Push method, device, equipment and medium based on multi-level regular matching

Abstract

The present invention relates to the field of artificial intelligence and big data technology. The present invention discloses a push method, device, equipment and medium based on multi-level regular matching. The method includes: receiving a push instruction containing user information of a user to be recommended, inputting the user information into a preset multi-level regular matching model containing multiple layers of permission matching layers; in each permission matching layer, performing regular matching on the user information and each permission regular expression to obtain the matching results between the user information and each piece of information to be pushed; obtaining the matching results between the user information and each piece of information to be pushed; determining the user group to which the user to be recommended belongs based on the user information, and obtaining the recommendation values of other users in the user group for each piece of information to be pushed; determining the item to be recommended from all the information to be pushed, and pushing the item to be recommended to the user to be recommended. The present invention improves the push efficiency and push accuracy of information pushed to the user to be recommended.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent recommendation of big data, and in particular to a push method, device, equipment and medium based on multi-level regular matching. Background Art

[0002] With the development of big data technology, scenarios where information is pushed to users often occur, and when information is pushed to users, the information is pushed in a targeted manner. For example, in the financial field, some financial products require restrictions on customer qualifications, identities, whether they fall within the regulatory scope, etc., and cannot be sold to these customers. Therefore, the push information corresponding to these financial products cannot be sold even if it is pushed to restricted customers; for example, when a user visits a website (such as an employee logging into the company website), some push information will also determine whether it needs to be pushed based on factors such as user level, group, and whether it has a correlation with the push information.

[0003] The recommendation algorithms existing in the prior art usually push different push information to a group with full permissions. The shortcomings of this solution are: since the above solution requires detailed calculation of whether each user meets the restricted requirements of each push information for each push information, from the perspective of big data calculation, since the restricted factors of different push information are different and varied, and users (such as new users logging in) will also be updated at any time; therefore, the data calculation amount and machine load of this solution are too large, the calculation process is complicated, and the push efficiency is low; and if targeted recommendations are not made through the above method, it will cause disorder in information push (for example, the user who receives the push information does not have the permission to receive the information, or does not have the permission to purchase the product corresponding to the push information), and seriously reduce the user experience. Summary of the Invention

[0004] Based on this, it is necessary to provide a push method, device, equipment and medium based on multi-level regular matching to address the above technical problems, so as to solve the problems of low push efficiency or disordered information push in the existing technology.

[0005] A push method based on multi-level regular expression matching, characterized by comprising:

[0006] Receive a push instruction containing user information of a user to be recommended, and input the user information into a preset multi-level regular matching model comprising multiple permission matching layers; wherein each permission matching layer corresponds to a piece of information to be pushed containing at least one permission regular expression;

[0007] In each permission matching layer of the preset multi-level regular matching model, regular matching is performed on the user information and each permission regular expression to obtain a matching result between the user information and the information to be pushed corresponding to each permission matching layer;

[0008] Determine the user group to which the user to be recommended belongs based on the user information, and obtain the recommendation values of other users in the user group for each item of information to be pushed;

[0009] According to the matching result and the recommendation value, an item to be recommended is determined from all the information to be pushed, and the item to be recommended is pushed to the user to be recommended.

[0010] A push device based on multi-level regular expression matching, characterized by comprising:

[0011] a receiving module configured to receive a push instruction containing user information of a user to be recommended, and input the user information into a preset multi-level regular matching model comprising multiple permission matching layers; wherein each permission matching layer corresponds to a piece of information to be pushed containing at least one permission regular expression;

[0012] A result acquisition module is used to perform regular matching on the user information and each permission regular expression in each permission matching layer of the preset multi-level regular matching model, and obtain a matching result between the user information and the information to be pushed corresponding to each permission matching layer;

[0013] A recommendation value acquisition module is used to determine the user group to which the recommended user belongs based on the user information, and obtain the recommendation values of other users in the user group for each item of information to be pushed;

[0014] The push module is used to determine the item to be recommended from all the information to be pushed according to the matching result and the recommendation value, and push the item to be recommended to the user to be recommended.

[0015] A computer device includes a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the above-mentioned push method based on multi-level regular matching is implemented.

[0016] One or more readable storage media storing computer-readable instructions, wherein when the computer-readable instructions are executed by one or more processors, the one or more processors execute the above-mentioned push method based on multi-level regular expression matching.

