Method and device for determining purchase intention of user, electronic equipment and storage medium

Abstract

The invention provides a user purchase intention determination method and device, electronic equipment and a storage medium, and the method comprises the steps: determining various effective behavior data of a user for each target product; wherein the effective behavior data is data which can reflect an intentional product of the user, and the correlation between the effective behavior data and the purchase behavior data meets requirements; calculating a behavior weight of each type of effective behavior data; wherein the behavior weight is used for indicating the importance of the effective behavior data in the various effective behavior data; constructing an intention index model based on the behavior weight and the various effective behavior data, and determining the purchase intention of the user for the specified series of products according to the intention index model; wherein the intention index model is used for determining the attention ratio or mental ratio of the user to the specified series of products of the specified type under the target brand in the target industry.

Description

Technical Field

[0001] The present disclosure relates to the technical field of computers, and in particular, to a method, device, electronic device, and storage medium for determining user purchase intention. Background Art

[0002] The purchase process is a long one. For example, purchasing important or expensive products involves a lengthy process. During this process, users are exposed to a vast amount of offline and online information, repeatedly comparing and analyzing it. Consequently, identifying a user's purchasing intent and preferences for a particular product line can be challenging.

[0003] Existing technology can estimate user purchase intentions through survey data. However, analysis shows that due to the influence of user interview awareness and the interview environment, there is a significant deviation between the interview results and the actual decision-making behavior, which cannot accurately reflect the user's actual purchase intentions. Summary of the Invention

[0004] The embodiments of the present disclosure at least provide a method, device, electronic device, and storage medium for determining user purchase intention.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for determining a user's purchase intention, comprising:

[0006] Determine multiple valid behavior data of the user for each target product; wherein the valid behavior data is data that can reflect the user's intended product and has a correlation with the purchase behavior data that meets the requirements;

[0007] Calculating a behavior weight for each type of valid behavior data; wherein the behavior weight is used to indicate the importance of the valid behavior data among the multiple types of valid behavior data;

[0008] An intention index model is constructed based on the behavior weights and the multiple valid behavior data, and the user's purchase intention for a specified series of products is determined according to the intention index model; wherein the intention index model is used to indicate the user's attention share or mental share to a specified series of products of a specified style under a target brand within a target industry.

[0009] In an optional implementation, the calculating of the behavior weight of each type of valid behavior data includes:

[0010] Performing dimension-removing and quantifying processing on each effective behavior data of each target product to obtain dimension-removed effective behavior data;

[0011] Determine the behavior weight of each type of effective behavior data after dimensioning.

[0012] In an optional implementation, determining the behavior weight of each type of dimensionless effective behavior data includes:

[0013] A linear regression fitting is performed on the attention ratio of each valid behavior data and the corresponding de-dimensionalized valid behavior data to obtain the behavior weight; wherein the attention ratio is the ratio between the statistical index of each valid behavior data and the statistical index of the user's valid behavior data.

[0014] In an optional implementation, the constructing of the intention index model based on the behavior weight and the multiple valid behavior data includes:

[0015] Performing weighted summation on the de-dimensionalized effective behavior data of each target product according to the behavior weight to obtain a first result;

[0016] Performing a weighted summation of each comprehensive behavior data of all the target products according to the behavior weight to obtain a second result; wherein the comprehensive behavior data is the sum of the dedimensionalized effective behavior data of the same type of all the target products;

[0017] The intention index model is determined based on a ratio of the first result to the second result.

[0018] In an optional embodiment, after constructing the intention index model based on the behavior weights and the multiple valid behavior data, the method further includes:

[0019] After the advertising material for the first product is delivered, determining changes in the target user's purchase intention for the first product based on the intention index model;

[0020] determining a correlation between the target user's actual online behavior with respect to the second product and the target user's purchase intention for the second product;

[0021] Based on the correlation and / or the change, the validity of the intention index model is verified, and if the validity verification fails, the behavior weight is adjusted.

[0022] In an optional implementation, determining multiple valid user behavior data for each target product includes:

[0023] Obtaining the user's initial behavior data for each of the target products;

[0024] Screening the initial behavior data for which the correlation with the purchase behavior data is greater than a preset threshold from the initial behavior data to obtain candidate behavior data;

[0025] Determine candidate behavior data similar to the intention-indicating behavior of high-intention users to obtain the valid behavior data; wherein, the high-intention user is a user who performs purchase behavior data for the target product, and the intention-indicating behavior can reflect the intended product of the high-intention user.

[0026] In an optional implementation, after determining the user's purchase intention for a specified series of products according to the intention index model, the method further includes:

[0027] Determining intention data for the designated product series based on the purchase intention; wherein the intention data is used to indicate the purchase intention of each user for the designated product series;

[0028] and / or

[0029] A competitive product analysis is performed on the designated product series based on the purchase intention to obtain a competitive product analysis result; wherein the competitive product analysis result is used to indicate competitive products of the designated product series and / or users who follow the competitive products.

[0030] In a second aspect, an embodiment of the present disclosure provides a device for determining a user's purchase intention, comprising:

[0031] A first determining unit is configured to determine a plurality of valid behavior data of the user for each target product; wherein the valid behavior data is data that can reflect the user's intended product and has a correlation with the purchase behavior data that meets the requirements;

[0032] a calculation unit, configured to calculate a behavior weight of each of the valid behavior data; wherein the behavior weight is used to indicate the importance of the valid behavior data among the multiple valid behavior data;

[0033] The second determination unit is used to construct an intention index model based on the behavior weight and the multiple valid behavior data, and determine the user's purchase intention for the specified series of products according to the intention index model; wherein the intention index model is used to indicate the user's attention share or mental share to the specified series of products of the specified style under the target brand in the target industry.

