Information push method, device, terminal equipment and storage medium

Abstract

The embodiment of the present invention discloses an information push method, apparatus, terminal device and storage medium, wherein the method comprises: in response to a visit event of a user terminal, determining the matching degree between the user terminal and the push object; judging whether the user terminal meets the hit state according to the matching degree and the latest determined circle selection threshold; wherein the circle selection threshold is adjusted in real time based on the expected value and the actual value of the preset push evaluation dimension in the preset time period corresponding to the visit event; if so, the push object is sent to the user terminal so that the user terminal displays the push object. By adjusting the circle selection threshold in real time according to the actual value and the expected value of the preset push evaluation dimension in the preset time period corresponding to the visit event of the user terminal, it is possible to realize real-time circle selection of online people and real-time adjustment of the object exposure speed, thereby ensuring that the indicator quantity required by the business is met.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of computer technology, and in particular to an information push method, apparatus, terminal equipment and storage medium. Background Art

[0002] In the e-commerce scenario, tagging refers to a method of selecting people who are strongly related to the recommended object and exposing them to information push of the recommended object, where the recommended object includes, for example, activities, advertisements, and products. By tagging, the click-through rate and order volume of the recommended object can be increased.

[0003] In the prior art, the method of tagging people mainly involves selecting a group of people based on the tagging command input by the user; and exposing the recommended objects to the selected group of people.

[0004] In the process of implementing the present invention, the inventors found that there are at least the following technical problems in the prior art: since there is a certain degree of randomness in whether the people selected by the offline circle visit the e-commerce application, there is a situation where the indicators required by the business (such as exposure, visits, etc.) are not met. Summary of the invention

[0005] The embodiments of the present invention provide an information push method, apparatus, terminal device and storage medium, which can select online people in real time to ensure that the indicators required by the business are met.

[0006] In a first aspect, an embodiment of the present invention provides an information push method, including:

[0007] In response to a visit event of a user terminal, determining a matching degree between the user terminal and a push object;

[0008] According to the matching degree and the latest determined circle selection threshold, it is judged whether the user terminal meets the hit status; wherein the circle selection threshold is adjusted in real time based on the expected value and actual value of the preset push evaluation dimension within the preset time period corresponding to the visit event;

[0009] If so, the push object is sent to the user terminal so that the user terminal displays the push object.

[0010] In a second aspect, an embodiment of the present invention provides an information push device, including:

[0011] A matching degree determination module, for determining the matching degree between the user terminal and the push object in response to the visit event of the user terminal;

[0012] A judgment module, used to judge whether the user terminal meets the hit status according to the matching degree and the latest determined circle selection threshold; wherein the circle selection threshold is adjusted in real time based on the expected value and actual value of the preset push evaluation dimension within the preset time period corresponding to the visit event;

[0013] The push module sends the push object to the user terminal if the user terminal meets the hit status, so that the user terminal displays the push object.

[0014] In a third aspect, an embodiment of the present invention provides a terminal device, including:

[0015] one or more processors;

[0016] A memory for storing one or more programs;

[0017] When the one or more programs are executed by the one or more processors, the one or more processors implement the information pushing method as described in any embodiment of the present invention.

[0018] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the information push method as described in any embodiment of the present invention.

[0019] An information push method, apparatus, terminal device and storage medium provided by the embodiment of the present invention, in response to a visit event of a user terminal, determines the matching degree between the user terminal and the push object; judges whether the user terminal meets the hit status according to the matching degree and the latest determined circle selection threshold; wherein the circle selection threshold is adjusted in real time based on the expected value and the actual value of the preset push evaluation dimension in the preset time period corresponding to the visit event; if so, the push object is sent to the user terminal so that the user terminal displays the push object. By adjusting the circle selection threshold in real time according to the actual value and the expected value of the preset push evaluation dimension in the preset time period corresponding to the visit event of the user terminal, it is possible to realize real-time circle selection of online people and real-time adjustment of the object exposure speed, thereby ensuring that the indicator quantity required by the business is met. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1A flow chart of an information push method provided by Embodiment 1 of the present invention is shown;

[0022] Figure 2 A flow chart of an information push method provided by Embodiment 3 of the present invention is shown;

[0023] Figure 3 A schematic diagram of the structure of an information push device provided in Embodiment 4 of the present invention is shown;

[0024] Figure 4 A schematic diagram of the hardware structure of a terminal device provided in Embodiment 5 of the present invention is shown. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described through implementation methods with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. In the following embodiments, each embodiment provides optional features and examples at the same time. The various features recorded in the embodiments can be combined to form multiple optional solutions. Each numbered embodiment should not be regarded as only one technical solution.