[0017] The above-mentioned push method, device, equipment and medium based on multi-level regular matching receive a push instruction containing the user information of the user to be recommended, input the user information into a preset multi-level regular matching model containing a multi-level permission configuration layer, and perform regular matching on the user information and each permission regular expression in each permission matching layer of the preset multi-level regular matching model to obtain the matching result of the user information and the information to be pushed corresponding to each permission matching layer; determine the user group to which the user to be recommended belongs based on the user information, and obtain the recommendation value of other users in the user group for each information to be recommended; determine the recommendation item from all the information to be recommended based on the matching result and the recommendation value, and push the recommendation to the user to be recommended. The present invention can directly obtain the matching results between a user information and all the information to be pushed by presetting a multi-level regular matching model. The calculation process is simple and the calculation efficiency is high, so that the recommended information pushed to the recommended user is targeted (the recommended users to whom a piece of information to be pushed is pushed are all matched with the information to be pushed), and the push accuracy is high; at the same time, after determining the matching result, according to the matching result and the recommendation value of the user group corresponding to the user to be recommended, a joint decision is made to make a targeted recommendation, which further improves the push accuracy and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0019] Figure 1 This is a schematic diagram of an application environment of a push method based on multi-level regular expression matching in one embodiment of the present invention;

[0020] Figure 2 This is a flow chart of a push method based on multi-level regular expression matching in one embodiment of the present invention;

[0021] Figure 3 1 is a flow chart of step S1 of a push method based on multi-level regular expression matching in one embodiment of the present invention;

[0022] Figure 4 1 is a flow chart of step S2 of a push method based on multi-level regular expression matching in one embodiment of the present invention;

[0023] Figure 5 2 is a flow chart of step S2 of a push method based on multi-level regular expression matching in another embodiment of the present invention;

[0024] Figure 61 is a flow chart of step S3 of the push method based on multi-level regular expression matching in one embodiment of the present invention;

[0025] Figure 7 4 is a flow chart of step S4 of the push method based on multi-level regular expression matching in one embodiment of the present invention;

[0026] Figure 8 1 is a structural diagram of a push device based on multi-level regular matching in one embodiment of the present invention;

[0027] Figure 9 FIG. 1 is a schematic diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0029] The push method based on multi-level regular matching provided in this embodiment can be applied in Figure 1 In an application environment, a client communicates with a server. Clients include, but are not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server can be implemented as a standalone server or a server cluster consisting of multiple servers.

[0030] In one embodiment, if Figure 2 As shown, a push method based on multi-level regular matching is provided, which is applied in Figure 1 The server in the example is used as an example, and the steps are as follows:

[0031] Step S1: Receive a push instruction containing user information of a user to be recommended, and input the user information into a preset multi-level regular matching model containing multiple permission matching layers; wherein each permission matching layer corresponds to a piece of information to be pushed containing at least one permission regular expression.

[0032] Among them, the preset multi-level regular matching model includes multiple layers of permission matching layers, wherein each permission matching layer can match the input user information through the permission regular expressions corresponding to the layer (one or more are acceptable), and one regular expression corresponds to one regular matching result item (the regular matching result item represents whether the input user information matches the regular expression). Based on all the regular matching result items obtained in the permission matching layer, the matching result of the user information matched by the permission matching layer can be obtained.

[0033] In one embodiment, if Figure 3 As shown, before step S1, that is, before receiving the recommendation instruction containing the user information of the user to be recommended and inputting the user information into the preset multi-level regular matching model containing multiple layers of authority matching layers, the following steps are also included:

[0034] Step S101: Obtain all information to be pushed, and obtain the permission information and preset priority corresponding to each of the information to be pushed; wherein, permission information refers to the text content corresponding to each information to be pushed, and is used to characterize the limited characteristics of the users to whom the information to be pushed can be pushed. The preset priority refers to the setting order of multiple permission matching layers in the multi-level regular matching model. The permission matching layers are set from first to last according to the preset priority in the multi-level regular matching model. After the user information is input into the multi-level regular matching model, the matching order of each permission matching layer is preferably determined in the order of the preset priority (that is, the earlier the preset priority, the earlier the permission matching layer will match the user information).

[0035] Step S102: extract at least one permission keyword from each item of the permission information; wherein, the permission keyword needs to be extracted from the above permission information, and the permission keyword refers to a word, field or sentence that can be used to clearly indicate the limiting characteristics of the user and can construct a permission regular expression.

[0036] Step S103: construct at least one permission regular expression corresponding to the information to be pushed based on all the permission keywords corresponding to the information to be pushed; understandably, the permission regular expression can be obtained after setting preset permission restriction rules for the permission keywords.

[0037] Step S104: constructing a permission matching layer corresponding to each item of information to be pushed based on all the permission regular expressions corresponding to each item of information to be pushed;

[0038] Step S105: Constructing a preset multi-level regular matching model based on all the permission matching layers and their corresponding preset priorities.

[0039] In this embodiment, permission keywords are first extracted, and then permission regular expressions are constructed based on the permission keywords. After establishing the permission matching layer corresponding to the information to be pushed, a preset multi-level regular matching model is constructed based on all the permission matching layers and their corresponding preset priorities. The multiple permission matching layers in the preset multi-level regular matching model of the present invention have priority settings. It is understandable that when the matching results corresponding to each permission matching layer obtained by the multi-level regular matching model are inconsistent, the matching result corresponding to the highest priority will prevail.