[0034] In a third aspect, an embodiment of the present disclosure further provides an electronic device comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the above-mentioned first aspect or any possible implementation of the first aspect are performed.

[0035] In a fourth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned first aspect or any possible implementation of the first aspect are executed.

[0036] The disclosed embodiments provide a method, device, electronic device, and storage medium for determining a user's purchase intention. In the disclosed embodiments, first, multiple valid behavioral data of the user for each target product is determined; wherein the valid behavioral data is data that can reflect the user's intended car series and has a correlation with the purchase behavior data that meets the requirements, and the purchase behavior data can indicate the user's behavior of having a purchase intention for the car series; then, the behavioral weight of each valid behavioral data is calculated; wherein the behavioral weight is used to indicate the importance of the valid behavioral data among multiple valid behavioral data; finally, an intention index model is constructed based on the behavioral weight and multiple valid behavioral data, and the user's purchase intention for a specified series of products is determined based on the intention index model; wherein the intention index model is used to indicate the user's attention share or mental share for a specified series of products of a specified model under a target brand within a target industry.

[0037] In the above implementation, by combining the user's effective behavior data on the terminal and the behavior weight of each effective behavior data to construct an intention index model for determining the user's purchase intention for the product, the user's intention degree for each series of products can be quantified, thereby more intuitively, accurately, and targetedly reflecting the user's true purchase intention; at the same time, the technical solution disclosed in the present invention can cover a wide range of user groups, thereby further meeting the service needs of more users.

[0038] In order to make the above-mentioned objectives, features and advantages of the present disclosure more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following briefly introduces the drawings required for use in the embodiments. The drawings herein are incorporated into and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and, together with the specification, are used to illustrate the technical solutions of the present disclosure. It should be understood that the following drawings only illustrate certain embodiments of the present disclosure and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without inventive effort.

[0040] Figure 1 A flowchart of a method for determining a user's purchase intention provided by an embodiment of the present disclosure is shown;

[0041] Figure 2 A flowchart showing another method for determining user purchase intention provided by an embodiment of the present disclosure is shown;

[0042] Figure 3 A schematic diagram showing a device for determining user purchase intention provided by an embodiment of the present disclosure is shown;

[0043] Figure 4 A schematic diagram of an electronic device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solutions and advantages of the embodiments of the present disclosure clearer, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. The components of the embodiments of the present disclosure generally described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure provided in the drawings is not intended to limit the scope of the disclosure for which protection is sought, but merely represents selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present disclosure.

[0045] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0046] The term "and / or" herein simply describes an association relationship, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, the simultaneous existence of A and B, and the existence of B alone. In addition, the term "at least one" herein refers to any combination of at least two of any one or more of a plurality of items. For example, "at least one of A, B, and C" can represent any one or more elements selected from the set consisting of A, B, and C.

[0047] Industry media outlets have reported that determining users' actual purchasing intentions earlier has long been a technical challenge. For example, the purchase of a car is a typically lengthy decision-making process. During this process, users are exposed to a vast amount of offline and online information, repeatedly comparing and analyzing it. Furthermore, their decisions are influenced by information from various dimensions and brands. Consequently, users' decisions fluctuate rapidly, with their indecision fluctuating across multiple car models. This makes it difficult to determine their true purchasing intentions from a platform or brand perspective.

[0048] Existing technology can estimate users' purchase intentions through survey data. However, analysis shows that due to the influence of users' interview awareness and interview environment, there is a significant deviation between the interview results and actual decision-making behavior, making it impossible to accurately reflect users' actual purchase intentions.

[0049] Based on the above research, the present disclosure provides a method, device, electronic device, and storage medium for determining user purchase intentions. In an embodiment of the present disclosure, first, multiple valid behavior data of the user for each target product is determined; wherein the valid behavior data is data that can reflect the user's intended car series and has a correlation with the purchase behavior data that meets the requirements, and the purchase behavior data can indicate the user's behavior of having purchase intentions for the car series; then, the behavior weight of each valid behavior data is calculated; wherein the behavior weight is used to indicate the importance of the valid behavior data among multiple valid behavior data; finally, an intention index model is constructed based on the behavior weight and multiple valid behavior data, and the user's purchase intention for a specified series of products is determined based on the intention index model; wherein the intention index model is used to indicate the user's attention share or mind share for a specified series of products of a specified model under a target brand within a target industry.

[0050] In the above implementation, by combining the user's effective behavior data on the terminal and the behavior weight of each effective behavior data to construct an intention index model for determining the user's purchase intention for the product, the user's intention degree for each series of products can be quantified, thereby more intuitively, accurately, and targetedly reflecting the user's true purchase intention; at the same time, the technical solution disclosed in the present invention can cover a wide range of user groups, thereby further meeting the service needs of more users.

[0051] To facilitate understanding of this embodiment, we first provide a detailed description of a method for determining user purchase intentions disclosed in this embodiment. This method is typically executed by an electronic device with certain computing capabilities. In some possible implementations, this method can be implemented by a processor invoking computer-readable instructions stored in a memory.

[0052] See also Figure 1 FIG. 1 is a flow chart of a method for determining a user's purchase intention provided by an embodiment of the present disclosure, the method comprising steps S101 to S103, wherein:

[0053] S101: Determine a plurality of valid behavior data of a user for each target product; wherein the valid behavior data is data that can reflect the user's intended product and has a correlation with the purchase behavior data that meets the requirements.

[0054] Here, purchasing behavior data can include user store visits and / or account retention, as well as other behavioral data reflecting user purchasing behavior. Target products can include various types of products, such as cars, home appliances, and electronic devices. Store visits can include visits to offline stores or online stores. Account retention can include account retention at offline stores or reservations at online stores, and similar behaviors.

