CN117668349A

CN117668349A - Information recommendation method, electronic equipment and server

Info

Publication number: CN117668349A
Application number: CN202211048901.9A
Authority: CN
Inventors: 赵洋; 刘辉; 肖兆鸾
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2022-08-30
Filing date: 2022-08-30
Publication date: 2024-03-08
Also published as: WO2024046081A1

Abstract

An information recommendation method, electronic equipment and a server relate to the technical field of information recommendation and can realize safe and reliable information recommendation, the method is applied to the electronic equipment or a component (such as a chip system) supporting the functions of the electronic equipment, and the method comprises the following steps: and acquiring first behavior data generated when the first user uses the electronic equipment, and sending a first desensitization request to a server, wherein the first desensitization request comprises the first behavior data and does not carry an identification of the first user. Thereafter, a first interest tag is received from the server, the first interest tag being associated with the first row of data. And then, sending the first interest tag and the identification of the first user to the server, so that the server updates portrait information of the first user according to the first interest tag and the identification of the first user.

Description

Information recommendation method, electronic equipment and server

Technical Field

The application relates to the technical field of terminals, in particular to an information recommendation method, electronic equipment and a server.

Background

Currently, a user may perform a search and browse service of content information through an Application (APP) such as a browser installed in a terminal device, and a server may recommend service content suitable for the user to the user according to a search keyword of the user.

However, when the user uses the browser, the terminal may report personal behavior data related to the privacy of the user to the server, resulting in revealing the privacy of the user to the server. Therefore, how to provide a secure information recommendation method is a technical problem to be solved.

Disclosure of Invention

The application provides an information recommendation method, electronic equipment and a server, which can realize safe and reliable information recommendation.

In order to achieve the above purpose, the embodiment of the present application provides the following technical solutions:

in a first aspect, the present application provides an information recommendation method applied to an electronic device or a component (such as a chip system) supporting functions of the electronic device, where the method includes:

and acquiring first behavior data generated when the first user uses the electronic equipment, and sending a first desensitization request to the server, wherein the first desensitization request comprises the first behavior data and does not carry the identification of the first user. Thereafter, a first interest tag is received from the server, the first interest tag being associated with the first row of data. And then, the first interest tag and the identification of the first user are sent to the server, so that the server updates portrait information of the first user according to the first interest tag and the identification of the first user.

Illustratively, as shown in fig. 7, the mobile phone collects first behavior data (such as searching for "price of the world" and related click, browse data) generated when the first user uses the mobile phone, and sends a first desensitization request to the server, where the first desensitization request includes the first behavior data and does not carry the identification of the first user. The handset then receives a first interest tag (e.g., category tag: SUV car, keyword tag: hua as a question) from the server. And then, the mobile phone sends a first interest tag (such as a classification tag: SUV automobile, keyword tag: hua as a boundary) and the identification of the first user to the server, so that the server updates the portrait information of the first user according to the first interest tag and the identification of the first user.

When the electronic device reports the first row of data to the server, the electronic device does not correlate with the identifier of the first user, so that the server cannot obtain the private data of a specific user. Subsequently, the electronic device may acquire a first interest tag obtained by desensitizing the first data, associate the first interest tag with the first user identifier, and report the first interest tag and the associated first user identifier to the server. In this way, the server can update the representation information of the first user based on the desensitized first interest tag and the identification of the first user. On the one hand, the first interest tag does not contain the privacy data of the first user after desensitization, so that the disclosure of the privacy of the user can be avoided, and the safety of the user information is improved. On the other hand, when the server determines the user portrait, the server can be associated with a specific first user according to the identification of the first user, so that the more accurate portrait can be determined for the first user in a targeted manner according to the characteristics of the first user.

In one possible design of the first aspect, after sending the first interest tag and the identification of the first user to the server, the method further comprises:

acquiring first recommendation information from a server, wherein the first recommendation information is determined by the server according to the updated portrait information of the first user;

and displaying the first recommendation information.

The server determines that the user preference is a boundary and the recommendation information of the new energy automobile tag according to the updated portrait information of the first user, and returns the recommendation information to the mobile phone. The mobile phone displays recommendation information 701 and 702 as in fig. 6 (b) according to the recommendation information.

Therefore, the server can accurately determine the recommendation information of the user preference according to the accurate user portrait, and then returns the recommendation information to the electronic equipment, so that the electronic equipment can recommend the recommendation information for the user, the interaction requirement of the user is met, and the human-computer interaction experience is improved.

In one possible design of the first aspect, displaying the first recommendation information includes:

and displaying the first recommendation information in a recommendation module of the application program, or displaying the first recommendation information in a recommendation card of the negative screen, or displaying a system recommendation message, wherein the recommendation message comprises the first recommendation information.

Illustratively, as in fig. 4 (b), the handset displays the first recommendation information 501 in the recommendation interface 50 of the browser. Alternatively, as in fig. 4 (c), the mobile phone displays the first recommendation information 901 in the recommendation card of the negative one screen 90. Alternatively, as shown in fig. 4 (d), the mobile phone displays a system recommendation message 1001, the recommendation message 1001 including first recommendation information. Therefore, the electronic equipment can recommend information to the user in different modes so as to meet the interaction requirement of the user and improve the man-machine interaction efficiency.

the updated portrait information of the first user is acquired from the server.

In one possible design of the first aspect, the first behavior data and the privacy data of the first user are uncorrelated. Illustratively, as in (a) of FIG. 4, the first user enters the keyword "pet dog," and subsequently, the user may click on, browse the search entries in the search results page, which data does not relate to the first user's privacy data.

In one possible design of the first aspect, sending the first interest tag and the identification of the first user to the server includes:

The method comprises the steps of sending a first interest tag, a second interest tag and an identifier of a first user to a server, so that the server updates portrait information of the first user according to the first interest tag, the second interest tag and the identifier of the first user; the second interest tag is an interest tag corresponding to the historical portrait information of the first user.

For example, as shown in fig. 7, the mobile phone may send a first interest tag (such as a classification tag: SUV car, keyword tag: hua as a question), a second interest tag (such as a history classification tag: new energy car, history classification tag: middle-large car) and an identification of the first user to the server, so that the server updates the portrait information of the first user according to the first interest tag, the second interest tag and the identification of the first user. The updated first user representation is as follows: classification label { New energy automobile: 0.65}; classification tag: { medium and large cars: 0.5}; classification tag: { SUV automobile: 0.6}; keyword tag: { Hua is a boundary: 0.6}.

Therefore, the user portrait can be updated by utilizing the historical search interest tags and the search interest tags corresponding to the current behaviors, and the precision of the user portrait is improved.

In one possible design of the first aspect, before the first row of data generated when the first user uses the electronic device is collected, a first interface is displayed, where the first interface includes second recommendation information, and the second recommendation information is different from the first recommendation information. Illustratively, the cell phone may recommend recommendation information (second recommendation information) such as SUV car to the user before the user inputs "price of the world" in the search box 601 as shown in fig. 6 (a). After the user inputs "price of the world" in the search box 601 shown in fig. 6 (a), the mobile phone interacts with the server, and the server can update the user's portrait according to the user's search interest tag, and return new first recommendation information (such as recommendation information 701 and 702 shown in fig. 6 (b)) to the mobile phone according to the updated user portrait.

In one possible design of the first aspect, the method further comprises:

collecting second behavior data generated when the first user uses the electronic equipment;

sending a second desensitization request to the server, wherein the second desensitization request comprises second behavior data and does not carry the identification of the first user;

a return message is obtained from the server for the second desensitization request, the return message indicating that the second behavior data is related to the privacy data of the first user.

Illustratively, as in fig. 10, the handset sends a second desensitization request to the server, the second desensitization request including "depression" related search data (second behavioral data) that relates to user privacy. The handset then receives a return message from the server. Optionally, the return message comprises an update identity (first field) indicating that no updated interest tag is present, since the second behavior data relates to user privacy.