[0040] In one embodiment, after step S105, that is, after the preset multi-level regular matching model is constructed, the method further includes:

[0041] A layer update instruction including layer update information and an update position is received, and a new permission matching layer is inserted at the update position of the preset multi-layer regular matching model according to the layer update information, or the permission matching layer at the update position is deleted.

[0042] It is understandable that all the permission matching layers in the preset multi-level regular matching model in this embodiment can be deleted (delete the permission matching layer at the update position), or a new permission matching layer can be added to the original preset multi-level regular matching model (insert a new permission matching layer at the update position of the preset multi-level regular matching model). In addition, in the present invention, the information to be pushed corresponding to all the permission matching layers in the preset multi-level regular matching model can also be updated, such as updating part or all of the permission regular expressions therein.

[0043] Step S2: In each permission matching layer of the preset multi-level regular matching model, regular matching is performed on the user information and each permission regular expression to obtain a matching result between the user information and the information to be pushed corresponding to each permission matching layer.

[0044] That is, the preset multi-level regular matching model includes multiple layers of permission matching layers, wherein each permission matching layer can match the input user information through the permission regular expressions corresponding to the layer (one or more are acceptable). After each regular expression in the permission matching layer is regularly matched with the input user information, a regular matching result item will be obtained. Based on all the regular matching result items obtained in the permission matching layer, the matching result of the user information matched by the permission matching layer can be obtained.

[0045] In one embodiment, if Figure 4 As shown, step S2 includes:

[0046] Step S21: determining all the permission matching layers in the preset multi-level regular matching model that are in an enabled state;

[0047] Step S22: performing regular matching on each permission regular expression in the information to be pushed and the user information corresponding to each permission matching layer in the enabled state, obtaining regular matching result items of each permission regular expression and the user information, and associating all regular matching result items corresponding to each permission matching layer with the information to be pushed corresponding to the permission matching layer;

[0048] Step S23: When all the regular matching result items associated with the information to be pushed are matched by the permission regular expression and the user information, the matching result of the information to be pushed is recorded as the information to be pushed matches the user information; the information to be pushed matches the user information, which means that the matched information to be pushed can be recommended to the user to be recommended. At this time, the matching result can also be expressed as the need to recommend the information to be pushed to the user to be recommended.

[0049] Step S24: If any of the regular matching results associated with the information to be pushed indicates that the permission regular expression does not match the user information, the matching result of the information to be pushed is recorded as the information to be pushed does not match the user information. If the information to be pushed does not match the user information, it is considered that the unmatched information to be pushed cannot be recommended to the user to be recommended. In this case, the matching result can also be expressed as not recommending the information to be pushed to the user to be recommended.

[0050] In a multi-level regular matching model, user information is used to traverse all the permission matching layers in the multi-level regular matching model that are in an open state. The regular matching result items of each permission regular expression and the user information in each permission matching layer can be obtained first, and then the matching results corresponding to each permission matching layer can be determined based on all the regular matching result items.

[0051] In one embodiment, after the above-mentioned step S105, that is, after the said construction of the preset multi-level regular matching model, it also includes: receiving a startup instruction containing opening information, and controlling each permission matching layer in the preset multi-level regular matching model to be in an open state or a closed state according to the opening information. That is, in this embodiment, since each information to be recommended may have a certain recommendation time limit (such as the promotion time limit of a product or the freshness time limit of news information, etc.), therefore, after the multi-level regular matching model is constructed, the permission matching layer corresponding to each information to be recommended can be set to an open state or a closed state according to the recommendation time limit of each information to be recommended. It can be understood that the opening information represents the opening state or the closed state of each permission matching layer in the preset multi-level regular matching model. Only when a permission matching layer is in the open state will the user information be regularly matched in the permission matching layer. Otherwise, when a permission matching layer is in the closed state, the permission matching layer is deemed not to participate in the regular matching process of the user information corresponding to each push instruction.

[0052] In one embodiment, the plurality of permission matching layers are arranged according to a preset priority; and the information to be pushed corresponding to the permission matching layers at different layers is the same or different.

[0053] like Figure 5As shown, after step S2, that is, after obtaining the matching results between the user information and each piece of information to be pushed, the process further includes:

[0054] Step S201: recording the different permission matching layers corresponding to the same information to be pushed as the same type of matching layers; at this time, the same type of matching layers refers to the different permission matching layers corresponding to the same information to be pushed.