[0055] In an embodiment of the present disclosure, behavioral data with a high correlation with store-going behavior and / or capital-leaving behavior is screened out from a large amount of user behavioral data. Specifically, data sampling can be performed from a large amount of behavioral data, and then the correlation between the sampled behavioral data and the purchase behavior data can be determined from multiple dimensions; wherein the dimensions include the type of behavioral data, the number of target products, the type of target products, and the like. For example, the correlation between a single piece of behavioral data and store-going behavior or capital-leaving behavior can be determined; for another example, the correlation between a single piece of behavioral data of a specified product and store-going behavior or capital-leaving behavior can be determined; for another example, the correlation between multiple pieces of behavioral data of a specified product and store-going behavior or capital-leaving behavior can be determined.

[0056] The behavioral data with high correlation that have been screened out need to be further screened to screen out behavioral data that can reflect the user's intended products, and the behavioral data finally screened out are determined as valid behavioral data.

[0057] Here, the data units of different effective behavior data may be different. For example, the data units of effective behavior data may be statistical units such as percentage, frequency, duration, positive or negative emotions, etc.

[0058] S102: Calculate a behavior weight of each type of valid behavior data; wherein the behavior weight is used to indicate the importance of the valid behavior data among the multiple types of valid behavior data.

[0059] Here, the behavior weight of each valid behavior data can indicate the importance of the valid behavior data, and the importance can be used to determine the importance of different behavior data in the intention index model calculation.

[0060] In the embodiment of the present disclosure, the behavior weights of different valid behavior data are the same or different, and the sum of the behavior weights of all valid behavior data is a value of 1.

[0061] S103: Constructing an intention index model based on the behavior weights and the multiple valid behavior data, and determining the user's purchase intention for a specified series of products based on the intention index model; wherein the intention index model is used to determine the user's attention share or mental share for a specified series of products of a specified style under a target brand within a target industry.

[0062] After obtaining the behavior weight of each valid behavior data, the behavior weight and the valid behavior data can be aggregated to obtain an intention index model, thereby determining the user's purchase intention for a specified series of products.

[0063] In the disclosed embodiment, purchase intention can also be understood as the share of user attention (SOA). The share of user attention can be used to identify the user's purchase decision intention through the user's deep user behavior within a specified period of time in the relevant industry media. Among them, the SOA indicator is used to indicate the user's attention share or mental share of a specified series of products of a specified style under a target brand in the target industry. The SOA indicator can represent the degree of concentration of the purchasing user during the key decision window period of the purchase. The higher the SOA of a specified series of products, the higher the possibility that the user will purchase the specified series of products.

[0064] Since users' active in-depth behavioral data can usually truly reflect their minds, users' purchasing preferences can be mapped through active in-depth behavioral data. Based on this, the method of constructing an intention index model through a variety of effective behavioral data of users can reflect users' real purchasing intentions in a targeted manner. By combining the behavioral weights of effective behavioral data to construct an intention index model, it is possible to quantify the user's intention for each series of products, thereby more intuitively and accurately reflecting the user's purchasing intentions; at the same time, the technical solution disclosed in this disclosure can cover a wide range of user groups, further meet the service needs of more users, and can also effectively help advertisers identify high-intention purchasing users earlier.

[0065] The above steps will be introduced below in conjunction with specific implementation methods.

[0066] In the embodiment of the present disclosure, the above step S101 determines a plurality of valid behavior data of the user for each target product, and specifically includes the following steps:

[0067] Step S11: Obtaining the user's initial behavior data for each of the target products;

[0068] Step S12: screening the initial behavior data for initial behavior data having a correlation with the purchase behavior data greater than a preset threshold, to obtain candidate behavior data;

[0069] Step S13: Determine candidate behavior data that is similar to the intention indication behavior of high-intention users to obtain the valid behavior data; wherein, the high-intention user is a user who performs purchase behavior data for the target product, and the intention indication behavior can reflect the intended product of the high-intention user.

[0070] In an embodiment of the present disclosure, various types of initial behavior data of users for each target product can be obtained from relevant industry media. Then, the initial behavior data having a correlation with store visit behavior and / or information retention behavior greater than a preset threshold is screened in a preset manner as candidate behavior data.

[0071] In one possible embodiment, the initial behavior data can be input into a pre-trained neural network model for processing, thereby outputting the correlation between each type of initial behavior data and store visit behavior and / or capital retention behavior, and then, a preset number of initial behavior data greater than a preset threshold are screened from multiple correlations as candidate behavior data.

[0072] Here, training samples can be collected in advance, wherein the user's actual behavior data and actual purchase behavior can be collected in advance, and the correlation between the actual behavior data and the actual purchase behavior can be analyzed, and the correlation can be used as a label for the training sample. Then, the neural network model is trained using the training samples.

[0073] In another possible implementation, candidate behavior data may be screened out from a variety of initial behavior data based on preset analysis data, wherein the preset analysis data includes at least one of the following:

[0074] Correlation study between the SOA indicator (a single user's single touchpoint) and store visit behavior. This correlation study can determine the degree of correlation between the SOA indicator value of each user's single behavior data and store visit behavior.

[0075] Correlation study between the SOA indicator of a single user's single touchpoint and retention behavior. This correlation study can determine the degree of correlation between the SOA indicator value of each customer's single behavioral data and retention behavior.

[0076] Correlation study between SOA, a multi-core touchpoint indicator for sampled products, and retention behavior. This correlation study can determine the degree of correlation between SOA values and retention behavior across various user behavior data for a specific product.

[0077] A correlation study between SOA, a multi-core touchpoint indicator, and retention behavior across all products. This correlation study can determine the degree of correlation between SOA values and retention behavior across various behavioral data points across all products.