In one possible design of the first aspect, the return message does not include an interest tag associated with the second behavior data. Illustratively, as in FIG. 10, the return message includes an update identification, but does not include an interest tag. Therefore, the mobile phone can determine that the updated interest tag does not exist according to the update identification, and then the search interest tag for constructing the user portrait is not reported to the server any more, so that the privacy of the user is prevented from being revealed in the process of constructing the user portrait.

In one possible design of the first aspect, the method further comprises:

displaying third recommendation information before the second behavior data are acquired;

displaying fourth recommendation information after acquiring the return message from the server; the third recommendation information is the same as the fourth recommendation information. For example, before the user inputs "symptoms of depression" in the search box 801 shown in fig. 9 (a), the mobile phone may recommend recommendation information (third recommendation information) such as new energy automobiles, a warrior, and the like to the user. After the user inputs "symptoms of depression" in the search box 801 shown in fig. 9 (a), the mobile phone still displays fourth recommendation information 701 and 702 identical to the third recommendation information since there is no updated search interest tag as in fig. 9 (b).

In one possible design of the first aspect, the first row of data includes one or more of the following information: click behavior related information, search behavior related information, browse behavior related information.

In one possible design of the first aspect, in view of the fact that the above desensitization process (calculating the search interest tag) does not need to be related to the user person, but the end user portrayal calculation and personalized recommendation service needs to be related to the user person, in order to avoid revealing user privacy during user portrayal processing, the data between the desensitization module and the user portrayal module may be set to be isolated from each other, and not revealed from each other, so as to improve the security of the user data.

In a second aspect, there is provided an information recommendation method applied to a server or a component (such as a chip system) supporting a server function, the method comprising:

receiving a first desensitization request from the electronic device, wherein the first desensitization request comprises first behavior data generated when a first user uses the electronic device and does not carry an identification of the first user;

according to the first behavior data, a first interest tag is sent to the electronic equipment, and the first interest tag is associated with the first behavior data;

receiving a first interest tag from an electronic device and an identification of a first user;

And updating the portrait information of the first user according to the first interest tag and the identification of the first user.

In one possible design of the second aspect, the method further comprises sending the first interest tag to the electronic device:

determining at least one target keyword matched with the first row of data in the white list according to the white list;

and carrying out cluster analysis on at least one target keyword to obtain a first interest tag.

Illustratively, as shown in fig. 7, the mobile phone sends "price of the question" related search data (first row of data) to the server, and the server determines, according to the keyword white list 1022, that there is a target keyword "question" in the white list, where the target keyword matches the search data of the user at this time. The server can perform cluster analysis on the target keywords to obtain classification labels: SUV automobile and keyword tag: the bloom is a boundary.

In one possible design of the second aspect, the whitelist includes a plurality of keywords, the plurality of keywords not related to privacy terms.

Thus, the privacy data can be filtered out from the white list word stock, and the desensitization of the first row of data is realized. By maintaining and updating the white list word stock, the precision requirement of the recommendation system on the user portrait can be met to the maximum extent while the privacy data is desensitized. For example, the white list word stock is updated periodically, and non-sensitive keywords with high service recommendation value are put into the white list word stock.

In a third aspect, an information recommendation apparatus is provided, applied to an electronic device or a corresponding component (such as a chip system), the apparatus comprising:

the processing module is used for collecting first behavior data generated when the first user uses the electronic equipment;

the communication module is used for sending a first desensitization request to the server, wherein the first desensitization request comprises first behavior data and does not carry the identification of a first user;

the communication module is further used for receiving a first interest tag from the server, wherein the first interest tag is associated with the first row of data;

and the communication module is also used for sending the first interest tag and the identification of the first user to the server so that the server updates the portrait information of the first user according to the first interest tag and the identification of the first user.

In one possible design of the third aspect, the communication module is further configured to obtain first recommendation information from the server, where the first recommendation information is recommendation information determined by the server according to the updated portrait information of the first user;

the state further includes: and the display module is used for displaying the first recommendation information.

In one possible design of the third aspect, displaying the first recommendation information includes:

In one possible design of the third aspect, the communication module is further configured to, after sending the first interest tag and the identification of the first user to the server, obtain the updated representation information of the first user from the server.

In one possible design of the third aspect, the first behavior data and the privacy data of the first user are uncorrelated.

In one possible design of the third aspect, sending the first interest tag and the identification of the first user to the server includes:

In one possible design of the third aspect, the communication module is further configured to display a first interface before acquiring the first data generated when the first user uses the electronic device, where the first interface includes second recommendation information, and the second recommendation information is different from the first recommendation information.

In one possible design of the third aspect, the processing module is further configured to collect second behavior data generated when the first user uses the electronic device;

the communication module is also used for sending a second desensitization request to the server, wherein the second desensitization request comprises second behavior data and does not carry the identification of the first user;

the communication module is further configured to obtain a return message from the server for the second desensitization request, the return message indicating that the second behavior data is related to the privacy data of the first user.

In one possible design of the third aspect, the return message does not include an interest tag associated with the second behavior data.

In one possible design of the third aspect, the display module is further configured to:

displaying fourth recommendation information after acquiring the return message from the server; the third recommendation information is the same as the fourth recommendation information.

In one possible design of the third aspect, the return message includes a first field, a value of the first field being used to indicate whether an updated interest tag exists.

In one possible design of the third aspect, the first row of data includes one or more of the following information: click behavior related information, search behavior related information, browse behavior related information.

In a fourth aspect, an information recommendation apparatus is provided, for application to a server or related component (such as a chip system), the apparatus comprising:

the communication module is used for receiving a first desensitization request from the electronic equipment, wherein the first desensitization request comprises first behavior data generated when a first user uses the electronic equipment and does not carry an identification of the first user;

the communication module is further used for sending a first interest tag to the electronic equipment according to the first behavior data, and the first interest tag is associated with the first behavior data;

the communication module is also used for receiving the first interest tag and the identification of the first user from the electronic equipment;

and the processing module is used for updating the portrait information of the first user according to the first interest tag and the identification of the first user.

In one possible design of the third aspect, the processing module is further configured to determine, before sending the first interest tag to the electronic device, at least one target keyword in the white list that matches the first behavior data according to the white list; and carrying out cluster analysis on at least one target keyword to obtain a first interest tag.

In one possible design of the third aspect, the whitelist includes a plurality of keywords, the plurality of keywords not related to privacy terms.

In a fifth aspect, the present application provides an electronic device, including: an input device, a display screen, one or more processors, memory, and one or more computer programs; wherein the processor is coupled to the input device, the processor and the memory, the one or more computer programs being stored in the memory, the processor executing the one or more computer programs stored in the memory when the electronic device is running, to cause the electronic device to perform the method of any of the designs of the first aspect.

In a sixth aspect, the present application provides an apparatus comprising a processor and a memory for storing computer program code, the computer program code comprising computer instructions which, when executed by the processor, perform a method as in any of the possible designs of the first or second aspects of the present application.

In a seventh aspect, a technical solution of the present application provides a computer readable storage medium, including computer instructions, which when executed on an electronic device, cause the electronic device to perform the method in any one of the possible designs of the first aspect.

In an eighth aspect, the present application provides a computer readable storage medium, including computer instructions, which when executed on a server, cause the server to perform the method in any one of the possible designs of the second aspect.

In a ninth aspect, the present application provides a computer program product for causing an electronic device to perform the method of any one of the possible designs of the first aspect, when the computer program product is run on the electronic device.

In a tenth aspect, the present application provides a computer program product, which when run on a server causes the server to perform the method of any one of the possible designs of the second aspect.

In an eleventh aspect, the present application provides an information recommendation system, where the system includes an electronic device in any one of the possible designs of the first aspect and a server in any one of the possible designs of the second aspect.