[0055] Step S202: When the matching results obtained from matching the various similar matching layers differ, the matching result obtained from matching the similar matching layer with the highest preset priority is recorded as the matching result for all similar matching layers. That is, when the matching results corresponding to all similar matching layers differ, it indicates that the matching result obtained in at least one similar matching layer differs from the matching results of the other similar matching layers; that is, the matching results corresponding to all similar matching layers corresponding to the same information to be pushed differ (that is, the matching results corresponding to the permission matching layers obtained by the multi-level regular matching model are inconsistent). At this time, the permission matching layers are arranged in descending order according to the preset priority in the multi-level regular matching model. Therefore, when the matching results corresponding to all similar matching layers corresponding to the same information to be pushed differ, the matching result corresponding to the highest priority will prevail. For example, the information to be recommended corresponding to the first and second permission matching layers is the same (but the permission regular expressions are not the same, otherwise only one layer is used for matching). If the preset priority of the first layer is higher than that of the second layer, at this time, if the matching result of the first permission matching layer is that the information to be pushed matches the user information, it is necessary to recommend the information to be pushed to the user to be recommended. The matching result of the second permission matching layer is that the information to be pushed does not match the user information, and there is no need to recommend the information to be pushed to the user to be recommended. At this time, the matching result of the first layer shall prevail. Specifically, all matching results corresponding to the information to be recommended can be regarded as matching results corresponding to the highest priority (here, the matching results of the first layer) by recording the matching results of the first layer as the matching results of the second layer, or deleting the matching results of the second layer.

[0056] Furthermore, an example is used to illustrate the above-mentioned preset multi-level regular matching model. The preset multi-level regular matching model includes the following permission matching layers (and the preset priority order is layer 1 > layer 2 > layer 3):

[0057] The information to be pushed corresponding to the first level permission matching layer is numbered A001 and contains the following regular expression: the user information contains "AA" but does not contain "BB";

[0058] The information to be pushed corresponding to the second level permission matching layer is numbered A002 and contains one of the following regular expressions: contains "BB" but does not contain "CC";

[0059] The information to be pushed corresponding to the third level permission matching layer is numbered A001 and contains the following regular expression: containing "DDD". According to the above preset multi-level regular matching model:

[0060] (1) If the first push instruction is received, and the user information contained in it is "XX AA Co., Ltd.", then:

[0061] In the first permission matching layer, it is considered that the information to be pushed matches the user information, and the matching result is: A001 needs to be recommended.

[0062] The second-level permission matching layer considers that the information to be pushed does not match the user information, and the matching result is: no recommendation A002 is required.

[0063] The third permission matching layer determines that the information to be pushed does not match the user information, and the matching result obtained is: no recommendation information A001 is needed. However, since both the first and third layers are permission matching layers corresponding to the information to be recommended numbered A001, and the preset priority of the first layer is higher than that of the second layer, the matching result of the third layer should be modified to the matching result of the first layer (recommendation A001 is needed), or the matching result of the third layer should be deleted and recorded as a null value.

[0064] (2) If the first push instruction is received, and the user information contained in it is "BBAA Co., Ltd.", then:

[0065] The first permission matching layer considers that the information to be pushed does not match the user information, and the matching result is: no recommendation A001 is required.

[0066] The second layer permission matching layer considers that the information to be pushed matches the user information, and the matching result is: A002 needs to be recommended.

[0067] The third permission matching layer determines that the information to be pushed matches the user information, and the matching result is: Recommendation A001 is not required. Since both the first and third layers are permission matching layers corresponding to the information to be recommended numbered A001, and the first layer has a higher preset priority than the second layer, and the matching results in both the first and third layers are that recommendation A001 is not required, there is no need to modify the matching result in the third layer.

[0068] Step S3: determining the user group to which the user to be recommended belongs based on the user information, and obtaining recommendation values of other users in the user group for each item of information to be pushed.

[0069] By using user information to determine the user group to which the user to be recommended belongs (the user group with the same behavioral preferences as the user to be recommended), the recommendation values of other users in the user group for each piece of information to be pushed are obtained. Furthermore, the recommendation values of other users in the user group with the same preferences and behaviors for each piece of information to be pushed are used as a further reference for whether to push the information to be pushed to the user, which can further improve the accuracy of the push.

[0070] like Figure 6 As shown, in step S3, the user group to which the user to be recommended belongs is determined based on the user information, and recommendation values of other users in the user group for each item of information to be pushed are obtained, including:

[0071] Step S31: obtaining user matching data of the user to be recommended based on the user information, wherein the user matching data includes user name, user attributes and user historical behavior data;

[0072] Step S32: determining the user group to which the user to be recommended belongs based on the user matching data, wherein the user group is a different group obtained by clustering the user-related data of all users to whom the information to be pushed has been recommended;

[0073] Step S33: obtaining the recommendation value corresponding to each item of information to be pushed for each user in the determined user group.

[0074] In this embodiment, the process of determining the user group is: first determine the users who have been recommended at least one of the above-mentioned information to be pushed (at least one of all the information to be pushed corresponding to each permission matching layer of the preset multi-level regular matching model), and then obtain different user groups by clustering the user-related data of all determined users. Since the user-related data includes user name, user attributes and user historical behavior data, the user historical behavior data can include user preferences related to the information to be recommended, etc., therefore, the users in the same user group finally obtained have the same preferences for the information to be pushed. Therefore, after finding the user group to which the user to be recommended belongs, the recommendation value of each information to be pushed by other users with the same preferences and behaviors in the user group can be used as a further reference for whether to push the information to be pushed to the user, which can further improve the accuracy of the push.

