CN111666309A

CN111666309A - Data processing method and device, electronic equipment and computer readable storage medium

Info

Publication number: CN111666309A
Application number: CN202010515174.7A
Authority: CN
Inventors: 姜谷雨
Original assignee: Rajax Network Technology Co Ltd
Current assignee: Rajax Network Technology Co Ltd
Priority date: 2020-06-08
Filing date: 2020-06-08
Publication date: 2020-09-15

Abstract

The embodiment of the disclosure discloses a data processing method, a data processing device, an electronic device and a computer readable storage medium, wherein the data processing method comprises the following steps: acquiring historical behavior data; determining multidimensional behavior characteristics according to the historical behavior data, wherein the multidimensional behavior characteristics at least comprise behavior operator characteristics, behavior related party characteristics and behavior related object characteristics; and executing preset operation based on the multi-dimensional behavior characteristics. According to the technical scheme, the extraction dimensionality of the interactive behavior features is increased, the characteristics of the interactive behavior can be comprehensively reflected, effective and correct data support can be provided for operations such as follow-up interactive behavior prediction and service data adjustment, improvement of service quality, increase of service opportunities and improvement of service efficiency are facilitated.

Description

Data processing method and device, electronic equipment and computer readable storage medium

Technical Field

The present disclosure relates to the field of data processing technologies, and in particular, to a data processing method and apparatus, an electronic device, and a computer-readable storage medium.

Background

With the development of internet technology, more and more service providers provide services and information through internet platforms, and the characteristics of interaction behaviors among service receivers, service providers and the internet platforms are mined, so that the service quality can be improved, the service opportunities are increased, and the service efficiency is improved. In the prior art, the interactive behavior is analyzed or the interactive behavior is predicted and the like by mining the service receiver, such as the user class characteristics, and the service provider, such as the merchant class characteristics, but the extraction dimension of the interactive behavior characteristics in the prior art is low, so that the characteristics of the interactive behavior cannot be comprehensively reflected, effective and correct data support cannot be provided for subsequent operations such as interactive behavior prediction and service data adjustment, and the improvement of service quality, service opportunity and service efficiency is not facilitated.

Disclosure of Invention

The embodiment of the disclosure provides a data processing method and device, electronic equipment and a computer-readable storage medium.

In a first aspect, an embodiment of the present disclosure provides a data processing method.

Specifically, the data processing method includes:

acquiring historical behavior data;

determining multidimensional behavior characteristics according to the historical behavior data, wherein the multidimensional behavior characteristics at least comprise behavior operator characteristics, behavior related party characteristics and behavior related object characteristics;

and executing preset operation based on the multi-dimensional behavior characteristics.

With reference to the first aspect, in a first implementation manner of the first aspect, the determining a multidimensional behavior feature according to the historical behavior data includes:

determining multidimensional initial behavior characteristics according to the historical behavior data;

and carrying out classification correction on the initial behavior characteristics to obtain the multidimensional behavior characteristics.

With reference to the first aspect and the first implementation manner of the first aspect, in a second implementation manner of the first aspect, the determining a multidimensional initial behavior feature according to the historical behavior data includes:

acquiring historical behavior data attribute information, wherein the historical behavior data attribute information at least comprises behavior operator attribute information, behavior related party attribute information and behavior related object attribute information;

classifying the historical behavior data according to the historical behavior data attribute information to obtain behavior operator data, behavior related party data and behavior related object data;

and extracting the behavior operator characteristics, the behavior related party characteristics and the behavior related object characteristics based on the behavior operator data, the behavior related party data and the behavior related object data.

With reference to the first aspect, the first implementation manner of the first aspect, and the second implementation manner of the first aspect, in a third implementation manner of the first aspect, the performing classification and correction on the initial behavior feature to obtain the multidimensional behavior feature includes:

classifying the initial behavior features to obtain a first class of initial behavior features, a second class of initial behavior features and a third class of initial behavior features;

performing first correction processing on the first-class initial behavior characteristics to obtain first-class behavior characteristics;

performing second correction processing on the second type initial behavior characteristics to obtain second type behavior characteristics;

and combining the first category of behavior characteristics, the second category of behavior characteristics and the third category of initial behavior characteristics to obtain the multi-dimensional behavior characteristics.

With reference to the first aspect, the first implementation manner of the first aspect, the second implementation manner of the first aspect, and the third implementation manner of the first aspect, in a fourth implementation manner of the first aspect, the performing a first modification process on the first category initial behavior features to obtain the first category behavior features includes:

acquiring attribute information of the first-class initial behavior characteristics, and determining time correction factors corresponding to the first-class initial behavior characteristics according to the attribute information of the first-class initial behavior characteristics;

and determining a first category behavior characteristic corresponding to the first category initial behavior characteristic based on the time correction factor.