[0078] The above preset analysis data can be used to determine the correlation between single behavioral data and store visit behavior or capital retention behavior, and can also determine the correlation between a combination of multiple behavioral data and store visit behavior or capital retention behavior, thereby providing a basis for screening candidate behavioral data.

[0079] After the candidate behavior data are determined, candidate behavior data similar to the intention-indicated behavior of the high-intention user may be determined from the candidate behavior data, thereby obtaining valid behavior data.

[0080] Here, high-intent users can be understood as users who have actually visited the store and / or left information for each product. Intent-indicating behaviors can be understood as behaviors that reflect high-intent users' product preferences; in other words, intent-indicating behaviors are prominent behavioral data that can distinguish high-intent users' primary intentions.

[0081] Based on research into user behavioral data from relevant industry media, this disclosed technical solution enables in-depth analysis of the correlation between user behavioral data and behaviors such as in-store retention and conversion. Based on this correlation, it extracts effective behavioral data for various statistical units (e.g., the frequency and duration of key user behaviors, positive and negative attitudes toward car models, and decision-making cycles). This effective behavioral data can be used to predict users' actual purchase intentions in real time, thereby improving the accuracy and confidence of the intention index model.

[0082] In the embodiment of the present disclosure, after determining a plurality of valid behavior data, the behavior weight of each valid behavior data of each target product may be calculated, specifically including the following steps:

[0083] Step S21: performing de-dimensionalization and quantization processing on each effective behavior data of each target product to obtain de-dimensionalized effective behavior data;

[0084] Step S22: determining the behavior weight of each type of the dimensionless effective behavior data.

[0085] Since different users have significant differences in their investment in different types of behavioral data, in order to build a comprehensive intention index model, it is necessary to integrate the different online behaviors of users. However, there are magnitude differences between behavioral data of different dimensions. Therefore, in order to eliminate the magnitude differences between different physical quantities so that behavioral data can be compared and analyzed on the same scale, it is necessary to de-dimensionalize each valid behavioral data of each target product (for example, each car series) to obtain de-dimensionalized valid behavioral data. Here, the de-dimensionalization algorithm selected is an algorithm that can reasonably integrate valid behavioral data of different statistical units. The technical solution disclosed in this disclosure does not make specific limitations on the de-dimensionalization algorithm, and is mainly based on the ability to integrate.

[0086] In the embodiment of the present disclosure, the function f(x k ) De-dimensionalize the effective behavior data. The de-dimensionalization function can normalize the absolute values of the effective behavior data of different dimensions, so that the extreme values and distribution of the absolute values of the various effective behavior data after de-dimensionalization are close, thereby not affecting the weighted accumulation.

[0087] After obtaining the effective behavioral data after dimensioning, the behavioral weight of each effective behavioral data after dimensioning can be determined.

[0088] Through the above implementation, the integration of effective behavioral data from different statistical units can be achieved, thereby providing a data basis for the construction of the intention index model.

[0089] In the embodiment of the present disclosure, the above steps of determining the behavior weight of each type of dimensionless effective behavior data specifically include the following steps:

[0090] Perform linear regression fitting on the attention ratio of each type of effective behavior data and the corresponding de-dimensionalized effective behavior data to obtain the behavior weight; wherein, the attention ratio is the ratio between the statistical index of each type of effective behavior data and the statistical index of the user's effective behavior data.

[0091] In the disclosed embodiment, statistical indicators for each type of valid behavior data of each user can be determined, where the statistical indicators can include frequency, duration, positive or negative emotions, etc. Then, the ratio of the statistical indicators for each type of valid behavior data to the statistical indicators of all valid behavior data of the user for each target product is calculated, and finally, the ratio is determined as the attention ratio of the valid behavior data.

[0092] After obtaining the attention ratio of each user's various effective behavior data, a linear regression fitting algorithm can be used to perform linear regression fitting on the attention ratio of each effective behavior data and the dedimensionalized effective behavior data, thereby obtaining the behavior weights of various effective behavior data.

[0093] For example, assume there are users 1 and 2. User 1's valid behavior data includes data 11, data 12, and data 13, while user 2's valid behavior data includes data 21, data 22, and data 23. For user 1, the ratios of the statistical indicators of data 11, data 12, and data 13 to the statistical indicators of user 1's total valid behavior data can be calculated, thereby obtaining the attention share 11, attention share 12, and attention share 13, respectively. For user 2, the ratios of the statistical indicators of data 21, data 22, and data 23 to the statistical indicators of user 2's total valid behavior data can be calculated, thereby obtaining the attention share 21, attention share 22, and attention share 23, respectively. Finally, the behavioral weight of this type of valid behavior data is determined based on the ratio of attention share 11 to the dedimensionalized data 11, and the ratio of attention share 21 to the dedimensionalized data 21. The behavioral weight of this type of valid behavior data is determined based on the ratio of attention share 12 to the dedimensionalized data 12, and the ratio of attention share 22 to the dedimensionalized data 22. The behavioral weight of this type of valid behavioral data is determined based on the attention ratio of 13 and the dedimensionalized data of 13, as well as the attention ratio of 23 and the dedimensionalized data of 23.

[0094] Through the above processing method, the weight of each valid behavioral data in the calculation can be analyzed more comprehensively, the user's purchasing intention can be reflected more intuitively and accurately, and a wide range of user groups can be covered, thereby further meeting the service needs of more users.

[0095] In the embodiment of the present disclosure, after determining the behavior weight of each valid behavior data, an intention index model can be constructed based on the behavior weight and the multiple valid behavior data, specifically including the following steps:

[0096] First, weighted summation is performed on each de-dimensionalized valid behavior data of each target product according to the behavior weight to obtain a first result;

[0097] Secondly, performing weighted summation on each comprehensive behavior data of all the target products according to the behavior weight to obtain a second result; wherein the comprehensive behavior data is the sum of the dedimensionalized effective behavior data of the same type of all the target products;

[0098] Finally, the intention index model is determined based on the ratio of the first result to the second result.