Drawings

Fig. 1 is a schematic architecture diagram of an information recommendation system according to an embodiment of the present application;

FIG. 2 is a schematic diagram of interaction between devices according to an embodiment of the present disclosure;

fig. 3 is a schematic structural diagram of an electronic device or a server according to an embodiment of the present application;

FIG. 4 is a set of interface schematic diagrams provided in an embodiment of the present application;

fig. 5 is a flow chart of an information recommendation method according to an embodiment of the present application;

FIG. 6 is a set of interface schematic diagrams provided in an embodiment of the present application;

fig. 7 is a schematic diagram of interaction between an electronic device and a server according to an embodiment of the present application;

Fig. 8 is a flow chart of an information recommendation method according to an embodiment of the present application;

FIG. 9 is a set of interface schematic diagrams provided in an embodiment of the present application;

fig. 10 is a schematic diagram of interaction between an electronic device and a server according to an embodiment of the present application;

fig. 11 is a schematic structural diagram of an information recommendation device provided in an embodiment of the present application;

fig. 12 is a schematic structural diagram of a chip system according to an embodiment of the present application.

Detailed Description

It should be noted that the terms "first" and "second" and the like in the description and the drawings of the present application are used for distinguishing between different objects or for distinguishing between different processes of the same object. The words "first," "second," and the like may distinguish between identical or similar items that have substantially the same function and effect. For example, the first device and the second device are merely for distinguishing between different devices, and are not limited in their order of precedence. It will be appreciated by those of skill in the art that the words "first," "second," and the like do not limit the amount and order of execution, and that the words "first," "second," and the like do not necessarily differ.

"at least one" means one or more, and "a plurality" means two or more.

"and/or", describes an association relationship of an association object, and indicates that there may be three relationships, for example, a and/or B, and may indicate: a alone, a and B together, and B alone, wherein a, B may be singular or plural. The character "/" generally indicates that the context-dependent object is an "or" relationship. "at least one of" or the like means any combination of these items, including any combination of single item(s) or plural items(s). For example, at least one (one) of a, b, or c may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be single or plural.

Furthermore, references to the terms "comprising" and "having" and any variations thereof in the description of the present application are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those listed but may optionally include other steps or elements not listed or inherent to such process, method, article, or apparatus.

It should be noted that, in the embodiments of the present application, words such as "exemplary" or "such as" are used to mean serving as an example, instance, or illustration. Any embodiment or design described herein as "exemplary" or "for example" should not be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "such as" is intended to present related concepts in a concrete fashion.

In the related art, a terminal may collect an original search term of a user through a browser, and upload desensitized user data to a server through desensitization processing, where the server recommends a corresponding service, such as recommending an advertisement, for the user based on the desensitized user data. Among them, desensitization processing methods include, but are not limited to, variable identifiers, joint group learning (federated learning of cohorts, FLOC), data obfuscation processing, and the like.

Although the method can protect the privacy of the user to a certain extent, as the user data obtained by the server is the data after desensitization, the user data after desensitization cannot be directly related to the user, so that the server cannot provide personalized services for different users.

In order to improve the wedge degree of service recommendation, an embodiment of the present application provides an information recommendation method. The technical scheme of the embodiment of the application can be applied to a service recommendation system. An exemplary architecture of a service recommendation system is shown in fig. 1. The system may comprise a terminal 101, a service server 102.

Optionally, the service recommendation system may further comprise an engine server 103 (not shown in fig. 1).

The terminal 101 may be connected to the service server 102 and the engine server 103 through a network, and in this embodiment of the present application, a communication protocol on which connection depends is not limited in a manner of establishing connection between the terminal 101 and the service server 102 and the engine server 103. Fig. 2 shows a partial interaction flow among the terminal 101, the service server 102, and the engine server 103.

Alternatively, applications such as a browser, news information and the like may be installed in the terminal 101, and a user may search through the applications to browse various information contents. Taking the example of searching and browsing by the user through the browser, the terminal 101 may collect the search word input by the user in the browser and send the search word to the engine server 103, and the engine server 103 may return the search result to the terminal 101. The search results may include one or more search terms. The terminal 101 may present the search results through a browser and collect browsing, clicking and collecting information of the user on the search items in the search results. Thereafter, the terminal 101 may anonymously upload the search data of the user to the content desensitization analysis service 1021 of the service server 102 through the browser. The user's search data includes, but is not limited to, the user's search terms, information of the search terms clicked/browsed by the user. The terminal 101 may also receive a search interest tag (which may also be referred to as a search behavior tag, a search data feature tag, a search feature tag, etc.) corresponding to the search data from the service content desensitization analysis service 1021, and upload the search interest tag and the user identification to the user portrayal service 1023 of the service server 102, so that the user portrayal service 1023 determines a user portrayal (which may also be referred to as a user search interest portrayal, etc.) according to the search interest tag and the user identification.

User portrayal, namely user information tagging, can be used for carrying out attribute depiction on users or product features by collecting various dimension data such as user social attributes, consumption habits, preference features and the like, so that the information overall view of a user is abstracted. The user Z registers an account number on a certain electronic commerce platform, a series of behaviors such as browsing, searching, collecting commodities, ordering and shopping are carried out on the web end/APP end of the electronic commerce platform for a period of time, the electronic commerce platform database records the behaviors of the user on the platform in the whole course, and the user Z is labeled according with the characteristics of the user Z through a series modeling algorithm. After that, the user Z can find the commodity which he wants to buy on the relevant recommended edition of the e-commerce platform.

Wherein, the user tag as in table 1-1 may generally comprise: tag id, tag name, the number of user actions corresponding to the tag (e.g., the keyword "big data" is searched twice), action type (different action types correspond to different wishles of the user on the commodity, e.g., purchase of a commodity > collection of a commodity > browse of a commodity), action time (the more distant time has less influence on the current user, e.g., search a book of college entrance examination 5 years ago, but now search a book of college entrance examination).

TABLE 1-1

Tag ID	Label name	Number of behaviours	Behavior type ID	Behavior type	Time of action	Tag weights
							111	Big data	2	1	Searching	2022-01-01	2.5
221	Pet dog	2	2	Browsing	2022-03-01	5.6
							331	Automobile	1	3	Purchasing	2022-04-01	3.2

In this embodiment of the present application, the terminal 101 anonymously reports the search data, which means that: the terminal 101 reports the search data and does not report the self-related or user-related identification to the content desensitization analysis service 1021. In this way, the content desensitization analysis service 1021 cannot know the specific terminal that reports the search data, and thus the privacy of the terminal or the user of the terminal cannot be revealed.

Depending on the division of functions and whether sensitive data processing is involved, the traffic server 102 may include services as shown in tables 1-2 below:

TABLE 1-2

In the embodiment of the present application, the content desensitization analysis service 1021 and the user portrayal service 1023 are used for processing different data, and the data between the content desensitization analysis service 1021 and the user portrayal service 1023 are isolated from each other so as to reduce the probability of user privacy disclosure. Optionally, the content desensitization analysis service 1021 is configured to obtain user data (such as a search term input by a user in a browser) anonymously reported by the terminal 101, and perform desensitization processing on the anonymous user data from the terminal 101, so as to obtain a desensitized search interest tag corresponding to the user data.

Optionally, the service server 102 may further include a keyword white list 1022, where the keyword white list 1022 includes a huge number (such as millions) of keywords. These keywords match the business recommendation scope of the business recommendation service 1025, and the keywords in the keyword whitelist 1022 do not contain privacy related keywords. For example, a user searches for health care related information in a browser, and the business recommendation service 1025 core provides entertainment news and other vertical information, does not provide health care related information content, and the health care related information belongs to sensitive data, which increases the risk of privacy disclosure once entering the user portrayal service. Thus, keyword whitelist 1022 contains entertainment news related keywords and does not contain health care related keywords (e.g., depression).

