the attention top-end stage.
therefore, an attention trend curve is determined according to the attention degree of the target object of the users determined from multiple dimensions and multiple angles, an inflection point coefficient preferred by each user to the target object can be calculated, parameters of the inflection point coefficients are reserved, and the preference degree is objectively divided: in practical application, an inflection point of the initial full period is used as a threshold for identifying user preference, and whether a reinforced regression value (logistic (t) and/or logistic (f)) is greater than zero can be used as a judgment standard for identifying the user preference, so that the influence of human subjectivity can be greatly reduced, and the identification of the user preference and the stability of preference degree subdivision are improved.
Finally, referring to S405, accordingly, following the embodiment in S404, the user corresponding to the first attention stage is determined as the user corresponding to the target attention stage; or determining the user corresponding to the first attention stage and the user corresponding to the second attention stage as the users corresponding to the target attention stage.
The attention peak period can be determined as the user corresponding to the target attention stage, namely, the initial full period inflection point is used as the threshold for identifying the preference of the user, so that the influence of artificially and subjectively setting the threshold can be greatly reduced, and the identification of the preference of the user and the stability of the subdivision of the preference degree are improved. For example, when the target object focused by the user is "middle school", the user corresponding to the second focusing stage may be selected to be determined as the user corresponding to the target focusing stage, because generally, information related to middle school is needed by the user urgently, and therefore, when the user focuses attention to "buying room", the user may be determined as the user corresponding to the target focusing stage.
The concern prosperous period may also be determined as the user corresponding to the target concern stage, so as to increase the diversity of the user identification result, for example, in the case that the target object concerned by the user is "house buying", the user corresponding to the first concern stage may be selected to be determined as the user corresponding to the target concern stage, because house buying generally needs to go through a process, which is not urgently needed by the user, but is possibly needed by the user in a future period of time, so that in the case that the user slightly concerns about "house buying", the user may be determined as the user corresponding to the target concern stage.
Therefore, which user group is the user corresponding to the target attention stage can be flexibly selected according to different scenes.
As another implementation manner of the present application, in order to improve the recommendation effect, after S405, the following step may be further included:
determining a plurality of first attention degrees according to attention days of a user corresponding to the target attention stage to a plurality of target objects; determining a plurality of second attention degrees according to the attention times of the user corresponding to the target attention stage to the target objects; and determining the target object of which the sum of the first attention and the second attention is greater than or equal to a first threshold value and the absolute value of the difference between the first attention and the second attention is greater than or equal to a second threshold value as the first target object of the user corresponding to the target attention stage.
Considering that extreme situations occur, such as a situation with a few days but a particularly high number of times or a situation with a few days but a high number of times, which may interfere with the determination of the first target object of the user, the general processing method is divided into two types, i.e., deletion or continuous retention, the embodiment of the present invention adopts a deletion mode.
Optionally, in a possible embodiment, according to a customer interest preference scoring system, each interest is respectively distinguished to determine whether the customer prefers a certain interest point;
the extreme value discrimination rule is as follows:
rule one is as follows: logistic (t) + logistic (f) > (1.0);
rule two: | logistic (t) -logistic (f) | > | 0.8;
according to the first rule, part of users preferred to the attention point can be determined according to the characteristics of more days and more internet surfing times, and part of users preferred to the attention point can be determined according to the characteristics of higher days, more times, higher days and more times. However, the rule one contains partial extreme points, which affects the judgment, so the rule two is introduced to remove the situation of the extreme points.
TABLE 4
The parenthesized data in table 4 is the processed data.
As shown in the first diagram of fig. 9, the user who has a large number of attention days and a high number of attention days likes the target object very much; the users with more attention days and low attention times probably like the target object; users with less attention days and high times probably like the target object; the user who has low attention frequency and few days does not like the target object.
In this way, the extreme case (as shown in the second diagram in fig. 9) can be removed by the first rule and the second rule, and the target object after further filtering (as shown in the shaded area in the third diagram in fig. 9) is obtained, so that the accuracy of identifying the preference degree of the user for the target object can be further improved.
In summary, according to the user determination method provided by the embodiment of the invention, the attention trend curve of the user to the target object is determined according to the online behavior data of the plurality of users, and then the user corresponding to the target attention stage is determined according to the attention trend curve, so that the accuracy of determining the user meeting the preset condition can be improved.