[0099] In the embodiment of the present disclosure, each user's effective behavior data after de-dimensionalization for each target product and its corresponding behavior weight can be weighted and summed to obtain the first result. For example, the formula Indicates that, f(x k )series-act represents the kth dimensionless effective behavior data of each target product, C k It represents the behavior weight of the kth type of valid behavior data, and n is the number of valid behavior data.

[0100] Next, the de-dimensionalized effective behavior data of the same type of all target products can be summed up to obtain a variety of comprehensive behavior data, where f(x k ) all-act The kth comprehensive behavior data is obtained by summing up the kth effective behavior data after de-dimensionalization of all target products. Each effective behavior data after de-dimensionalization corresponds to a comprehensive behavior data. At this time, the behavior weight and the comprehensive behavior data can be weighted and summed to obtain the second result. The second result can be obtained by the formula express.

[0101] Finally, the ratio of the first result to the second result can be determined as an intention index model; wherein the intention index model can be expressed as:

[0102] Among them, SOA is the purchase intention calculated according to the intention index model (or, called the comprehensive attention ratio, or the comprehensive SOA index). The SOA is used to determine the user's attention ratio or mental ratio for a specified series of products of a specified model under a target brand in the target industry. SOA is defined as the ratio of a user's comprehensive behavior towards a certain series of products (for example, the sum of multiple valid behavior data for a certain car series) to the user's comprehensive behavior towards the entire series of products (for example, the sum of multiple valid behavior data for the entire car series).

[0103] By combining the behavioral weights of effective behavioral data to construct an intention index model, it is possible to quantify the degree of user intention for each product series, thereby more intuitively and accurately reflecting the user's purchase intention; at the same time, the technical solution disclosed in this disclosure can cover a wide range of user groups, thereby further meeting the service needs of more users, and can also effectively help advertisers identify high-intention purchasing users earlier.

[0104] In an embodiment of the present disclosure, after constructing the intention index model based on the behavior weights and the multiple valid behavior data, the method further includes the following steps:

[0105] Step S31: After the advertising material for the first product is released, determining changes in the target user's purchase intention for the first product based on the intention index model;

[0106] Step S32: determining the correlation between the target user's actual online behavior regarding the second product and the target user's purchase intention for the second product;

[0107] Step S33: Based on the correlation and / or the change, verify the validity of the intention index model, and adjust the behavior weight if the validity verification fails.

[0108] In the disclosed embodiment, after constructing the intention index model based on the behavior weights and multiple valid behavior data, the effectiveness of the intention index model can be tested using the following test scenarios. Here, assuming that the first product is a first car series and the second product is a second car series, the specific test scenarios are as follows:

[0109] Test plan 1:

[0110] Calculate the correlation between the comprehensive SOA index of each car series and the behavior of leaving information in the store. Specifically, you can select a target user and obtain the collection permission of the target user's relevant data; wherein, the target user is a user who has a purchase intention for the second car series. Next, obtain the target user's valid behavior data for the second car series, and substitute the valid behavior data into the intention index model for calculation, so as to obtain the target user's purchase intention for the second car series. At the same time, you can also obtain the target user's actual online behavior for the second car series under the condition that the target user allows. Afterwards, determine the correlation between the actual online behavior and the purchase intention. For example, if the target user's purchase intention for the second car series is high, then the target user's actual online behavior for the second car series may include a behavior of leaving information or a behavior of visiting the store. At this time, it is considered that the correlation between the actual online behavior and the purchase intention is high, and the intention index model is reasonable; otherwise, the intention index model is considered unreasonable.

[0111] Here, the second car series may be a car series with specified characteristics selected from all car series. For example, the second car series may be a popular car series, a classic car series, a car series with a high reputation, etc.

[0112] Test plan 2:

[0113] In the disclosed embodiment, advertising materials may be placed for the first car series. Then, during the placement of the advertising materials, changes in the target user's purchase intention for the first car series may be determined based on the intention index model.

[0114] Here, the target users in Test Scenario 1 and Test Scenario 2 can be the same or different. Similarly, the target users are users who have purchasing intentions for the first vehicle series. Test Scenario 2 still requires permission to collect data related to the target users. The first and second vehicle series can be the same or different, and this disclosure does not impose specific limitations on this.

[0115] Here, you can choose to deliver advertising materials only to target users, or to a designated user group. This disclosure does not specifically limit the scope of users to whom advertising materials are delivered, and what can be achieved is the standard.

[0116] In the embodiment of the present disclosure, the comprehensive SOA index calculated after the advertisement is delivered can be compared with the comprehensive SOA index calculated before the advertisement is delivered to obtain the change situation; the comprehensive SOA index can also be counted at preset intervals starting from the advertisement delivery and during the advertisement delivery cycle, and the change situation of the comprehensive SOA index obtained by counting can be determined.

[0117] If it is determined that the changes in the comprehensive SOA index can reflect the changes in the user's mind as expected, then the intention index model validity test is determined to have passed; if it is determined that the changes in the comprehensive SOA index cannot reflect the changes in the user's mind as expected, then the intention index model validity test is determined to have failed.

[0118] In addition to the two test schemes mentioned above, the validity of the intention index model can also be verified according to test scheme 1 and test scheme 2. If the validity verification of the intention index model fails, the behavior weights of each valid behavior data need to be adjusted.

[0119] In specific implementations, valid behavioral data can be re-screened from the user's initial behavioral data for each target product, and a behavioral weight can be determined for each new valid behavioral data. An intention index model can then be constructed based on the new behavioral weight and the new valid behavioral data. The newly constructed intention index model can then be validated, and if the validation passes, the user's purchase intention for each target product can be determined based on the intention index model.

