Integration of user declarations in advertisement personalization

By merging users' declared preferences and inferred preferences, generating a merged preference profile and selecting matching advertising components, solving the problem of mismatch between advertising and user interests in the prior art, and achieving efficient personalized output and resource saving of advertising.

CN118901075APending Publication Date: 2024-11-05GOOGLE LLC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202380012139.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-03-02
Publication Date
2024-11-05

AI Technical Summary

Technical Problem

When providing advertisements to users, the prior art usually ignores the user's preferences, resulting in the advertisements that do not match the user's interests and the inability to effectively personalize the user experience.

Method used

By merging user's declared preferences and inferred preferences, generate a merged preference profile, select digital components that match user preferences, and prioritize user declaration preferences when conflicts to ensure personalized output of ads.

Benefits of technology

It improves the degree of personalization of advertisements, increases the probability of users accepting and using advertisements, reduces unnecessary network bandwidth and resource consumption, and improves the computing efficiency and consistency of advertising choices.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118901075A_ABST
    Figure CN118901075A_ABST
Patent Text Reader

Abstract

The techniques are generally directed to personalizing digital components of a user based on a merged preference profile. A merged preference profile may be generated based on declared preferences and inferred preferences of the user. The declared preferences and inferred preferences may be preferences related to brands, topics, product types, etc. The merged preference profile may be used to personalize digital components output to the user. For example, when a digital component to be selected is identified, at least some of the digital components may be selected based on a user's merged preference profile. At least a portion of the selected digital components may be digital components having topics that match preferences of the user, thereby improving the digital components output to the user.
Need to check novelty before this filing date? Find Prior Art

Description

Background Art

[0001] When a user performs a search or browses for content, digital content such as advertisements typically appear on a publisher's website or mobile app along with organic search results. Ads are typically selected based on optimizing the user's engagement in an attempt to increase clicks or conversions, with little regard for the user's ad preferences, such as the user's explicit preferences. This results in users being served ads that are not necessarily consistent with their preferences. Summary of the invention

[0002] The technology is generally directed to improving the output of digital components on a publisher's website or mobile application based on a merged preference profile of a user. The merged preference profile can be determined based on the user's declared preferences and inferred preferences. The declared preferences and inferred preferences can be preferences related to the subject matter of the content (such as brand, topic, product type, etc.). The merged preference profile can be used to identify digital components to be selected and provided for output to the user. At least some of the selected digital components can be digital components with a subject matter that matches the user's preferences, thereby improving the digital components output to the user.

[0003] One aspect of the present technology is directed to a method, comprising: receiving, by one or more processors, a request for at least one digital component for display on a page of the publisher from a publisher, determining, by the one or more processors, a user's declared preferences, by the one or more processors, determining a user's inferred preferences, by the one or more processors, merging, by the one or more processors, the user's declared preferences and inferred preferences into a merged preference profile, wherein, when a conflict exists between at least one declared preference and at least one inferred preference, selecting, by the one or more processors, the at least one declared preference to be included in the merged preference profile, identifying, by the one or more processors, at least one digital component associated with the merged preference profile, and providing, by the one or more processors, at least one selected digital component to the publisher for output.

[0004] The method may also include discarding, by the one or more processors, at least one inferred preference that conflicts with the at least one declared preference.

[0005] The method may also include reserving, by the one or more processors, display space for one or more selected digital components associated with the merged preference profile.

[0006] The user's stated preferences may be related to at least one subject matter of a digital component previously provided for output. The stated preferences may include at least one of a pure preference statement or an influenced preference statement. The inferred preferences may be determined by identifying one or more behaviors regarding a subject matter of content previously provided for output or one or more search queries for the subject matter of the content.

[0007] Another aspect of the present technology is directed to a computer-readable storage medium storing instructions that, when executed by one or more processors, cause the one or more processors to receive a request from a publisher for at least one digital component for display on the publisher's page, determine a user's declared preferences, determine the user's inferred preferences, merge the user's declared preferences and inferred preferences into a merged preference profile, wherein when there is a conflict between at least one declared preference and at least one inferred preference, select the at least one declared preference to be included in the merged preference profile, identify at least one digital component associated with the merged preference profile, and provide at least one selected digital component to the publisher for output.

[0008] Yet another aspect of the present technology is generally directed to a system including one or more processors. The one or more processors may be configured to receive a request from a publisher for at least one digital component for display on a page of the publisher, determine a user's declared preferences, determine an inferred preference of the user, merge the user's declared preferences and the inferred preferences into a merged preference profile, wherein when there is a conflict between at least one declared preference and at least one inferred preference, select at least one declared preference to be included in the merged preference profile, identify at least one digital component associated with the merged preference profile, and provide the at least one selected digital component to the publisher for output. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 are example screenshots illustrating example content according to aspects of the present disclosure.

[0010] Figure 2 is a flow chart of an example method of generating a merged preference profile according to aspects of the present disclosure.

[0011] Figure 3 is an example screenshot illustrating example input for declaring a pure preference in accordance with aspects of the present disclosure.

[0012] Figure 4 is an example screenshot illustrating example input for declaring a pure preference in accordance with aspects of the present disclosure.

[0013] Figure 5 is an example screenshot illustrating example input for declaring affected preferences in accordance with aspects of the present disclosure.

[0014] Figure 6 is an example screenshot illustrating example content of inferred preferences according to aspects of the present disclosure.

[0015] Figure 7 is an example flow chart of an example method of resolving conflicts according to aspects of the present disclosure.

[0016] Figure 8 is an example of a merged preference profile according to aspects of the present disclosure.

[0017] Fig. 9 is an example flow chart for providing advertisements according to aspects of the present disclosure.

[0018] Fig.10 is a block diagram of an example system according to aspects of the present disclosure.

[0019] Fig.11 is a flow chart of an example method of selecting a digital component according to aspects of the present disclosure. DETAILED DESCRIPTION

[0020] The technology is generally directed to improving a user's digital components based on a merged preference profile. In some examples, improving a user's digital components may include personalizing the user's digital components. The merged preference profile may be generated based on the user's declared preferences and inferred preferences. The declared preferences and inferred preferences may be preferences related to brands, topics, product types, etc. The merged preference profile may be used to personalize the digital components output to the user. For example, when selecting a digital component, there may be multiple digital components to be selected from. At least some of the digital components may be selected based on the user's merged preference profile. The selected digital components may match the user's preferences. At least a portion of the selected digital components may be digital components with a theme that matches the user's preferences, thereby personalizing the digital components output to the user.

