A method and apparatus for determining a user tag weight

Abstract

The application provides a method and device for determining a user label weight, wherein the method comprises: obtaining a behavior frequency of a target behavior of a target object in an initial period and a behavior frequency of the target behavior in a termination period; determining a behavior attenuation coefficient of the target behavior according to the behavior frequency in the initial period and the behavior frequency in the termination period; and determining a user label weight of the target object when the target object labels a user as a user label according to the behavior attenuation coefficients of multiple target behaviors of the target object. The above scheme solves the technical problem that the existing behavior attenuation coefficient cannot be accurately determined, resulting in inaccurate labeling results of the user, and achieves the technical effects of accurately determining the behavior attenuation coefficient, accurately determining the user label weight, and accurately labeling the user.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of electric digital data processing, and particularly relates to a method and device for determining user label weight. BACKGROUND

[0002] In order to realize crowd delineation, the user habit and behavior are predicted, and a user portrait is often formed, so as to accurately and efficiently classify and identify the user based on the user portrait. For example, in the medical field, the user's label for goods and the user's label for diseases are established based on the user's behavior data, and when labeling the user, the influence of the user's behavior data on the weight needs to be considered. In the process of weight fitting, the user label weight needs to be determined based on the behavior decay coefficient.

[0003] At present, the decay coefficient is generally determined by Newton's cooling law. However, in actual use, it is found that the accuracy of measuring the decay of user behavior in the e-commerce field by Newton's cooling law is not high.

[0004] At present, there is no effective solution to how to accurately measure the decay of user behavior in the e-commerce field by time to form an accurate decay coefficient to accurately label the user. SUMMARY

[0005] The present application aims to provide a method and device for determining user label weight, which can accurately determine the influence of time on user behavior decay, so as to accurately determine the user label weight.

[0006] The present application provides a method and device for determining user label weight, which is realized as follows:

[0007] A method for determining user label weight comprises:

[0008] Obtaining the number of behaviors of a target behavior of a target object in an initial period and the number of behaviors in a termination period;

[0009] Determining the behavior decay coefficient of the target behavior according to the number of behaviors in the initial period and the number of behaviors in the termination period;

[0010] According to the behavior decay coefficients of a plurality of target behaviors of the target object, determining the user label weight when the target object is labeled as a user label for a user.

[0011] In one embodiment, the behavior decay coefficient of the target behavior is determined according to the number of behaviors in the initial period and the number of behaviors in the termination period, comprising:

[0012] Determining the number of periods from the initial period to the termination period as the square root number;

[0013] calculating a ratio of the number of behaviors in the termination period to the number of behaviors in the initial period as a base value;

[0014] performing the square root operation on the base value to obtain a result value;

[0015] taking the result value as the behavior attenuation coefficient.

[0016] In one embodiment, determining a user tag weight of a target object as a user tag for a target user according to behavior attenuation coefficients of a plurality of target behaviors of the target object comprises:

[0017] obtaining a behavior type weight of each target behavior in the plurality of target behaviors, a number of behaviors of the target user in each target behavior from the initial period to the termination period, and an objective weight value of the target object as a tag;

[0018] determining a user tag weight of the target object as a user tag for the target user according to the behavior attenuation coefficients of each target behavior in the plurality of target behaviors, the behavior type weight of each target behavior, the number of behaviors of the target user in each target behavior from the initial period to the termination period, and the objective weight value of the target object as a tag for the target user.

[0019] In one embodiment, determining a user tag weight of the target object as a user tag for the target user according to the behavior attenuation coefficients of each target behavior in the plurality of target behaviors, the behavior type weight of each target behavior, the number of behaviors of the target user in each target behavior from the initial period to the termination period, and the objective weight value of the target object as a tag for the target user comprises:

[0020] performing the following operation on each target behavior respectively to obtain an intermediate term of each target behavior: calculating a product of the behavior attenuation coefficient, the behavior type weight, and the number of behaviors of the target user in the initial period to the termination period of the current target behavior as the intermediate term of the current target behavior;

[0021] accumulating the intermediate terms of each target behavior to obtain an accumulated value;

[0022] obtaining the user tag weight of the target object as a user tag for the target user by multiplying the accumulated value by the objective weight value of the target object as a tag for the target user.

[0023] In one embodiment, obtaining a number of behaviors of a target behavior of a target object in an initial period and a number of behaviors in a termination period comprises:

[0024] calling log files of a plurality of statistical platforms;

[0025] The behavior records in the log file are matched to obtain the behavior times of the target behavior of the target object in the initial period and the behavior times of the target behavior of the target object in the termination period.

[0026] In an embodiment, the method further comprises, after determining the user tag weight of the target object as a user tag for tagging users according to the behavior attenuation coefficients of the multiple target behaviors of the target object:

[0027] Tagging users according to the determined user tag weight;

[0028] Forming a user portrait according to the tagging result.

[0029] In an embodiment, the target object is a chronic disease.

[0030] A method for determining a user tag weight, comprising:

[0031] Obtaining the behavior times of a target behavior of a target object on an e-commerce platform in an initial period and the behavior times of the target behavior of the target object in a termination period;

[0032] Determining a behavior attenuation coefficient of the target behavior according to the behavior times in the initial period and the behavior times in the termination period;

[0033] Determining a user tag weight of the target object as a user tag for tagging users according to the behavior attenuation coefficients of the multiple target behaviors of the target object.