Fig. 10 is a schematic structural diagram of a user determination apparatus according to an embodiment of the present invention.
As shown in fig. 10, the apparatus 100 may include:
the obtaining module 1010 is configured to obtain online behavior data of a plurality of users within a preset time period.
A first determining module 1020, configured to determine, according to the online behavior data, a degree of attention of each user to the target object.
And a second determining module 1030, configured to determine, according to the online behavior data of the multiple users and the attention degrees corresponding to the online behavior data, an attention trend curve of each user for the target object.
And a third determining module 1040, configured to determine, according to the attention trend curve of each user for the target object, an attention stage of each user for the target object.
The fourth determining module 1050 is configured to determine, from the multiple users, a user corresponding to the target attention stage according to the attention stage of each user to the target object.
As an example, the second determining module 1030 is specifically configured to perform the following steps for the online behavior data of each user: preprocessing the online behavior data of the user to obtain preprocessed online behavior data; determining the attention of a user to a target object according to the preprocessed online behavior data; carrying out normalization processing on the attention degree to obtain the attention degree after the normalization processing; and determining an attention trend curve according to the preprocessed on-line behavior data and the attention degree after the normalization processing.
As an example, the second determining module 1030 is specifically configured to calculate, based on a ralada criterion, on-line behavior data corresponding to a target object, and determine a data distribution range of the on-line behavior data; determining a third threshold and a fourth threshold corresponding to the target object according to the data distribution range and the preset value-taking interval; replacing the on-line behavior data smaller than the third threshold value in the on-line behavior data corresponding to the target object with the third threshold value; and replacing the on-line behavior data which is larger than the fourth threshold value in the on-line behavior data corresponding to the target object with the fourth threshold value.
As an example, the third determining module 1040 is specifically configured to perform derivation calculation on the attention trend curve, and determine a first characteristic value and a second characteristic value of the attention trend curve, where the first characteristic value is smaller than the second characteristic value; determining the user corresponding to the online behavior data which is greater than the first characteristic value and smaller than the second characteristic value as the user corresponding to the first attention stage; determining the user corresponding to the on-line behavior data larger than the second characteristic value as the user corresponding to the second attention stage; correspondingly, determining the user corresponding to the first attention stage as the user corresponding to the target attention stage; or determining the user corresponding to the first attention stage and the user corresponding to the second attention stage as the users corresponding to the target attention stage.
The related online behavior data comprise the attention days of each user to the target object and/or the attention times of each user to the target object.
The fourth determining module 1050 is further configured to determine a plurality of first attention degrees according to the attention days of the user to the plurality of target objects corresponding to the target attention stage; determining a plurality of second attention degrees according to the attention times of the user corresponding to the target attention stage to the target objects; and determining the target object of which the sum of the first attention and the second attention is greater than or equal to a first threshold value and the absolute value of the difference between the first attention and the second attention is greater than or equal to a second threshold value as the first target object of the user corresponding to the target attention stage.
In summary, the user determination device provided in the embodiment of the present invention determines the attention trend curve of the user to the target object according to the online behavior data of the plurality of users, and then determines the user corresponding to the target attention stage according to the attention trend curve, so that the accuracy of determining the user meeting the preset condition can be improved.
Fig. 11 is a diagram illustrating an exemplary hardware architecture provided by an embodiment of the present invention.
The computing device may include a processor 1101 and a memory 1102 storing computer program instructions.
Specifically, the processor 1101 may include a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or may be configured as one or more Integrated circuits implementing embodiments of the present invention.
The processor 1101 implements any of the user determination methods in the above embodiments by reading and executing computer program instructions stored in the memory 1102.
In one example, the positioning device can also include a communication interface 1103 and a bus 1110. As shown in fig. 11, the processor 1101, the memory 1102, and the communication interface 1103 are connected via a bus 1110 to complete communication therebetween.
The communication interface 1103 is mainly used for implementing communication between modules, apparatuses, units and/or devices in the embodiment of the present invention.
The processing device may execute the user determination method in the embodiment of the present invention, thereby implementing the user determination method described in conjunction with fig. 4.
In addition, in combination with the user determination method in the foregoing embodiment, the embodiment of the present invention may be implemented by providing a computer storage medium. The computer storage medium having computer program instructions stored thereon; the computer program instructions, when executed by a processor, implement any of the user determination methods in the above embodiments.