[0120] Through the effectiveness testing scheme described above, the intention index model can be verified and tested from multiple perspectives, thereby further ensuring the accuracy and effectiveness of the intention index model.

[0121] In the embodiment of the present disclosure, after determining the user's purchase intention for each target product according to the intention index model, the purchase intention based on the intention index model may also be determined. Specific applications include the following scenarios:

[0122] Scenario 1: Determining intention data for a specified product series based on the purchase intention; wherein the intention data is used to indicate each user's purchase intention for the specified product series;

[0123] In this scenario, for example, in the automotive industry, purchase intentions can be organized to determine the purchase intentions of users for each car series. Based on this intention data, high-intent users for a particular car series can be identified, and relevant data for that car series can be delivered to these high-intent users.

[0124] Scenario 2: Based on the purchase intention, a competitive product analysis is performed on the specified product series to obtain a competitive product analysis result; wherein the competitive product analysis result is used to indicate competitive products of the specified product series and / or users who follow the competitive products.

[0125] In this scenario, taking the automotive industry as an example, for a single car series, other car series with the same users as the current car series can be identified as competing car series based on purchase intention. Next, intention data for the competing car series is determined, namely the users who follow the competing car series and the purchase intention of each following user for the competing car series. The competitive product analysis results can serve as the basis for generating competitive strategies for the car series. For example, relevant information about the current car series can be pushed to users who follow the competing car series but do not follow the current car series.

[0126] The technical solution disclosed herein implements an intention index model for indicating user purchase intention based on the correlation between store visits or information retention. By combining a large amount of user behavior characteristics in relevant industry media, it fits a single user's intention index model for a specific product. This can not only fully reflect the user's purchase intention, but also clearly judge the degree and extent of the user's intention for a single series of products, thereby achieving early prediction, quantification, influence and measurement of the user's purchase intention, and at the same time providing a basis for advertisers to formulate marketing strategies.

[0127] The following combination Figure 2 The above process is described in general. In this embodiment, the target product is introduced by taking a car series as an example. The process specifically includes the following modules:

[0128] A preliminary research module on user attention ratio.

[0129] The module may screen candidate behavior data from a plurality of initial behavior data based on preset analysis data, wherein the preset analysis data includes at least one of the following:

[0130] Research on the correlation between the single-user single-touchpoint indicator SOA and store visit behavior; research on the correlation between the single-user single-touchpoint indicator SOA and store retention behavior; research on the correlation between the multi-core touchpoint indicator SOA and store retention behavior under the sampled car series; and research on the correlation between the multi-core touchpoint indicator SOA and store retention behavior under the full car series.

[0131] Basic data processing module.

[0132] The module can determine candidate behavior data similar to the intention-indicated behavior of high-intention users from the candidate behavior data, thereby obtaining valid behavior data.

[0133] Here, high-intent users can be understood as those who have actually visited the store and / or left information for each car series. Intention-indicating behavior can reflect the car series that high-intent users prefer; in other words, intention-indicating behavior is the prominent behavioral data that can distinguish the primary intention of high-intent users.

[0134] Index development module.

[0135] Here, the index development module includes a dedimensionalization module, a linear regression fitting module and a SOA aggregation indicator module; among them, the dedimensionalization module is used to dedimensionalize the effective behavioral data, the linear regression fitting module is used to determine the behavioral weight of each effective behavioral data after dedimensionalization, and the SOA aggregation indicator module is used to construct an intention index model based on the behavioral weight and multiple effective behavioral data.

[0136] Here, the linear regression fitting module can perform linear regression fitting on the attention ratio of each valid behavior data and the corresponding de-dimensionalized valid behavior data to obtain the behavior weight; wherein the attention ratio is the ratio between the statistical index of each valid behavior data and the statistical index of the user's valid behavior data.

[0137] Validity testing module.

[0138] Here, the effectiveness testing module includes Test Plan 1 and Test Plan 2. Test Plan 1 tests the correlation between the comprehensive SOA index of a single car series and in-store information retention behavior. Specifically, it can determine the correlation between the target user's actual online behavior for the second car series and the target user's purchase intention for the second car series, and then conduct an effectiveness test based on this correlation. Test Plan 2 conducts an effectiveness test based on the effectiveness of the advertising material delivery. Specifically, after the advertising material for the first car series is delivered, the intention index model is used to determine the changes in the target user's purchase intention for the first car series, and then the effectiveness test is conducted based on this change.

[0139] In the embodiment of the present disclosure, if the validity test of the intention index model fails, it is necessary to return to the basic data processing module, reselect valid behavior data, and reprocess the newly selected valid behavior data through the index development module until the intention index model that passes the validity test is obtained.

[0140] After the intention index model passes the validity test, it can be applied in specified application scenarios to obtain users' purchase intentions for a single car series. Specific application scenarios include the following:

[0141] High-intent user selection: In this scenario, you can organize purchase intentions to identify users with specific car models. Based on this intention data, you can identify high-intent users for a specific car model and then target them with relevant data for that model.

[0142] Competitive analysis scenario based on car series purchase intention. In this scenario, for a single car series, other car series with the same users as the current car series can be identified as competing car series based on purchase intention. Next, intention data for the competing car series is determined, namely, the users who follow the competing car series and the purchase intention of each following user for the competing car series. The competitive analysis results can serve as the basis for generating competitive strategies for the car series. For example, relevant information about the current car series can be pushed to users who follow the competing car series but do not follow the current car series.