[0034] In an embodiment, the method for determining the behavior attenuation coefficient of the target behavior according to the behavior times in the initial period and the behavior times in the termination period comprises:

[0035] Determining the number of periods from the initial period to the termination period as a square root number;

[0036] Calculating the ratio of the behavior times in the termination period to the behavior times in the initial period as a base value;

[0037] Performing a square root operation on the base value to obtain a result value;

[0038] Taking the result value as the behavior attenuation coefficient.

[0039] In an embodiment, the method for determining the user tag weight of the target object as a user tag according to the behavior attenuation coefficients of the multiple target behaviors of the target object comprises:

[0040] Obtaining the behavior type weight of each target behavior, the behavior times of each target behavior of a target user from the initial period to the termination period, and the objective weight value of the target object as a tag;

[0041] According to the behavior attenuation coefficient of each target behavior in the plurality of target behaviors, the behavior type weight of each target behavior, the behavior number of the target user in each target behavior from the initial period to the termination period, and the objective weight value of the target commodity as the label of the target user, the user label weight when the target commodity as the user label is labeled to the target user is determined.

[0042] In one embodiment, according to the behavior attenuation coefficient of each target behavior in the plurality of target behaviors, the behavior type weight of each target behavior, the behavior number of the target user in each target behavior from the initial period to the termination period, and the objective weight value of the target commodity as the label of the target user, the user label weight when the target commodity as the user label is labeled to the target user is determined, comprising:

[0043] For each target behavior, the following operations are performed to obtain the intermediate term of each target behavior: the product of the behavior attenuation coefficient, the behavior type weight, and the behavior number of the target user in the current target behavior from the initial period to the termination period is calculated as the intermediate term of the current target behavior;

[0044] The intermediate terms of each target behavior are accumulated to obtain an accumulated value;

[0045] The user label weight when the target commodity as the user label is labeled to the target user is obtained by multiplying the accumulated value by the objective weight value of the target commodity as the label of the target user.

[0046] In one embodiment, the behavior number of the target behavior of the target commodity on the e-commerce platform in the initial period and the behavior number in the termination period are obtained, comprising:

[0047] The log files of a plurality of statistical platforms are called;

[0048] Each behavior record in the log file is matched and counted to obtain the behavior number of the target behavior of the target commodity in the initial period and the behavior number in the termination period.

[0049] In one embodiment, after the user label weight when the target commodity as the user label is labeled to the user is determined according to the behavior attenuation coefficient of the plurality of target behaviors of the target commodity, the method further comprises:

[0050] According to the determined user label weight, the user is labeled;

[0051] According to the labeling result, a user portrait is formed.

[0052] In one embodiment, the target commodity is a chronic disease medicine.

[0053] A user label weight determination device, comprising:

[0054] The acquisition module is configured to acquire a behavior frequency of a target behavior of a target object in an initial period and a behavior frequency of the target behavior in a termination period.

[0055] The first determination module is configured to determine a behavior attenuation coefficient of the target behavior according to the behavior frequency in the initial period and the behavior frequency in the termination period.

[0056] The second determination module is configured to determine a user tag weight when the target object is tagged as a target user according to the behavior attenuation coefficients of a plurality of target behaviors of the target object.

[0057] An electronic device includes a processor and a memory storing processor-executable instructions that, when executed by the processor, implement the steps of the above method.

[0058] A computer-readable storage medium stores computer programs / instructions that, when executed by a processor, implement the steps of the above method.

[0059] The method and device for determining a user tag weight provided in the present application determine a behavior attenuation coefficient of a target behavior through a behavior frequency of the target behavior in an initial period and a behavior frequency of the target behavior in a termination period. The behavior attenuation coefficient is determined based on a comparison of the behavior frequencies in the two statistical periods, and thus is more accurate and reasonable. Furthermore, if there are a plurality of behavior types for a target object, the attenuation coefficients of the behaviors can be determined respectively, and the user tag weight when the target object is tagged as a target user based on the attenuation coefficients of the plurality of behaviors. The above solution solves the technical problem of inaccurate tagging of a user caused by an inaccurate behavior attenuation coefficient, and achieves the technical effect of accurately determining a behavior attenuation coefficient and thus accurately determining a user tag weight to accurately tag a user. BRIEF DESCRIPTION OF DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0061] Figure 1 is a method flowchart of an embodiment of the method for determining a user tag weight provided in the present application;

[0062] Figure 2 is a schematic diagram of the correspondence between a user and a tag provided in the present application;

[0063] Figure 3 is a flowchart of a process for determining a user tag weight of a drug or disease provided by the present application;

[0064] Figure 4 is a hardware structure block diagram of an electronic device for a user tag weight determination method provided by the present application;

[0065] Figure 5 is a module structure diagram of an embodiment of a user tag weight determination apparatus provided by the present application. DETAILED DESCRIPTION

[0066] In order to enable personnel in the technical field to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor should fall within the scope of protection of the present application.

[0067] It is considered that in the application direction based on medical health, for example: the inquiry platform, insurance, disease tracing and e-commerce and other applications generally exist the demand of determining the circle selection target group through diseases, in order to realize the purpose of accurate circle selection, generally need to label the user, for example, user-goods label, user-disease label, and need to fit to obtain each user label weight. There is a relationship between the user label weight and the time decay of behavior data. The existing way of determining the decay coefficient through Newton cooling law mathematical model is not very accurate in determining the label weight of users, goods and diseases in the direction of medical health, because in the actual application scene, the decay of user operation behavior on diseases and e-commerce products often does not conform to the Newton cooling law mathematical model.

