A method and related device for analyzing user behavior data

Abstract

Embodiments of the present application disclose a method and related device for analyzing user behavior data, which are used to improve the quality of notification content pushed to users. The method of the embodiments of the present application includes: obtaining user information, where the user information includes the user's identity information and the user's message notification click record; obtaining notification information according to the user information, where the notification information is the notification information confirmed to be clicked by the user; performing clustering analysis on the notification information according to an unsupervised algorithm to obtain notification data, where the notification data includes the word segmentation results of all the notification information of the user and the notification information; establishing the relationship between the notification data and the notification template to obtain a message template; and binding the message template with the user information to obtain a binding result.

Description

Technical Field

[0001] Embodiments of the present application relate to the field of data processing, and in particular, to a method for analyzing user behavior data and related devices. Background Art

[0002] With the development of society, smart phones and intelligent portable terminals have a greater and greater impact on people. To improve the efficiency of information promotion, a message push system will analyze the behavior of the target users to be pushed, so as to achieve the purpose of accurate message delivery.

[0003] In the prior art, message push analyzes according to user behavior, and based on the user's behavior data, that is, the operation records of software on the intelligent portable terminal by the user recently, analyzes the user's behavior, and integrates the user's interests according to the analysis results, so as to achieve the purpose of targeted message push.

[0004] Since the user's behavior data refers to the operation behavior of the user on the smart phone and the portable intelligent terminal, that is, the integration of all data such as software usage records, web browsing history, and software push click situations, etc., the quality of message push is reduced due to the overly large amount of user behavior data obtained. Summary of the Invention

[0005] Embodiments of the present application provide a method for analyzing user behavior data and related devices, which are used to improve the quality of notification content pushed to users.

[0006] The first aspect of the present application provides a method for analyzing user behavior data, including:

[0007] Obtain user information, where the user information includes the user's identity information and the user's message notification click record;

[0008] Obtain notification information according to the user information, where the notification information is the notification information confirmed by the user's click;

[0009] Perform clustering analysis on the notification information according to an unsupervised algorithm to obtain notification data, where the notification data includes the word segmentation results of all the user's notification information and the notification information;

[0010] Establish the relationship between the notification data and the notification template to obtain a message template;

[0011] Bind the message template to the user information to obtain a binding result.

[0012] Optionally, after obtaining the user information, the method further includes:

[0013] Obtain a rule table, where the rule table is used to store the analysis format of the user information;

[0014] Determine whether there is matching data for the user information in the rule table;

[0015] If there is, determine the message template that matches the user information through the rule table.

[0016] Optionally, after binding the message template to the user information, the method further includes:

[0017] Synchronize the binding result to the rule table.

[0018] Optionally, establishing the relationship between the notification data and the notification template to obtain the message template includes:

[0019] Perform word segmentation on the notification data;

[0020] Calculate the weight of each word segment in the notification data to generate a word segment weight ratio;

[0021] Obtain the collection of notification templates;

[0022] Match the word segment weight ratio with the notification templates in the collection of notification templates to obtain a matching result;

[0023] Obtain the matching result with the highest degree of matching to obtain the target matching result;

[0024] Determine the target notification template according to the target matching result;

[0025] Merge the target notification template and the notification data to obtain the message template.

[0026] Optionally, merging the target document and the notification data to obtain the message template includes:

[0027] Merge the target document and the notification data to generate a merge result;

[0028] Obtain the template data;

[0029] Determine whether the format of the merge result matches the template data;

[0030] If they match, determine the merge result as the message template.

[0031] The second aspect of the present application provides a device for analyzing user behavior data, including:

[0032] A first acquisition unit, configured to acquire user information, where the user information includes the user's identity information and the user's message notification click record;

[0033] A second acquisition unit, configured to acquire notification information according to the user information, where the notification information is the notification information that confirms the user's click;

[0034] A clustering analysis unit, configured to perform clustering analysis on the notification information according to an unsupervised algorithm to obtain notification data;

[0035] An establishment unit, configured to establish a relationship between the notification data and a notification template to obtain a message template;

[0036] A binding unit, configured to bind the message template to the user information to obtain a binding result.

[0037] Optionally, the device further includes:

[0038] A third acquisition unit, configured to acquire a rule table, where the rule table is used to store the analysis format of the user information;

[0039] A judgment unit, configured to judge whether there is matching data for the user information in the rule table;

[0040] A determination unit, configured to, when the judgment result of the judgment unit is yes, determine a message template that matches the user information through the rule table.

[0041] Optionally, the device further includes:

[0042] A synchronization unit, configured to synchronize the binding result to the rule table.