[0026] Embodiment 1

[0027] Figure 1 The flowchart of an information push method provided by the first embodiment of the present invention is shown. The information push method provided by the embodiment of the present invention can be applied to the case of information push, such as the case of exposing the push object to the user end in the e-commerce scenario. The method can be executed by an information push device, which is implemented in software and / or hardware, and is preferably configured in a terminal device, such as a server.

[0028] like Figure 1 As shown, the information push method provided in the embodiment of the present invention includes the following steps:

[0029] S110 : In response to a visit event of a user terminal, determining a matching degree between the user terminal and a push object.

[0030] In the embodiment of the present invention, the user terminal can be considered as the user version of the application software (Application, APP) installed in the terminal device; the user terminal visit event, for example, can be the user terminal start event, or the user terminal account online event, etc. Among them, the push object can be the object to be promoted, such as activities, advertisements and commodities, etc., and the number of push objects can be one or more.

[0031] The backend server of the user terminal (hereinafter referred to as the server) can monitor the visit events of each user terminal. Moreover, when the server monitors the visit event of any user terminal, it can determine the matching degree between the visiting user terminal and at least one push object.

[0032] Among them, the way in which the server determines the matching degree between the user terminal and the pushed object can be, for example, that the server obtains user characteristics of the user terminal (such as portrait tags and / or behavioral characteristics, etc.); parses the object attributes of each pushed object; determines the degree of association between the user terminal and each pushed object based on the user characteristics and each object attribute; and determines the matching degree between the user terminal and each pushed object based on the determined degree of association.

[0033] S120: judging whether the user terminal meets the hit status according to the matching degree and the latest determined circle selection threshold.

[0034] For each push object, the server can determine whether the user terminal meets the hit status of the current push object by comparing the degree of match between the user terminal and the current push object and the latest determined circle selection threshold corresponding to the current push object. For example, when the degree of match between the user terminal and the current push object is greater than or equal to the latest determined circle selection threshold corresponding to the current push object, it can be considered that the current push object can be pushed to the user terminal, that is, the user terminal meets the hit status for the current push object; when the degree of match between the user terminal and the current push object is less than the latest determined circle selection threshold corresponding to the current push object, it can be considered that there is no need to push the current push object to the user terminal, that is, the user terminal does not meet the hit status for the current push object.

[0035] In this embodiment, the circle selection threshold is adjusted in real time based on the expected value and actual value of the preset push evaluation dimension within the preset time period corresponding to the visit event.

[0036] Among them, the preset push evaluation dimension can be considered as a dimension used to evaluate the push effect, for example, it may include exposure, number of unique visitors and / or click-through rate, etc. Among them, before executing the push method disclosed in this embodiment, the operator can use the server to pre-set the push evaluation dimension according to the business requirements of the business party of the push object. For example, assuming that the advertiser requires a basketball advertisement to be exposed 5 million times a day, before pushing the advertisement, the operator can use the server to set the exposure as the push evaluation dimension of the advertisement.

[0037] Among them, although the visit events of the user end have a certain degree of randomness, the overall fluctuation characteristics remain basically the same. For example, on the whole, the number of user-end visit events in the morning is relatively small, and the number of user-end visit events in the afternoon is relatively large. Before executing the push method disclosed in this embodiment, the server may also divide the business cycle (i.e., the cycle in the business requirements, such as every day) into multiple preset time periods according to the overall fluctuation characteristics of the records, wherein each preset time period may have a more obvious visit number characteristic, for example, some preset time periods may be visit low trough periods, and some time periods may be visit peak periods. Among them, the time period length of each time period may be equal or unequal, for example, 5 minutes, 10 minutes, 30 minutes or 60 minutes, etc.

[0038] Among them, after dividing each preset time period, the server can also set corresponding expected values ​​for each preset time period according to the indicator quantity in the business requirements of each push object. For example, assuming that the advertiser requires a basketball advertisement to be exposed 5 million times a day, the indicator quantity in the business requirements of the advertisement is "exposure 5 million times / day", and the indicator quantity can be allocated to each preset time period to specify the indicator quantity of each preset time period. Generally, the expected value of each preset time period can be set to be greater than the indicator quantity of each preset time period, so as to increase the probability of completing the indicator quantity required by the business.

[0039] After dividing each preset time period, the server may also set an initial selection threshold for each preset time period. The initial selection threshold for each preset time period may be set based on experience or experimental values, and the initial selection threshold for each preset time period may be the same or different.