[0075] In the present invention, the recommendation value of each item of information to be pushed by a user in a user group refers to the actual evaluation value of the information to be pushed after the information is pushed. For example, the recommendation value of user AE for items numbered A001 and A002 is shown in Table 1 below.

[0076] Table 1 Recommended values of different information to be pushed by users in the same user group

[0077] A001 A002 User A 5 2 User B 10 6 User C 9 3 User D 8 7 User E 5 5

[0078] Step S4: Based on the matching result and the recommendation value, a recommended item is determined from all the information to be pushed, and the recommended item is pushed to the user to be recommended. That is, after obtaining a matching result, a recommended item can be further determined based on the recommendation value from all the information to be pushed that matches the matching result, and the recommended item is pushed to the user to be recommended. The combined decision of the matching result and the recommendation value allows for targeted recommendation of the information to be pushed, further improving push accuracy.

[0079] The present invention can directly obtain the matching results between a user information and all the information to be pushed by presetting a multi-level regular matching model. The calculation process is simple and the calculation efficiency is high, so that the recommended information pushed to the recommended user is targeted (the recommended users to whom a piece of information to be pushed is pushed are all matched with the information to be pushed), and the push accuracy is high; at the same time, after determining the matching result, a joint decision is made to make a targeted recommendation based on the matching result and the recommendation value of the user group corresponding to the user to be recommended, which further improves the push accuracy and user experience.

[0080] In one embodiment, if Figure 7 As shown, step S4, i.e., determining a to-be-recommended item from all the to-be-pushed information based on the matching result and the recommendation value, and pushing the to-be-recommended item to the to-be-recommended user, includes:

[0081] Step S41: All information to be pushed whose matching result is that the information to be pushed matches the user information is recorded as matched information; that is, all information to be pushed whose matching result is a match is determined to be information to be pushed that meets the authority requirements and can be pushed, and is recorded as matched information.

[0082] Step S42: Obtain the preset restricted push conditions associated with the user information, and determine the matched information that does not meet the preset restricted push conditions as a to-be-recommended item; that is, the preset restricted push conditions may refer to specific conditions that cannot be pushed, for example, the preset restricted push conditions are: if the message to be pushed has been pushed to the user to be pushed before the current moment, then the preset restricted push conditions are met. For another example, the preset restricted push conditions are: if the user to be pushed has set a setting not to receive messages of a certain information type (each message to be pushed has an information type) within a preset time period, and the current time point is within the preset time period, and the message to be pushed belongs to this information type, then the preset restricted push conditions are met. For another example, the preset restricted push conditions are: if the user to be pushed has purchased the product corresponding to the message to be pushed, then the preset restricted push conditions are met.

[0083] Step S43: Determine the final recommendation value of each item to be recommended based on the recommendation value of each user in the user group and the preset weight. It is understandable that the preset weight can be set as needed, and can be set based on the user's compatibility with the user group. For example, if the compatibility with the user group is 0.8, the preset weight is set to 1.8; if the compatibility with the user group is 1, the preset weight is set to 2; if the compatibility with the user group is 0, the preset weight is set to 1; if the compatibility with the user group is 0.5, the preset weight is set to 1.5. In other words, the compatibility is between 0 and 1, but the preset weight fluctuates between 1 and 2. The higher the compatibility, the higher the corresponding preset weight. In other words, the closer the user's behavior preferences are to the user to be recommended, the higher the weight of the recommendation value is set, which can improve the accuracy of the recommendation. The final recommendation value can be the product of the preset weight and the recommendation value. When there are multiple users in the matched user group who have been pushed the same item to be recommended (information to be recommended), that is, each user in the obtained user group will have multiple (for example, N) recommendation values corresponding to the item to be recommended. At this time, N / 3 of the highest recommendation values can be removed (if N / 3 is not an integer, the value before the decimal point can be taken as the number of the highest recommendation values to be removed), and then N / 3 of the lowest recommendation values can be removed (if N / 3 is not an integer, the value before the decimal point can be taken as the number of the lowest recommendation values to be removed), and the average of all the remaining intermediate recommendation values can be taken as the final recommendation value.

[0084] For example, the recommendation value corresponding to each user in Table 1 above needs to be further adjusted by the preset weight (the preset weight has been set and can be determined based on the user's compatibility with the user group). At this time, it is first necessary to determine which information to be pushed corresponding to which part of the users needs to be adjusted according to the preset weight. In the present invention, the user data of each user in the user group (corresponding to the user information of the user to be recommended) can be input into the preset multi-level regular matching model containing multiple layers of permission matching layers to be traversed, and the matching results of the user data of each user and the information to be pushed corresponding to each permission matching layer are confirmed. When the user data of a user matches the matching result of the information to be pushed, the information to be pushed corresponding to the user needs to be determined as recommended, otherwise it is not recommended.