[0068] Therefore, in the present example, a user label weight determination method calculates the decay coefficient in an objective and practical application manner, so that the calculated user label weight is more accurate.

[0069] Figure 1is a method flowchart of an embodiment of the method for determining the weight of a user tag provided by the present application. Although the present application provides the method operation steps or device structures as described in the following embodiments or drawings, more or fewer operation steps or module units can be included in the method or device based on conventional or non-inventive labor. There is no necessary causal relationship between the steps or structures in terms of logic, and the execution order of the steps or the module structure of the device is not limited to the execution order or module structure described in the embodiments and shown in the drawings. When the method or module structure is applied to the actual device or terminal product, it can be sequentially executed or executed in parallel (for example, in a parallel processor or multi-thread processing environment, or even a distributed processing environment) according to the method or module structure shown in the embodiments or drawings.

[0070] Specifically, as shown in Figure 1 The method for determining the weight of a user tag can include the following steps:

[0071] Step 101: Obtain the number of behaviors of a target object in an initial period and the number of behaviors in a termination period;

[0072] That is, considering the number of behaviors of a user on a target object (for example: a disease, a commodity) in a period of time, the stickiness between the user and the target object can be reflected, and the stickiness value reflects the relationship label between the user and the target object. However, for the operation behavior of the user on the target object, some behaviors will be continuously weakened by time, and the farther the behavior time is from now, the less meaningful the behavior data is to the user at present. Therefore, the decay coefficient can be used to measure the value of a certain behavior to the current application.

[0073] In this example, the behavior time is used as an influencing factor to measure the influence of historical behavior on the present, that is, to determine the number of times the historical behavior will occur at present. Specifically, the retention rate of n statistical periods can be used to represent the decay coefficient, for example, taking days as the period, then the behavior retention rate of n days is equal to the nth power of the behavior retention rate x of each day. If it is necessary to determine the decay coefficient to measure the influence of the user's historical behavior on the present, the result of raising the behavior retention rate of the initial and final periods to the power of n can be used to represent the decay coefficient, where n is the time interval between the initial and final periods.

[0074] Therefore, in this example, the number of behaviors in the initial period and the number of behaviors in the termination period are obtained. The number of behaviors can be the number of behaviors of all users on the target object in the target platform. When counting, the period can be used as a unit, for example, the period is "day", then the number of behaviors on the initial day and the number of behaviors on the termination day are obtained, where the termination period is the current period, that is, the period at the end of the calculation.

[0075] In actual implementation, other time metrics can also be used as the period, for example, a period of "week" can be used, that is, 7 days as a statistical period, or a period of "month" can be used, that is, a month as a statistical period. The specific selection of the period metric can be selected according to actual conditions, and the application does not limit this.

[0076] Step 102: determining the behavior attenuation coefficient of the target behavior according to the number of behaviors in the initial period and the number of behaviors in the termination period;

[0077] After obtaining the number of behaviors in the initial period and the number of behaviors in the termination period, the number of periods from the initial period to the termination period can be determined as the square root number; the ratio of the number of behaviors in the termination period to the number of behaviors in the initial period is calculated as a base value; the base value is subjected to the square root operation of the square root number to obtain a result value; and the result value is used as the behavior attenuation coefficient.

[0078] For example, the number of behaviors one year ago and the current number of behaviors are counted as data basis, that is, it is determined how many behaviors remain one year later, and the period is 365 days. The number of behaviors in the initial period is the number of behaviors 365 days ago, and the number of behaviors in the termination period is the current number of behaviors. Accordingly, the behavior attenuation coefficient can be calculated according to the following formula:

[0079]

[0080] In this example, the method for determining the behavior attenuation coefficient can be applied to the determination of the label weight of, for example, "chronic disease" and "chronic disease drug", because chronic disease is a disease that needs to be treated and medicated continuously. For this disease, the behavior data of the user is persistent. That is, in this example, the target object is an object whose corresponding target behavior is persistent and periodic.

[0081] Step 103: determining the user label weight when the target object is used as a user label to label a user according to the behavior attenuation coefficients of the multiple target behaviors of the target object.

[0082] Specifically, after the behavior attenuation coefficient is determined, the user label weight can be determined according to the behavior attenuation coefficient. In this example, the operation behavior of the target object is of multiple behavior types, rather than a single behavior type. Therefore, in order to improve the accuracy of the determined user label weight, the attenuation coefficients of multiple behavior types can be accumulated, which is more accurate than the user label obtained by a single behavior type.

[0083] To this end, determining the user tag weight of the target object as a user tag for the target user according to the behavior decay coefficients of the plurality of target behaviors of the target object can include:

[0084] S1: obtaining the behavior type weight of each target behavior in the plurality of target behaviors, the behavior frequency of the target user in each target behavior from the initial period to the termination period, and the objective weight value of the target object as a tag;

[0085] S2: determining the user tag weight of the target object as a user tag for the target user according to the behavior decay coefficients of each target behavior in the plurality of target behaviors, the behavior type weight of each target behavior, the behavior frequency of the target user in each target behavior from the initial period to the termination period, and the objective weight value of the target object as a tag for the target user.

[0086] Specifically, determining the user tag weight of the target object as a user tag for the target user according to the behavior decay coefficients of each target behavior in the plurality of target behaviors, the behavior type weight of each target behavior, the behavior frequency of the target user in each target behavior from the initial period to the termination period, and the objective weight value of the target object as a tag for the target user can include: performing the following operations on each target behavior respectively to obtain an intermediate term of each target behavior: calculating the product of the behavior decay coefficient, the behavior type weight, and the behavior frequency of the target user in the current target behavior from the initial period to the termination period as the intermediate term of the current target behavior; accumulating the intermediate terms of each target behavior to obtain an accumulated value; and multiplying the accumulated value by the objective weight value of the target object as a tag for the target user to obtain the user tag weight of the target object as a user tag for the target user.