[0021] Declared preferences may be preferences that have been selected by a user, preferences determined based on user responses to survey questions, or preferences determined based on a user's light engagement with the content, such as a user's reaction to the content, e.g., "liking," sharing, "hearting," filtering, etc. Inferred preferences may be determined based on user behavior with respect to the content, such as a search query submitted by the user, an ad click or view, a viewing threshold time period, or other user interactions with the content. According to some examples, the content may include, be associated with, or be identified by an embedding. The embedding may be a representation of a brand, topic, product, etc. of the content. In some examples, embeddings may be used to identify a user's preferences. An embedding may be a numeric, alphabetic, or alphanumeric representation of a user's preferences and / or a subject of the content expressed as a vector. Declared preferences and inferred preferences may be merged into a merged preference profile.

[0022] According to some examples, when merging the declared preferences and the inferred preferences, there may be a conflict between the user's declared preferences and the inferred preferences. For example, a user may declare that they do not want to see any digital components related to sports, but the user's inferred preferences may indicate that they have searched for sporting goods stores. When merging the declared preferences and the inferred preferences, conflicts such as the described conflicts may be resolved by selecting the declared preferences instead of the inferred preferences.

[0023] The merged preference profile can be used to target and retrieve digital components to be output to the user. For example, a publisher can submit a request for one or more digital components to be output to the user. When determining which digital components to select, at least a portion of the selected digital components is reserved for digital components associated with the user's merged preference profile.

[0024] In order to determine whether the digital component under consideration matches the user's preferences, the brand, topic, product, etc. (collectively referred to herein as "theme") of the digital component under consideration may be compared with the user's merged preference profile. The theme of the digital component may be determined based on the content of the digital component, a classifier associated with an entity that paid for the digital component, a description of the digital component, a landing page of the digital component, etc. According to some examples, the theme of the digital component may be determined based on an embedding associated with the digital component. Once the theme of the digital component is determined, it may be compared with the theme(s) in the merged preference profile. At least one of the digital components having a threshold match with the theme of the merged preference profile may be a selected digital component and thus output to the user. Additionally or alternatively, the embedding associated with the digital component may be compared with the embedding associated with the merged preference profile. For example, the coordinates of the embedding of the digital component may be compared with the coordinates of the embedding associated with the merged preference profile. A threshold match between the digital component and the merged preference profile may be that the digital component has an embedding within a threshold of the coordinates of the embedding associated with the merged preference profile.

[0025] According to some examples, the subject of the digital component can be used to ask the user a survey question. The survey question can be used to determine the user's stated preferences, to reconcile conflicts between stated preferences and inferred preferences, etc. For example, if the subject of the digital component is "basketball" and the user previously stated that they are not interested in "sports", the survey question can be "Would you like to see more digital components related to basketball?" Based on the response, the stated preferences can be updated, and thus the merged preference profile can be updated.

[0026] The digital space reserved for digital components associated with the merged preference profile ensures that the user is presented with personalized digital components that correspond to their stated preferences and inferred preferences. In addition, conflicts between the user's stated preferences and inferred preferences are resolved by selecting the stated preferences, filtering out irrelevant digital components from the digital components considered for selection. This can ensure that the user is presented with digital components that are relevant to their preferences, rather than digital components that the digital component system may deem optimal for the user. For example, if the user has stated that they want to see fewer digital components related to electric vehicles ("EVs"), then the stated preference to see fewer EV digital components will override any inference that they are to select EV digital components for output to the user.

[0027] Identifying digital components to include during the selection process and retaining at least a portion of the selected digital components to be associated with the affirmative preferences indicated in the merged preference profile can ensure that the digital components provided to the user are personalized to the interests and preferences of the user. By using the merged preference profile to select digital components, the computational efficiency of the process for selecting digital components is increased compared to determining user preferences when the publisher requests the digital components. In addition, by generating the merged preference profile before the publisher requests the digital components, the consistency and accuracy of the declaration is improved.

[0028] By providing a digital component that will be accepted and used by the user, network bandwidth and other resources (such as processor and memory overhead) associated with sending unwanted digital components to the publisher and / or user device are not wasted because one or more replacement digital components do not need to be sent to the user after the unwanted digital component. In addition, additional display space at the publisher and / or user device does not need to be reserved, also saving resources at the publisher / user device.

[0029] Figure 1 is example content output on a device. The subject or focus of content 114 may be determined and used to determine user preferences. In some examples, embeddings associated with the subject of the content may be determined and used to determine user preferences. According to some examples, content 114 may be used to identify digital components that match a user's merged preference profile during selection of digital components. The subject of content 114 may include, for example, a brand, topic, product type, etc. The subject may be associated with the embedding. Content 114 may include, for example, digital components, videos, images, information, etc.

[0030] According to some examples, the content 114 may be a video advertisement for a retail store that sells shoes. The subject matter of the content 114 may be determined based on the brand 110, topic, product 106, etc. in the advertisement, such as a retail store, a type of shoe, a brand 110 of shoe, etc. The determined subject matter may be used to survey the user about the retail store 116, the type of shoe, or the brand 110 of shoe. In some instances, the determined subject matter may be used to determine the user's inferred preferences based on the user's engagement or behavior with the content.

[0031] According to some examples, the subject of the content 114 may be determined based on the description 106, the landing page, the categories associated with the content 114, etc. Continuing with the above example, a video of a retail store selling shoes may include a description of the type or brand 110 of shoes, information related to the sale, etc. In some examples, clicking on the video may lead to a landing page for an associated web page for the retail store or a web page for the brand of shoes in the video. Additionally or alternatively, they may be categories associated with the content 114, such as tags related to the subject of the content 114. Continuing with the example provided, the categories may be "shoes", a specific brand 110, a specific type of shoes, etc.

[0032] In some examples, a subject matter of the content 114 may be provided to be output with the content 114. For example, an indicator may be included such that, when selected, a pop-up window 102 identifying the subject matter 104 of the content 114 may be output. This may provide the user with more clarity regarding the subject matter 104 of the content 114 than if the publisher provided the content 114. According to some examples, the subject matter 104 of the content 114 may identify the brand 110, topic, product 112 promoted by the content 114, as opposed to the payment entity. Continuing with the video of the retail store 116 selling shoes, the brand 110 of the shoes and the type of shoes may be identified, rather than identifying the retail store 116.

[0033] Figure 2 is an example flow diagram of an example method of generating a merged preference profile 212. The merged preference profile 212 can be generated based on the declared preferences 202, the inferred preferences 206, and the contextual preferences 208. The declared preferences 202, the inferred preferences 206, and / or the contextual preferences 208 can be determined based on the user's engagement, interaction, viewing, behavior, etc. with respect to the content. Therefore, the declared preferences 202, the inferred preferences 206, and / or the contextual preferences 208 can be changed, updated, confirmed, etc. based on the user's engagement, interaction, search, selection, viewing, etc.