It is to be understood that the embodiments of the invention are not limited to the particular configurations and processes described above and shown in the drawings. A detailed description of known methods is omitted herein for the sake of brevity. In the above embodiments, several specific steps are described and shown as examples. However, the method processes of the embodiments of the present invention are not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications and additions or change the order between the steps after comprehending the spirit of the embodiments of the present invention.
The functional blocks shown in the above-described structural block diagrams may be implemented as software, and the elements of the embodiments of the present invention are programs or code segments used to perform desired tasks. The program or code segments may be stored in a machine-readable medium or transmitted by a data signal carried in a carrier wave over a transmission medium or a communication link. A "machine-readable medium" may include any medium that can store or transfer information. Examples of a machine-readable medium include circuits, semiconductor memory devices, ROM, flash memory, Erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, Radio Frequency (RF) links, and so forth. The code segments may be downloaded via a computer line, such as the internet, an intranet, etc.
It should also be noted that the exemplary embodiments mentioned in this patent describe some methods or systems based on a series of steps or devices. However, the embodiments of the present invention are not limited to the order of the above steps, that is, the steps may be performed in the order mentioned in the embodiments, may be performed in an order different from the order in the embodiments, or may be performed simultaneously.
As described above, only the specific embodiments of the present invention are provided, and it can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the system, the module and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. It should be understood that the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the present invention, and these modifications or substitutions should be covered within the scope of the present invention.
Claims (9)
1. A method for user determination, the method comprising:
acquiring online behavior data of a plurality of users within a preset time period;
determining the attention of each user to the target object according to the on-line behavior data;
determining attention trend curves of the multiple users to the target object according to the online behavior data of the multiple users and the attention degrees corresponding to the online behavior data;
determining the attention stage of each user to the target object according to the attention trend curve of each user to the target object;
and determining a user corresponding to the target attention stage from the plurality of users according to the attention stage of each user to the target object.
2. The method according to claim 1, wherein the determining the attention stage of each user to the target object according to the attention trend curve comprises:
performing derivation calculation on the attention trend curve, and determining a first characteristic value and a second characteristic value of the attention trend curve, wherein the first characteristic value is smaller than the second characteristic value;
determining the user corresponding to the on-line behavior data which is greater than the first characteristic value and smaller than the second characteristic value as the user corresponding to the first attention stage;
determining the user corresponding to the on-line behavior data larger than the second characteristic value as the user corresponding to the second attention stage;
correspondingly, the determining the user corresponding to the target attention stage from the plurality of users includes:
determining the user corresponding to the first attention stage as the user corresponding to the target attention stage; or, determining the user corresponding to the first attention stage and the user corresponding to the second attention stage as the users corresponding to the target attention stage.
3. The method of claim 1, wherein the online behavior data comprises days of interest of the each user for the target object and/or times of interest of the each user for the target object.
4. The method of claim 3, wherein after said determining the user corresponding to the target attention stage from the plurality of users, the method further comprises:
determining a plurality of first attention degrees according to attention days of the user corresponding to the target attention stage to the target objects;
determining a plurality of second attention degrees according to the attention times of the user corresponding to the target attention stage to the target objects;
and determining a target object of which the sum of the first attention and the second attention is greater than or equal to a first threshold value and the absolute value of the difference between the first attention and the second attention is greater than or equal to a second threshold value as a first target object of the user corresponding to the target attention stage.
5. The method according to claim 1, wherein determining attention trend curves of a plurality of users for the target object according to online behavior data of the plurality of users and the attention degrees corresponding to the online behavior data comprises:
the following steps are performed for each user's online behavior data:
preprocessing the online behavior data of the user to obtain preprocessed online behavior data;
determining the attention of the user to the target object according to the preprocessed online behavior data;
carrying out normalization processing on the attention degree to obtain the attention degree after the normalization processing;
and determining the attention trend curve according to the preprocessed on-line behavior data and the attention degree after the normalization processing.
6. The method of claim 5, wherein the preprocessing the online behavior data of the user to obtain the preprocessed online behavior data comprises:
calculating on-line behavior data corresponding to the target object based on a Lauda criterion, and determining a data distribution range of the on-line behavior data;
determining a third threshold and a fourth threshold corresponding to the target object according to the data distribution range and a preset value-taking interval;
replacing the on-line behavior data smaller than the third threshold in the on-line behavior data corresponding to the target object with the third threshold;
and replacing the on-line behavior data which is larger than the fourth threshold value in the on-line behavior data corresponding to the target object with the fourth threshold value.