[0043] Optionally, the establishment unit includes:

[0044] A word segmentation module, configured to perform word segmentation on the notification data;

[0045] A calculation module, configured to calculate the weight of each word segment in the notification data to generate a word segment weight ratio;

[0046] A first acquisition module, configured to acquire the notification template collection;

[0047] A matching module, configured to match the word segment weight ratio with the notification templates in the notification template collection to obtain a matching result;

[0048] A second acquisition module, configured to acquire the matching result with the highest matching degree to obtain a target matching result;

[0049] A determination module, configured to determine a target document according to the target matching result;

[0050] A merging module, configured to merge the target document and the notification data to obtain a message template.

[0051] Optionally, the merging module is further configured to:

[0052] Merge the target document and the notification data to generate a merge result;

[0053] Obtain template data;

[0054] Determine whether the format of the merge result matches the template data;

[0055] When the judgment result of the judgment sub-module is a match, determine the merge result as the message template.

[0056] A third aspect of the present application provides a device for analyzing user behavior data, including:

[0057] A processor, a memory, an input / output unit, and a bus;

[0058] The processor is connected to the memory, the input / output unit, and the bus;

[0059] The processor specifically performs the same operations as those in the foregoing first aspect.

[0060] From the above technical solutions, it can be seen that the present application performs clustering analysis on the data included in the user information according to an unsupervised algorithm, and then establishes the relationship between the analyzed notification data and the notification template to obtain the message template, and binds the message template to the user information to obtain the binding result, so that the message template applicable to the user can be searched through the binding result, thereby improving the quality of the message pushed to the user. Description of the Drawings

[0061] Figure 1 It is a schematic flowchart of an embodiment of the method for analyzing user behavior data in an embodiment of the present application;

[0062] Figure 2 It is a schematic flowchart of another embodiment of the method for analyzing user behavior data in an embodiment of the present application;

[0063] Figure 3 It is a schematic structural diagram of an embodiment of the device for analyzing user behavior data in an embodiment of the present application;

[0064] Figure 4 It is a schematic structural diagram of another embodiment of the device for analyzing user behavior data in an embodiment of the present application;

[0065] Figure 5 It is a schematic structural diagram of another embodiment of the device for analyzing user behavior data in an embodiment of the present application. Detailed Embodiments

[0066] Embodiments of the present application provide a method and related device for analyzing user behavior data, which are used to improve the quality of notification content pushed to users.

[0067] Next, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0068] The execution entities in the embodiments of the present application include, but are not limited to, devices such as terminals, servers, and systems that all have logical computing and operating capabilities. Specifically, no limitation is made here. The embodiments of the present application will be described by taking the terminal as an example.

[0069] Please refer to Figure 1 , an embodiment of the method for analyzing user behavior data provided by the embodiments of the present application includes:

[0070] 101. Obtain user information, where the user information includes the user's identity information and the user's message notification click record;

[0071] In the embodiments of the present application, after the terminal obtains the user information, the terminal will parse the user information and parse out the notifications received by the user on the same day, the notifications clicked by the user on the same day, and the user's ID information from the user information. The ID information includes, but is not limited to, flag data such as the user's mobile phone number or the independent serial number of the user's smart device that can be used to determine the user's identity. Specifically, no limitation is made here.

[0072] Among them, the data included in the user information includes the notification data clicked by the user, enabling the terminal to accurately analyze the message notifications that the user has an interest in clicking.

[0073] 102. Obtain notification information according to the user information, where the notification information is the notification information that confirms the user's click;

[0074] After the terminal obtains the user information, the terminal will trace back the content of the notification information of the message notification clicked by the user according to the data carried in the user information and obtain the specific text content in the notification information, so that the terminal can perform data analysis on the specific notification text.

[0075] 103. Perform clustering analysis on the notification information according to the unsupervised algorithm to obtain notification data, where the notification data includes the word segmentation results of all the notification information of the user and the notification information;

[0076] Unsupervised algorithms are used to cluster data samples in the input algorithm. The clustering is specifically as follows: for a given dataset of M samples, given the number of clusters K (K < M), initialize the category to which each sample belongs, and then continuously iterate and re-partition the categories of the dataset (change the category relationship between samples and clusters) according to certain rules, so that each partition is better than the previous one.

[0077] Specifically, after the terminal obtains the notification information clicked by the user on the same day, the terminal will cluster the notification information through an unsupervised algorithm based on the text content of the notification information and the label attributes carried in the notification information, so that the terminal can classify the obtained notification information and then analyze each classified notification information one by one, thereby improving the value of the terminal's analysis of the notification information for subsequent binding of notification labels to the user.