[0034] When determining the declared preferences 202 and the inferred preferences 206, conflicts may arise between the declared preferences 202 and the inferred preferences 206. The conflicts may be resolved via conflict resolution 210. The conflict resolution 210 may utilize a hierarchy, a weighting system, a sliding scale, a direct comparison, etc. to resolve conflicts between preferences. For example, if the declared preferences 202 are positive for sports, but the inferred preferences 206 are negative for hockey, a conflict may arise. Hockey is a sport, and therefore, there is a conflict between the positive declared preferences 202 for sports and the negative inferred preferences 206 for hockey. To resolve the conflict, the conflict resolution 210 may give more weight to the declared preferences 202 than to the inferred preferences 206. In some examples, the conflict resolution 210 may consider other declared preferences 202 and inferred preferences 206 when resolving the conflict. In some examples, the conflict resolution 210 may resolve conflicts between the contextual preferences 208 and the declared preferences 202 and / or inferred preferences 206. Similar to resolving conflicts between stated preferences 202 and inferred preferences 206, conflict resolution 210 can use a hierarchy, weights, populations, etc. of preferences to resolve conflicts involving contextual preferences 208. Once any conflicts are resolved, a merged preference profile 212 can be generated.

[0035] The declared preferences 202 may include pure preference statements 214 and affected preference statements 204. Pure preference statements 214 may be preferences determined based on explicit user indications related to topics, brands, interests, etc. Topics, brands, interests, etc. may correspond to the subject of the content. The affected preference statements 204 may be statements based on the user's light involvement with the subject of the content. The content may include, for example, published web pages, advertisements, videos, images, etc. According to some examples, the declared preferences 202 may be determined based on content, digital components, advertisements, etc. previously provided for output. The declared preferences 202, inferred preferences 206, and contextual preferences 208 may be associated with embeddings. For example, a user interaction with the content may be associated with an embedding, which is then included as a pure preference statement 214 and / or an affected preference statement 204, depending on the type of interaction. Although not a human-understandable classification, embeddings may be used to identify similar content and / or based on embeddings. According to some examples, embeddings may be used to identify similar topics, personalized content, personalized digital components, etc., with embeddings within a threshold coordinate distance of a given embedding.

[0036] Figure 3An example user interface 310 is shown. The user interface 310 may include an input 302 that allows a user to switch between receiving personalized ads based on the merged preference profile 212, receiving ads based only on the contextual preferences 208, or not receiving personalized ads at all. Although the input 302 for the user to select to receive personalized ads is shown as a drop-down menu that allows the user to select "on" or "off", the input 302 may be a radio button, a sliding scale, a checkbox, etc. Therefore, the input 302 of the user interface 310 as shown is an example and is not intended to be limiting.

[0037] Pure preference statement 214 may be determined based on input received via user interface 310. Input received from a user may indicate a preference for a given advertising topic 308. The preference may be to increase or decrease the frequency of advertisements related to the topic 308. In some examples, the preference may be to increase 306 or decrease 304 the likelihood of displaying advertisements related to the topic 308 to the user. As shown, advertising topics 308 may be "handbags and wallets," "hotels, motels and resorts," "clothing," and the like.

[0038] Input received via user interface 310 may provide an indication to increase 306 or decrease 304 the frequency of advertisements related to “handbags and purses,” “hotels, motels, and resorts,” and “apparel.” In such an example, pure preference statement 214 may be to increase 306 or decrease 304 the frequency and / or likelihood of displaying advertisements related to “handbags and purses,” “hotels, motels, and resorts,” and “apparel.” Although inputs for increasing or decreasing the frequency or likelihood of displaying advertisements for a particular topic 308 are illustrated as “plus” and “minus” buttons, the inputs may be sliding scales, radio buttons, text boxes, and the like. Thus, user interface 310 is merely one example for receiving pure preference statement 214 and is not intended to be limiting.

[0039] Figure 4Another example for determining pure preference statement 214 is shown. For example, instead of receiving input via user interface 310, input 404 may be received in the form of a response to survey question 402. Survey question 402 may be related to the subject matter of digital component 406. For example, digital component 406 may be directed to shoes, such as work boots and dress shoes. In conjunction with digital component 406, survey question 402 may be provided to confirm the user's interest in advertisements related to work boots. If input 404 corresponding to a selection of "yes" is received, it may be determined that pure preference statement 214 is an affirmative statement of advertisements related to work boots. If input 404 corresponding to a selection of "no" is received, it may be determined that pure preference statement 214 is a negative statement of advertisements related to work boots. According to some examples, an affirmative statement may correspond to an increase in the frequency and / or likelihood of advertisements related to a given subject (e.g., work boots), while a negative statement may correspond to a decrease in the frequency and / or likelihood of advertisements related to a given subject.

[0040] The affected preference statement 204 may be determined based on a user's light engagement with content such as social media content, web page content, advertisements, and the like. Light engagement may include a user reaction to the content. The reaction may include "liking" the content, "hearting" or any other emoticon reaction, such as a thumbs up, a smiley face, an angry face, and the like. For example, if a user reacts to a social media post related to an insulated travel mug by "liking" the post, the affected preference statement 204 may be a positive statement for an insulated travel mug. In some instances, if a user reacts to web page content related to a disposable travel mug with an angry face emoticon, the affected preference statement 204 may be determined as a negative statement for a disposable travel mug. A positive statement or preference may correspond to a brand, topic, product, and the like that the user is interested in seeing more of, while a negative statement or preference may correspond to a brand, topic, product, and the like that the user is interested in seeing less of or having no interest in seeing. According to some examples, the affected preference statement 204 may be determined based on user engagement (such as light engagement) with content, advertisements, digital components, and the like that were previously provided for output to the user.

[0041] Light engagement may be determined and used after a user provides the system with authorization to access or receive information related to the user's light engagement with content. According to some examples, authorization is provided via user engagement with content. Any use of the user's light engagement is authorized by the user. For example, a user may provide authorization to an application by setting certain permissions for the application to generate a merged preference profile. Authorization may be used for an application or system to access one or more databases or sub-databases in the memory of a device. According to one example, a user may select a specific sub-database that an application is authorized to access. For example, a user may authorize access to a search history database, but not to a video or content observation database.

[0042] Figure 5 Example output 506 is shown. Output 506 may include content 510, survey questions 520, and responses 516. Output 506 additionally includes information about retailer 512, description 504 of content 510, indication of subject matter 502 of content 510, and input 522 for responding to survey questions 520.

[0043] As shown, content 510 includes an advertisement for work boots. The work boots have a given brand 508, and content 510 is provided by a retailer 512. Content 510 may be a video, video still, image, gif, or any form of media.