As one possible implementation manner, the content desensitization analysis service 1021 performs desensitization processing on the private data, which may be implemented as follows: according to the search data searched by the user, searching a keyword white list 1022, if keywords matched with the search data of the user exist in the keyword white list 1022, the content desensitization analysis service may perform grouping clustering on the search data to obtain a search interest tag corresponding to the search data, and return the desensitized search interest tag to the terminal 101. Otherwise, if the search data searched by the user does not belong to the keyword white list 1022, the content desensitization analysis service 1021 does not return a corresponding search interest tag to the terminal 101.

In this way, the content desensitization analysis service can identify data having actual value for the service recommendation service from massive search data of users through the keyword white list 1022, and complete privacy desensitization of the search data and validity matching of the service recommendation service. Wherein, for the search data of the keywords matched with the keyword white list 1022, the search data is matched with the business of the business recommendation service. For search data for which there are no matching keywords in the keyword white list 1022, the search data relates to user privacy and does not match the business of the business recommendation service. Thus, the user portraits are not computed using this type of search data that relates to user privacy, so as not to compromise user privacy. Therefore, by setting the keyword white list 1022, the data for user portrait calculation can be ensured, privacy sensitive data is not contained, and the data is in the effective range of the service recommendation service, so that the probability of calculating the user portrait by adopting invalid data is reduced, the processing efficiency of the user portrait calculation is improved, and the risk of revealing the user privacy data can be reduced.

For example, assuming that the user inputs the search keyword "price of the world" in the browser as in (a) of fig. 6, the browser may transmit the search keyword to the engine server 103. The engine server 103 returns relevant search results to the browser. The browser presents a search results interface 60 as shown in fig. 6 (a). The user may click on and browse the search entries in the interface 60, and the browser may collect user browsing and clicking information on the search entries and send the user search data anonymously to the content desensitization analysis service 1021. The content desensitization analysis service 1021 may perform desensitization processing on the search data of the user, and generate a corresponding search interest tag. For example, the content desensitization analysis service performs cluster analysis on the search data, determines one or more classifications corresponding to the search data, and generates a classification label { SUV car }: 0.3} (shown in bold font), etc. The SUV is a characteristic value of a classification label, which indicates that the classification corresponding to the search data of the user is SUV, and 0.3 is a weight corresponding to the classification. The weight of the tag may be used to characterize the degree of similarity or association between the user interest and the tag. The greater the weight of the tag, the more interesting the user is to the information (such as articles) associated with the tag. Conversely, the smaller the weight of the tag, the less interested the user is in the tag-associated information (e.g., the tag-associated push message).

Alternatively, the weight of the tag may be calculated by using TF-IDF algorithm or other algorithms, and the embodiment of the present application does not limit the specific calculation manner of the weight.

For another example, the content desensitization analysis service generates keyword tags { boundary: 0.8} (shown in bold font). Wherein the keyword labels are generated based on the keyword whitelist 1022, without regard to user privacy.

Optionally, the browser may report not only the search interest tag corresponding to the current search behavior of the user to the user portrait service 1023, but also the search interest tag in the history. For example, the user historically searches for search keywords related to automobiles and new energy vehicles, and obtains corresponding search interest tags from the service server 102. For example, as in FIG. 2, the historic search interest tag includes: classification label { New energy vehicle: 0.7} (shown underlined), class label { medium-large car: 0.6} (shown underlined).

The user portrait service 1023 is configured to obtain the search interest tag and the user identifier reported by the terminal 101, and construct a user portrait according to the search interest tag and the user identifier. Wherein the user representation is used to characterize the user's interest preferences in the service. The user portrayal service 1023 may also be used to send the generated user portrayal to the business recommendation service 1025. In some examples, the user portrait may include one or more search interest tags for the user. Illustratively, as in FIG. 2, the user's historical search interest tags include: classification label { New energy vehicle: 0.7} (shown underlined), class label { medium-large car: 0.6} (shown underlined). The user portrayal service 1023 can search for interest tags based on the user's history, determining a historical user portrayal: user 1 classification label { new energy vehicle: 0.7}, class label { medium and large car: 0.6}. The interest label of the current search of the user comprises the following steps: classification tag { SUV automobile: 0.3} (shown in bold font), keyword tag { boundary: 0.8} (shown in bold font). The user portrait service 1023 may determine the latest user portrait of the user based on the user's current search interest tag.

In consideration of the fact that compared with the current search interest tag, the influence degree of the historical search interest tag on the user portrait is small, the weight corresponding to each characteristic value in the historical search interest tag can be attenuated. For example, as shown in fig. 2, the interest tag { new energy vehicle ] is searched for histories in historic user portraits: 0.7} (shown in underline), the weight corresponding to the feature value "new energy car" is 0.7, and the user portrait service 1023 may attenuate the weight from 0.7 to 0.65 to obtain an updated classification label { new energy car, 0.65}, and the updated user portrait includes the updated classification label { new energy car, 0.65}. Similarly, for the history search interest tag { medium and large cars: 0.6} (shown in underline), user portrayal service 1023 decays the weight corresponding to the feature value "medium-large car" from 0.6 to 0.5 and updates the user portrayal according to the updated class label.

The service recommendation service 1025 is used for receiving the user portrait from the user portrait service 1023, determining the target service of interest to the user according to the user portrait, and sending the information of the target service to the terminal 101.

On the one hand, since the content desensitization analysis service of the service server 102 has already performed desensitization processing on the user data, the user portrait service determines the user portrait using the search interest tag after desensitization, without revealing the privacy of the user. On the other hand, the user portrayal is determined based on the user identification, i.e. the user portrayal can be associated to a specific user, so that personalized services with higher fit with the corresponding user can be recommended for different users.

Alternatively, the content desensitization analysis service may also be referred to as a desensitization service, or a desensitization module. The user portrayal service may also be referred to as a user portrayal management service, or a user portrayal management module.

Alternatively, the terminal 101 may be a mobile phone, tablet, personal computer, vehicle-mounted terminal, or the like. The embodiment of the present application does not limit the device configuration of the terminal 101.

The above is mainly exemplified by the content desensitization analysis service and the user portrayal service being located in the same service server 102, and in other embodiments, the content desensitization analysis service and the user portrayal service may be located in different servers.

Alternatively, the terminal and the server in the embodiments of the present application may be implemented by the electronic device in fig. 3. Fig. 3 is a schematic hardware structure of an electronic device according to an embodiment of the present application. The electronic device 200 comprises at least one processor 201, a memory 202 and at least one transceiver 203.

The processor 201 may be a general purpose central processing unit (central processing unit, CPU), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of programs in accordance with aspects of the present application.

Communication lines may be included between the components for communicating information between the components.

A transceiver 203 for communicating with other devices. In the embodiments of the present application, the transceiver may be a module, a circuit, a bus, an interface, or other devices capable of implementing a communication function, for communicating with other apparatuses. Alternatively, the transceiver may be a separately provided transmitter that is operable to transmit information to other devices, or a separately provided receiver that is operable to receive information from other devices. The transceiver may also be a component that integrates the functions of transmitting and receiving information, and the embodiments of the present application do not limit the specific implementation of the transceiver.

The memory 202 may be, but is not limited to, a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (random access memory, RAM) or other type of dynamic storage device that can store information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc-only memory (compact disc read-only memory) or other optical disk storage, a compact disc storage (including compact disc, laser disc, optical disc, digital versatile disc, blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. The memory may be stand alone and be coupled to the processor via a communication line. The memory may also be integrated with the processor.

The memory 202 is used for storing computer-executable instructions for implementing the embodiments of the present application, and is controlled by the processor 201 to execute the instructions. The processor 201 is configured to execute computer-executable instructions stored in the memory 202, thereby implementing the methods provided in the embodiments described below.