[0078] 104. Establish the relationship between the notification data and the notification template to obtain a message template;

[0079] Specifically, a large number of notification templates are stored in the terminal. The notification templates are obtained through message push of an independent software. After the terminal clusters the notification information, it will search for similar templates in the notification templates according to the common labels carried in the information stored in each category, and bind the notification data to the template, so that when the terminal determines that the user clicks on the notification information pointed to by the notification data, it can obtain the corresponding similar message template for information generation, thereby increasing the user's interest in the notification information.

[0080] In actual situations, the amount of notification template data is huge. The labels contained in the notification data can greatly reduce the working time for the terminal to search for similar notification templates according to the notification data when searching for similar notification templates in the terminal, thereby reducing the consumption of the terminal's computing power.

[0081] 105. Bind the message template to the user information to obtain a binding result.

[0082] Specifically, when the terminal determines the relationship between the notification data and the notification template through the notification data, it will directly bind the user information to the message template, so that when the terminal receives an instruction to push a notification to the user, it can directly extract the corresponding notification template, thereby improving the response time of the terminal to the notification push instruction.

[0083] As can be seen from the above technical solutions, in this application, clustering analysis is performed on the data included in the user information according to an unsupervised algorithm, and the notification data obtained after the analysis is then used to establish the relationship between the notification data and the notification template to obtain a message template. The message template is bound to the user information to obtain a binding result, enabling the message template applicable to the user to be searched through the binding result, thereby improving the quality of the messages pushed to the user.

[0084] Please refer to Figure 2 , another embodiment of the method for analyzing user behavior data provided by an embodiment of this application includes:

[0085] 201. Obtain user information, where the user information includes the user's identity information and the user's message notification click record;

[0086] Step 201 in this embodiment is similar to step 101 in the foregoing embodiment, and will not be elaborated here.

[0087] 202. Obtain a rule table, where the rule table is used to store the analysis format of the user information;

[0088] After the terminal obtains the user information, the terminal will first extract the processed data set, which is the rule table. The data information included in the rule table includes, but is not limited to, rule names, rule tags, user IDs, rule generation times, etc., which are parameters required for the terminal to more quickly search for the corresponding notification template of the user information. The specific parameter format is not limited here.

[0089] 203. Determine whether there is matching data for the user information in the rule table;

[0090] After the terminal obtains the rule table, the terminal will search in the rule table according to the ID information carried in the user information and the tag of the notification. If it is determined that there is data in the rule table that matches the user information, step 204 is executed; if there is no data in the rule table that matches the user information, step 205 is executed.

[0091] 204. If there is, determine the message template that matches the user information through the rule table.

[0092] When the judgment result of step 203 is that there is data that matches the user information, the terminal will directly search through the rules in the rule table to find the message template corresponding to the user ID information and the notification-related tags in the user information, and directly generate a push message sent to the user through the notification template.

[0093] 205. Obtain notification information according to the user information, where the notification information is the notification information for which the user's click is confirmed;

[0094] 206. Perform clustering analysis on the notification information according to the unsupervised algorithm to obtain notification data, where the notification data includes the word segmentation results of all the user's notification information and the notification information.

[0095] Steps 205 to 206 in this embodiment are similar to steps 102 to 103 in the foregoing embodiment, and will not be elaborated here.

[0096] 207. Perform word segmentation on the notification data.

[0097] When the terminal obtains user information that cannot be matched in the rule table, the terminal will trace back the notification content of the user according to the user information and the notification label, so as to obtain the notification data of this notification. The notification data includes the text content of this notification. The terminal will perform clustering analysis of the unsupervised algorithm on the text content. When performing clustering analysis, the terminal will perform word segmentation on the text content of the notification, so that the terminal can perform word segmentation clustering on the text content of the notification.

[0098] 208. Calculate the weight of each word segment in the notification data to generate a word segment weight ratio.

[0099] When the terminal performs word segmentation on the text content of the notification, the terminal will calculate the weight ratio of each word segment state according to the word segmentation situation, and match the word segment data with a preset notification template in the terminal whose weight ratio is higher than the preset weight ratio value.

[0100] 209. Obtain the collection of notification templates.

[0101] The collection of notification templates contains all the preset notification templates in the terminal. The terminal can match according to the collection of notification templates and the word segment weight ratio summarized from the notification data.

[0102] 210. Match the word segment weight ratio with the notification templates in the collection of notification templates to obtain a matching result.

[0103] When the terminal obtains the collection of notification templates, the terminal will match the word segment weight ratio with the notification templates in the collection of notification templates. In actual situations, a single notification may match more than one notification template, so word segmentation of the notification data may generate more than one matching result at a time.