[0087] That is, the user tag weight can be calculated using the following formula:

[0088] User tag weight = ∑(behavior type weight x behavior decay coefficient x behavior frequency) x objective weight of the tag

[0089] Wherein, the user behavior can include but is not limited to: search, browse, collect, add to cart, order, payment, scan code, etc. Different behaviors have different importance, and the behavior type weight is used to represent such importance. In this example, the behavior type weight can be the proportion of payment brought by the behavior.

[0090] For example, the behavior type weight can be obtained according to the following formula:

[0091]

[0092] The behavior number is the number of behaviors of the target user in the current statistical period (for example, the current day, the current month, etc.) of the behavior, that is, the more the number of behaviors of the target object corresponding to the target user in the current statistical period, the greater the influence of the target object on the user, and therefore, the higher the relative label weight.

[0093] The objective weight value of the label can be the product of the importance of the label to the target user and the importance of the label among all labels.

[0094] The importance of the label to the target user can be the proportion of the number of occurrences of the label in the number of occurrences of all labels of the target user, for example, as shown in Figure 2 The user is user 1, user 2, and user 3, and the label is label A, label B, label C, and label D. Specifically, if user 1 has label A 5 (i.e., the user has 2 interactions with A), label B 2 (i.e., the user has 2 interactions with B), and label C 1 (i.e., the user has 1 interaction with C), the importance of label A on user 1 is 5 / (5+2+1).

[0095] The importance of the label among all labels can be the scarcity of the label among all labels, for example, by dividing the cumulative sum of the number of users tagged with the label by the cumulative sum of the number of labels tagged by all users, that is, the proportion of the number of occurrences of the current label among all labels.

[0096] The user label weight can be updated periodically, for example, the user label weight is updated once a day.

[0097] When obtaining the behavior number of the target behavior of the target object in the initial period and the behavior number in the termination period, the log files of multiple statistical platforms can be called; each behavior record in the log file is matched and counted to obtain the behavior number of the target behavior of the target object in the initial period and the behavior number in the termination period. For example, if the user tagging scenario based on user behavior in the e-commerce platform is used, the log files can be obtained from multiple statistical platforms, and the behavior number of the target behavior of the target object in the initial period and the behavior number in the termination period can be obtained based on the time dimension and the user identifier dimension. Specifically, a unified user label for multiple platforms can be formed, that is, the final user tagging result can be shared by multiple platforms. For example, the user tagging result obtained by the inquiry platform, insurance, traceability code, and e-commerce can be shared to achieve effective identification and matching of users.

[0098] After the user tag weight when the target object is tagged as a user tag for a user is obtained, the user can be tagged according to the determined user tag weight; and a user portrait is formed according to the tagging result. Specifically, the user can be tagged based on the determined user tag weight, that is, the greater the weight value, the larger the size of the tag on the corresponding user, and the smaller the weight value, the smaller the size of the tag on the corresponding user. In this way, the user portrait can be formed, and the user is tagged based on the weight value on the user portrait.

[0099] The target object described above can be a chronic disease, a drug for treating a chronic disease, or other objects with persistent and periodic operations.

[0100] In the above example, the decay coefficient of the target behavior is determined by the number of behaviors of the target object in the initial period and the number of behaviors in the termination period. Since the comparison is based on the number of behaviors in two statistical periods, the determined behavior decay coefficient is more accurate and reasonable. Further, for the target object, if there are multiple behavior types, the decay coefficients of each behavior can be determined respectively, and the user tag weight when the target object is tagged as a user tag for a user is determined based on the decay coefficients of multiple behaviors. The above solution solves the technical problem that the user tagging result is inaccurate due to the inability to accurately determine the behavior decay coefficient, and achieves the technical effect of accurately determining the behavior decay coefficient, thereby accurately determining the user tag weight and accurately tagging the user.

[0101] In this example, a method for determining a user tag weight is also provided, which is applied to the processing of a target commodity in an e-commerce platform to tag users of the e-commerce platform, and can include the following steps:

[0102] S1: Obtain the number of behaviors of a target behavior of a target commodity in an initial period and the number of behaviors in a termination period on an e-commerce platform, wherein the target commodity can be a chronic disease drug;

[0103] S2: Determine the behavior decay coefficient of the target behavior according to the number of behaviors in the initial period and the number of behaviors in the termination period;

[0104] S3: Determine the user tag weight when the target commodity is tagged as a user tag for a user according to the behavior decay coefficients of multiple target behaviors of the target commodity.

[0105] The above method will be described in a specific scenario below. However, it should be noted that the specific embodiments are only used to better illustrate the present application and do not constitute an improper limitation on the present application.

[0106] Taking medical, health, and e-commerce platforms as examples, the relationship between users and goods needs to be determined, and the relationship between users and diseases needs to be determined. To this end, the application scenarios of the consultation platform, insurance, traceability code, and e-commerce can be connected to build a unified patient identification capability. To this end, user-goods tags and user-disease tags can be formed, and the weight fitting can be performed to obtain the label weight of each target object as the patient label when labeling the user.