[0040] In the process of executing the push method disclosed in this embodiment, the server can obtain the expected value of each push object in its preset push evaluation dimension within the current preset period; it can also determine in real time the actual value of each push object in its preset push evaluation dimension within the current preset period; and it can perform iterative regulation based on the initial selection threshold according to the actual value and the expected value. Among them, the way to regulate the selection threshold is, for example, when the deviation between the actual value and the expected value is large, the selection threshold is lowered; when the deviation between the actual value and the expected value is small, the selection threshold is maintained or increased.

[0041] By adjusting the selection threshold in real time according to the actual value and expected value of the preset push evaluation dimension within the preset time period, the latest adjusted selection threshold can be used to determine whether the user meets the hit status, thereby achieving speed control of object exposure to online visiting users to ensure that the indicator quantity required by the business is met.

[0042] S130: If yes, the pushed object is sent to the user terminal, so that the user terminal displays the pushed object.

[0043] When the server determines that the visiting user terminal meets the hit status of the push object, the server can push the corresponding push object to the user terminal, so that the user terminal displays the push object, thereby realizing exposure of the push object.

[0044] By adjusting the selection threshold in real time and judging the hit status of the user terminal according to the latest adjusted selection threshold, it is possible to realize real-time selection of the online user terminal and adjust the object exposure speed in real time, thereby ensuring that the actual value of the preset evaluation dimension reaches the expected value, that is, the indicator quantity required by the business is achieved.

[0045] The information push method provided in this embodiment can be applied to e-commerce scenarios; accordingly, the user end can be an e-commerce application APP, the server can be a server of an e-commerce platform, and the push object can be commodities, activities, and advertisements pushed by the e-commerce platform, etc. The e-commerce platform server executes the information push method provided in any embodiment, and can realize the selection threshold based on real-time regulation, and perform real-time selection on the online user end, which is conducive to achieving the index quantity of the push object required by the business party.

[0046] An information push method provided by an embodiment of the present invention determines the matching degree between the user terminal and the push object in response to the visit event of the user terminal; judges whether the user terminal meets the hit state according to the matching degree and the latest determined circle selection threshold; wherein the circle selection threshold is adjusted in real time based on the expected value and the actual value of the preset push evaluation dimension in the preset time period corresponding to the visit event; if so, the push object is sent to the user terminal so that the user terminal displays the push object. By adjusting the circle selection threshold in real time according to the actual value and the expected value of the preset push evaluation dimension in the preset time period corresponding to the visit event of the user terminal, it is possible to realize real-time circle selection of online people and real-time adjustment of the object exposure speed, thereby ensuring that the indicator quantity required by the business is met.

[0047] Embodiment 2

[0048] This embodiment can be combined with each optional scheme in the information push method provided in the above embodiment. The information push method provided in this embodiment optimizes the control step of the circle selection threshold. Based on the PID algorithm, the circle selection threshold can be controlled in real time according to the expected value and the real-time actual value, so as to achieve the speed control of the object exposure for the online visiting user terminal. In addition, the step of determining the expected value is also optimized. By adjusting the initial expected value according to the difference between the actual value accumulated in each time period before the preset time period within the business cycle and the accumulated expected value, the accuracy of the threshold control of the current preset time period can be improved.

[0049] In some optional embodiments, the step of regulating the circle selection threshold includes: determining an expected value of a preset push evaluation dimension within a preset time period, and determining an actual value of the preset push evaluation dimension at every preset time interval; determining a current numerical deviation value based on the currently determined actual value and the expected value each time the actual value is determined; determining a threshold adjustment value based on the current numerical deviation value, the accumulated numerical deviation value within the preset time period, and the difference between the current numerical deviation value and the numerical deviation value determined last time; regulating the circle selection threshold determined last time according to the threshold adjustment value; wherein, the initial circle selection threshold of the preset time period is taken when the current time reaches the starting time of the preset time period.

[0050] Among them, the server can obtain the preset initial selection threshold and the expected value of the preset push evaluation dimension when the current time reaches the starting time of the preset time period; and can obtain the actual value of the preset push evaluation dimension at every preset time interval to achieve real-time acquisition of the actual value.

[0051] Among them, the selection threshold can be iteratively regulated from the initial selection threshold based on the proportion (P), integral (I), and differential (D) algorithms according to the expected value and the real-time actual value, that is, it is regulated based on the selection threshold determined last time. Indicatively, the selection threshold can be regulated by the following formula:

[0052]

[0053] Among them, for each regulation process, Thr i is the selection threshold after the current adjustment, Thr i-1 is the selection threshold determined previously; real index is the actual value currently determined, exp index is the expected value for the preset period; e i is the current numerical deviation value, which can represent the actual value real index and expected value exp index Deviation; is the accumulated numerical deviation value within the preset period; (e i -e i-1) is the difference between the current numerical deviation value and the numerical deviation value determined last time; P, I and D are the proportional coefficient, integral coefficient and differential coefficient respectively, among which P can be used to quickly respond to the current deviation, which may cause the circle selection threshold to oscillate; I can be used to eliminate the accumulated deviation to achieve control accuracy and can slow down the above oscillation; D can be used to predict future deviations and control them to improve the stability of the circle selection threshold and achieve rapid adjustment of the circle selection threshold; K is the connection factor that converts the deviation determined according to the PID algorithm into the threshold adjustment value.