[0104] 211. Obtain the matching result with the highest matching degree to get the target matching result.

[0105] When the terminal obtains more than one matching result, the terminal will extract the matching result with the highest matching degree for processing compared with other matching results. The matching result with the highest matching degree is the target matching result.

[0106] 212. Determine a target notification template based on the target matching result;

[0107] After the terminal obtains the target matching result, the terminal will determine the word segmentation situation included in the target matching result and the message notification in the message notification template, so that the terminal can convert the segmented notification data back into the original notification data text content.

[0108] 213. Merge the target document and the notification data to generate a merge result;

[0109] The terminal performs text merging on the target document and the notification data. The purpose of text merging is for the terminal to extract the keywords in the notification data and the target notification template from the same text, so that the target notification template can be filled based on the custom settings for this user or the keywords in the notification data. The merged target notification template and notification data are called the merge result.

[0110] 214. Obtain template data;

[0111] After merging into the merge result, the terminal will obtain the template data again. The template data is used to determine whether the carried information in the merge result is complete. The purpose of this step is to verify the data integrity of the merge result.

[0112] 215. Determine whether the formats of the merge result and the template data match;

[0113] Specifically, after the terminal obtains the template data, it will match the data in the merge result with the data in the template data one by one. Only when the template data completely matches the merge result will step 216 be executed. If they do not match, the terminal will report an error, so as to notify the staff to process the information in the merge result. The specific processing method can be to manually fill in the missing information or end the current process and restart the process to process the notification data.

[0114] 216. If they match, determine the merge result as the message template.

[0115] Only when the merge result completely matches the template data will the terminal mark the matching result as the message template.

[0116] 217. Bind the message template to the user information to obtain a binding result.

[0117] Step 217 in this embodiment is similar to step 105 in the foregoing embodiment, and will not be elaborated here.

[0118] 218. Synchronize the binding result to the rule table.

[0119] Specifically, to reduce the working intensity of the terminal, after each analysis of the notification, the terminal will synchronize the analyzed binding result to the rule table, so that when the same notification label of the same user is encountered again, the terminal can directly generate a push message for the user according to the rules pre-stored in the rule table.

[0120] Please refer to Figure 3 , an embodiment of the apparatus for analyzing user behavior data provided by the embodiment of the present application includes:

[0121] The first acquisition unit 301 is used to acquire user information, and the user information includes the identity information of the user and the message notification click record of the user;

[0122] The second acquisition unit 302 is used to acquire notification information according to the user information, and the notification information is the notification information confirmed to be clicked by the user;

[0123] The clustering analysis unit 303 is used to perform clustering analysis on the notification information according to an unsupervised algorithm to obtain notification data;

[0124] The establishment unit 304 is used to establish the relationship between the notification data and the notification template to obtain a message template;

[0125] The binding unit 305 is used to bind the message template with the user information to obtain a binding result.

[0126] In this embodiment, the functions of each unit correspond to the steps in the foregoing Figure 1 shown embodiment, and will not be described in detail here.

[0127] Please refer to Figure 4 , another embodiment of the apparatus for analyzing user behavior data provided by the embodiment of the present application includes:

[0128] The first acquisition unit 401 is used to acquire user information, and the user information includes the identity information of the user and the message notification click record of the user;

[0129] The third acquisition unit 402 is used to acquire a rule table, and the rule table is used to store the analysis format of the user information;

[0130] The judgment unit 403 is used to judge whether there is matching data of the user information in the rule table;

[0131] The determination unit 404 is used to determine the message template matching the user information through the rule table when the judgment result of the judgment unit is existence.

[0132] The second acquisition unit 405 is used to acquire notification information according to the user information, and the notification information is the notification information confirmed to be clicked by the user;

[0133] The clustering analysis unit 406 is configured to perform clustering analysis on the notification information according to an unsupervised algorithm to obtain notification data;

[0134] The establishing unit 407 is configured to establish a relationship between the notification data and a notification template to obtain a message template;

[0135] The binding unit 408 is configured to bind the message template with user information to obtain a binding result.

[0136] The synchronization unit 409 is configured to synchronize the binding result to the rule table.

[0137] In the embodiment of the present application, the establishing unit 407 includes:

[0138] The word segmentation module 4071 is configured to perform word segmentation on the notification data;

[0139] The calculation module 4072 is configured to calculate the weight of each word segment in the notification data to generate a word segment weight ratio;

[0140] The first obtaining module 4073 is configured to obtain the set of notification templates;

[0141] The matching module 4074 is configured to match the word segment weight ratio with the notification templates in the set of notification templates to obtain a matching result;

[0142] The second obtaining module 4075 is configured to obtain the matching result with the highest matching degree to obtain a target matching result;

[0143] The determining module 4076 is configured to determine a target document according to the target matching result;

[0144] The merging module 4077 is configured to merge the target document and the notification data to obtain a message template.