[0107] Specifically, as shown in Figure 3 , the e-commerce drug ID and the disease ID are obtained, the user-e-commerce drug behavior is summarized to obtain a user-e-commerce drug behavior set, the user-disease e-commerce behavior is summarized to obtain a user-disease behavior set, then the user-e-commerce drug behavior set and the user-disease behavior set are aggregated to obtain a behavior type set, then the medical e-commerce, consultation platform, traceability code, and other platforms are taken as data sources to obtain behavior data related to each behavior type in the behavior type set, and then the behavior data is analyzed to calculate the user label weight.

[0108] Specifically, the user label weight can be determined by the following formula:

[0109] User label weight = ∑(behavior type weight x behavior decay coefficient x behavior frequency) x objective weight of label

[0110] 1) To determine the user-goods weight, the data shown in Table 1 below can be obtained:

[0111] Table 1

[0112]

[0113] The user-goods weight can be calculated according to the following formula:

[0114] User label weight = ∑(behavior type weight x behavior decay coefficient x behavior frequency) x objective weight of label

[0115] Among them, the behavior type weight can be obtained by fitting the user-goods behavior weight, for example, the payment frequency of goods in the past 365 days can be taken as the basis, and the proportion of the behavior frequency converted to payment and the behavior frequency of search / browse / collection / add to cart / order / scan code and the like can be obtained. As shown in Table 2 below, the behavior type weight of each behavior type obtained is as follows:

[0116] Table 2

[0117]

[0118] Wherein, the behavior decay coefficient is the discount of the historical behavior times to the actual effect of the present, for this, the remaining rate of the behavior times of all users on the goods after one year can be based on the different behavior times of all users on the goods before one year, and the true number obtained by taking the remaining rate as the base number and 365 as the power represents the behavior decay coefficient, and whether the disease type of the disease is a chronic disease can be distinguished, specifically, as shown in Table 3:

[0119] Table 3

[0120]

[0121]

[0122] Wherein, the objective weight of the label can be calculated by multiplying TF by IDF, wherein, TF represents the importance of the commodity label to the user, which can be taken as the proportion of the sum of the user's behavior times on the commodity to the sum of the user's behavior times on all commodities; IDF represents the importance of the commodity label among all commodity labels, which can be taken as the sum of the behavior times of all users on the commodity divided by the sum of all behavior times, wherein, the sum of all behavior times can represent the total number of times of searching, browsing, collecting, adding to cart, ordering, paying and scanning code of the user on the commodity.

[0123] 2) In order to determine the user disease weight, the data shown in Table 4 can be obtained:

[0124] Table 4

[0125]

[0126] The user disease weight can be calculated according to the following formula:

[0127] User label weight = ∑(behavior type weight x behavior decay coefficient x behavior times) x label objective weight

[0128] Wherein, the behavior type weight can be obtained by comparing the behavior type of the medical e-commerce with the payment times guided by the behavior of the medical e-commerce, and the behavior type of the consultation platform can be obtained by comparing the behavior type of the consultation platform with the payment times guided by the behavior of the consultation platform, specifically, as shown in Table 5:

[0129] Table 5

[0130]

[0131]

[0132] The behavior attenuation coefficient represents that the influence of the historical behavior times on the present needs to be discounted, and therefore, the behavior attenuation coefficient can be obtained by taking the survival rate as the base number and 365 as the power of the survival rate based on the behavior times of all users for the different diseases one year ago and how many of the behavior times remain one year later. Further, the behavior attenuation coefficient of each behavior type can be shown in Table 6 as follows:

[0133] Table 6

[0134]

[0135] The objective weight of the label can be calculated by multiplying the TF and the IDF, where the TF represents the importance of the disease label to the user, and can be taken as the proportion of the sum of the behavior times of the user for the disease / the sum of the behavior times of the user for all diseases; and the IDF represents the importance of the disease label in all disease labels, and can be taken as the sum of the behavior times of all users for the disease / the sum of all behavior times, where the sum of all behavior times can represent the total number of searches, browses, collections, adds to shopping carts, orders, payments, and inquiries of the user for the disease.

[0136] The label weight of the user disease label can be obtained based on the real-time update of the label weight of the user commodity label, and the user label can be updated in time, the user amount covered by the label and the coverage rate of each commodity and disease can be determined based on the label weight of the user commodity label and the label weight of the user disease label, and the effective population division of the user can be performed based on the label weight of the user commodity label and the label weight of the user disease label.

[0137] The method embodiments provided by the above embodiments of the application can be executed in a mobile terminal, a computer terminal, a processor, a server, or a similar computing device. Taking the running on an electronic device as an example, Figure 4 is a hardware structure block diagram of an electronic device for determining a user label weight provided by the application. As Figure 4 indicated, the electronic device 10 can include one or more (only one is shown in the figure) processors 02 (the processor 02 can include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 04 for storing data, and a transmission module 06 for communication function. Those skilled in the art can understand that Figure 4 the structure shown is only schematic, and does not limit the structure of the above-mentioned electronic device. For example, the electronic device 10 can include more or fewer components than Figure 4 indicated, or have a different configuration from Figure 4 indicated.

[0138] The memory 04 can be used to store software programs of application software and modules, such as program instructions / modules corresponding to the method for determining user tag weight in the embodiments of the present application, and the processor 02 executes various functional applications and data processing by running the software programs and modules stored in the memory 04, that is, implements the method for determining user tag weight of the application program as described above. The memory 04 can include a high-speed random access memory, and can also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 04 can further include memories disposed remotely with respect to the processor 02, which can be connected to the electronic device 10 through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0139] The transmission module 06 is used to receive or send data via a network. Specific examples of the above-mentioned network can include a wireless network provided by a communication provider of the electronic device 10. In one example, the transmission module 06 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to be able to communicate with the Internet. In one example, the transmission module 06 can be a radio frequency (Radio Frequency, RF) module, which is used to communicate with the Internet in a wireless manner.