[0044] Reactions 516 may provide inputs, such as affirmations 518 and negations 514, for reacting to or engaging with content 510. Reactions 516 may include affirmations 518 or negations 514 indications associated with brands 508, products, retailers 512, and the like. As shown, when input corresponding to the selection of a "check mark" is received, an affirmation 518 reaction 516 may be determined. When input corresponding to the selection of a "cancel" symbol is received, a mild engagement 516 of a negation 514 may be determined. Although a "check mark" is shown as corresponding to an affirmation 518 reaction 516, and a "cancel" symbol is shown as corresponding to a negation 514 reaction 516, reactions 516 may be any symbol, word, input, and the like. Thus, affirmations 518 and negations 514 are merely some examples and are not intended to be limiting.

[0045] According to some examples, reaction 516 can be associated with an embed. For example, instead of confirming a positive 518 reaction to a brand, product, retailer, etc., an embed having coordinates at or near the brand, product, retailer, etc. can be identified in response to receiving an input corresponding to a selection of the positive 518 reaction. In some examples, instead of confirming a negative reaction 514, an embed having coordinates opposite to the brand, product, retailer, etc. can be identified as something that may be of interest to the user. Additionally or alternatively, in response to the negative reaction 514, an embed having coordinates corresponding to the brand, product, retailer, etc. can be identified. Embeds associated with negative reactions can be used to ensure that digital components with themes having embeds close to negative reaction embeds are not provided to users.

[0046] The affected preference statement 204 may be determined based on the input corresponding to the selection of affirmative 518 or negative 514. Reactions 516 or selections of affirmative 518 or negative 514 may be received with respect to content 510, brand 508, retailer 512, topic 502, etc. Although reactions 516 with respect to content 510 are shown, additional reactions 516, such as affirmative 518 and negative 514, may be provided with respect to brand 508, retailer 512, topic 502, etc.

[0047] The inferred preference 206 can be determined based on user interactions with the content. The user interactions can include, for example, search queries for subjects (e.g., brands, topics, products, etc.), clicks, observations, etc. Clicks can correspond to selections of content, and observations can correspond to viewings of videos, video clips, gifs, etc. According to some examples, if the system receives a search query for left-handed scissors, receives input corresponding to selection of an advertisement for left-handed scissors, and / or determines that content for left-handed scissors has been viewed, the system can infer a positive preference for left-handed scissors.

[0048] The affected preference statements 204 and / or inferred preferences 206 may be associated with an embedding. For example, rather than identifying the subject of the affected preference statements 204 and / or inferred preferences 206, an embedding may be identified. The embedding may map the affected preference statements 204 and / or inferred preferences 206 to coordinates such that digital components having embeddings that are nearby or within a threshold may be selected and provided for output to the user. Thus, the embedding may allow the system to identify digital components that are similar to the embeddings of the user's preferences. This may allow the system to personalize the digital components provided to the user.

[0049] According to some examples, the inferred preferences 206 may be determined based on the user's engagement or behavior with content, advertisements, digital components, etc., previously provided for output to the user. After the user provides authorization for the system to access or receive information related to user interaction with the content, the inferred preferences may be determined and used. Any use of the user interaction is authorized by the user.

[0050] Figure 6 An example output of search results in response to a search query is shown. For example, a publisher 608 may receive a search query 602 for "shoes." In response to the search query 602, the publisher 608 may provide content 610 for output on a display of a user device 612. The content 610 may include one or more search results 606a, 606b, 606c, 606d, 606e, 606f. The search results may be products, such as products 1-6. The search results may be related to the search query 602. For example, because the search query 602 is for shoes, products 1-6 may be different types of shoes.

[0051] The inferred preference 206 can be determined based on the search query 602. For example, the inferred preference 206 can be determined as shoes. Additionally or alternatively, the inferred preference 206 can be determined based on clicks. For example, if an input 604 corresponding to a selection of a given product 1-6 is received, the inferred preference 206 can be determined based on the input 604 (e.g., the selected product).

[0052] Although content 610 is shown as search results for retailers, content 610 may be any content, such as a video, a news article, an advertisement, etc. Thus, inferred preference 206 may be determined based on a click, view, hover, selection, etc. of any type of content.

[0053] According to some examples, contextual preferences 208 may be determined based on the content 610 being viewed and / or the context of the content 610 without any user history. Contextual preferences 208 may be determined based on the content 610 currently being output to the user, while inferred preferences 206 may be determined based on the current and / or historical content 610 output to the user. Figure 6 , the search query 602 is for shoes, so that the contextual preference 208 can be determined to be shoes in addition to the inferred preference 206. If no other information about the user's merged preference profile 212 is available, a digital component related to the search query 602, such as shoes, can be provided. In examples where the user's merged preference profile 212 is available, the digital component provided for output can be related to the search query 602 and the user's stated preferences 202 and inferred preferences 206.

[0054] After the user provides the system with authorization to access or receive information related to user interaction with content and / or viewing of content, contextual preferences may be determined and used. Any use of user interaction is authorized by the user.

[0055] According to some examples, there may be a conflict between declared preferences and inferred preferences. In order to resolve conflicts between declared preferences, inferred preferences, and contextual preferences, the system can use a hierarchy. For example, when resolving conflicts, pure preference statements can be given the greatest weight compared to contextual preferences. According to some examples, the hierarchy from the greatest weight to the least amount of weight can be pure preference statements, affected preference statements, inferred preferences, and contextual preferences. The conflict will be resolved by selecting a topic (e.g., topic, brand, product, etc.) based on the preference with the greatest weight. Although the preferences are provided as a hierarchy from pure statements to contextual preferences, the described hierarchy is merely an example, as the hierarchy can be provided in any order, giving any preference the greatest weight.

[0056] According to some examples, the freshness of declared preferences, inferred preferences, and contextual preferences can be used to resolve conflicts between declared preferences, inferred preferences, and contextual preferences. The freshness of the preference can be, for example, the timing of the preference, the amount of freshness, etc. For example, when resolving a conflict, the system can determine whether a declared negative preference for a topic was received after the user participated in the topic for an extended period of time. In another example, when resolving a conflict, the system can determine that the preference for a topic was received some time before the most recent participation in the same topic. Therefore, when resolving a conflict, freshness, such as the timing of the received preference, the timing of participation with the topic, etc., can be used compared to the received preference.

[0057] Figure 7 An example of resolving conflicts between stated preferences, inferred preferences, and contextual preferences is shown. For example, pure preference statement 702 may be negative for content related to cars, inferred preference 704 may be positive for content related to electric vehicles, and contextual preference 706 may be for content related to pickup trucks. In such an example, there is a conflict between stated preferences and inferred preferences and contextual preferences. That is, pure preference statement 702 indicating a negative statement for content related to cars contradicts inferred preference 707 for electric vehicles and contextual preference 706 for pickup trucks.