7. A user identification device, comprising:
the acquisition module is used for acquiring online behavior data of a plurality of users within a preset time period;
the first determining module is used for determining the attention degree of each user to the target object according to the online behavior data;
a second determining module, configured to determine, according to the online behavior data of the multiple users and the attention degree corresponding to the online behavior data, an attention trend curve of each user for the target object;
the third determination module is used for determining the attention stage of each user to the target object according to the attention trend curve of each user to the target object;
and the fourth determining module is used for determining the user corresponding to the target attention stage from the plurality of users according to the attention stage of each user to the target object.
8. A computing device, the device comprising: a processor and a memory storing computer program instructions;
the processor, when executing the computer program instructions, implements the user determination method of any of claims 1-6.
9. A computer storage medium having computer program instructions stored thereon which, when executed by a processor, implement the user determination method of any one of claims 1-6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010363873.4A CN113590925A (en) | 2020-04-30 | 2020-04-30 | User determination method, device, equipment and computer storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010363873.4A CN113590925A (en) | 2020-04-30 | 2020-04-30 | User determination method, device, equipment and computer storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113590925A true CN113590925A (en) | 2021-11-02 |
Family
ID=78237330
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010363873.4A Pending CN113590925A (en) | 2020-04-30 | 2020-04-30 | User determination method, device, equipment and computer storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113590925A (en) |
Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102681999A (en) * | 2011-03-08 | 2012-09-19 | 阿里巴巴集团控股有限公司 | Method and device for collecting and sending user action information |
US20170011420A1 (en) * | 2015-07-10 | 2017-01-12 | The Nielsen Company (Us), Llc | Methods and apparatus to analyze and adjust age demographic information |
CN106682225A (en) * | 2017-01-04 | 2017-05-17 | 成都四方伟业软件股份有限公司 | Big data collecting and storing method and system |
CN107391712A (en) * | 2017-07-28 | 2017-11-24 | 王亚迪 | A kind of network public opinion trend prediction analysis method |
CN107679916A (en) * | 2017-10-12 | 2018-02-09 | 北京京东尚科信息技术有限公司 | For obtaining the method and device of user interest degree |
CN107784390A (en) * | 2017-10-19 | 2018-03-09 | 北京京东尚科信息技术有限公司 | Recognition methods, device, electronic equipment and the storage medium of subscriber lifecycle |
CN107943953A (en) * | 2017-11-24 | 2018-04-20 | 福建中金在线信息科技有限公司 | List recommends method, apparatus, electronic equipment and computer-readable recording medium |
CN107992601A (en) * | 2017-12-14 | 2018-05-04 | 上海宽全智能科技有限公司 | Trend prediction analysis method, equipment and storage medium |
CN108833458A (en) * | 2018-04-02 | 2018-11-16 | 腾讯科技(深圳)有限公司 | A kind of application recommended method, device, medium and equipment |
CN109241133A (en) * | 2018-08-14 | 2019-01-18 | 北京粉笔未来科技有限公司 | Data monitoring method, calculates equipment and storage medium at device |
CN109800483A (en) * | 2018-12-29 | 2019-05-24 | 北京城市网邻信息技术有限公司 | A kind of prediction technique, device, electronic equipment and computer readable storage medium |
CN109840788A (en) * | 2017-11-27 | 2019-06-04 | 北京京东尚科信息技术有限公司 | For analyzing the method and device of user behavior data |
CN110442801A (en) * | 2019-07-26 | 2019-11-12 | 新华三信息安全技术有限公司 | A kind of determination method and device of the concern user of object event |
CN111078386A (en) * | 2019-12-30 | 2020-04-28 | 北京奇艺世纪科技有限公司 | Control method and control device of distributed scheduling system |
-
2020
- 2020-04-30 CN CN202010363873.4A patent/CN113590925A/en active Pending
Patent Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102681999A (en) * | 2011-03-08 | 2012-09-19 | 阿里巴巴集团控股有限公司 | Method and device for collecting and sending user action information |
US20170011420A1 (en) * | 2015-07-10 | 2017-01-12 | The Nielsen Company (Us), Llc | Methods and apparatus to analyze and adjust age demographic information |
CN106682225A (en) * | 2017-01-04 | 2017-05-17 | 成都四方伟业软件股份有限公司 | Big data collecting and storing method and system |