[0085] For example, assuming that the preset weight of each user in Table 1 is 1.2, only user A needs to recommend the information to be pushed with number A001; user C needs to recommend the information to be pushed with number A002. The recommendation values corresponding to the above two users need to be multiplied by the preset weights for adjustment. That is, at this time, the recommendation values in Table 1 should be adjusted and updated as shown in Table 2 below.

[0086] Table 2 Adjusted recommendation values for different information to be pushed by each user in the same user group

[0087]

[0088]

[0089] Understandably, in one embodiment, the recommended value of the information to be pushed corresponding to each user can be adjusted in the user group according to the preset restriction push condition (the preset restriction push condition has been described in the above embodiment. For example, the preset restriction push condition is: the user to be pushed has set not to receive messages of a certain type of information within a preset time period, and the current time point is within the preset time period, and the message to be pushed belongs to this type of information; then the preset restriction push condition is met. For another example, the preset restriction push condition is: the user to be pushed has purchased the product corresponding to the message to be pushed, then the preset restriction push condition is met). For example, if the information to be pushed corresponding to a user meets the preset restriction push condition, the recommended value that meets the preset restriction push condition is uniformly adjusted to a preset value, such as "-1". At this time, when calculating the final recommended value, it is also necessary to adjust it according to the recommended value after the adjustment. For example, in Table 2 above, if the information to be pushed A002 corresponding to user D meets the preset restriction push condition, its recommended value "7" is adjusted to "-1". At this time, when calculating the final recommended value, it can be included in the recommended value base for calculating the final recommended value, or it can be first screened out and not considered, and the final recommended value can be calculated using other recommended values after screening.

[0090] Step S44: Push the recommended items to the user to be recommended based on the final recommendation value. That is, in this embodiment, each ultimately obtained recommended item has a corresponding final recommendation value. Therefore, the information to be pushed (the recommended item) corresponding to the highest final recommendation value can be pushed to the user to be recommended. Alternatively, all recommended items can be sorted by their final recommendation values and pushed to the user to be recommended. Other methods can also be used, such as selecting and pushing some recommended items based on demand.

[0091] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0092] In one embodiment, a push device based on multi-level regular matching is provided, and the push device based on multi-level regular matching corresponds to the push method based on multi-level regular matching in the above embodiment. Figure 8As shown, the push device based on multi-level regular matching includes a receiving module 10, a result acquisition module 20, a recommendation value acquisition module 30, and a push module 40. The functional modules are described in detail as follows:

[0093] A receiving module 10 is configured to receive a push instruction containing user information of a user to be recommended, and input the user information into a preset multi-level regular matching model comprising multiple permission matching layers; wherein each permission matching layer corresponds to a piece of information to be pushed containing at least one permission regular expression;

[0094] A result acquisition module 20 is configured to perform regular matching on each permission regular expression in the information to be pushed and the user information corresponding to each permission matching layer in the preset multi-level regular matching model, and obtain a matching result between the user information and each item of information to be pushed;

[0095] The recommendation value acquisition module 30 is used to determine the user group to which the user to be recommended belongs based on the user information, and obtain the recommendation values of other users in the user group for each item of information to be pushed;

[0096] The push module 40 is configured to determine an item to be recommended from all the information to be pushed according to the matching result and the recommendation value, and push the item to be recommended to the user to be recommended.

[0097] The present invention can directly obtain the matching results between a user information and all the information to be pushed by presetting a multi-level regular matching model. The calculation process is simple and the calculation efficiency is high, so that the recommended information pushed to the recommended user is targeted (the recommended users to whom a piece of information to be pushed is pushed are all matched with the information to be pushed), and the push accuracy is high; at the same time, after determining the matching result, according to the matching result and the recommendation value of the user group corresponding to the user to be recommended, a joint decision is made to make a targeted recommendation, which further improves the push accuracy and user experience.

[0098] In one embodiment, the device further includes: an information acquisition module, configured to acquire all information to be pushed, and acquire permission information and a preset priority corresponding to each information to be pushed;

[0099] an extraction module, configured to extract at least one permission keyword from each item of the permission information;

[0100] An expression construction module, configured to construct at least one permission regular expression corresponding to the information to be pushed based on all the permission keywords corresponding to the information to be pushed;

[0101] A matching layer construction module, configured to construct a permission matching layer corresponding to each item of information to be pushed based on all the permission regular expressions corresponding to each item of information to be pushed;

[0102] The model building module is used to build a preset multi-level regular matching model based on all the permission matching layers and their corresponding preset priorities.

[0103] In one embodiment, the model building module is further used to: receive a start instruction including start information, and control each permission matching layer in the preset multi-level regular matching model to be in an open state or a closed state according to the start information.