[0058] To resolve conflicts such as the above, conflict resolution 210 can determine the preference with the greatest weight in the hierarchy, such as pure preference statement 702. In this example, the preference with the greatest weight in the hierarchy can be pure preference statement 214. Since the stated preference is negative for content related to cars, the conflict can be resolved by following the stated preference and ignoring the inferred preference and contextual preference. Continuing with this example, the conflict can be resolved by indicating a negative preference for content related to cars and ignoring the positive preference for electric vehicles and pickup trucks.

[0059] In some examples, conflict resolution can be determined based on responses to survey questions and / or natural language input. Continuing with the car-related example, if the stated preference is negative for car-related content, but the inferred preference is positive for electric vehicle-related content, the system can send a survey question related to the user's statement. For example, the survey question can be "Do you want to change your stated preference?" or "Do you want to see car-related content?" Input received in response to the survey question can be used to update the stated preference and, therefore, to resolve conflicts between user preferences. According to some examples, the natural language input can be an automated chat that provides survey questions asking users to declare their preferences.

[0060] According to some examples, conflicts can be resolved by inferring new preferences based on a combination of data. For example, a preference for EV pickup trucks can be inferred based on a combination of the inferred preference for electric vehicles 704 and the contextual preference for pickup trucks 706. In some examples, a preference for vehicles other than sedans with internal combustion engines can be inferred based on the inferred preference for electric vehicles 704 and the contextual preference for pickup trucks 706.

[0061] Figure 8 2 is an example merged preference profile 212. As discussed above and herein, the merged preference profile 212 may be determined based on the stated preferences 202 and the inferred preferences 206. According to some examples, the merged preference profile may be the user's preferences after resolving conflicts between the user's stated preferences, inferred preferences, and contextual preferences. The resulting merged preference profile may be used to target retrieval of digital components to be output to the user, such as advertisements.

[0062] The declared preferences 202 include pure preference statements 214 and affected preference statements 204. The pure preference statements 214 can be determined based on the user's explicit declaration of the subject matter of the content. Figure 3 and Figure 4 As discussed, explicit statements can be determined based on input indicating a preference for a given topic or input in response to a survey question. As shown, pure preference statement 214 includes an indication that no digital components or advertisements related to cars are displayed. Pure preference statement 214 may additionally or alternatively include an indication to reduce the number of digital components related to "handbags and wallets", while including an indication to increase the number of digital components related to "hotel H" and "clothing". According to some examples, the greater the value of the positive indication, the greater the frequency of display components related to the topic should be shown. In some examples, the greater the value of the negative indication, the lower the frequency of display components related to the topic should be shown. Although the indication of increasing or decreasing frequency is shown as positive and negative values, the indication can be provided as a sliding scale, a check box (such as "yes" or "no" or "display" or "do not display", a radio button, etc.). In addition, in some examples, the indication may be provided or may not be provided for output. Therefore, the indication shown is only an example and is not intended to be limiting.

[0063] The affected preference statement 204 may be determined based on the user's light engagement with the content. Figure 5 As discussed, light engagement with content may include user reactions to the content, such as “like”, “heart”, “anger”, etc. Figure 8As shown, the affected preference statements 204 include work boots and the brand "Sportz". According to some examples, the affected preference statements can include instructions to increase or decrease the frequency of providing digital components related to a given topic. These instructions can be, for example, positive or negative values, such as those shown in the pure preference statement 214, a sliding scale, a check box, a radio button, etc.

[0064] Inferred preferences 206 may be determined based on user interactions with or behavior regarding the content. Figure 6 As discussed, user interactions may include, for example, search queries, clicks corresponding to selections of content, observations corresponding to viewing of content, page dwell time, long clicks versus short clicks, long observations versus short observations, and the like. Long clicks may refer, for example, to user interactions with search results in response to a search query, while short clicks may, for example, indicate little to no user interaction with the search results. In some examples, short clicks may indicate that the search results do not include content of interest to the user. Long observations may refer, for example, to user observation time of a given content, such as a video. In contrast, short observations may indicate that the user spent little to no time observing the content.

[0065] like Figure 8 As shown, the inferred preferences 206 may include electric vehicles, earbuds, video games, and basketball. According to some examples, the inferred preferences 206 may include an indication to increase or decrease the frequency of providing digital components related to a given topic. In some examples, a survey question may be output to the user asking whether the inferred preferences 206 should be a pure preference statement 214.

[0066] According to some examples, a merged preference profile may be generated offline in a user personalization system ("UPS"), rather than generating the merged preference profile when determining which advertisements to include in an auction. For example, generating the merged preference profile offline may include generating the merged preference profile based on the user's past history and / or activity rather than when an ad server receives a request for an advertisement. The UPS may generate personalized information for the user to be included in the merged preference profile. According to some examples, the UPS may operate offline and / or when new activity of the user is detected. The UPS may update the merged preference profile based on the generated personalized information.

[0067] When an ad server receives a request from a publisher for an ad that fills a display space, the ad server may identify a user who is viewing the publisher's website or mobile app. The ad server may receive a merged preference profile of the user. For example, at the time of an ad request, the ad server may have a complete context about the request, such as who, where, when, what, etc. According to some examples, who may be a user, where may be a publisher's website or app, when may be a day of the week, a time of day, etc., and "what" may be an action of the user. For example, when a user is watching a video, after watching a video sequence, after searching for comments on a topic, etc., "what" may be included as an ad request. According to some examples, the ad server may additionally or alternatively identify the context of the request for an ad. The context may include who, what, where, when, etc., such as a time of day, a day of the week, what content the user is watching, etc. The ad server may use the merged preference profile and / or content to identify relevant ads to be included in the auction. In some examples, the merged preference profile generated offline may improve the efficiency of the ad server when responding to ad requests. For example, the ad server may use a merged preference profile generated offline, including user declarations, to identify relevant ads given the context of the request.

[0068] Generating a merged preference profile prior to an auction can ensure consistency of declared preferences across all applications. According to some examples, generating a merged preference profile on the build side and implementing the merged preference profile on the server side can allow multiple ad serving stacks to read one source. Reading one source, such as a merged preference profile, can allow the system to consistently apply preferences in the merged preference profile.

[0069] Multiple separate ad delivery stacks can operate based on shared data. The shared data can be, for example, personalized data. According to some examples, the merged preference profile can be used to train a machine learning model across users. For example, the merged preference profiles of multiple users can be used to train a machine learning model to predict the interests or preferences of other users based on the preferences of users with similar interests. In another example, the merged preference profile can be used as a ground truth for a model that attempts to predict the interests of other users. In yet another example, the merged preference profile can be used as an input to a recommendation model. The recommendation model can be, for example, a model that recommends content. The recommendation of content can be, for example, in response to a search query.