[0054] Among them, K can be set according to different preset push evaluation dimensions and the range of the circled threshold. Assuming that when the range of the circled threshold is [0,1] and the preset push evaluation dimension is exposure, since the deviation of exposure is large, in order to convert it into a threshold adjustment value of smaller magnitude, the value of the connection factor K needs to be very small; for example, when the preset push evaluation dimension is click-through rate, since the deviation of click-through rate is small, in order to convert it into a threshold adjustment value of slightly larger magnitude, the value of K can be slightly larger.

[0055] In these optional implementations, by adjusting the selection threshold in real time based on the PID algorithm according to the expected value and the real-time actual value, it is possible to lower the selection threshold when the actual value deviates greatly from the expected value; and maintain or increase the selection threshold when the actual value deviates from the expected value, thereby achieving speed control of object exposure for online visiting users.

[0056] In some optional embodiments, the step of determining the expected value includes: obtaining an initial expected value of a preset push evaluation dimension within a preset time period; wherein the initial expected value is determined based on historical visit data; and adjusting the initial expected value based on the difference between the actual values ​​accumulated in each time period before the preset time period within the business cycle to which the preset time period belongs and the accumulated expected value.

[0057] Before executing the push method disclosed in this embodiment, the server can divide the business cycle (i.e., the cycle in the business requirements, such as every day) into multiple preset time periods according to the overall fluctuation characteristics of the recorded visit data; and can determine the preset push evaluation dimension for the business requirements of each push object, and set the corresponding initial expected value for each preset time period. Accordingly, when the current time reaches the starting time of the preset time period, the server can obtain the initial expected value of the preset push evaluation dimension within the preset current preset time period.

[0058] Since there may be a difference between the accumulated actual data and the accumulated expected data in the period before the current preset period in the business cycle, the referenced expected value can be adjusted according to the above difference during the threshold control process of the current preset period. For example, if the accumulated actual values ​​of the previous periods are greater than the accumulated expected values, the initial expected value of the current preset period can be adaptively adjusted downward; if the accumulated actual values ​​of the previous periods are less than the accumulated expected values, the initial expected value of the current preset period can be adaptively adjusted upward.

[0059] In these optional implementations, by adjusting the initial expected value based on the difference between the actual values ​​accumulated in each time period before the preset time period within the business cycle and the accumulated expected value, the accuracy of the threshold control for the current preset time period can be improved to ensure that the indicator quantity required by the business is met.

[0060] Based on the above embodiments, the embodiment of the present invention optimizes the control step of the circle selection threshold. Based on the PID algorithm, the circle selection threshold can be controlled in real time according to the expected value and the real-time actual value, so as to achieve the speed control of the object exposure of the online visiting user terminal. In addition, the step of determining the expected value is also optimized. By adjusting the initial expected value according to the difference between the actual value accumulated in each time period before the preset time period within the business cycle and the accumulated expected value, the accuracy of the threshold control of the current preset time period can be improved. In addition, the information push method proposed in the embodiment of the present invention and the above embodiments belong to the same inventive concept. The technical details not described in detail in this embodiment can be referred to the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0061] Embodiment 3

[0062] This embodiment can be combined with the various optional solutions in the information push method provided in the above embodiment. The information push method provided in this embodiment optimizes the step of determining the matching degree between the user terminal and the push object, and can comprehensively determine the matching degree by preference measurement and seed measurement, which is conducive to automatic selection of online people, and the selection accuracy is also improved.

[0063] Figure 2 FIG. 1 shows a flow chart of an information push method provided by Embodiment 3 of the present invention. Figure 2 , the information push method comprises the following steps:

[0064] S210: In response to a visit event of a user terminal, determine a preference metric of the user terminal for a push object, and determine a seed metric between the user terminal and a seed population.

[0065] In this embodiment, the seed population can be considered as a collection of users that are strongly related to the push object. Among them, the preference metric can be considered as the preference degree of the user terminal for the push object, which can be determined by the user terminal characteristics and the push object attributes; the seed metric can be considered as the similarity between the online user terminal and the user terminals in the seed population, which can be determined by the online user terminal characteristics and the characteristics of the user terminals in various sub-populations.