CN107391712A (en) * | 2017-07-28 | 2017-11-24 | 王亚迪 | A kind of network public opinion trend prediction analysis method |
CN107679916A (en) * | 2017-10-12 | 2018-02-09 | 北京京东尚科信息技术有限公司 | For obtaining the method and device of user interest degree |
CN107784390A (en) * | 2017-10-19 | 2018-03-09 | 北京京东尚科信息技术有限公司 | Recognition methods, device, electronic equipment and the storage medium of subscriber lifecycle |
CN107943953A (en) * | 2017-11-24 | 2018-04-20 | 福建中金在线信息科技有限公司 | List recommends method, apparatus, electronic equipment and computer-readable recording medium |
CN109840788A (en) * | 2017-11-27 | 2019-06-04 | 北京京东尚科信息技术有限公司 | For analyzing the method and device of user behavior data |
CN107992601A (en) * | 2017-12-14 | 2018-05-04 | 上海宽全智能科技有限公司 | Trend prediction analysis method, equipment and storage medium |
CN108833458A (en) * | 2018-04-02 | 2018-11-16 | 腾讯科技(深圳)有限公司 | A kind of application recommended method, device, medium and equipment |
CN109241133A (en) * | 2018-08-14 | 2019-01-18 | 北京粉笔未来科技有限公司 | Data monitoring method, calculates equipment and storage medium at device |
CN109800483A (en) * | 2018-12-29 | 2019-05-24 | 北京城市网邻信息技术有限公司 | A kind of prediction technique, device, electronic equipment and computer readable storage medium |
CN110442801A (en) * | 2019-07-26 | 2019-11-12 | 新华三信息安全技术有限公司 | A kind of determination method and device of the concern user of object event |
CN111078386A (en) * | 2019-12-30 | 2020-04-28 | 北京奇艺世纪科技有限公司 | Control method and control device of distributed scheduling system |
Non-Patent Citations (1)
Title |
---|
周琦等: "基于WPF的水槽控制及数据采集系统设计与实现", 《实验技术与管理》, 22 March 2018 (2018-03-22), pages 78 - 83 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107633265B (en) | Data processing method and device for optimizing credit evaluation model | |
CN112258093B (en) | Data processing method and device for risk level, storage medium and electronic equipment | |
WO2017143919A1 (en) | Method and apparatus for establishing data identification model | |
CN111914090B (en) | Method and device for enterprise industry classification identification and characteristic pollutant identification | |
Kočišová et al. | Discriminant analysis as a tool for forecasting company's financial health | |
CN111079941B (en) | Credit information processing method, credit information processing system, terminal and storage medium | |
CN106164896B (en) | Multi-dimensional recursion method and system for discovering counterparty relationship | |
CN113222149B (en) | Model training method, device, equipment and storage medium | |
CN112150298A (en) | Data processing method, system, device and readable medium | |
CN110634060A (en) | User credit risk assessment method, system, device and storage medium | |
CN113704389A (en) | Data evaluation method and device, computer equipment and storage medium | |
CN113553319A (en) | LOF outlier detection cleaning method, device and equipment based on information entropy weighting and storage medium | |
CN110990384B (en) | Big data platform BI analysis method | |
CN116719714A (en) | Training method and corresponding device for screening model of test case | |
CN113590925A (en) | User determination method, device, equipment and computer storage medium | |
US20230385664A1 (en) | A computer-implemented method for deriving a data processing and inference pipeline | |
CN117035472A (en) | Enterprise operation health assessment method, model construction method and device | |
CN118333737A (en) | Method for constructing retail credit risk prediction model and consumer credit business Scorebetai model | |
CN116187739A (en) | Visitor risk assessment supervision method, system and equipment based on BP neural network | |
CN116843473A (en) | Stock data acquisition method, system, equipment and storage medium based on cloud computing | |
CN118071482A (en) | Method for constructing retail credit risk prediction model and consumer credit business Scorebetad model | |
CN117217522A (en) | Financial pre-billing risk management and control system based on artificial intelligence and operation method thereof | |
CN117081947A (en) | Method, device, processing equipment and storage medium for identifying unsatisfied users | |
CN118333738A (en) | Method for constructing retail credit risk prediction model and credit card service Scorealpha model | |
CN114596152A (en) | Method, device and storage medium for predicting debt subject default based on unsupervised model |
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 |