With reference to the first aspect, the first implementation manner of the first aspect, the second implementation manner of the first aspect, the third implementation manner of the first aspect, and the fourth implementation manner of the first aspect, in a fifth implementation manner of the first aspect, the performing a second correction process on the second category initial behavior feature to obtain a second category behavior feature includes:

determining a confidence coefficient correction factor, and acquiring the total number of statistical samples corresponding to the second category initial behavior characteristics;

and calculating to obtain second category behavior characteristics corresponding to the second category initial behavior characteristics based on the confidence coefficient correction factor and the total number of the statistical samples.

With reference to the first aspect, the first implementation manner of the first aspect, the second implementation manner of the first aspect, the third implementation manner of the first aspect, the fourth implementation manner of the first aspect, and the fifth implementation manner of the first aspect, in a sixth implementation manner of the first aspect, the combining the first category behavior feature, the second category behavior feature, and the third category initial behavior feature to obtain the multidimensional behavior feature includes:

combining the first category behavior characteristics, the second category behavior characteristics and the third category initial behavior characteristics to obtain a behavior characteristic set;

and acquiring attribute information of the first category behavior feature, the second category behavior feature and the third category initial behavior feature, and dividing the behavior feature set into a behavior operator feature, a behavior related party feature and a behavior related object feature according to the attribute information of the first category behavior feature, the second category behavior feature and the third category initial behavior feature to obtain the multidimensional behavior feature.

With reference to the first aspect, the first implementation manner of the first aspect, the second implementation manner of the first aspect, the third implementation manner of the first aspect, the fourth implementation manner of the first aspect, the fifth implementation manner of the first aspect, and the sixth implementation manner of the first aspect, in a seventh implementation manner of the first aspect, the executing a preset operation based on the multidimensional behavior feature includes:

determining a target behavior related party and/or a target behavior related object based on the multi-dimensional behavior characteristics, and sending the target behavior related party and/or the target behavior related object to the target operator; and/or the presence of a gas in the gas,

determining a target behavior related party and/or a target behavior related object based on the multi-dimensional behavior characteristics, and executing preset processing on the target behavior related party and/or the target behavior related object; and/or the presence of a gas in the gas,

and training the multi-dimensional behavior characteristics as training data to obtain a behavior prediction model, and predicting future behavior data based on the behavior prediction model and the current behavior data.

In a second aspect, a data processing apparatus is provided in an embodiment of the present disclosure.

Specifically, the data processing apparatus includes:

an acquisition module configured to acquire historical behavior data;

a determining module configured to determine a multi-dimensional behavior feature according to the historical behavior data, wherein the multi-dimensional behavior feature at least comprises a behavior operator feature, a behavior related party feature, and a behavior related object feature;

and the execution module is configured to execute preset operation based on the multi-dimensional behavior characteristics.

With reference to the second aspect, in a first implementation manner of the second aspect, the determining module is configured to:

With reference to the second aspect and the first implementation manner of the second aspect, in a second implementation manner of the second aspect, the determining, according to the historical behavior data, the multidimensional initial behavior feature part is configured to:

With reference to the second aspect, the first implementation manner of the second aspect, and the second implementation manner of the second aspect, in a third implementation manner of the second aspect, the classifying and modifying the initial behavior feature to obtain the multidimensional behavior feature is configured to:

With reference to the second aspect, the first implementation manner of the second aspect, the second implementation manner of the second aspect, and the third implementation manner of the second aspect, in a fourth implementation manner of the second aspect, in an embodiment of the present disclosure, the performing a first modification process on the first category initial behavior feature to obtain a first category behavior feature is configured to:

With reference to the second aspect, the first implementation manner of the second aspect, the second implementation manner of the second aspect, the third implementation manner of the second aspect, and the fourth implementation manner of the second aspect, in a fifth implementation manner of the second aspect, in the embodiment of the present disclosure, the part that performs the second correction processing on the second category initial behavior feature to obtain the second category behavior feature is configured to:

With reference to the second aspect, the first implementation manner of the second aspect, the second implementation manner of the second aspect, the third implementation manner of the second aspect, the fourth implementation manner of the second aspect, and the fifth implementation manner of the second aspect, in a sixth implementation manner of the second aspect, the combining the first category behavior feature, the second category behavior feature, and the third category initial behavior feature to obtain the multidimensional behavior feature is configured to:

With reference to the second aspect, the first implementation manner of the second aspect, the second implementation manner of the second aspect, the third implementation manner of the second aspect, the fourth implementation manner of the second aspect, the fifth implementation manner of the second aspect, and the sixth implementation manner of the second aspect, in a seventh implementation manner of the second aspect, the execution module is configured to:

In a third aspect, the disclosed embodiments provide an electronic device, comprising a memory and at least one processor, wherein the memory is configured to store one or more computer instructions, and wherein the one or more computer instructions are executed by the at least one processor to implement the method steps of the above data processing method.