[0070] The merged preference profile 212 may be used to target and retrieve advertisements to be included in the selection process. For example, as part of the selection process, advertisers may bid to have their advertisements displayed on a publisher's website or mobile application. The advertisements included in the selection may be based at least in part on the user's merged preference profile 212. According to some examples, advertisements with topics associated with negative preferences may be filtered out so that they are limited in number or not included in the selection process. At least a portion of the selected advertisements may be reserved for advertisements with topics associated with positive preferences in the user's merged preference profile 212.

[0071] Fig. 9 An example flow diagram of a selection process is shown. A publisher 908, such as "retailer.com," may receive a search query 906 for "shoes." In response to receiving the search query 906, the publisher 908 may send a request to an ad server for one or more advertisements to be presented to the user in addition to the search results. The request may be based on the search query for "shoes." For example, the request may include a request for advertisements corresponding to or associated with "shoes."

[0072] The ad server may receive a request including the terms of a search query (e.g., "shoes") and identify one or more available ads 902 that are relevant to the request from the publisher and / or the terms of the search query. Fig. 9 As shown, available advertisements 902 may include advertisements for products 1-5.

[0073] When selecting one or more available advertisements 902 to be returned in response to a request from a publisher 908, the advertisement server may select advertisements that are related to the preferences in the merged preference profile 212. For example, the advertisement server may compare brands, topics, products, etc. that are identified as having positive or negative preferences with the brands, topics, products, etc. of the advertisement. According to some examples, if the merged preference profile 212 includes a positive preference for sports brands and the advertisement theme is sports, the system may determine that the advertisement matches the merged preference profile. In another example, if the merged preference profile 212 includes a negative preference for clothing brands and the advertisement theme is clothing, the system may determine that the advertisement does not match the merged preference profile 212. In an example where the advertisement server determines that the advertisement matches the merged preference profile 212, the advertisement may be included in the selection process. In an example where the system determines that the advertisement does not match the merged preference profile 212, the advertisement may be excluded from the selection process. This may ensure that the advertisement provided for output to the user substantially corresponds to the preferences indicated in the merged preference profile 212.

[0074] like Fig. 9As shown, one or more available advertisements 902 may be identified. Available advertisements 902 may be for products 1-6. According to some examples, products 1-6 may be related or relevant to the request from the publisher. The ad server may compare available advertisements 902 to the preferences of the merged preference profile 212. For example, the merged preference profile 212 may indicate that there is a positive preference for work boots, and the ad server may ensure that an advertisement for product 4 is included in the selection process. The winning or selected advertisement 904 may be provided for output by the publisher 908 along with the search results.

[0075] In some examples, a machine learning ("ML") model may be used to identify advertisements to be included in the selection process. The ML model may be trained to determine whether the subject matter of the advertisement matches the preferences indicated in the merged preference profile 212. Each training example may consist of an advertisement. The input features of the ML model may be the subject matter of the advertisement, such as a topic, brand, product, etc., the positive preferences indicated in the merged preference profile 212, and the negative preferences indicated in the merged preference profile 212. According to some examples, additionally or alternatively, the input features of the ML model may be embeddings associated with the preferences indicated in the merged preference profile and the advertisement. The ML model may use the input features to more accurately determine whether the subject matter of the advertisement matches the preferences in the merged preference profile 212. In some examples, the ML model may use the input features to more accurately determine whether the embedding of the advertisement matches the embedding in the merged preference profile. The output of the ML model may be a determination of whether the advertisement should be included in the selection process. In some examples, the system may ask for feedback from the user, such as by providing survey questions related to the brand, topic, product, etc. in the advertisement. For example, a user may be asked to confirm whether an advertisement related to a given brand, topic, or product should be provided. The user may provide feedback indicating whether the determination is correct, such as yes or no.

[0076] Fig.10 An example system is shown in which the features described above and herein can be implemented. It should not be considered to limit the scope of the present disclosure or the usefulness of the features described herein. In this example, system 1000 includes device 1016, ad server 1018, ad server storage system 1022, publisher server 1020, publisher storage system 1026, and network 1024.

[0077] Device 1016 may include one or more processors 1002, memory 1004, data 1006, and instructions 1008. Device 1016 may also include input 1010, output 1012, and communication interface 1014. The device may be, for example, a smartphone, a tablet, a laptop, a smart watch, an AR / VR headset, a smart helmet, a home assistant, etc.

[0078] The device's memory 1004 may store information accessible by the processor 1002. The memory 1004 may also include data that can be retrieved, manipulated, or stored by the processor 1002. The memory 1004 may be any type capable of storing information accessible by the processor 1002, including computer-readable media that may be non-transitory, or other media that store data that can be read with the aid of an electronic device, such as a hard drive, a memory card, a read-only memory ("ROM"), a random access memory ("RAM"), an optical disk, and other writable and read-only memories. The memory 1004 may store information accessible by the processor, including instructions and data 1006 that may be executed by the processor 1002.

[0079] Data 1006 may be retrieved, stored, or modified by processor 1002 according to instructions 1008. For example, although the present disclosure is not limited to a particular data structure, the data may be stored in a computer register, in a relational database as a table with a plurality of different fields and records, an XML document, or a flat file. The data may also be formatted in a computer readable format, such as, but not limited to, binary values, ASCII, or Unicode. Further, by way of example only, data 1006 may include information sufficient to identify the relevant information, such as numbers, descriptive text, proprietary codes, pointers, references to data stored in other storage (including other network locations), or information used by a function to calculate the relevant data.

[0080] Instructions 1008 may be any set of instructions executed directly by a processor (such as machine code) or any set of instructions executed indirectly (such as a script). In this regard, the terms "instructions," "applications," "steps," and "programs" may be used interchangeably herein. Instructions may be stored in an object code format for direct processing by a processor, or in any other computing device language, including scripts or collections of independent source code modules that are interpreted on demand or pre-compiled. The functions, methods, and routines of instructions are explained in more detail below.

[0081] The one or more processors 1002 may include any conventional processor, such as a commercially available CPU or microprocessor. Alternatively, the processor 1002 may be a dedicated component, such as an ASIC or other hardware-based processor. Although not required, the device 1016 may include a dedicated hardware component for performing specific computing functions faster or more efficiently.

[0082] although Fig.10The processor 1002, memory 1004, and other elements of the device 1016 are functionally shown as being within the same respective boxes, but one of ordinary skill in the art will appreciate that the processor 1002 or memory 1004 may actually include multiple processors or memories that may or may not be stored within the same physical housing. Similarly, the memory 1004 may be a hard drive or other storage medium located in a housing different from that of the device 1016. Thus, references to a processor or device will be understood to include references to a collection of processors or devices or memories that may or may not operate in parallel.