[0140] At the software level, the device for determining user tag weight can include, as shown in Figure 5

[0141] The acquisition module 501 is used to acquire the number of behaviors of a target object in an initial period and the number of behaviors of the target object in a termination period.

[0142] The first determination module 502 is used to determine a behavior attenuation coefficient of a target behavior according to the number of behaviors in the initial period and the number of behaviors in the termination period.

[0143] The second determination module 503 is used to determine a user tag weight of a target object as a user tag for a user according to behavior attenuation coefficients of multiple target behaviors of the target object.

[0144] In one embodiment, the first determination module 502 can specifically determine the number of periods from the initial period to the termination period as a square root number; calculate a ratio of the number of behaviors in the termination period to the number of behaviors in the initial period as a base value; perform a square root operation on the base value to obtain a result value; and take the result value as the behavior attenuation coefficient.

[0145] ​In an embodiment, the second determining module 503 can specifically acquire the behavior type weight of each target behavior in the plurality of target behaviors, the behavior frequency of the target user in each target behavior from the initial period to the termination period, and the objective weight value of the target object as a label; and determine the user label weight when the target object as a user label labels the target user according to the behavior decay coefficient of each target behavior in the plurality of target behaviors, the behavior type weight of each target behavior, the behavior frequency of the target user in each target behavior from the initial period to the termination period, and the objective weight value of the target object as a label of the target user.

[0146] In an embodiment, the second determining module 503 can specifically acquire the behavior type weight of each target behavior in the plurality of target behaviors, the behavior frequency of the target user in each target behavior from the initial period to the termination period, and the objective weight value of the target object as a label; and determine the user label weight when the target object as a user label labels the target user according to the behavior decay coefficient of each target behavior in the plurality of target behaviors, the behavior type weight of each target behavior, the behavior frequency of the target user in each target behavior from the initial period to the termination period, and the objective weight value of the target object as a label of the target user.

[0147] In an embodiment, the acquiring module 501 can specifically call the log file of the platform to be counted; and match and count each behavior record in the log file to obtain the behavior frequency of the target behavior of the target object in the initial period and the behavior frequency in the termination period.

[0148] In an embodiment, the above-mentioned user label weight determining device can further label the user according to the determined user label weight after determining the user label weight when the target object as a user label labels the user according to the behavior decay coefficient of the plurality of target behaviors of the target object; and form a user portrait according to the labeling result.

[0149] In this example, a user label weight determining device is also provided, which can include an acquiring module configured to acquire the behavior frequency of a target behavior of a target object in an initial period and the behavior frequency in a termination period on an e-commerce platform; a first determining module configured to determine a behavior decay coefficient of the target behavior according to the behavior frequency in the initial period and the behavior frequency in the termination period; and a second determining module configured to determine a user label weight when the target object as a user label labels a user according to the behavior decay coefficient of the plurality of target behaviors of the target object.

[0150] In an embodiment, the target commodity can be a chronic disease medicine.

[0151] Embodiments of the present application also provide a specific implementation of an electronic device capable of implementing all steps of the method for determining the user tag weight in the above embodiments. The electronic device specifically includes a processor, a memory, a communications interface, and a bus. The processor, memory, and communications interface communicate with each other through the bus. The processor is configured to call a computer program in the memory. When the processor executes the computer program, all steps of the method for determining the user tag weight in the above embodiments are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0152] Step 1: Obtain the number of behaviors of a target behavior of a target object in an initial period and the number of behaviors in a termination period.

[0153] Step 2: Determine a behavior decay coefficient of the target behavior according to the number of behaviors in the initial period and the number of behaviors in the termination period.

[0154] Step 3: Determine a user tag weight of the target object as a user tag for a user according to the behavior decay coefficients of multiple target behaviors of the target object.

[0155] As can be seen from the above description, embodiments of the present application determine the decay coefficient of a target behavior by the number of behaviors of the target behavior of a target object in an initial period and the number of behaviors in a termination period. The behavior decay coefficient is determined based on a comparison of the number of behaviors in the initial period and the number of behaviors in the termination period, and thus is more accurate and reasonable. Further, for a target object, if there are multiple behavior types, the decay coefficients of the behaviors can be determined respectively, and the user tag weight of the target object as a user tag for a user can be determined based on the decay coefficients of the multiple behaviors. The above solution solves the technical problem of inaccurate user tagging results caused by the inability to accurately determine the behavior decay coefficient, and achieves the technical effect of accurately determining the behavior decay coefficient and thus accurately determining the user tag weight to accurately tag users.

[0156] Embodiments of the present application also provide a computer-readable storage medium capable of implementing all steps of the method for determining the user tag weight in the above embodiments. The computer-readable storage medium stores a computer program. When the processor executes the computer program, all steps of the method for determining the user tag weight in the above embodiments are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0157] Step 1: obtaining the behavior times of a target behavior of a target object in an initial period and the behavior times of the target behavior in a termination period;

[0158] Step 2: determining a behavior decay coefficient of the target behavior according to the behavior times in the initial period and the behavior times in the termination period;

[0159] Step 3: determining a user tag weight of the target object when the target object tags a user as a user tag according to the behavior decay coefficients of multiple target behaviors of the target object.