[0143] As can be seen from the above description, the disclosed technical solution conducts in-depth research on the characteristics of relevant industry media and massive amounts of user behavior data, analyzes the correlation between in-depth user behavior data and behaviors such as retention conversion, and extracts effective behavioral data from various statistical units, allowing real-time prediction of users' true purchase intentions. This solution is suitable for product industries with long decision-making cycles, can cover a wide range of user groups, and has a clear and quantified level of intention. It is highly representative and indicative of users' true purchase intentions, thereby enhancing the representativeness and confidence of users' purchase intentions. At the same time, it can also determine users' purchase intentions for individual car models in real time, helping platforms to predict users' purchase intentions in advance and implement advertisers' advertising material delivery strategies.

[0144] Those skilled in the art will understand that in the above-mentioned method of the specific implementation method, the writing order of each step does not mean a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0145] Based on the same inventive concept, the embodiment of the present disclosure also provides a device for determining user purchase intention corresponding to the method for determining user purchase intention. Since the principle of solving the problem by the device in the embodiment of the present disclosure is similar to the method for determining user purchase intention in the above-mentioned embodiment of the present disclosure, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0146] Reference Figure 3FIG. 1 is a schematic diagram of a device for determining user purchase intention provided by an embodiment of the present disclosure, wherein the device includes: a first determining unit 10, a calculating unit 20, and a second determining unit 30; wherein,

[0147] The first determining unit 10 is configured to determine a plurality of valid behavior data of the user for each target product; wherein the valid behavior data is data that can reflect the user's intended product and has a correlation with the purchase behavior data that meets the requirements;

[0148] A calculation unit 20 is configured to calculate a behavior weight of each of the valid behavior data; wherein the behavior weight is used to indicate the importance of the valid behavior data among the multiple valid behavior data;

[0149] The second determination unit 30 is used to construct an intention index model based on the behavior weight and the multiple valid behavior data, and determine the user's purchase intention for the specified series of products based on the intention index model; wherein the intention index model is used to indicate the user's attention share or mental share to the specified series of products of the specified style under the target brand in the target industry.

[0150] In the above implementation, by combining the user's effective behavior data on the terminal and the behavior weight of each effective behavior data to construct an intention index model for determining the user's purchase intention for the product, the user's intention degree for each series of products can be quantified, thereby more intuitively, accurately, and targetedly reflecting the user's true purchase intention; at the same time, the technical solution disclosed in the present invention can cover a wide range of user groups, thereby further meeting the service needs of more users.

[0151] In one possible implementation, the computing unit is further configured to:

[0152] Performing dimension-removing and quantifying processing on each effective behavior data of each target product to obtain dimension-removed effective behavior data;

[0153] Determine the behavior weight of each type of effective behavior data after dimensioning.

[0154] In one possible implementation, the computing unit is further configured to:

[0155] A linear regression fitting is performed on the attention ratio of each valid behavior data and the corresponding de-dimensionalized valid behavior data to obtain the behavior weight; wherein the attention ratio is the ratio between the statistical index of each valid behavior data and the statistical index of the user's valid behavior data.

[0156] In a possible implementation manner, the second determining unit is further configured to:

[0157] Performing weighted summation on the de-dimensionalized effective behavior data of each target product according to the behavior weight to obtain a first result;

[0158] Performing a weighted summation of each comprehensive behavior data of all the target products according to the behavior weight to obtain a second result; wherein the comprehensive behavior data is the sum of the dedimensionalized effective behavior data of the same type of all the target products;

[0159] The intention index model is determined based on a ratio of the first result to the second result.

[0160] In a possible implementation manner, the device is further used for:

[0161] After the advertising material for the first product is delivered, determining changes in the target user's purchase intention for the first product based on the intention index model;

[0162] determining a correlation between the target user's actual online behavior with respect to the second product and the target user's purchase intention for the second product;

[0163] Based on the correlation and / or the change, the validity of the intention index model is verified, and if the validity verification fails, the behavior weight is adjusted.

[0164] In a possible implementation manner, the first determining unit is further configured to:

[0165] Obtaining the user's initial behavior data for each of the target products;

[0166] Screening the initial behavior data for which the correlation with the purchase behavior data is greater than a preset threshold from the initial behavior data to obtain candidate behavior data;

[0167] Determine candidate behavior data similar to the intention-indicating behavior of high-intention users to obtain the valid behavior data; wherein, the high-intention user is a user who performs purchase behavior data for the target product, and the intention-indicating behavior can reflect the intended product of the high-intention user.

[0168] In a possible implementation manner, the device is further used for:

[0169] Determining intention data for the designated product series based on the purchase intention; wherein the intention data is used to indicate the purchase intention of each user for the designated product series;

[0170] and / or

[0171] A competitive product analysis is performed on the designated product series based on the purchase intention to obtain a competitive product analysis result; wherein the competitive product analysis result is used to indicate competitive products of the designated product series and / or users who follow the competitive products.

[0172] For descriptions of the processing flow of each module in the device and the interaction flow between each module, reference can be made to the relevant descriptions in the above method embodiment, which will not be described in detail here.

[0173] Corresponding to Figure 1 The present disclosure also provides an electronic device 400, such as Figure 4 FIG. 4 is a schematic diagram of the structure of an electronic device 400 provided in an embodiment of the present disclosure, including:

[0174] Processor 41, memory 42, and bus 43; memory 42 is used to store execution instructions, including internal memory 421 and external memory 422; memory 421 herein, also referred to as internal memory, is used to temporarily store operation data in processor 41, as well as data exchanged with external memory 422 such as a hard disk. Processor 41 exchanges data with external memory 422 via internal memory 421. When electronic device 400 is running, processor 41 communicates with memory 42 via bus 43, causing processor 41 to execute the following instructions:

[0175] Determine multiple valid behavior data of the user for each target product; wherein the valid behavior data is data that can reflect the user's intended product and has a correlation with the purchase behavior data that meets the requirements;

[0176] Calculating a behavior weight for each type of valid behavior data; wherein the behavior weight is used to indicate the importance of the valid behavior data among the multiple types of valid behavior data;

[0177] An intention index model is constructed based on the behavior weights and the multiple valid behavior data, and the user's purchase intention for a specified series of products is determined according to the intention index model; wherein the intention index model is used to determine the user's attention share or mental share to a specified series of products of a specified style under a target brand within a target industry.