[0083] Input 1010 may be, for example, a mouse, keyboard, touch screen, microphone, or any other type of input. The input may receive keywords for searching on a publisher's website or mobile application. According to some examples, input 1010 may receive indications of affirmative and negative statements about a given topic. Additionally or alternatively, input 1010 may receive a light engagement with the content. In some examples, input 1010 may receive a user interaction with the content, such as a click.

[0084] Output 1012 may be a display, such as a monitor with a screen, a touch screen, a projector, or a television. Display 1012 of a device may electronically display information to a user via a graphical user interface (“GUI”) or other type of user interface. For example, the display may electronically display product information, digital components, and advertisements on a publisher's website or mobile application, etc.

[0085] Despite Fig.10 Only one device is depicted in the figure, but it should be understood that a typical system may include one or more computing devices 1016, each of which is located at a different node of the network. The device may be able to communicate directly and indirectly with other nodes of the network. The network and intermediate nodes described herein can be interconnected using various protocols and systems, so that the network can be part of the Internet, the World Wide Web, a specific intranet, a wide area network, or a local network. The network can utilize one or more company-specific standard communication protocols, such as WiFi, Bluetooth, 4G, 5G, etc. Although certain advantages are obtained when sending or receiving information as described above, other aspects of the subject matter described herein are not limited to any particular transmission method.

[0086] System 1000 may include one or more server computing devices, such as ad server 1018 and publisher server 1020. The server computing device may be, for example, a load balancing server group that exchanges information with different nodes of a network for the purpose of receiving data from other computing devices, processing data, and sending data to other computing devices. For example, ad server 1018 and publisher server 1020 may be web servers capable of communicating with devices via a network. In addition, ad server 1018 and publisher server 1020 may use the network to send and provide information to users of device 1016. Ad server 1018 and publisher server 1020 may include one or more processors, memories, data, instructions, etc. These components operate in the same or similar manner as those described above with respect to device 1016.

[0087] Ad server 1018 can manage digital components such as advertisements and provide various services to advertisers, publishers, and devices. According to some examples, ad server 1018 can receive ad campaigns from one or more advertisers. Ad campaigns can include campaign information such as products, bidding strategies, targeting information, duration, etc. Ad campaigns can be stored in the memory of ad server 1018 and / or in ad server storage system 1022.

[0088] According to some examples, publisher server 1020 may receive a search query from device 1016. For example, the search query may be for specific content, such as product information. In response to the request, publisher server 1020 may retrieve relevant content from publisher storage system 1026. Publisher server 1020 may send a content page or other presentation, representation, or characterization of the content to requesting device 1016. The content page may include, for example, a digital component, such as an advertisement.

[0089] According to some examples, in response to the received search query, the publisher server 1020 may send a request to the ad server 1018. The request may be for one or more digital components, such as advertisements. The ad server 1018 may facilitate identifying relevant advertisements for distribution to the publisher. The relevant advertisements may be determined based at least in part on the user's merged preference profile. For example, the ad server may identify available advertisements in response to the request from the publisher server. The ad server may identify which available advertisement matches the preference in the merged preference profile. At least one of the advertisements selected to be returned to the publisher may correspond to the preference in the merged preference profile.

[0090] Device 1016 may present the search query results integrated with one or more of the advertisements provided by the ad server in a viewer such as a browser, mobile application, or other content display system.

[0091] Fig.11 An example method for providing at least one winning advertisement in response to a request from a publisher is shown. The following operations do not have to be performed in the exact order described below. Instead, various operations may be processed in a different order or simultaneously, and operations may be omitted.

[0092] In block 1102, a request for at least one digital component for display on a page of a publisher may be received. The request may be received from a publisher. The digital component may be an advertisement, such that the request from the publisher is for one or more advertisements to be displayed.

[0093] In block 1104, a user's stated preferences may be determined. The user's stated preferences may be related to at least one subject of a digital component previously provided for output. The subject of a digital component may include, for example, a brand, a topic, or a product. In some examples, the stated preferences may be determined based on user input received in conjunction with a digital component previously provided for output.

[0094] According to some examples, the declared preferences may include at least one of a pure preference statement or an affected preference statement. The pure preference statement may be determined based on at least one of a user profile or input received in response to one or more survey questions. The affected preference statement may be determined based on a user interaction with at least one of content or a digital component previously provided for output.

[0095] In block 1106, an inferred preference of the user may be determined. The inferred preference may be determined by identifying one or more engagements with a subject matter of content previously provided for output or one or more search queries for the subject matter of the content. The content may include a digital component, such as an advertisement, and the subject matter of the content includes at least one of a brand, a topic, or a product.

[0096] The user's stated preferences and inferred preferences may be merged into a merged preference profile in block 1108. The merged preference profile may include stated preferences, such as pure preference statements and / or influenced preference statements, inferred preferences, and / or contextual preferences.

[0097] In block 1110, when there is a conflict between at least one declared preference and at least one inferred preference, at least one declared preference may be selected for inclusion in the merged preference profile. According to some examples, at least one inferred preference that conflicts with at least one declared preference may be discarded.

[0098] In block 1112, at least one digital component associated with the merged preference profile may be identified. For example, in response to receiving a request for at least one digital component, a plurality of available digital components may be identified. Based on the plurality of available digital components, the at least one digital component may be identified as being associated with the merged preference profile. For example, the at least one digital component identified as being associated with the merged preference profile may have a theme corresponding to a preference indicated in the merged preference profile.

[0099] In block 1114, one or more digital components may be selected. At least one of the one or more digital components is at least one identified digital component associated with the merged preference profile. According to some examples, display space may be reserved for the selected one or more digital components associated with the merged preference profile.

[0100] In block 1116, the selected one or more digital components may be provided to the publisher for output. The publisher may output the digital components for display on its website, mobile application, etc. For example, the publisher may output the digital components along with search results or other content (such as a video, article, etc.).

[0101] By providing digital components based on the user's consolidated preference profile, the digital components are likely to be accepted and used by the user. Providing digital components that are accepted and used by the user can reduce network bandwidth and other computer-based resources, such as associated processor and memory overhead. For example, by providing digital components that are accepted and used by the user, there is no need to send unwanted and / or replacement digital components to the user to replace previously sent unwanted digital components. In addition, display space for the publisher can be saved because the publisher does not need to reserve additional display space for unwanted digital components.

[0102] Furthermore, the digital components are likely to be accepted and used by the user given the preferences indicated in the user's merged preference profile. Using the merged preference profile to select the digital components that will be accepted and used by the user increases the computational efficiency of selecting the digital components compared to determining the user preferences at the publisher's request. Furthermore, selecting the digital components based on the merged preference profile increases the accuracy and consistency of selecting the digital components that will be accepted and used by the user.