[0066] When the server starts to push the push information, it has not yet determined the user terminals that are strongly related to the push object, that is, the number of seed population is zero. Therefore, in the early stage of push, the server can select the user terminals that meet the hit status to push the push object according to the preference measurement of the visiting user terminals for the push object and the latest determined circle selection threshold.

[0067] When these pushed user terminals give feedback to the pushed object, for example, when they visit the details page of the pushed object, feedback data can be generated. If the user terminal generates feedback data, the user terminal can be regarded as a strongly related user terminal that is interested in the pushed object. Furthermore, the server can select strongly related user terminals from the user terminals that have been judged to be in a hit state based on real-time feedback data, and use them as seed populations. It can be considered that the seed population is determined based on the feedback data of the user terminals that have been judged to be in a hit state, and the number of seed populations can continue to grow with the push process. In theory, the larger the seed population, the better the effect of seed measurement, but the increase in the seed population will increase the calculation cost of population measurement, and in practice, when the population increases to a certain extent, the growth rate of the effect of population measurement slows down, so in practice, the seed population only needs to grow to a preset equilibrium value.

[0068] When the number of seed populations is not zero, the server may determine, in response to the visit event of the client, both the preference metric of the client for the push object and the seed metric between the client and the seed population.

[0069] S220: Determine a fusion metric of the user terminal according to the preference metric and the seed metric, and use the fusion metric as a matching degree between the user terminal and the push object.

[0070] Since the real-time seed metric has a better selection effect than the preference metric, but the seed population used in the seed metric depends on the preference metric, the complementarity between the metrics can be achieved by integrating the two metrics. By using real-time feedback data to update the seed population, determining the seed metric based on the real-time seed population, and combining the preference metric and the seed metric to determine the matching degree, the accuracy of information push can be improved.

[0071] In some optional implementations, a fusion metric of the user terminal is determined based on the preference metric and the seed metric, including: determining the fusion metric of the user terminal based on the preference metric, the seed metric, the fusion factor of the preference metric and the fusion factor of the seed metric; wherein the fusion factor of the preference metric and the fusion factor of the seed metric are adjusted based on actual values ​​and expected values ​​within a historical business cycle.

[0072] Indicatively, the fusion metric can be determined by the following formula:

[0073] S mul (obj, seeds, a, b) = a*S pref (obj)+b*S sml (seeds);

[0074] Among them, S mul is the fusion metric at the user end, S pref (obj) is the preference measure, S sml (seeds) is the seed metric; a and b are the fusion factors of the preference metric and the seed metric, respectively.

[0075] Since the number of seed populations is zero at the initial stage of push, a and b can be set to 1 and 0 respectively at the initial stage. As the number of seed populations increases, their metric weights can gradually increase, so a can gradually decrease to 0, b can gradually increase to 1, and the sum of a and b is 1.

[0076] The server may adjust a and b based on the actual values ​​and expected values ​​in the historical business cycle. For example, if after adjusting a and b, the actual values ​​and expected values ​​in the current business cycle show a better selection effect than the actual values ​​and expected data in the historical business cycle, then the next adjustment of a and b may be made based on the adjustment trend of a and b this time; if after adjusting a and b, the actual values ​​and expected values ​​in the current business cycle show a worse selection effect than the actual values ​​and expected data in the historical business cycle, then the adjustment trend of a and b this time may be weakened, and the next adjustment of a and b may be made based on the weakened adjustment trend.

[0077] In these optional implementations, the push effect can be optimized by adjusting the fusion factor of the preference metric and the seed metric according to the circle selection effect, and fusing the preference metric and the seed metric based on the fusion factor.

[0078] S230: Determine whether the user terminal meets the hit status based on the matching degree and the latest determined circle selection threshold.

[0079] Among them, the circle selection threshold is adjusted in real time based on the expected value and actual value of the preset push evaluation dimension within the preset time period corresponding to the visit event.

[0080] S240: If yes, the pushed object is sent to the user terminal, so that the user terminal displays the pushed object.

[0081] In the traditional method of tagging people, the tagging instruction input by the user usually carries a manually selected user portrait label, so that the server can select the group of people offline based on the user portrait label. Since the determination process of the user portrait label is limited by the personal ability and experience of the operator, and the low recall rate of the segmentation label in the current user portrait technology, it is easy to cause the accuracy of the selected group to be low. The information push method provided in this embodiment can combine the preference metric and the seed metric to determine the above-mentioned matching degree, which is conducive to the automatic selection of online people, and the selection accuracy is also improved.