In a fourth aspect, the disclosed embodiments provide a computer-readable storage medium for storing computer instructions for a data processing apparatus, which contains computer instructions for executing the data processing method described above as a data processing apparatus.

The technical scheme provided by the embodiment of the disclosure can have the following beneficial effects:

according to the technical scheme, the multi-dimensional behavior characteristics capable of comprehensively reflecting the characteristics of the interactive behaviors are extracted according to the historical behavior data, and then the preset operation is executed based on the multi-dimensional behavior characteristics. According to the technical scheme, the extraction dimensionality of the interactive behavior features is increased, the characteristics of the interactive behavior can be comprehensively reflected, effective and correct data support can be provided for operations such as follow-up interactive behavior prediction and service data adjustment, improvement of service quality, increase of service opportunities and improvement of service efficiency are facilitated.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

Drawings

Other features, objects, and advantages of the present disclosure will become more apparent from the following detailed description of non-limiting embodiments when taken in conjunction with the accompanying drawings. In the drawings:

FIG. 1 shows a flow diagram of a data processing method according to an embodiment of the present disclosure;

FIG. 2 illustrates a multi-dimensional behavioral characteristic diagram according to an embodiment of the present disclosure;

FIG. 3 illustrates a multi-dimensional behavioral characteristic processing and acquisition diagram according to an embodiment of the present disclosure;

FIG. 4 shows a block diagram of a data processing apparatus according to an embodiment of the present disclosure;

FIG. 5 shows a block diagram of an electronic device according to an embodiment of the present disclosure;

FIG. 6 is a schematic block diagram of a computer system suitable for use in implementing a data processing method according to an embodiment of the present disclosure.

Detailed Description

Hereinafter, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily implement them. Also, for the sake of clarity, parts not relevant to the description of the exemplary embodiments are omitted in the drawings.

In the present disclosure, it is to be understood that terms such as "including" or "having," etc., are intended to indicate the presence of the disclosed features, numbers, steps, behaviors, components, parts, or combinations thereof, and are not intended to preclude the possibility that one or more other features, numbers, steps, behaviors, components, parts, or combinations thereof may be present or added.

It should be further noted that the embodiments and features of the embodiments in the present disclosure may be combined with each other without conflict. The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

According to the technical scheme provided by the embodiment of the disclosure, the multidimensional behavior characteristics capable of comprehensively embodying the characteristics of the interactive behaviors are extracted according to the historical behavior data, and then the preset operation is executed based on the multidimensional behavior characteristics. According to the technical scheme, the extraction dimensionality of the interactive behavior features is increased, the characteristics of the interactive behavior can be comprehensively reflected, effective and correct data support can be provided for operations such as follow-up interactive behavior prediction and service data adjustment, improvement of service quality, increase of service opportunities and improvement of service efficiency are facilitated.

Fig. 1 shows a flowchart of a data processing method according to an embodiment of the present disclosure, which includes the following steps S101 to S103, as shown in fig. 1:

in step S101, historical behavior data is acquired;

in step S102, determining a multidimensional behavior feature according to the historical behavior data, wherein the multidimensional behavior feature at least includes a behavior operator feature, a behavior related party feature, and a behavior related object feature;

in step S103, a preset operation is performed based on the multidimensional behavior feature.

As mentioned above, with the development of internet technology, more and more service providers provide services and information through internet platforms, and mining the characteristics of interaction between a service receiver, a service provider, and an internet platform can be beneficial to improving the service quality, increasing the service opportunities, and improving the service efficiency. In the prior art, the interactive behavior is analyzed or the interactive behavior is predicted and the like by mining the service receiver, such as the user class characteristics, and the service provider, such as the merchant class characteristics, but the extraction dimension of the interactive behavior characteristics in the prior art is low, so that the characteristics of the interactive behavior cannot be comprehensively reflected, effective and correct data support cannot be provided for subsequent operations such as interactive behavior prediction and service data adjustment, and the improvement of service quality, service opportunity and service efficiency is not facilitated.

In view of the above drawbacks, in this embodiment, a data processing method is provided, which extracts a multidimensional behavior feature that can comprehensively embody characteristics of an interactive behavior according to historical behavior data, and then performs a preset operation based on the multidimensional behavior feature. According to the technical scheme, the extraction dimensionality of the interactive behavior features is increased, the characteristics of the interactive behavior can be comprehensively reflected, effective and correct data support can be provided for operations such as follow-up interactive behavior prediction and service data adjustment, improvement of service quality, increase of service opportunities and improvement of service efficiency are facilitated.