[0160] From the above description, the embodiments of the present application determine the decay coefficient of the target behavior through the behavior times of the target behavior of the target object in the initial period and the behavior times in the termination period. The behavior decay coefficient is determined based on the comparison of the behavior times in the two statistical periods, i.e., the behavior times in the initial period and the behavior times in the termination period, and thus the behavior decay coefficient is more accurate and reasonable. Further, for the target object, if there are multiple behavior types, the decay coefficients of the behaviors can be determined respectively, and the user tag weight of the target object when the target object tags a user as a user tag is determined based on the decay coefficients of the multiple behaviors. The above solution solves the technical problem that the behavior decay coefficient cannot be accurately determined, resulting in inaccurate tagging of the user, and achieves the technical effect of accurately determining the behavior decay coefficient and accurately determining the user tag weight to accurately tag the user.

[0161] Each of the embodiments in the specification is described in a progressive manner, and the same or similar parts of each embodiment can be referred to each other. Each embodiment focuses on the difference from other embodiments. In particular, for the hardware+program type embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the part of the method embodiment.

[0162] The above describes specific embodiments of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be executed in an order different than the order in the embodiments and still achieve the desired result. In addition, the processes depicted in the figures do not necessarily require the particular order shown or sequential order to achieve the desired results. In certain implementations, multitasking and parallel processing can be advantageous.

[0163] Although the present application provides method operations steps as recited in the embodiments or flowcharts, more or less operation steps can be included based on routine or non-creative labor. The order of steps recited in the embodiments is only one of the many step execution orders, and does not represent the only execution order. In actual device or client product execution, the method order can be executed in sequence or in parallel (for example, in a parallel processor or multi-thread processing environment) as shown in the embodiments or drawings.

[0164] The system, device, module or unit illustrated in the above embodiments can be specifically implemented by a computer chip or entity, or by a product with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, an in-vehicle human-computer interaction device, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0165] Although the present application provides method operations steps as recited in the embodiments or flowcharts, more or less operation steps can be included based on routine or non-creative labor. The order of steps recited in the embodiments is only one of the many step execution orders, and does not represent the only execution order. In actual device or client product execution, the method order can be executed in sequence or in parallel (for example, in a parallel processor or multi-thread processing environment, even in a distributed data processing environment) as shown in the embodiments or drawings. The term "comprise", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that processes, methods, products or devices including a series of elements not only include those elements, but also include other elements not explicitly listed or inherent to such processes, methods, products or devices. Without more limitations, it does not exclude the presence of other same or equivalent elements in the processes, methods, products or devices including the elements.

[0166] For ease of description, the above apparatus is described in various modules with functions respectively. Of course, functions of the modules can be implemented in one or more software and / or hardware in implementing the embodiments of the present application, and the modules with the same functions can be implemented by combinations of a plurality of sub-modules or sub-units, etc. The apparatus embodiments described above are only schematic, for example, the division of the units is only a logical function division, and there can be another division manner in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0167] Those skilled in the art will appreciate that, in addition to implementing the controller in the form of a purely computer-readable program code, it is also possible to implement the controller in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers and embedded microcontrollers, etc. by logically programming the method steps to perform the same functions. Therefore, such a controller can be considered as a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0168] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as combinations of the flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus generate a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The means for implementing each flow or multiple flows and / or blocks Figure 1 The means for implementing the functions specified in each block or multiple blocks.

[0169] These computer program instructions can also be stored in a computer-readable memory capable of directing the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction means, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The means for implementing each flow or multiple flows and / or blocks Figure 1 The means for implementing the functions specified in each block or multiple blocks.

[0170] These computer program instructions can also be loaded into computer or other programmable data processing devices to cause a series of operational steps to be performed on the computer or other programmable devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable devices provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 Figure 1

[0171] In one typical configuration, the computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0172] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) about which the computer stores information about the operating environment. The memory can also include non-volatile memory, such as read only memory (ROM), EPROM, and / or flash RAM, about which the computer stores information, such as firmware for the computing device. Thus, the memory is an example of computer readable storage media.

[0173] Computer readable media includes permanent and non-permanent, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital versatile discs (DVDs) or other optical storage, magnetic cassettes, magnetic disks storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to computing devices. According to the definition herein, computer readable media does not include transitory media, such as modulated data signals and carrier waves.

[0174] Those skilled in the art will appreciate that embodiments of the present specification can be provided as methods, systems or computer program products. Therefore, embodiments of the present specification can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present specification can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk memory, CD-ROMs, optical storage media, etc.) having computer usable program code embodied therein.

[0175] ​​The embodiments of the present specification can be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like, that perform particular tasks or implement particular abstract data types. The embodiments of the present specification can also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules can be located in both local and remote computer storage media including memory storage devices.

[0176] Each of the embodiments in the present specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each of the embodiments focuses on the difference from other embodiments. In particular, for the system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the description of the method embodiments. In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the embodiments of the present specification. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the present specification and the features of the different embodiments or examples without contradiction.

[0177] The above only describes the embodiments of the embodiments of the present specification, and is not intended to limit the embodiments of the present specification. The embodiments of the present specification can have various changes and variations for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiments of the present specification shall be included in the scope of claims of the embodiments of the present specification.