[0082] In some optional implementations, determining the user terminal's preference metric for the push object includes: parsing the attributes of the push object; determining the user terminal's preference value for the attribute in a preset time dimension based on the user terminal's historical behavior data; and determining the user terminal's preference metric for the push object based on each preference value.

[0083] The attributes of different push objects may be different. For example, when the push object is a product, the attributes may include the product category, brand, etc.; when the push object is an activity, the attributes may include the activity concept (such as preferential policies), etc.

[0084] In some further implementations, before parsing the attributes of the push object, the information of the push object may be collected based on a preset interface. The preset interface may include, but is not limited to, a message queue (MQ) or an application programming interface (API). The information of the push object may include, but is not limited to, text, pictures and / or videos.

[0085] The push object can be parsed by a parsing method corresponding to the information type of the push object to obtain the attributes of the push object. For example, when the information is text, the attributes of the push object can be parsed by a text recognition method; when the information is a picture, the attributes of the push object can be parsed by an image recognition method. The specific parsing method is not limited here.

[0086] The historical behavior data of the user terminal may be, for example, the search data, access data, favorite data and / or order data of the user terminal. The preset duration dimensions may be, for example, long-term (for example, one week), short-term (for example, three days) and real-time (for example, one hour) dimensions. The preference value of the attribute of the user terminal in the preset duration dimension is determined based on the historical behavior data of the user terminal. For each push object, the server may determine the preference value of the user terminal in each preset duration dimension for each attribute of the current push object based on the historical behavior data of the user terminal through a regression model.

[0087] Indicatively, the following Sigmod formula can be used to determine the user's preference metric for the push object based on each preference value:

[0088]

[0089] Among them, S pref (obj) is the preference measure; x i They represent the user's preference values ​​for each attribute of the current push object in each preset time dimension, x i The value range of can be (0,1); i The coefficient of S pref The value range of (obj) is (0,1).

[0090] In some optional implementations, determining the seed metrics of the user terminal and the seed population includes: determining the seed metrics of the user terminal and the seed population based on the feature values ​​of the user terminal and each seed user terminal in the seed population in each preset feature dimension.

[0091] The preset feature dimensions may include but are not limited to user portrait features (such as gender, age, etc.) and behavioral features (such as behavioral preference features, etc.). The seed metric may be determined by synthesizing the similarity between the values ​​of the preset feature dimensions of the visiting user terminal and the values ​​of the preset feature dimensions represented by the seed population.

[0092] Indicatively, the seed metrics of the user end and the seed population can be determined by the Information Value (IV) formula:

[0093]

[0094] Among them, S sml (seeds,user) is the seed metric; K is the total number of dimensions of the preset feature dimension; f i,j is the value of the jth preset feature dimension of the i-th visiting user terminal; where p j is the characteristic proportion of the j-th dimension element in the seed population, and can be obtained through Determine, where S is the data set of the seed population, f′ i,j is the value of the jth preset feature dimension of the i-th seed user terminal; where q j is the feature proportion of the j-th dimension element in the entire user population, and can be obtained through Determine, where U is the data set of all active people, f″ i,j is the value of the jth preset feature dimension of the i-th active user terminal.

[0095] Based on the above embodiment, the calculation logic required to determine whether the user terminal meets the hit status can be schematically disassembled by the following formula:

[0096]

[0097] Among them, S mul is the fusion metric of the user end, Thr i is the most recently determined circle selection threshold; f hit Whether the user terminal meets the hit status and can mul <Thr i When it does not meet the hit status, in S mul ≥Thr i The other parameters can be referred to the above embodiment and will not be described in detail here.

[0098] Based on the above embodiment, the embodiment of the present invention optimizes the step of determining the matching degree between the user terminal and the push object, and can determine the above matching degree by combining the preference metric and the seed metric, which is conducive to the automatic selection of online people, and the selection accuracy is also improved. In addition, the information push method proposed in the embodiment of the present invention and the above embodiment belongs to the same inventive concept, and the technical details not described in detail in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0099] Embodiment 4

[0100] Figure 3 The schematic diagram of the structure of an information push device provided by the fourth embodiment of the present invention is shown. The information push device provided by the embodiment of the present invention can be applied to the situation of information push, such as the situation of exposing the push object to the user end in the e-commerce scenario. The information push method provided by the above embodiment can be implemented by the information push device provided by the present invention.

[0101] like Figure 3 As shown, the information push device in the embodiment of the present invention includes:

[0102] A matching degree determination module 310, for determining a matching degree between the user terminal and the push object in response to a visit event of the user terminal;

[0103] The judgment module 320 is used to judge whether the user terminal meets the hit status according to the matching degree and the latest determined circle selection threshold; wherein the circle selection threshold is adjusted in real time based on the expected value and actual value of the preset push evaluation dimension within the preset time period corresponding to the visit event;

[0104] The push module 330 sends the push object to the user terminal if the user terminal meets the hit status, so that the user terminal displays the push object.