In an optional implementation manner of this embodiment, the historical behavior data refers to behavior data generated in a preset historical time period, where the behavior may be, for example, an interaction behavior such as a click behavior, a collection behavior, a transaction behavior, and the like between a service receiver such as a user receiving a service, a device or apparatus receiving a service, and an internet platform such as a merchant providing a service, a device or apparatus providing a service, and the like, and the internet platform refers to a network platform providing data interaction and communication for the service receiver and the service provider. For convenience of description, the following explains and explains the technical solution of the present disclosure by taking the behaviors as interaction behaviors among a service receiver, a service provider and an internet platform, the service receiver is a user, and the service provider is a merchant.

In an optional implementation manner of this embodiment, the behavior feature refers to a feature that can embody characteristics of the behavior data, for example, a feature that can embody characteristics of a user related to a certain behavior or a merchant related to a certain behavior. As mentioned above, the interactive behavior features mined in the prior art include a service receiver such as a user class feature and a service provider such as a merchant class feature, that is, the interactive behavior features mined in the prior art only have a user dimension and a merchant dimension, but in many interactive scenes, a user can obtain some information desired by the user by inputting a search object such as a search word, and in these scenes, a certain association relationship exists between the search object and a search result, and a certain association relationship exists between the search result and the service provider, so that if the search object features can be extracted, the search object features are combined with the user class feature and the merchant class feature, and more effective and correct data support can be provided for subsequent operations such as interactive behavior prediction, service data adjustment, and the like. Therefore, in an optional implementation manner of this embodiment, the multidimensional behavior feature at least includes a behavior operator feature, a behavior related party feature, and a behavior related object feature, where:

the behavior operator characteristics may be, for example, user characteristics, and the user characteristics may further include user attribute characteristics such as user age, occupation, and whether the behavior operator characteristics are members; user statistical characteristics such as the next order amount in a user preset time period, the click rate in the user preset time period, the conversion rate in the user preset time period and the like; and user behavior characteristics such as a merchant list clicked by the user, a merchant list listed by the user, a merchant list collected by the user and the like. The click rate refers to a ratio of the number of times that a certain content displayed to a user is clicked to the number of times that the content is displayed, and if the content is merchant information display content, the click rate can be obtained by dividing the number of merchants clicked by the user on a certain website or a certain webpage by the total number of merchants that the user can see on the website or the webpage; the conversion rate refers to a ratio of the number of times that a certain content is placed or traded and the number of times that the content is displayed, and if the content is merchant information display content, the conversion rate can be obtained by dividing the number of merchants where the user places an order or trades on a certain website or a certain webpage by the total number of merchants that the user can see on the website or the webpage; the click through rate and conversion rate can be used to characterize the degree to which the merchant is engaging the user.

The behavior-related party characteristics may be, for example, merchant characteristics, and the merchant characteristics may further include merchant attribute characteristics such as merchant business state, merchant city, and whether the merchant is a sole merchant; and the merchant statistical characteristics such as the exposure in the merchant preset time period, the clicked amount in the merchant preset time period, the clicked rate in the merchant preset time period and the like.

The behavior related object characteristics may be characteristics of a search object, such as a search word used by a user, and the search object characteristics may further include at least statistical characteristics of the number of times of occurrence of the search object, a click rate of the search object, a conversion rate of the search object, and the like.

The multi-dimensional behavior characteristics can comprehensively reflect the characteristics of the interactive behavior, not only can provide effective and correct data support for operations such as subsequent interactive behavior prediction and service data adjustment, but also is beneficial to improving the service quality, increasing the service opportunity and improving the service efficiency.

After the multi-dimensional behavior feature is obtained, a subsequent preset operation can be executed based on the multi-dimensional behavior feature.

In an optional implementation manner of this embodiment, the step S102, that is, the step of determining the multidimensional behavior feature according to the historical behavior data, may include the following steps:

Considering that the behavior of the user usually has periodicity, some of the behavior features are periodically characterized, for example, user statistical features such as a next order amount within a user preset time period, a click rate within the user preset time period, a conversion rate within the user preset time period, and other merchant statistical features such as an exposure amount within a merchant preset time period, a clicked amount within the merchant preset time period, and a clicked rate within the merchant preset time period. If the above-mentioned features with periodicity characteristics are simply and directly used as the data basis for subsequent operations such as interactive behavior prediction, service data adjustment, etc., without considering the periodicity factor, the accuracy of the subsequent operation result may be reduced. For example, if the order quantity of the merchant in a certain month is to be counted, the prior art scheme is to directly sum up the order quantities of the merchant a in the month, and assume that the order quantity data of the merchant a in the certain month is: the order quantity 1 is 10, the order quantity 15 is 20, the order quantity 30 is 15, and the order quantity of the merchant A in the month is 45; suppose that the order quantity data of the monthly merchant B is: the order quantity 1 is 5, the order quantity 15 is 5, the order quantity 30, the order quantity of the merchant B in the month is 40, if the merchant information pushed to the user is determined simply based on the above order quantity statistical data, it is obvious that the merchant a with the larger order quantity in the month is more advantageous, but it can be found by carefully analyzing the data of the merchant a and the merchant B, although the amount of the orders of the merchant B in the month is not as much as that of the merchant A, the merchant B improves the sales mode when the number of the merchant B is 30, increases the discount strength, therefore, the order size of 30 # is rapidly increased, and by combining the above analysis, the information of the merchant B should be pushed to the user, the deviation of the pushed information is just because the periodic time factor is not considered when the characteristics are used, and therefore, the characteristics with the periodic characteristics need to be corrected in time factor.