Alternatively, the computer-executable instructions in the embodiments of the present application may be referred to as application code, instructions, computer programs, or other names, and the embodiments of the present application are not limited in detail.

In a particular implementation, as one embodiment, processor 201 may include one or more CPUs, such as CPU0 and CPU1 of FIG. 2.

In a particular implementation, as one embodiment, the electronic device 200 may include multiple processors. Each of these processors may be a single-core (single-CPU) processor or may be a multi-core (multi-CPU) processor. A processor herein may refer to one or more devices, circuits, and/or processing cores for processing data (e.g., computer program instructions).

The electronic device may be any device having data processing functions, for example, a computer device having data processing functions. An exemplary block diagram of an electronic device is shown in fig. 3. It should be understood that the illustrated electronic device is only one example, and that in actual practice the electronic device may have more or fewer components than shown in fig. 3, may combine two or more components, or may have a different configuration of components.

Illustratively, as in (a) of fig. 4, the handset displays a browser interface 40, wherein the browser interface 40 includes a search box 401. The user may enter the keyword "pet dog" in search box 401 and click on search button 404. In response to a user clicking on the search button 404, the cell phone may send a content request to a search engine, which returns search entries related to "pet dog" to the cell phone according to the keyword "pet dog". As in fig. 4 (a), the handset may display search entries 402, 403, etc. returned by the search engine.

In the embodiment of the application, the service content related to the search behavior can be recommended to the user based on the search behavior of the user, so that the pertinence and the accuracy of service recommendation are improved. As a possible implementation manner, as shown in fig. 4, after detecting the operation of the user on the search item (such as clicking on the search item 402, 403), the mobile phone may anonymously send "pet dog" related search data to the content desensitization analysis service, and the content desensitization analysis service performs desensitization processing on the search data to obtain a desensitized search interest tag corresponding to the search data.

As one possible implementation manner, the content desensitization analysis service may search whether the keyword whitelist 1022 has a word similar to "pet dog", and if so, the keyword whitelist 1022 has a word similar to or similar to "pet dog", which indicates that the service corresponding to the search data related to "pet dog" belongs to the service recommendation range of the service recommendation service, and the search data related to "pet dog" does not relate to user privacy. Then, the content desensitization analysis service may perform a grouping clustering process on the search data related to the "pet dog" to obtain a search interest tag corresponding to the search data.

Optionally, the search interest tags include category tags and keyword tags. The category labels are used to characterize the search categories to which the search data matches. For example, as shown in fig. 4, assuming that the search types matched with the search data related to "pet dog" are "animal" and "pet" through the clustering process, the content desensitization analysis service may use "animal" and "pet" as the classification tags of the search data related to "pet dog". The content desensitization analysis service may return a category label for "pet dog" related search data to the cell phone.

As shown in fig. 4, after the content desensitization analysis service performs desensitization processing on the search data of the user to obtain the search interest tags (such as the classification tags and the keyword tags) corresponding to the search data, the search interest tags corresponding to the search data may be returned to the mobile phone.

In the interaction process of the mobile phone and the content desensitization analysis service, the mobile phone anonymously reports the search data of the user, so that the content desensitization analysis service cannot acquire the specific identity of the mobile phone or the mobile phone user reporting the search data, and the probability of privacy disclosure of the user can be reduced.

Then, as shown in fig. 4 (b), the mobile phone can associate the search interest tag (classified tag: animal/pet; keyword tag: pet dog) returned by the content desensitization analysis service with the user identification, and report the desensitized search interest tag and the associated user identification to the user portrait service, and the user portrait service determines the user portrait according to the search interest tag and the user identification.

As one possible implementation, to avoid the user portrayal service from retrieving the user's search data from the content desensitization analysis service, the content desensitization analysis service may be data isolated from the user portrayal service. Optionally, a firewall may be provided between the content desensitization analysis service and the user portrayal service to prevent data interaction between the user portrayal service and the content desensitization analysis service.

After the user portrayal service determines the user portrayal, the user portrayal service may transmit the determined user portrayal to a service recommendation service, which determines a target service matching the user portrayal according to the user portrayal, and returns information of the target service to the mobile phone, as shown in fig. 4. For example, the service recommendation service determines that the user likes animal, pet and dog related contents according to the user portrait, and the service recommendation service can provide related personalized information contents for the user. As shown in fig. 4 (b), the handset may display information 501 such as animal, pet, dog related information on the recommendation interface 50.

In the scheme, in the process of determining the user portrait by the user portrait service, the search interest tag acquired by the user portrait is data subjected to desensitization, so that the user portrait acquired based on the desensitization data does not generally relate to user privacy, and the risk of user privacy leakage in the process of constructing the user portrait can be reduced. In addition, the data between the user portrait service and the content desensitization analysis service are mutually isolated, so that the user portrait service cannot acquire search data possibly related to the user privacy from the content desensitization analysis service, and the leakage risk of the user privacy can be further reduced.

The above is mainly taken as an example that the mobile phone presents the related recommended service in the browser, and in other embodiments, the mobile phone may present the related recommended service in other application programs.

Illustratively, as in fig. 4 (a), the user inputs a search keyword "pet dog" in the browser, the mobile phone transmits the search keyword to the search engine, and receives a search result determined according to the search keyword from the search engine. As in fig. 4 (a), the handset may display the search entries 402, 403, etc. in the search results in the browser. The user may browse and click on the search term. The mobile phone can anonymously report the user search data (including search keywords, search items browsed and clicked by the user, and the like) to a content desensitization analysis service on the server side, and the content desensitization analysis service performs desensitization processing on the search data to generate and return a desensitized search interest label to the mobile phone. The mobile phone can send the search interest label returned by the mobile phone and the associated user identification to the user portrait service on the server side, and the user portrait service can determine the user portrait of the specific user according to the desensitized search interest label and the user identification. The business recommendation service can determine a target service which the user wants to use based on the user portrait of the specific user, and return information of the target service to the mobile phone. After the handset receives the information 901 for the target service from the server, the information 901 for the target service may be presented in the negative one-screen interface 90, as in fig. 4 (c).

As another example, as shown in fig. 4 (d), after the mobile phone receives the information of the target service from the server, as shown in fig. 4 (c), a push message 1001 may be displayed, where the push message includes the information of the target service (such as a feeding method of the pet dog).

The technical solution of the embodiments of the present application is described in detail below by taking a terminal as an example of a mobile phone. Referring to fig. 5, an exemplary flow of the information recommendation method according to the embodiment of the present application is shown. The method may comprise the steps of:

s101, the mobile phone browser receives keywords input by a user.

Illustratively, as in fig. 6 (a), the handset displays a browser interface 60, where the browser interface 60 includes a search box 601. The user may enter a search keyword in the search box 601 and click the search button 604. Such as input: but is the price of the question and clicks the search button 604.

S102, the mobile phone sends the information of the keywords to the search engine.

It should be appreciated that after the handset receives the keyword entered by the user in the search box 601, the keyword may be sent to the server of the search engine.

S103, the search engine returns search items corresponding to the keywords to the mobile phone.

And the server of the search engine returns the search result corresponding to the keyword to the mobile phone according to the search keyword 'price of being a question' input by the user. The search results may include a plurality of search terms. The mobile phone can display a plurality of search items corresponding to the keywords. For example, as shown in fig. 6 (a), the mobile phone displays a search entry 602 and a search entry 603 corresponding to the keyword "price of the world" and the like.

S104, the mobile phone determines search data according to the operation of the user on the browsing items.

The mobile phone browser displays the search items, and the user can browse and click the search items displayed by the mobile phone. The mobile phone can determine search data related to the searching, browsing and other actions of the user according to clicking, browsing and other operations of the user. Tables 1-3 illustrate examples of search data, as examples. It should be appreciated that the search data may also include more or less data than tables 1-3, and the embodiments of the present application do not limit the type, amount, etc. of search data.