[0105] In some optional implementations, the information push device further includes:

[0106] The threshold control module is used to determine the expected value of the preset push evaluation dimension within a preset time period, and to determine the actual value of the preset push evaluation dimension at every preset time interval; each time the actual value is determined, the current value deviation value is determined according to the currently determined actual value and the expected value; the threshold adjustment value is determined according to the current value deviation value, the accumulated value deviation value within the preset time period, and the difference between the current value deviation value and the value deviation value determined last time; according to the threshold adjustment value, the previously determined circle selection threshold is regulated; wherein, the initial circle selection threshold of the preset time period is obtained when the current time reaches the starting time of the preset time period.

[0107] In some optional implementations, the information push device further includes:

[0108] The expected value determination module is used to obtain the initial expected value of the preset push evaluation dimension within a preset time period; wherein the initial expected value is determined based on historical visit data; the initial expected value is adjusted based on the difference between the actual values ​​accumulated in each time period before the preset time period within the business cycle to which the preset time period belongs and the accumulated expected value.

[0109] In some optional implementations, the matching degree determination module is specifically used to:

[0110] Determine the user's preference metric for the push object, and determine the seed metric between the user and the seed population; wherein the seed population is determined based on the feedback data of the user that has been determined to be in a hit state;

[0111] According to the preference metric and the seed metric, the fusion metric of the user end is determined, and the fusion metric is used as the matching degree between the user end and the push object.

[0112] In some optional implementations, the matching degree determination module includes:

[0113] The preference metric determination unit is used to analyze the attributes of the push object; determine the user's preference value for the attribute in a preset time dimension based on the user's historical behavior data; and determine the user's preference metric for the push object based on each preference value.

[0114] In some optional implementations, the matching degree determination module further includes:

[0115] The collection interface unit is used to collect information of the push object based on a preset interface before parsing the attributes of the push object.

[0116] In some optional implementations, the matching degree determination module includes:

[0117] The seed metric determination unit is used to determine the seed metrics of the user terminal and the seed population according to the feature values ​​of the user terminal and each seed user terminal in the seed population in each preset feature dimension.

[0118] In some optional implementations, the matching degree determination module includes:

[0119] A fusion metric determination unit, configured to determine a fusion metric of a user terminal according to the preference metric, the seed metric, the fusion factor of the preference metric and the fusion factor of the seed metric;

[0120] Among them, the fusion factor of the preference metric and the fusion factor of the seed metric are adjusted based on the actual values ​​and expected values ​​in the historical business cycle.

[0121] The information push device provided in the embodiment of the present invention belongs to the same inventive concept as the information push method applied to the server provided in the above embodiment. The technical details not fully described in the embodiment of the present invention can be referred to the above embodiment, and the embodiment of the present invention has the same beneficial effects as the above embodiment.

[0122] Embodiment 5

[0123] Figure 4 The hardware structure diagram of a terminal device provided by Embodiment 5 of the present invention is shown. The terminal device in the embodiment of the present invention may include but is not limited to mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0124] like Figure 4As shown, the terminal device 400 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 401, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 402 or a program loaded from a storage device 408 to a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the terminal device 400 are also stored. The processing device 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0125] Typically, the following devices may be connected to the I / O interface 405: an input device 406 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 407 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 408 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 409. The communication device 409 may allow the terminal device 400 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 4 The terminal device 400 is shown to have various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0126] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 409, or installed from the storage device 408, or installed from the ROM 402. When the computer program is executed by the processing device 401, the above-mentioned functions defined in the information push method provided in the embodiment of the present invention are executed.

[0127] The terminal provided in the embodiment of the present invention and the information push method provided in the above embodiment belong to the same inventive concept. The technical details not fully described in the embodiment of the present invention can be referred to the above embodiment, and the embodiment of the present invention has the same beneficial effects as the above embodiment.

[0128] Embodiment 6

[0129] An embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the information push method provided in the above embodiment is implemented.

[0130] It should be noted that the computer-readable storage medium in the embodiment of the present invention may be a computer-readable signal medium, or a computer-readable storage medium, or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM) or a flash memory (FLASH), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0131] In an embodiment of the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device, or device. In an embodiment of the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in combination with an instruction execution system, device, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0132] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0133] The computer-readable storage medium may be included in the terminal device, or may exist independently without being installed in the terminal device.