[0178] The present disclosure also provides a computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program executes the steps of the method for determining user purchase intentions described in the above method embodiment. The storage medium may be a volatile or non-volatile computer-readable storage medium.

[0179] The embodiments of the present disclosure also provide a computer program product, which carries program code. The instructions included in the program code can be used to execute the steps of the method for determining user purchase intention described in the above method embodiment. For details, please refer to the above method embodiment and will not be repeated here.

[0180] The computer program product may be implemented in hardware, software, or a combination thereof. In one embodiment, the computer program product is implemented as a computer storage medium. In another embodiment, the computer program product is implemented as a software product, such as a software development kit (SDK).

[0181] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems and devices described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. In the several embodiments provided in the present disclosure, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0182] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0183] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0184] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present disclosure, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0185] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present disclosure, which are used to illustrate the technical solutions of the present disclosure, rather than to limit them. The scope of protection of the present disclosure is not limited thereto. Although the present disclosure has been described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily conceive of changes to the technical solutions described in the above-described embodiments within the technical scope disclosed in the present disclosure, or replace some of the technical features therein with equivalents. Such modifications, changes, or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the scope of protection of the present disclosure. Therefore, the scope of protection of the present disclosure shall be subject to the scope of protection of the claims.

Claims

1. A method for determining user purchase intention, characterized in that: include: Determine multiple valid behavior data of the user for each target product; wherein the valid behavior data is data that can reflect the user's intended product and has a correlation with the purchase behavior data that meets the requirements; Calculating a behavior weight for each type of valid behavior data; wherein the behavior weight is used to indicate the importance of the valid behavior data among the multiple types of valid behavior data; An intention index model is constructed based on the behavior weights and the multiple valid behavior data, and the user's purchase intention for a specified series of products is determined according to the intention index model; wherein the intention index model is used to determine the user's attention share or mental share to a specified series of products of a specified style under a target brand within a target industry.

2. The method according to claim 1, characterized in that The calculating of the behavior weight of each type of valid behavior data includes: Performing dimension-removing and quantifying processing on each effective behavior data of each target product to obtain dimension-removed effective behavior data; Determine the behavior weight of each type of effective behavior data after dimensioning.

3. The method according to claim 2, characterized in that Determining the behavior weight of each type of dimensionless effective behavior data includes: A linear regression fitting is performed on the attention ratio of each valid behavior data and the corresponding de-dimensionalized valid behavior data to obtain the behavior weight; wherein the attention ratio is the ratio between the statistical index of each valid behavior data and the statistical index of the user's valid behavior data.

4. The method according to claim 1, wherein The constructing of the intention index model based on the behavior weight and the multiple valid behavior data includes: Performing weighted summation on the de-dimensionalized effective behavior data of each target product according to the behavior weight to obtain a first result; Performing a weighted summation of each comprehensive behavior data of all the target products according to the behavior weight to obtain a second result; wherein the comprehensive behavior data is the sum of the dedimensionalized effective behavior data of the same type of all the target products; The intention index model is determined based on a ratio of the first result to the second result.

5. The method according to claim 1, wherein After constructing the intention index model based on the behavior weights and the multiple valid behavior data, the method further includes: After the advertising material for the first product is delivered, determining changes in the target user's purchase intention for the first product based on the intention index model; determining a correlation between the target user's actual online behavior with respect to the second product and the target user's purchase intention for the second product; Based on the correlation and / or the change, the validity of the intention index model is verified, and if the validity verification fails, the behavior weight is adjusted.

6. The method according to claim 1, characterized in that The determination of multiple valid user behavior data for each target product includes: Obtaining the user's initial behavior data for each of the target products; Screening the initial behavior data for which the correlation with the purchase behavior data is greater than a preset threshold from the initial behavior data to obtain candidate behavior data; Determine candidate behavior data similar to the intention-indicating behavior of high-intention users to obtain the valid behavior data; wherein, the high-intention user is a user who performs purchase behavior data for the target product, and the intention-indicating behavior can reflect the intended product of the high-intention user.

7. The method according to claim 1, characterized in that After determining the user's purchase intention for a specified series of products according to the intention index model, the method further includes: Determining intention data for the designated product series based on the purchase intention; wherein the intention data is used to indicate the purchase intention of each user for the designated product series; and / or A competitive product analysis is performed on the designated product series based on the purchase intention to obtain a competitive product analysis result; wherein the competitive product analysis result is used to indicate competitive products of the designated product series and / or users who follow the competitive products.

8. A device for determining user purchase intention, characterized in that: include: A first determining unit is configured to determine a plurality of valid behavior data of the user for each target product; wherein the valid behavior data is data that can reflect the user's intended product and has a correlation with the purchase behavior data that meets the requirements; a calculation unit, configured to calculate a behavior weight of each of the valid behavior data; wherein the behavior weight is used to indicate the importance of the valid behavior data among the multiple valid behavior data; The second determination unit is used to construct an intention index model based on the behavior weight and the multiple valid behavior data, and determine the user's purchase intention for the specified series of products according to the intention index model; wherein the intention index model is used to indicate the user's attention share or mental share to the specified series of products of the specified style under the target brand in the target industry.

9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the method for determining user purchase intention as described in any one of claims 1 to 7 are performed.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, executes the steps of the method for determining user purchase intention according to any one of claims 1 to 7.

Images