Tables 1 to 3

/>

S105, the mobile phone sends a data desensitization request to the content desensitization analysis service.

In the embodiment of the application, accurate service recommendation can be provided for the user based on the searching behavior of the user. It should be appreciated that, in order to reduce the risk of privacy disclosure of user data during service recommendation, the handset may report a data desensitization request to the cloud-side service. The data desensitization request is for requesting desensitization processing of the search data. Optionally, the mobile phone sends a data desensitization request to a first interface of the content desensitization analysis service.

In the embodiment of the application, in order to reduce the risk of disclosure of user privacy, the data desensitization request does not carry the user identifier. The user identification may be data that can be used to correlate to personal information of the user, for example, the user identification may include, but is not limited to, any one or more of the following: a device identifier used by the user, a cell phone number, an internet protocol (internet protocol, IP) address of the device, a geographic location where the device is located, registration account information for device login, a cookie, login time, etc. That is, the data desensitization request is a request sent anonymously by the handset. The content desensitization analysis service cannot be associated to a specific user or handset through the data desensitization request.

TABLE 2

And S106, the content desensitization analysis service performs desensitization processing on the search data to obtain a corresponding search interest tag.

Wherein the search interest tags may be used to construct a user representation.

As a possible implementation manner, the content desensitization analysis service searches the keyword whitelist 1022, and if the keyword whitelist 1022 includes keywords matched with the search data, the content desensitization analysis service may analyze the features of the search data by using a content analysis algorithm model (such as a cluster analysis model) to obtain a search interest tag corresponding to the search data.

Optionally, the search interest tags include category tags and keyword tags. Each search data may correspond to one or more category labels. The category labels are used to characterize the type to which the search data belongs. The keyword tag is used to characterize keywords that match the user's search data.

Illustratively, as shown in fig. 7, the mobile phone anonymously reports search data corresponding to the current search behavior of the user to the content desensitization analysis service, wherein the search data comprises latest search data related to 'price of the price as a question'. The content desensitization analysis service is searched to determine that the keyword "Hua as a question" matching the search data exists in the keyword white list 1022, meaning that the search data of the user does not relate to private data. The content desensitization analysis service may perform a grouping cluster analysis on the search data to obtain a search interest tag matched with the search data and a weight corresponding to the corresponding search interest tag. For example, as shown in fig. 7, the content desensitization analysis service may perform a grouping cluster analysis on search data related to "price of a water as a boundary" through a grouping cluster model to generate a classification tag of a sport utility vehicle (sport utility vehicle, SUV) and a keyword tag of "water as a boundary". The weight corresponding to the SUV automobile classification label is 0.8, and the weight corresponding to the Chinese boundary keyword label is 0.6.

And S107, the content desensitization analysis service returns a desensitization result corresponding to the search data to the mobile phone.

Wherein the desensitization result includes updated search interest tags. Optionally, the desensitization result includes an update identifier, where the update identifier is used to indicate whether the search interest tag has an update. Illustratively, the desensitization results may include the parameters queryTag and isUpdate shown in Table 3-1. Wherein the queryTag is an updated searching interest tag, and the isUpdate is an update identifier.

TABLE 3-1

S108, the mobile phone sends the user identification and the search interest label corresponding to the search data to the user portrait service.

It can be appreciated that after the mobile phone receives the desensitized search interest tag from the content desensitization analysis service, the search interest tag and the user identifier can be associated, and the search interest tag and the associated user identifier can be reported to the user portrait service, so that the user portrait service builds a user portrait with higher degree of agreement with the user according to the search interest tag and the associated user identifier.

As a possible implementation, the mobile phone sends the search interest tag and the user identification to the second interface of the user portrait service.

Optionally, the search interest tag may include search data corresponding to the actions such as searching and browsing, and search interest tags corresponding to the actions such as searching historically. For example, a user historically inputs search data such as "new energy automobiles" in a mobile phone browser, and the mobile phone can receive search interest tags "new energy automobiles" and "medium-sized and large-sized cars" from a content desensitization analysis service. Alternatively, the cell phone may store the search interest tag. Optionally, in order to occupy the storage space of the mobile phone for a long time, the mobile phone can store the search interest tag in a preset period, and automatically delete the search interest tag after the preset period. Alternatively, the cell phone may delete the search interest tag according to the user's instruction. The embodiment of the application does not limit the time for storing the search interest labels, the time for deleting the search interest labels and the like.

Alternatively, different historical search interest tags may correspond to different weights. For example, considering that the influence degree of the historical search interest tag with earlier time on the current search interest of the user is smaller, the influence degree of the historical search interest tag with more recent time on the current search interest of the user is larger, a smaller weight value can be set for the historical search interest tag with earlier time, and a larger weight value can be set for the historical search interest tag with later time.

In the case of a stored history search interest tag (such as "new energy automobile/middle-large car"), the user searches for data related to "price of the world" in the browser of the mobile phone, and the mobile phone can determine that the history search interest tag corresponding to the data includes "new energy automobile/middle-large car". The mobile phone can carry the search data and the historical search interest tag corresponding to the search data in the data desensitization request (the request desensitizes the search data related to the price which is a boundary), so that the cloud side service updates the search interest tag of the user according to the historical search interest tag of the user and the search data input by the user.

After receiving the desensitization result from the first interface of the content desensitization analysis service, the mobile phone analyzes an update identifier isUpdate in the desensitization result to obtain that the value of the parameter is 1, which indicates that the search interest tag is updated, and then the mobile phone can report the search interest tag queryTag in the desensitization result and the user identifier to the second interface of the user portrait service, so that the user portrait service calculates the user portrait according to the search interest tag and the user identifier.

By way of example, table 3-2 shows examples of parameters reported by the second interface.

TABLE 3-2

/>

In table 3-2, the user identifier is exemplified by IP, and it should be understood that the user identifier may be any other identifier that may be used to be associated with a user and/or a user device, and the specific form of the user identifier is not limited in the embodiments of the present application.

S109, the user portrait service determines the user portrait according to the search interest tag and the user identification.

As one possible implementation, the user portrayal service calculates the user portrayal from the search interest tag queryTag received by the second interface and the user identification. Optionally, the user portrayal service may redetermine the weight of each search interest tag according to the time, the occurrence frequency, etc. of the search behavior, and determine the user portrayal according to the updated search interest tag. For example, as shown in fig. 7, the history search interest tag { new energy vehicle: 0.7}, the weight corresponding to the feature value "new energy automobile" is 0.7, and the user portrait service 1023 can attenuate the weight from 0.7 to 0.65 to obtain an updated classification label { new energy automobile, 0.65}, and the updated user portrait includes the updated classification label { new energy automobile, 0.65}. Similarly, for the history search interest tag { medium and large cars: 0.6, the user portrayal service 1023 attenuates the weight corresponding to the feature value "middle-large car" from 0.6 to 0.5, and updates the user portrayal according to the updated class label.

S110, the user portrait service returns the user portrait to the mobile phone.

S111, the user portrait service sends the user portrait to the business recommendation service.

S112, the service recommendation service determines target recommendation information matched with the user according to the user portrait.

Illustratively, assume that the user portrayal is: and the energy-saving type and the middle-large-sized automobile are favored, and the preference is a question of the automobile, so that the service recommendation information can recommend the service matched with the user preference for the user according to the user portrait and the candidate recommended materials. For example, other energy-saving automobiles, other middle-large-sized cars, or advertisements related to the world are recommended.

S113, the service recommendation service sends target recommendation information to the mobile phone.

S114, displaying target recommendation information by the mobile phone.