[0134] The terminal device stores and carries one or more programs. When the one or more programs are executed by the terminal device, the terminal device:

[0135] In response to a visit event of the user terminal, determine the matching degree between the user terminal and the pushed object; judge whether the user terminal meets the hit status based on the matching degree and the latest determined circle selection threshold; wherein the circle selection threshold is based on the expected value and actual value of the preset push evaluation dimension within the preset time period corresponding to the visit event, and is adjusted in real time; if so, send the pushed object to the user terminal so that the user terminal displays the pushed object.

[0136] Computer program code for performing the operations of the present invention may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0137] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the function pages marked in the square box can occur in a different order from the order marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0138] The units involved in the embodiments of the present invention may be implemented by software or hardware, wherein the name of a unit does not limit the unit itself in some cases.

[0139] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0140] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. An information push method, characterized in that: include: In response to a visit event of a user terminal, determining a matching degree between the user terminal and a push object; According to the matching degree and the latest determined circle selection threshold, it is judged whether the user terminal meets the hit status; wherein the circle selection threshold is adjusted in real time based on the expected value and actual value of the preset push evaluation dimension within the preset time period corresponding to the visit event; If yes, sending the pushed object to the user terminal so that the user terminal displays the pushed object; The step of determining the expected value includes: Obtaining an initial expected value of a preset push evaluation dimension within the preset time period; wherein the initial expected value is determined based on historical visit data; The initial expected value is adjusted according to the difference between the actual values ​​accumulated in each time period before the preset time period and the accumulated expected value within the business cycle to which the preset time period belongs.

2. The method according to claim 1, characterized in that: The step of adjusting and controlling the selection threshold comprises: Determining an expected value of a preset push evaluation dimension within the preset time period, and determining an actual value of the preset push evaluation dimension at every preset time interval; Each time an actual value is determined, the current value deviation value is determined according to the actual value currently determined and the expected value; Determine a threshold adjustment value according to a current numerical deviation value, a numerical deviation value accumulated within the preset time period, and a difference between the current numerical deviation value and a numerical deviation value determined last time; According to the threshold adjustment value, the previously determined circle selection threshold is adjusted; wherein the initial circle selection threshold of the preset time period is obtained when the current time reaches the starting time of the preset time period.

3. The method according to claim 1, characterized in that The determining the matching degree between the user terminal and the push object includes: Determine the preference metric of the user terminal for the push object, and determine the seed metric of the user terminal and the seed population; wherein the seed population is determined based on feedback data of the user terminal that has been determined to be in a hit state; According to the preference metric and the seed metric, a fusion metric of the user terminal is determined, and the fusion metric is used as a matching degree between the user terminal and the push object.

4. The method according to claim 3, characterized in that The determining the preference metric of the user terminal for the push object includes: Parse the attributes of the pushed object; Determining, based on the historical behavior data of the user terminal, a preference value of the user terminal for the attribute in a preset time dimension; According to each of the preference values, a preference metric of the user terminal for the push object is determined.

5. The method according to claim 4, characterized in that Before parsing the attributes of the pushed object, the method further includes: Based on the preset interface, the information of the push object is collected.

6. The method according to claim 3, characterized in that: The determining of the seed metrics of the user terminal and the seed population includes: The seed metrics of the user terminal and the seed population are determined according to the feature values ​​of the user terminal and each seed user terminal in the seed population in each preset feature dimension.

7. The method according to claim 3, characterized in that The determining the fusion metric of the user terminal according to the preference metric and the seed metric includes: Determining a fusion metric of the user terminal according to the preference metric, the seed metric, a fusion factor of the preference metric, and a fusion factor of the seed metric; The fusion factor of the preference metric and the fusion factor of the seed metric are adjusted based on actual values ​​and expected values ​​in a historical business cycle.

8. An information push device, characterized in that: include: A matching degree determination module, for determining the matching degree between the user terminal and the push object in response to the visit event of the user terminal; A judgment module, used to judge whether the user terminal meets the hit status according to the matching degree and the latest determined circle selection threshold; wherein the circle selection threshold is adjusted in real time based on the expected value and actual value of the preset push evaluation dimension within the preset time period corresponding to the visit event; A push module, if the user terminal meets the hit status, sends the push object to the user terminal, so that the user terminal displays the push object; The expected value determination module is used to obtain the initial expected value of the preset push evaluation dimension within the preset time period; wherein the initial expected value is determined based on historical visit data; and the initial expected value is adjusted based on the difference between the actual values ​​accumulated in each time period before the preset time period within the business cycle to which the preset time period belongs and the accumulated expected value.

9. A terminal device, characterized in that: The terminal comprises: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the information push method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the information push method as described in any one of claims 1 to 7 is implemented.

Images