In addition, considering that some of the behavior features are probabilistic features, for example, user statistical features such as click rate within a user preset time period, conversion rate within a user preset time period, and merchant statistical features such as clicked amount within a merchant preset time period, clicked rate within a merchant preset time period, and the like, and in many interaction scenes, types of merchants are relatively rich, for example, in a sales scene, various types of merchants such as flowers, medicines, supermarkets, and the like may exist, for some merchants, because of reasons of the related fields, frequencies of user click and order placement are relatively low, for example, flower and medical merchants, overall data of the type of merchants are relatively sparse, confidence of the statistical features is relatively low, if the features with the probabilistic features are simply and directly used as data bases for operations such as subsequent interactive behavior prediction, service data adjustment, and the like, regardless of the probabilistic factors, the accuracy of the results of the subsequent operations may be reduced. For example, the merchant information pushed to the user is determined based on an exposure conversion rate of a merchant in a preset time period, where the exposure conversion rate refers to a ratio obtained by dividing an order amount of a certain merchant in the preset time period by an exposure amount in the preset time period. Assuming that a merchant a in a supermarket type is exposed 1000 times in a week to generate 400 orders, the exposure conversion rate of the merchant a is 400/1000-40%, and a fresh flower merchant B is exposed 2 times in the week to generate 1 order, the exposure conversion rate of the merchant B is 1/2-50%, if the merchant information pushed to the user is determined simply based on the above exposure conversion statistical data, it is obvious that the merchant B with higher exposure conversion rate in the week is more advantageous, but if the statistical time in the week is prolonged to one month, it is found that after one month, the exposure conversion rate of the merchant a is relatively stable and basically floats around 40%, but the merchant B has 10 more exposures but the order amount is only one more, that is, the exposure conversion rate of the merchant B is reduced to 16.6% in the one month statistical time, and in the month, the two merchants have no change in operation and discount, and by combining the above analysis, the information of the merchant a, but not the information of the merchant B, should be pushed to the user at this time, and the deviation of the pushed information is just the reason that the probabilistic confidence factor is not considered when using the above features, that is, the reliability of the statistical data is not sufficient due to the insufficient sample basis, so that the features with probabilistic characteristics need to be corrected on the confidence factor.

In view of the above, in this implementation, it is necessary to perform further classified modification on the multidimensional behavior features obtained from the historical behavior data to obtain effective multidimensional behavior features serving as a data basis for subsequent operations such as interactive behavior prediction and service data adjustment.

In an optional implementation manner of this embodiment, the step of determining the multi-dimensional initial behavior feature according to the historical behavior data may include the following steps:

As mentioned above, the multidimensional behavior features obtained from the historical behavior data need to be further classified and modified to obtain effective multidimensional behavior features serving as data bases for subsequent operations such as interactive behavior prediction and service data adjustment, and therefore, in this implementation, the historical behavior data needs to be firstly classified to obtain information related to behavior operators, information related to behavior correlation parties, and information related to behavior correlation objects, that is, initial behavior features, and then the features needing modification in the initial behavior features need to be modified according to categories.

When obtaining an initial behavior feature, first obtaining historical behavior data attribute information, where the historical behavior data attribute information at least includes behavior operator attribute information, behavior related party attribute information, and behavior related object attribute information, where the attribute information is used to represent a related subject of the historical behavior data, and may be category information and/or identification information of the related subject, for example, the behavior operator attribute information may be category information and identification information of the behavior operator, such as a user class and a user ID, and is used to represent a related subject of some historical behavior data as the behavior operator; the attribute information of the behavior-related party may be category information and identification information of the behavior-related party, such as a merchant class and a merchant ID, and is used to represent that a related subject of some historical behavior data is the behavior-related party; the behavior related object attribute information may be category information and identification information of the behavior related object, such as a search object and search word content, and a related subject used for representing a certain historical behavior data is the behavior related object.