[0104] In one embodiment, the model building module is also used to: receive a hierarchical update instruction including hierarchical update information and an update position, and insert a new permission matching layer at the update position of the preset multi-level regular matching model according to the hierarchical update information, or delete the permission matching layer at the update position.

[0105] In one embodiment, the result acquisition module 20 includes:

[0106] A determination submodule, configured to determine all the permission matching layers in the preset multi-level regular matching model that are in an enabled state;

[0107] a matching submodule for performing regular matching on each permission regular expression in the information to be pushed and the user information corresponding to each permission matching layer in an enabled state, obtaining regular matching result items of each permission regular expression and the user information, and associating all regular matching result items corresponding to each permission matching layer with the information to be pushed corresponding to the permission matching layer;

[0108] a matching record submodule, configured to record the matching result of the information to be pushed as a match between the information to be pushed and the user information when all the regular matching result items associated with the information to be pushed are matches between the permission regular expression and the user information;

[0109] The mismatch recording submodule is used to record the matching result of the information to be pushed as the mismatch between the information to be pushed and the user information when any regular matching result item associated with the information to be pushed is that the permission regular expression does not match the user information.

[0110] The device further comprises:

[0111] A matching layer recording module, configured to record the permission matching layers of different layers corresponding to the same information to be pushed as the same type of matching layers;

[0112] The result recording module is configured to record the matching result obtained by matching the matching layer of the same type with the highest preset priority as the matching result of all matching layers of the same type when the matching results obtained by matching the matching layers of the same type are different.

[0113] In one embodiment, the recommendation value acquisition module 30 includes:

[0114] A data acquisition submodule is used to acquire user matching data of the user to be recommended based on the user information, wherein the user matching data includes user name, user attributes and user historical behavior data;

[0115] A group determination submodule is used to determine the user group to which the user to be recommended belongs based on the user matching data, wherein the user group is a different group obtained by clustering the user-related data of all users to whom the information to be pushed has been recommended;

[0116] The recommendation value acquisition submodule is used to acquire the recommendation value corresponding to each item of information to be pushed for each user in the determined user group.

[0117] In one embodiment, the push module 40 includes:

[0118] An information recording submodule, configured to record all information to be pushed, for which the matching result is that the information to be pushed matches the user information, as matched information;

[0119] A submodule for determining items to be recommended, configured to obtain a preset push restriction condition associated with the user information, and determine the matched information that does not meet the preset push restriction condition as an item to be recommended;

[0120] A final recommendation value determination submodule, configured to determine a final recommendation value for each of the items to be recommended based on the recommendation values of each user in the user group corresponding to each of the items to be recommended and a preset weight;

[0121] The push submodule is used to push the item to be recommended to the user to be recommended according to the final recommendation value.

[0122] For the specific limitations of the push device based on multi-level regular matching, please refer to the limitations of the push method based on multi-level regular matching above, which will not be repeated here. The various modules in the above-mentioned push device based on multi-level regular matching can be implemented in whole or in part by software, hardware, and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0123] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 9 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium and an internal memory. The readable storage medium stores an operating system, computer-readable instructions and a database. The internal memory provides an environment for the operation of the operating system and computer-readable instructions in the readable storage medium. The database of the computer device is used to store data involved in the push method based on multi-level regular matching. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer-readable instructions are executed by the processor, a push method based on multi-level regular matching is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.

[0124] In one embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the above-mentioned push method based on multi-level regular matching is implemented.

[0125] In one embodiment, one or more computer-readable storage media storing computer-readable instructions are provided. The computer-readable storage media provided in this embodiment include non-volatile readable storage media and volatile readable storage media. The computer-readable instructions stored on the readable storage media, when executed by one or more processors, implement the aforementioned push method based on multi-level regular expression matching.

[0126] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through computer-readable instructions. The computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When the computer-readable instructions are executed, they can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0127] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0128] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A push method based on multi-level regular matching, characterized in that: include: Receive a push instruction containing user information of a user to be recommended, and input the user information into a preset multi-level regular matching model comprising multiple permission matching layers; wherein each permission matching layer corresponds to a piece of information to be pushed containing at least one permission regular expression; In each permission matching layer of the preset multi-level regular matching model, regular matching is performed on the user information and each permission regular expression to obtain a matching result between the user information and the information to be pushed corresponding to each permission matching layer; Determine the user group to which the user to be recommended belongs based on the user information, and obtain recommendation values of other users in the user group for each item of information to be pushed. The user to be recommended has the same behavioral preferences as other users in the user group. The recommendation value refers to the actual evaluation value of the information to be pushed by other users in the user group; According to the matching result and the recommendation value, an item to be recommended is determined from all the information to be pushed, and the item to be recommended is pushed to the user to be recommended.