For example, before the user search shown in fig. 6 (a) is a question, the mobile phone may show some recommended information on the recommended interface, for example, SUV, medium-sized and large-sized automobiles as shown in table 3-2. After the user searches for the price of the world and the mobile phone reports the search data related to the search, the server side may update the user representation according to the technical scheme of the above embodiment, and if the weight of the new energy automobile tag and the world tag is high in the updated user representation, the mobile phone displays a recommendation interface 70, as shown in fig. 6 (b), and the recommendation interface 70 includes information 702 related to the world, information 701 related to the new energy automobile, and the like.

In the technical scheme of the embodiment of the application, the user portrait service can construct the user portrait based on the search interest tag and the user identification of the user received from the terminal. The search interest labels and the user identifications can be associated, so that different portraits can be more accurately built for different users, and the wedging degree of the user portraits and the users is higher. Moreover, the search interest labels are desensitized, so that the user privacy cannot be revealed, and the risk of revealing the user privacy in the process of constructing the user portrait can be reduced. Therefore, by using the technical scheme of the embodiment of the application, the accuracy of the user portrait can be improved under the condition of avoiding the user privacy disclosure.

In other scenarios, the keywords of the user search may relate to privacy, in which case the user portrayal service no longer builds a user portrayal based on the user's search data to avoid revealing the user's privacy data. Specifically, as shown in fig. 8, in a scenario where the keyword searched by the user relates to privacy, the technical solution in the embodiment of the present application includes the following steps:

s201, the mobile phone browser receives keywords input by a user.

Illustratively, as in fig. 9 (a), the handset displays a browser interface 80, where the browser interface 80 includes a search box 801. The user may enter the search keyword "symptoms of depression" in search box 801 and click on search button 802.

S202, the mobile phone browser sends search keywords of the user to a search engine.

And S203, the search engine returns the search items related to the keywords to the mobile phone browser according to the search keywords of the user.

S204, the mobile phone determines search data according to the operation of the user on the search items.

S205, the mobile phone sends a data desensitization request to the content desensitization analysis service.

For specific implementation manners of S201 to S205, reference may be made to S101 to S105 described above, and details are not repeated here.

For example, the parameters included in the data desensitization request may be as shown in fig. 10. For example, the data desensitization request includes search data related to "depression".

Illustratively, the data desensitization request may include parameters as shown in table 4:

TABLE 4 Table 4

S206, the content desensitization analysis service determines that keywords matched with the search data do not exist in the keyword white list.

Illustratively, as shown in fig. 10, the search keyword "depression" of the user relates to health information, belongs to user-sensitive data, and the service related to "depression" is not within the effective range of the service recommendation service, and thus, the keyword related to "depression" is not configured in the keyword whitelist 1022. Thus, the content desensitization analysis service cannot find the keywords related to the user search data in the keyword whitelist 1022, and the keyword matching fails.

S207, the content desensitization analysis service returns a desensitization result to the mobile phone browser.

Failure of keyword matching means that the search keywords entered by the user relate to private or sensitive data. In some embodiments, the content desensitization analysis service may carry an updated identification with a parameter value of null (0) in the desensitization result when returning the desensitization result to the handset. Therefore, the mobile phone can determine that the search interest tag for constructing the user portrait is not provided for the user portrait service any more according to the update identification with the null parameter value, so that the user privacy is prevented from being revealed in the process of constructing the user portrait.

Illustratively, the desensitization results may be as shown in fig. 10. The desensitization result includes a classification tag and an update identification. Wherein the category label is empty and the update flag may be 0 (indicating that the search interest label is not updated).

Illustratively, the desensitization results may also be expressed in the format shown in Table 5. As shown in table 5, the desensitization result may include an update identifier (isUpdate) field shown in table 5, where the value of the field is 0, which indicates that the search interest tag is not updated this time.

TABLE 5

It can be understood that after the mobile phone receives the desensitization result from the content desensitization analysis service, it is determined that the content desensitization analysis service does not update the search interest tag of the user according to the update identifier isUpdate (with a value of 0) carried by the desensitization result. This means that the search behavior of the user relates to private data, the search data of the user cannot be matched to keywords in the keyword white list, or the search data of the user is out of the service range of the business recommendation service. In this case, the handset does not report the search interest tag to the user portrayal service, so as to prevent sensitive search data from entering the user portrayal service. Thereby ensuring that the search data of the user cannot be revealed.

In other embodiments, S207 may be optional. That is, the content desensitization analysis service may not return the desensitization result to the handset when the search data of the user does not have a matching keyword. Correspondingly, if the mobile phone does not receive the desensitization result within a period of time, the mobile phone can determine that the content desensitization analysis service does not update the search interest tag of the user, and the mobile phone does not need to report the search interest tag to the user portrait service.

Illustratively, as in (b) of fig. 9, the user opens the recommendation interface 70, and the recommendation interface 70 may present information 701, 702, etc. that does not relate to user privacy, and does not present recommendation entries that relate to user privacy.

The system architecture and the service scenario described in the present application are for more clearly describing the technical solution of the present application, and do not constitute a limitation to the technical solution provided in the present application, and those skilled in the art can know that, with the evolution of the system architecture and the appearance of a new service scenario, the technical solution provided in the present application is also applicable to similar technical problems.

One or more of the interfaces described above are exemplary, and other interface designs are possible, and the specific design of the interface is not limited in this application.

The above embodiments may be combined and the combined solution may be implemented. Optionally, some operations in the flow of method embodiments are optionally combined, and/or the order of some operations is optionally changed. The order of execution of the steps in each flow is merely exemplary, and is not limited to the order of execution of the steps, and other orders of execution may be used between the steps. And is not intended to suggest that the order of execution is the only order in which the operations may be performed. Those of ordinary skill in the art will recognize a variety of ways to reorder the operations described herein. In addition, it should be noted that details of processes involved in a certain embodiment herein apply to other embodiments as well in a similar manner, or that different embodiments may be used in combination.

Moreover, some steps in method embodiments may be equivalently replaced with other possible steps. Alternatively, some steps in method embodiments may be optional and may be deleted in some usage scenarios. Alternatively, other possible steps may be added to the method embodiments.

Moreover, the method embodiments described above may be implemented alone or in combination.

Optionally, as shown in fig. 7, the data desensitization request sent by the handset to the content desensitization analysis service may also include a historical search interest tag. Such as new energy automobiles and search interest tags of medium-sized and large-sized cars.

Alternatively, as shown in fig. 7, if the data desensitization request received by the content desensitization analysis service from the mobile phone also carries the history search interest tag, the content desensitization analysis service may not process the history search interest tag, but directly return the original history search interest tag to the mobile phone.

Optionally, after receiving the updated search interest tag from the content desensitization analysis service, the mobile phone may store the search interest tag locally and may report the search interest tag when a desensitization request is subsequently anonymously reported. Therefore, the labels finally participating in the user portrait calculation comprise the history accumulation and the newly added search interest labels, and the accuracy of the user portrait is improved.

Other embodiments of the present application provide an apparatus that may be an electronic device (e.g., a mobile phone, etc.) or a server as described above. The apparatus may include: a display screen, a memory, and one or more processors. The display, memory, and processor are coupled. The memory is for storing computer program code, the computer program code comprising computer instructions. When the processor executes the computer instructions, the apparatus may perform the functions or steps performed by the mobile phone in the method embodiments described above. The structure of the apparatus may refer to the electronic device (such as a terminal) shown in fig. 3.

The core structure of the device may be represented as the structure shown in fig. 11, where the device includes: a processing module 1301, an input module 1302, a storage module 1303, and a display module 1304.

Processing module 1301 may include at least one of a Central Processing Unit (CPU), an application processor (Application Processor, AP), or a communication processor (Communication Processor, CP). Processing module 1301 may perform operations or data processing related to control and/or communication of at least one of the other elements of the consumer electronic device. Specifically, the processing module 1301 may be configured to control the content displayed on the home screen according to a certain trigger condition. The processing module 1301 is further configured to process the input instruction or data, and determine a display style according to the processed data.