[0145] In the embodiment of the present application, the merging module is further configured to:

[0146] Merge the target document and the notification data to generate a merging result;

[0147] Obtain template data;

[0148] Determine whether the format of the merging result matches the template data;

[0149] When the judgment result of the judgment sub-module is a match, determine the merging result as the message template.

[0150] In this embodiment, the functions of each unit correspond to the steps in the foregoing Figure 2 illustrated embodiment, and will not be elaborated herein.

[0151] Please refer to Figure 5 , another embodiment of the apparatus for analyzing user behavior data provided by the embodiments of the present application includes:

[0152] A processor 501, a memory 502, an input / output unit 503, and a bus 504;

[0153] The processor 501 is connected to the memory 502, the input / output unit 503, and the bus 504;

[0154] The processor 501 specifically executes Figures 1 to 2 the operations corresponding to the steps in the method of

[0155] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, apparatuses, and units can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0156] In several embodiments provided by the present application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of the apparatuses or units can be in electrical, mechanical, or other forms.

[0157] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0158] In addition, the functional units in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0159] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, read-only memory), random access memories (RAM, random access memory), magnetic disks, or optical discs that can store program codes.

Claims

1. A method for analyzing user behavior data, characterized in that, it includes: Obtain user information, where the user information includes the user's identity information and the user's message notification click record; Obtain notification information according to the user information, and the notification information is the notification information confirmed by the user's click; Perform clustering analysis on the notification information according to the unsupervised algorithm to obtain notification data, where the notification data includes the word segmentation results of all the user's notification information and the notification information; Establish the relationship between the notification data and the notification template to obtain a message template; Bind the message template to the user information to obtain a binding result, and the binding result carries a label for searching the message template; After obtaining the user information, the method further includes: Obtain a rule table, where the rule table is used to store the analysis format of the user information; Determine whether there is matching data for the user information in the rule table; If there is, determine the message template matching the user information through the rule table; After binding the message template to the user information, the method further includes: Synchronize the binding result to the rule table; The establishing the relationship between the notification data and the notification template to obtain a message template includes: Perform word segmentation on the notification data; Calculate the weight of each word segment in the notification data to generate a word segment weight ratio; Obtain the collection of notification templates; Match the word segment weight ratio with the notification templates in the collection of notification templates to obtain a matching result; Obtain the matching result with the highest matching degree to obtain a target matching result; Determine the target notification template according to the target matching result; Merge the target notification template and the notification data to obtain a message template; Merging the target document and the notification data to obtain a message template includes: Merge the target document and the notification data to generate a merge result; Obtain template data; Determine whether the format of the merge result matches the template data; If it matches, determine the merge result as the message template.

2. A device for analyzing user behavior data, characterized in that, it includes: A first obtaining unit for obtaining user information, where the user information includes the user's identity information and the user's message notification click record; A second obtaining unit for obtaining notification information according to the user information, and the notification information is the notification information confirmed by the user's click; A clustering analysis unit for performing clustering analysis on the notification information according to the unsupervised algorithm to obtain notification data, where the notification data includes the word segmentation results of all the user's notification information and the notification information; An establishing unit for establishing the relationship between the notification data and the notification template to obtain a message template; A binding unit for binding the message template to the user information to obtain a binding result; The device further includes: A third obtaining unit for obtaining a rule table, where the rule table is used to store the analysis format of the user information; A judging unit for judging whether there is matching data for the user information in the rule table; A determination unit, configured to determine, when the determination result of the judgment unit is "exists", a message template that matches the user information through the rule table; A synchronization unit, configured to synchronize the binding result to the rule table; The establishment unit includes: A word segmentation module, configured to perform word segmentation on the notification data; A calculation module, configured to calculate the weight of each word segment in the notification data, and generate a word segment weight ratio; A first acquisition module, configured to acquire the notification template set; A matching module, configured to match the word segment weight ratio with the notification templates in the notification template set, and obtain a matching result; A second acquisition module, configured to acquire the matching result with the highest matching degree to obtain a target matching result; A determination module, configured to determine a target document according to the target matching result; A merging module, configured to merge the target document and the notification data to obtain a message template; The merging module is further configured to: Merge the target document and the notification data to generate a merging result; Acquire template data; Determine whether the format of the merging result matches the template data; When the determination result of the judgment sub-module is "matched", determine the merging result as the message template.

Images