2. The push method based on multi-level regular expression matching according to claim 1, characterized in that: In the preset multi-level regular matching model, each permission regular expression in the information to be pushed and the user information corresponding thereto are subjected to regular matching by each permission matching layer to obtain a matching result between the user information and each item of information to be pushed, including: Determine all the permission matching layers in the preset multi-level regular matching model that are in an enabled state; Performing regular matching on each permission regular expression in the information to be pushed and the user information corresponding to each permission matching layer in the enabled state, obtaining regular matching result items of each permission regular expression and the user information, and associating all regular matching result items corresponding to each permission matching layer with the information to be pushed corresponding to the permission matching layer; When all the regular matching result items associated with the information to be pushed are matches between the permission regular expression and the user information, recording the matching result of the information to be pushed as a match between the information to be pushed and the user information; When any one of the regular matching result items associated with the information to be pushed is that the permission regular expression does not match the user information, the matching result of the information to be pushed is recorded as the information to be pushed does not match the user information.

3. The push method based on multi-level regular expression matching according to claim 1, characterized in that: The multiple layers of permission matching layers are arranged according to a preset priority; the information to be pushed corresponding to the permission matching layers at different layers is the same or different; After obtaining the matching results between the user information and each item of information to be pushed, the method further includes: Recording the authority matching layers of different levels corresponding to the same information to be pushed as the same type of matching layers; When the matching results obtained by matching the similar matching layers are different, the matching result obtained by matching the similar matching layer with the highest preset priority is recorded as the matching result of all similar matching layers.

4. The push method based on multi-level regular expression matching according to claim 1, characterized in that: The step of determining the user group to which the user to be recommended belongs based on the user information, and obtaining recommendation values of other users in the user group for each item of information to be pushed, includes: Acquire user matching data of the user to be recommended based on the user information, wherein the user matching data includes user name, user attributes and user historical behavior data; Determining the user group to which the user to be recommended belongs based on the user matching data, wherein the user group is a different group obtained by clustering user-related data of all users to whom the information to be pushed has been recommended; The recommendation value corresponding to each item of information to be pushed for each user in the determined user group is obtained.

5. The push method based on multi-level regular expression matching according to claim 4, characterized in that: The determining of a to-be-recommended item from all the to-be-pushed information according to the matching result and the recommendation value, and pushing the to-be-recommended item to the to-be-recommended user, includes: Record all the information to be pushed that matches the information to be pushed and the user information as matched information; Acquire a preset push restriction condition associated with the user information, and determine the matched information that does not meet the preset push restriction condition as an item to be recommended; Determining a final recommendation value for each of the items to be recommended based on the recommendation value of each user in the user group corresponding to each of the items to be recommended and a preset weight; The item to be recommended is pushed to the user to be recommended according to the final recommendation value.

6. The push method based on multi-level regular expression matching according to claim 1, characterized in that: Before receiving the recommendation instruction including the user information of the user to be recommended and inputting the user information into a preset multi-level regular matching model including multiple permission matching layers, the method further includes: Obtain all information to be pushed, and obtain the authority information and preset priority corresponding to each of the information to be pushed; Extracting at least one permission keyword from each item of permission information; Constructing at least one permission regular expression corresponding to the information to be pushed based on all the permission keywords corresponding to the information to be pushed; Constructing a permission matching layer corresponding to each item of information to be pushed based on all the permission regular expressions corresponding to each item of information to be pushed; According to all the permission matching layers and their corresponding preset priorities, a preset multi-level regular matching model is constructed.

7. The push method based on multi-level regular expression matching according to claim 6, characterized in that: After the preset multi-level regular matching model is constructed, the method further includes: receiving a start instruction including start information, and controlling each permission matching layer in the preset multi-level regular matching model to be in an open state or a closed state according to the start information; or A layer update instruction including layer update information and an update position is received, and a new permission matching layer is inserted at the update position of the preset multi-layer regular matching model according to the layer update information, or the permission matching layer at the update position is deleted.

8. A push device based on multi-level regular matching, characterized in that: include: a receiving module configured to receive a push instruction containing user information of a user to be recommended, and input the user information into a preset multi-level regular matching model comprising multiple permission matching layers; wherein each permission matching layer corresponds to a piece of information to be pushed containing at least one permission regular expression; A result acquisition module is used to perform regular matching on each permission regular expression in the information to be pushed and the user information corresponding to each permission matching layer in the preset multi-level regular matching model, and obtain the matching result between the user information and each item of information to be pushed; A recommendation value acquisition module is used to determine the user group to which the recommended user belongs based on the user information, and obtain the recommendation values of other users in the user group for each item of information to be pushed; The push module is used to determine the item to be recommended from all the information to be pushed according to the matching result and the recommendation value, and push the item to be recommended to the user to be recommended.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the push method based on multi-level regular matching according to any one of claims 1 to 7 is implemented.

10. One or more readable storage media, wherein the readable storage medium stores a computer program, characterized in that: When the computer program is executed by a processor, the push method based on multi-level regular matching according to any one of claims 1 to 7 is implemented.

Images