After obtaining the attribute information of the behavior operating party, the attribute information of the behavior related party and the attribute information of the behavior related object, the historical behavior data can be classified according to the attribute information of the historical behavior data, so that the behavior operating party data, the behavior related party data and the behavior related object data are obtained.

And finally, based on the behavior operator data, the behavior related party data and the behavior related object data, the behavior operator characteristics, the behavior related party characteristics and the behavior related object characteristics can be extracted and obtained.

In an optional implementation manner of this embodiment, the step of performing classification and correction on the initial behavior feature to obtain the multidimensional behavior feature may include the following steps:

As mentioned above, the behavior feature with periodic characteristics needs to be modified in terms of time, and the behavior feature with probabilistic characteristics needs to be modified in terms of confidence, so that when the initial behavior feature is classified and modified in this embodiment, the initial behavior feature is firstly classified according to the specific behavior feature to obtain a first category initial behavior feature, a second category initial behavior feature and a third category initial behavior feature, where the first category initial behavior feature refers to a behavior feature with periodic characteristics, the second category initial behavior feature refers to a behavior feature with probabilistic characteristics, and the third category initial behavior feature refers to other behavior features that do not need special processing; then, carrying out first correction processing on the first category initial behavior characteristics to obtain first category behavior characteristics, and carrying out second correction processing on the second category initial behavior characteristics to obtain second category behavior characteristics, wherein the first correction processing refers to correction based on time factors, and the second correction processing refers to correction based on confidence coefficient factors; finally, the processed first category behavior features and the second category behavior features are combined with the third category initial behavior features which do not need to be specially processed, so that the multidimensional behavior features can be obtained.

In an optional implementation manner of this embodiment, the step of performing a first modification process on the first-class initial behavior feature to obtain a first-class behavior feature may include the following steps:

The attribute information of the first-class initial behavior feature may include statistical duration of the first-class initial behavior feature, and may further include behavior operator attribute information, behavior related party attribute information, behavior related object attribute information, and other information, that is, it may be determined, through the behavior operator attribute information, the behavior related party attribute information, the behavior related object attribute information, and other information, whether the first-class initial behavior feature belongs to a feature of a behavior operator, a feature of a behavior related party, or a feature of a behavior related object.

In this implementation, the first category of behavior features may be represented as:

N＝N₀×W，

wherein N represents a first category behavior characteristic obtained after the first correction processing, and N₀Representing a first category of initial behavior characteristics before the first modification process, W being a temporal modification factor, in an alternative implementation of this embodiment, the temporal modification factor W is e^(-kt)The time correction factor is a half-decay time correction factor, where e is a natural number, k is a half-decay constant, and t is a statistical time length, that is, the time correction factor W is related to the statistical time length, and therefore, after the attribute information of the first-class initial behavior feature is obtained, the time correction factor corresponding to the first-class initial behavior feature can be determined according to the attribute information of the first-class initial behavior feature.

In this implementation, N may be made 0.5, N₀Determining the half-decay constant k as 1, and correcting the time by the expression W as e^(-kt)Substituting the above formula to obtain 0.5 ═ 1 × e^(-kt)Assuming that the statistical duration is 30 days, the half-decay constant k is 0.023. Namely, after the statistical time length is determined, the half-decay constant k and the value of the time correction factor W can be calculated, and then the first-class initial behavior characteristic N is used₀And multiplying the value of the time correction factor W to obtain a first-class behavior characteristic N corresponding to the first-class initial behavior characteristic after the first correction processing.

In an optional implementation manner of this embodiment, the step of performing a second correction process on the second category initial behavior feature to obtain a second category behavior feature may include the following steps:

In an optional implementation manner of this embodiment, the confidence correction factor may be 1.96, which represents a confidence interval of 95%.

In this implementation, the second category behavior characteristic may be calculated by:

wherein p represents a second category behavior characteristic corresponding to the second category initial behavior characteristic obtained after the second correction processing, and p₀Representing the second category initial behavior characteristics before the second correction processing, z representing a confidence coefficient correction factor, and n representing the total number of statistical samples corresponding to the second category initial behavior characteristics.

In an optional implementation manner of this embodiment, the step of combining the first category behavior feature, the second category behavior feature, and the third category initial behavior feature to obtain the multidimensional behavior feature may include the following steps:

The attribute information of the first category behavior feature, the second category behavior feature and the third category initial behavior feature at least comprises behavior operator attribute information, behavior related party attribute information and behavior related object attribute information, namely, the first category behavior feature, the second category behavior feature and the third category initial behavior feature can be determined to be respectively a feature subordinate to a behavior operator, a feature subordinate to a behavior related party or a feature subordinate to a behavior related object according to the behavior operator attribute information, the behavior related party attribute information, the behavior related object attribute information and other information.