[0103] Unless otherwise stated, the foregoing alternative examples are not mutually exclusive, but may be implemented in various combinations to achieve unique advantages. Since these and other variations and combinations of the features discussed above may be utilized without departing from the subject matter defined by the claims, the foregoing description of the examples should be made by way of illustration rather than by way of limitation of the subject matter defined by the claims. In addition, the provision of examples described herein and clauses phrased as "such as," "including," etc., should not be construed as limiting the subject matter of the claims to specific examples; rather, these examples are intended to illustrate only one of many possible implementations. In addition, the same reference numerals in different figures may identify the same or similar elements.

Claims

1. A method comprising: receiving, by one or more processors, a request from a publisher for at least one digital component for display on a page of the publisher; determining, by the one or more processors, a user's stated preferences; determining, by the one or more processors, an inferred preference of a user; merging, by the one or more processors, the user's stated preferences and the inferred preferences into a merged preference profile; wherein, when there is a conflict between at least one declared preference and at least one inferred preference, selecting, by the one or more processors, the at least one declared preference for inclusion in the merged preference profile; identifying, by the one or more processors, at least one digital component associated with the merged preference profile; and selecting, by the one or more processors, one or more digital components, wherein at least one of the one or more digital components is at least one identified digital component associated with the merged preference profile; and The selected one or more digital components are provided by the one or more processors to the publisher for output. 2 . The method of claim 1 , further comprising discarding, by the one or more processors, the at least one inferred preference that conflicts with the at least one declared preference.

3. The method of claim 1 or 2, further comprising reserving, by the one or more processors, display space for selected one or more digital components associated with the merged preference profile.

4. The method according to any one of claims 1 to 3, wherein: The user's stated preference is associated with at least one theme of a digital component previously provided for output.

5. The method according to claim 4, wherein: The at least one theme of the digital component includes a brand, a topic, or a product.

6. The method according to claim 4 or 5, wherein: The declared preference is determined based on user input received in conjunction with a digital component previously provided for output.

7. The method according to any one of claims 1 to 6, wherein: The declared preference comprises at least one of a pure preference statement or an affected preference statement.

8. The method according to claim 7, wherein: The pure preference statement is determined based on at least one of a user profile or input received in response to one or more survey questions, and The affected preference statements are determined based on a user interaction with at least one of the content or the digital component previously provided for output.

9. The method according to any one of claims 1 to 8, wherein: The inferred preferences are determined by identifying one or more behaviors related to the subject matter of content previously provided for output or one or more search queries for the subject matter of the content.

10. The method according to claim 9, wherein: The content includes a digital component, and the subject matter of the content includes at least one of a brand, a topic, or a product.

11. A computer-readable storage medium comprising instructions that, when executed by one or more processors, cause the one or more processors to: receiving a request from a publisher for at least one digital component for display on a page of the publisher; determining a user's stated preferences; Determine the user's inferred preferences; merging the user's stated preferences and inferred preferences into a merged preference profile; wherein, when there is a conflict between at least one declared preference and at least one inferred preference, selecting, by the one or more processors, the at least one declared preference for inclusion in the merged preference profile; identifying at least one digital component associated with the merged preference profile; selecting one or more digital components, wherein at least one of the one or more digital components is at least one identified digital component associated with the merged preference profile; and The selected one or more digital components are provided to the publisher for output.

12. The computer-readable storage medium of claim 11, wherein: The instructions further configure the one or more processors to discard, by the one or more processors, the at least one inferred preference that conflicts with the at least one declared preference.

13. The computer-readable storage medium according to claim 11 or 12, wherein: The instructions further configure the one or more processors to reserve, by the one or more processors, display space for the selected one or more digital components associated with the merged preference profile.

14. The computer-readable storage medium according to any one of claims 11 to 13, wherein: The user's stated preference is associated with at least one theme of a digital component previously provided for output.

15. The computer-readable storage medium of claim 14, wherein: The at least one theme of the digital component includes a brand, a topic, or a product.

16. The computer-readable storage medium according to claim 14 or 15, wherein: The declared preference is determined based on user input received in conjunction with a digital component previously provided for output.

17. The computer-readable storage medium according to any one of claims 11 to 16, wherein: The declared preference comprises at least one of a pure preference statement or an affected preference statement.

18. The computer-readable storage medium of claim 17, wherein: The pure preference statement is determined based on at least one of a user profile or input received in response to one or more survey questions, and The affected preference statements are determined based on a user interaction with at least one of the content or the digital component previously provided for output.

19. The computer-readable storage medium according to any one of claims 11 to 18, wherein: The inferred preferences are determined by identifying one or more behaviors related to the subject matter of content previously provided for output or one or more search queries for the subject matter of the content.

20. The computer-readable storage medium of claim 19, wherein: The content includes a digital component, and the subject matter of the content includes at least one of a brand, a topic, or a product.

21. A system comprising: One or more processors configured to: receiving a request from a publisher for at least one digital component for display on a page of the publisher; determining a user's stated preferences; Determine the user's inferred preferences; merging the user's stated preferences and inferred preferences into a merged preference profile; wherein, when there is a conflict between at least one declared preference and at least one inferred preference, selecting the at least one declared preference for inclusion in the merged preference profile; identifying at least one digital component associated with the merged preference profile; selecting one or more digital components, wherein at least one of the one or more digital components is at least one identified digital component associated with the merged preference profile; and The selected one or more digital components are provided to the publisher for output.

22. The system of claim 21, wherein: The one or more processors are further configured to discard, by the one or more processors, the at least one inferred preference that conflicts with the at least one declared preference.

23. The system according to claim 21 or 22, wherein: The one or more processors are further configured to reserve, by the one or more processors, display space for the selected one or more digital components associated with the merged preference profile.

24. A system according to any one of claims 21 to 23, wherein: The user's stated preference is associated with at least one theme of a digital component previously provided for output.

25. The system of claim 24, wherein: The at least one theme of the digital component includes a brand, a topic, or a product.

26. A system according to claim 24 or 25, wherein: The declared preference is determined based on user input received in conjunction with a digital component previously provided for output.

27. A system according to any one of claims 21 to 26, wherein: The declared preference comprises at least one of a pure preference statement or an affected preference statement.

28. The system of claim 27, wherein: The pure preference statement is determined based on at least one of a user profile or input received in response to one or more survey questions, and The affected preference statements are determined based on a user interaction with at least one of the content or the digital component previously provided for output.

29. A system according to any one of claims 21 to 28, wherein: The inferred preferences are determined by identifying one or more behaviors related to the subject matter of content previously provided for output or one or more search queries for the subject matter of the content.

30. The system of claim 29, wherein: The content includes a digital component, and the subject matter of the content includes at least one of a brand, a topic, or a product.