Alternatively, the processing module 1301 may be implemented as the processor 201 such as shown in fig. 3.

The input module 1302 is configured to obtain an instruction or data input by a user, and transmit the obtained instruction or data to other modules of the electronic device. Specifically, the input mode of the input module 1302 may include touch, gesture, proximity screen, or voice input. For example, the input module may be a screen of an electronic device, acquire an input operation of a user, generate an input signal according to the acquired input operation, and transmit the input signal to the processing module 1301.

The storage module 1303 may include volatile memory and/or nonvolatile memory. The storage module is used for storing at least one relevant instruction or data in other modules of the user terminal equipment.

Alternatively, the storage module 1303 may be implemented as the memory 202, such as shown in FIG. 3.

Display module 1304, which may include, for example, a Liquid Crystal Display (LCD), a Light Emitting Diode (LED) display, an Organic Light Emitting Diode (OLED) display, a microelectromechanical system (MEMS) display, or an electronic paper display. For displaying user viewable content (e.g., text, images, video, icons, symbols, etc.).

Optionally, a communication module 1305 is further included for supporting the personal terminal to communicate with other personal terminals (via a communication network). For example, the communication module may be connected to a network via wireless communication or wired communication to communicate with other personal terminals or network servers. The wireless communication may employ at least one of cellular communication protocols, such as Long Term Evolution (LTE), long term evolution-advanced (LTE-a), code Division Multiple Access (CDMA), wideband Code Division Multiple Access (WCDMA), universal Mobile Telecommunications System (UMTS), wireless broadband (WiBro), or global system for mobile communications (GSM). The wireless communication may include, for example, short-range communication. The short-range communication may include at least one of wireless fidelity (Wi-Fi), bluetooth, near Field Communication (NFC), magnetic Stripe Transmission (MST), or GNSS.

Alternatively, the communication module 1305 may be implemented as a transceiver 203 such as that shown in fig. 3.

It should be noted that each functional module of the apparatus may perform one or more steps of the above-described method embodiments.

Embodiments of the present application also provide a chip system, as shown in fig. 12, comprising at least one processor 1401 and at least one interface circuit 1402. The processor 1401 and the interface circuit 1402 may be interconnected by wires. For example, interface circuit 1402 may be used to receive signals from other devices (e.g., a memory of an electronic apparatus). For another example, interface circuit 1402 may be used to send signals to other devices (e.g., processor 1401). Illustratively, the interface circuit 1402 may read instructions stored in the memory and send the instructions to the processor 1401. The instructions, when executed by the processor 1401, may cause the electronic device to perform the various steps of the embodiments described above. Of course, the chip system may also include other discrete devices, which are not specifically limited in this embodiment of the present application.

The embodiment of the application also provides a computer storage medium, which comprises computer instructions, when the computer instructions run on the electronic device, the electronic device is caused to execute the functions or steps executed by the mobile phone in the embodiment of the method.

The present application also provides a computer program product, which when run on a computer, causes the computer to perform the functions or steps performed by the mobile phone in the above-mentioned method embodiments.

It will be apparent to those skilled in the art from this description that, for convenience and brevity of description, only the above-described division of the functional modules is illustrated, and in practical application, the above-described functional allocation may be performed by different functional modules according to needs, i.e. the internal structure of the apparatus is divided into different functional modules to perform all or part of the functions described above.

In the several embodiments provided in this application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative, and the division of modules or units, for example, is merely a logical function division, and there may be additional divisions when actually implemented, for example, multiple units or components may be combined or integrated into another apparatus, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and the parts displayed as units may be one physical unit or a plurality of physical units, may be located in one place, or may be distributed in a plurality of different places. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a readable storage medium. Based on such understanding, the technical solution of the embodiments of the present application may be essentially or a part contributing to the prior art or all or part of the technical solution may be embodied in the form of a software product stored in a storage medium, including several instructions for causing a device (may be a single-chip microcomputer, a chip or the like) or a processor (processor) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read Only Memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

The foregoing is merely a specific embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions within the technical scope of the present disclosure should be covered in the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. An information recommendation method, applied to an electronic device, comprising:

collecting first behavior data generated when a first user uses the electronic equipment;

sending a first desensitization request to a server, wherein the first desensitization request comprises the first behavior data and does not carry an identification of the first user;

receiving a first interest tag from the server, the first interest tag being associated with the first row of data;

and sending the first interest tag and the identification of the first user to the server, so that the server updates portrait information of the first user according to the first interest tag and the identification of the first user.

2. The method of claim 1, wherein after sending the first interest tag and the identification of the first user to the server, the method further comprises:

Acquiring first recommendation information from the server, wherein the first recommendation information is recommendation information determined by the server according to the updated portrait information of the first user;

and displaying the first recommendation information.

3. The method of claim 1 or 2, wherein displaying the first recommendation information comprises:

and displaying the first recommendation information in a recommendation module of the application program, or displaying the first recommendation information in a recommendation card of a negative screen, or displaying a system recommendation message, wherein the recommendation message comprises the first recommendation information.

4. A method according to any of claims 1-3, wherein after sending the first interest tag and the identification of the first user to the server, the method further comprises:

and acquiring the portrait information of the first user after updating from the server.

5. The method of any of claims 1-4, wherein the first behavior data and the first user's privacy data are uncorrelated.

6. The method of any of claims 1-5, wherein sending the first interest tag and the identification of the first user to a server comprises:

Sending the first interest tag, the second interest tag and the identification of the first user to the server, so that the server updates portrait information of the first user according to the first interest tag, the second interest tag and the identification of the first user; and the second interest tag is an interest tag corresponding to the historical portrait information of the first user.

7. The method of claim 2, wherein prior to collecting first behavior data generated by a first user using the electronic device,

displaying a first interface, wherein the first interface comprises second recommendation information, and the second recommendation information is different from the first recommendation information.

8. The method according to any one of claims 1-7, further comprising:

sending a second desensitization request to the server, the second desensitization request including the second behavior data and not carrying an identification of the first user;

a return message is obtained from the server for the second desensitization request, the return message indicating that the second behavior data is related to privacy data of the first user.

9. The method of claim 8, wherein the return message does not include an interest tag associated with the second behavior data.

10. The method according to claim 8 or 9, characterized in that the method further comprises:

displaying fourth recommendation information after the return message is acquired from the server; the third recommendation information is the same as the fourth recommendation information.

11. The method according to any of claims 8-10, wherein the return message comprises a first field, a value of the first field being used to indicate whether an updated interest tag is present.

12. The method of any one of claims 1-11, wherein the first row of data includes one or more of the following information: click behavior related information, search behavior related information, browse behavior related information.

13. An information recommendation method, applied to a server, comprising:

receiving a first desensitization request from an electronic device, wherein the first desensitization request comprises first behavior data generated when a first user uses the electronic device and does not carry an identification of the first user;

receiving the first interest tag and the identification of the first user from the electronic device;

and updating portrait information of the first user according to the first interest tag and the identification of the first user.

14. The method of claim 13, wherein prior to transmitting the first interest tag to the electronic device, the method further comprises:

determining at least one target keyword matched with the first action data in the white list according to the white list;

and carrying out cluster analysis on the at least one target keyword to obtain the first interest tag.

15. The method of claim 13 or 14, wherein the whitelist includes a plurality of keywords, the plurality of keywords not related to privacy terms.

16. An electronic device comprising a processor and a memory; the memory is configured to store computer-executable instructions that, when executed by the electronic device, cause the electronic device to perform the method of any of claims 1-12.

17. A server, wherein the server comprises a processor and a memory; the memory is configured to store computer-executable instructions that, when executed by the server, cause the server to perform the method of any of claims 13-15.

18. A computer readable storage medium comprising a program or instructions which, when executed, implement the method of any one of claims 1 to 12 or the method of any one of claims 13 to 15.