Considering that most subsequent preset processes need to be analyzed from the perspective of a behavior operator, a behavior related party and a behavior related object, the behavior features in the behavior feature set need to be divided into the behavior operator features, the behavior related party features and the behavior related object features according to the attribute information of the first category behavior features, the second category behavior features and the third category initial behavior features, and then the multidimensional behavior features composed of the behavior operator features, the behavior related party features and the behavior related object features are obtained.

In an optional implementation manner of this embodiment, the step S103, that is, the step of performing the preset operation based on the multidimensional behavior feature, may include the following steps:

The target behavior related party and/or the target behavior related object are related to the needs of practical application and the needs of subsequent operations, for example, if a business which is likely to be interested in the target behavior related party and/or the target behavior related object needs to be recommended to the user subsequently or a search word with a high hit rate is recommended to the user subsequently, evaluation can be performed according to the popularity of the multidimensional behavior characteristics for the business and/or the effectiveness of the search word, so as to obtain one or more businesses and/or search words which can be recommended to the user; for example, if the preset processing is to perform account clearing processing on the merchant with poor sales performance, the survival quality of the merchant can be evaluated according to the multidimensional behavior characteristics to determine the merchant with low survival quality, and account clearing processing is performed on the merchant; for another example, if the future behavior data needs to be predicted subsequently, the multi-dimensional behavior feature may be trained as training data to obtain a behavior prediction model, and the future behavior data is predicted based on the trained behavior prediction model and the obtained behavior data in the current preset time period.

The technical solution of the present disclosure is explained and explained with reference to fig. 2 and fig. 3 as an example. In this example, the multidimensional behavior features include user features, merchant features, and search term features, where the user features include user attribute features such as user age, user occupation, whether the user is a member, and the like; the user statistical characteristics such as the order placing quantity within 30 days of the user, the click rate within 30 days of the user, the conversion rate within 30 days of the user and the like have periodic characteristics and/or probabilistic characteristics; user behavior characteristics such as a merchant list clicked by the user, a merchant list listed by the user, a merchant list collected by the user and the like; the merchant characteristics comprise merchant attribute characteristics such as merchant business state, merchant city, whether the merchant is a sole merchant and the like; the merchant statistical characteristics such as the exposure of the merchant within 30 days, the clicked amount of the merchant within 30 days, the clicked rate of the merchant within 30 days and the like have periodic characteristics and/or probabilistic characteristics; the search term characteristics include search term statistical characteristics having periodic characteristics and/or probabilistic characteristics, such as the number of occurrences of a search term within 30 days, the click rate of a search term within 30 days, and the conversion rate of a search term within 30 days, as shown in fig. 2. In practical application, as shown in fig. 3, firstly, user attribute features and merchant attribute features are obtained through system data, then, user statistical features, user behavior features, merchant statistical features and search term statistical features are obtained based on network interaction behavior log data, and the obtained features are classified to obtain behavior features with periodic characteristics: user statistical characteristics such as order placing quantity within 30 days of a user, click rate within 30 days of the user, conversion rate within 30 days of the user and the like, merchant statistical characteristics such as exposure within 30 days of a merchant, clicked quantity within 30 days of the merchant, clicked rate within 30 days of the merchant and the like, and search term statistical characteristics such as search term occurrence times within 30 days, search term click rate within 30 days, search term conversion rate within 30 days and the like; behavioral characteristics with probabilistic characteristics: user statistical characteristics such as a click rate of a user within 30 days, a conversion rate of the user within 30 days, merchant statistical characteristics such as a clicked amount of a merchant within 30 days, a clicked rate of the merchant within 30 days, and search term statistical characteristics such as a search term click rate within 30 days and a search term conversion rate within 30 days; and other behavioral characteristics that do not require special handling: user attribute features, user behavior features, and merchant attribute features. And then, correcting time factors for the behavior characteristics with the periodic characteristics, correcting confidence factors for the behavior characteristics with the probabilistic characteristics, not specially processing other behavior characteristics, combining the three processed characteristics to obtain the multidimensional behavior characteristics, and distinguishing to obtain user characteristics, merchant characteristics and search word characteristics for subsequent execution of preset operation.

The following are embodiments of the disclosed apparatus that may be used to perform embodiments of the disclosed methods.

Fig. 4 shows a block diagram of a data processing apparatus according to an embodiment of the present disclosure, which may be implemented as part or all of an electronic device by software, hardware, or a combination of both. As shown in fig. 4, the data processing apparatus includes:

an obtaining module 401 configured to obtain historical behavior data;

a determining module 402 configured to determine a multi-dimensional behavior feature according to the historical behavior data, wherein the multi-dimensional behavior feature includes at least a behavior operator feature, a behavior related party feature, and a behavior related object feature;

an executing module 403 configured to execute a preset operation based on the multidimensional behavior feature.