Claims

1. A method for determining user tag weights, characterized in that, The method comprises: obtaining the behavior times of the target behavior of the target object in the initial period and the behavior times in the termination period; determining the behavior attenuation coefficient of the target behavior according to the behavior times in the initial period and the behavior times in the termination period; determining the user label weight of the target object as a user label for a user according to the behavior attenuation coefficients of multiple target behaviors of the target object; wherein the user label weight is calculated on the basis of the behavior type weight, and the behavior type weight comprises the weight of the behavior type of the medical e-commerce and the weight of the behavior type of the consultation platform; the weight of the behavior type of the medical e-commerce is obtained by comparing the behavior and payment times of the medical e-commerce on the basis of the fitting result of the behavior times guiding the payment times of the medical e-commerce; and the weight of the behavior type of the consultation platform is obtained by comparing the behavior times and payment times of the consultation platform on the basis of the fitting result of the behavior times guiding the payment times of the consultation platform.

2. The method of claim 1, wherein, According to the behavior times in the initial period and the behavior times in the termination period, the behavior attenuation coefficient of the target behavior is determined, comprising: determining the number of periods from the initial period to the termination period as the square root number; calculating the ratio of the behavior times in the termination period to the behavior times in the initial period as the base value; performing the square root operation on the base value to obtain a result value; the result value is taken as the behavior attenuation coefficient.

3. The method of claim 1, wherein, According to the behavior attenuation coefficients of multiple target behaviors of the target object, the user label weight of the target object as a user label is determined, comprising: obtaining the behavior type weight of each target behavior in multiple target behaviors, the behavior times of each target behavior of the target user in the initial period to the termination period, and the objective weight value of the target object as a label; determining the user label weight of the target object as a user label for the target user according to the behavior attenuation coefficients of each target behavior in multiple target behaviors, the behavior type weight of each target behavior, the behavior times of each target behavior of the target user in the initial period to the termination period, and the objective weight value of the target object as a label of the target user.

4. The method of claim 3, wherein, According to the behavior attenuation coefficients of each target behavior in multiple target behaviors, the behavior type weight of each target behavior, the behavior times of each target behavior of the target user in the initial period to the termination period, and the objective weight value of the target object as a label of the target user, the user label weight of the target object as a user label for the target user is determined, comprising: performing the following operation on each target behavior respectively to obtain the intermediate term of each target behavior: calculating the product of the behavior attenuation coefficient, the behavior type weight, and the behavior times of the target user in the initial period to the termination period of the current target behavior as the intermediate term of the current target behavior; accumulating the intermediate terms of each target behavior to obtain an accumulated value; multiplying the accumulated value by the objective weight value of the target object as a label of the target user to obtain the user label weight of the target object as a user label for the target user.

5. The method of claim 1, wherein, The behavior times of the target behavior of the target object in an initial period and in a termination period are obtained, including: Logs of a plurality of statistical platforms are called; Matching statistics are performed on each behavior record in the logs to obtain the behavior times of the target behavior of the target object in the initial period and in the termination period.

6. The method of claim 1, wherein, After determining the user tag weight of the target object as a user tag for tagging a user according to the behavior attenuation coefficients of a plurality of target behaviors of the target object, the method further includes: Tagging the user according to the determined user tag weight; Forming a user portrait according to the tagging result.

7. The method according to any one of claims 1 to 6, characterized in that, The target object is a chronic disease.

8. A method for determining user tag weights, characterized in that, The method includes: Obtaining the behavior times of the target behavior of the target object in an initial period and in a termination period on an e-commerce platform; Determining the behavior attenuation coefficient of the target behavior according to the behavior times in the initial period and in the termination period; Determining the user tag weight of the target object as a user tag for tagging a user according to the behavior attenuation coefficients of a plurality of target behaviors of the target object; The user tag weight is calculated on the basis of a behavior type weight, and the behavior type weight includes a weight of a behavior type for a medical e-commerce and a weight of a behavior type for a consultation platform; the weight of the behavior type for the medical e-commerce is obtained by comparing the behavior and payment times of the medical e-commerce on the basis of a fitting result of the behavior times guiding the payment times of the medical e-commerce; and the weight of the behavior type for the consultation platform is obtained by comparing the behavior times and payment times of the consultation platform on the basis of a fitting result of the behavior times guiding the payment times of the consultation platform.

9. The method of claim 8, wherein, The target object is a chronic disease.

10. A device for determining user tag weights, characterized in that, The method includes: An obtaining module is configured to obtain the behavior times of the target behavior of the target object in an initial period and in a termination period; A first determining module is configured to determine the behavior attenuation coefficient of the target behavior according to the behavior times in the initial period and in the termination period; A second determining module is configured to determine the user tag weight of the target object as a user tag for tagging a user according to the behavior attenuation coefficients of a plurality of target behaviors; The user tag weight is calculated on the basis of a behavior type weight, and the behavior type weight includes a weight of a behavior type for a medical e-commerce and a weight of a behavior type for a consultation platform; the weight of the behavior type for the medical e-commerce is obtained by comparing the behavior and payment times of the medical e-commerce on the basis of a fitting result of the behavior times guiding the payment times of the medical e-commerce; and the weight of the behavior type for the consultation platform is obtained by comparing the behavior times and payment times of the consultation platform on the basis of a fitting result of the behavior times guiding the payment times of the consultation platform.

11. An electronic device comprising a processor and a memory for storing processor-executable instructions, the electronic device characterized by: The processor executes the instructions to implement the steps of the method in any one of claims 1 to 7.

12. A computer readable storage medium having stored thereon computer programs / instructions, characterized in that, The computer program / instructions are executed by the processor to implement the steps of the method in any one of claims 1 to 7.

Images