In view of the above drawbacks, in this embodiment, a data processing apparatus is provided, which extracts a multidimensional behavior feature that can comprehensively embody characteristics of an interactive behavior from historical behavior data and then performs a preset operation based on the multidimensional behavior feature. According to the technical scheme, the extraction dimensionality of the interactive behavior features is increased, the characteristics of the interactive behavior can be comprehensively reflected, effective and correct data support can be provided for operations such as follow-up interactive behavior prediction and service data adjustment, improvement of service quality, increase of service opportunities and improvement of service efficiency are facilitated.

In an optional implementation manner of this embodiment, the determining module 402 may be configured to:

In an optional implementation manner of this embodiment, the determining the multidimensional initial behavior feature according to the historical behavior data may be configured to:

In an optional implementation manner of this embodiment, the classifying and modifying the initial behavior feature to obtain the multidimensional behavior feature may be configured to:

In an optional implementation manner of this embodiment, the performing a first modification process on the first category initial behavior feature to obtain a first category behavior feature may be configured to:

N＝N₀×W，

In an optional implementation manner of this embodiment, the performing a second modification process on the second category initial behavior feature to obtain a second category behavior feature may be configured to:

In an optional implementation manner of this embodiment, the execution module 403 may be configured to:

The present disclosure also discloses an electronic device, fig. 5 shows a block diagram of an electronic device according to an embodiment of the present disclosure, and as shown in fig. 5, the electronic device 500 includes a memory 501 and a processor 502; wherein the content of the first and second substances,

the memory 501 is used to store one or more computer instructions, which are executed by the processor 502 to implement the above-described method steps.

As shown in fig. 6, the computer system 600 includes a processing unit 601 which can execute various processes in the above-described embodiments according to a program stored in a Read Only Memory (ROM)602 or a program loaded from a storage section 608 into a Random Access Memory (RAM) 603. In the RAM603, various programs and data necessary for the operation of the system 600 are also stored. The processing unit 601, the ROM602, and the RAM603 are connected to each other via a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.

The following components are connected to the I/O interface 605: an input portion 606 including a keyboard, a mouse, and the like; an output portion 607 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage section 608 including a hard disk and the like; and a communication section 609 including a network interface card such as a LAN card, a modem, or the like. The communication section 609 performs communication processing via a network such as the internet. The driver 610 is also connected to the I/O interface 605 as needed. A removable medium 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 610 as necessary, so that a computer program read out therefrom is mounted in the storage section 606 as necessary. The processing unit 601 may be implemented as a CPU, a GPU, a TPU, an FPGA, an NPU, or other processing units.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowcharts or block diagrams may represent a module, a program segment, or a portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units or modules described in the embodiments of the present disclosure may be implemented by software or hardware. The units or modules described may also be provided in a processor, and the names of the units or modules do not in some cases constitute a limitation of the units or modules themselves.

As another aspect, the present disclosure also provides a computer-readable storage medium, which may be the computer-readable storage medium included in the apparatus in the above-described embodiment; or it may be a separate computer readable storage medium not incorporated into the device. The computer readable storage medium stores one or more programs for use by one or more processors in performing the methods described in the present disclosure.

The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention in the present disclosure is not limited to the specific combination of the above-mentioned features, but also encompasses other embodiments in which any combination of the above-mentioned features or their equivalents is possible without departing from the inventive concept. For example, the above features and (but not limited to) the features disclosed in this disclosure having similar functions are replaced with each other to form the technical solution.

Claims

1. A method of data processing, comprising:

acquiring historical behavior data;

2. The method of claim 1, the determining a multi-dimensional behavior feature from the historical behavior data, comprising:

3. The method of claim 2, the determining a multi-dimensional initial behavior feature from the historical behavior data, comprising:

4. The method according to claim 2 or 3, wherein the performing classification correction on the initial behavior feature to obtain the multidimensional behavior feature comprises:

5. A data processing apparatus comprising:

an acquisition module configured to acquire historical behavior data;

6. The apparatus of claim 5, the determination module configured to:

7. The apparatus of claim 6, the portion to determine multi-dimensional initial behavior features from the historical behavior data configured to:

8. The apparatus according to claim 6 or 7, wherein the classifying modification of the initial behavior feature to obtain the multi-dimensional behavior feature is configured to:

9. An electronic device comprising a memory and at least one processor; wherein the memory is to store one or more computer instructions, wherein the one or more computer instructions are to be executed by the at least one processor to implement the method steps of any one of claims 1-4.

10. A computer-readable storage medium having stored thereon computer instructions, characterized in that the computer instructions, when executed by a processor, carry out the method steps of any of claims 1-4.