Machine learning method for interface feature presentation across time zones or geographic regions

Abstract

The present disclosure relates to systems, methods, and non-transitory computer-readable media for flexibly and accurately utilizing machine learning models to intelligently determine and provide interface features for display via client devices located in different time zones or geographic regions. For example, the disclosed systems can utilize a feature visualization machine learning model to generate a graphical arrangement, graphical pairing, or other graphical visualization of one or more interface features in a target time zone (or target geographic region) based on client device interactions from other (e.g., leading) time zones or geographic regions. In certain embodiments, the disclosed systems also (or alternatively) determine a sequence of geographic regions for showcasing or presenting interface features based on similarities between the geographic regions and a comparison of performance metrics for a plurality of candidate sequences.

Description

Background Technology

[0001] Recent advancements in hardware and software platforms have spurred innovations in systems and methods for distributing digital content to client devices. For example, engineers have developed traditional digital content distribution systems that utilize analytics to distribute digital content to specific devices at specific times based on historical distribution patterns and results. However, despite these advances, many traditional digital content distribution systems still exhibit numerous shortcomings or deficiencies, particularly in terms of flexibility and efficiency.

[0002] For example, many traditional digital content distribution systems inflexibly provide or present graphical user interface features for display on client devices. Specifically, traditional systems typically distribute digital content in a rigid, uniform manner, presenting interface features in the same way, regardless of the time or location of the corresponding client device receiving the digital content. For instance, traditional systems often provide product websites that display the same layout and combination of interface features (such as digital images, banners, and product descriptions), regardless of which computing device accesses the product website.

[0003] At least partly due to their inflexibility, some existing digital content delivery systems inefficiently present rigid or navigation-intensive graphical user interfaces. More specifically, legacy systems often require excessive client device interaction to locate certain interface features, thus consuming unnecessarily large amounts of computational resources, such as processing power and memory. For example, some legacy systems maintain the nested positions of popular products within several interface layers of an application or website, requiring numerous client device interactions to navigate to and access the products from each client device using the application or website. Due to their inflexible nature, existing systems typically provide particularly popular interface features at locations nested within multiple layers of graphical user interfaces, requiring several client device interactions to locate and select them.

[0004] Therefore, traditional digital content distribution systems have drawbacks. Summary of the Invention

[0005] This disclosure describes one or more embodiments of systems, methods, and non-transient computer-readable media that address one or more of the aforementioned or other problems in the art. For example, the disclosed system utilizes a machine learning model to intelligently determine and provide interface features for display via client devices located in different time zones or geographic regions. When using such a machine learning model, the disclosed system can modify the location of graphical user interface features presented via a client device in one time zone (or geographic region) or modify different types of graphical user interface features based on the client device's interaction with interface features from other time zones (or geographic regions).

[0006] For illustration, in some cases, the disclosed system utilizes a feature visualization machine learning model to generate graphical arrangements, graphical combinations, or other graphical visualizations of one or more interface features in a target time zone (or target geographic region) based on client device interactions from other (e.g., leading) time zones or geographic regions. In some embodiments, the disclosed system also (or alternatively) determines a sequence of geographic regions for displaying or presenting interface features based on the similarity between geographic regions and a comparison of performance metrics of multiple candidate sequences. By utilizing a feature visualization machine learning model to generate graphical visualizations for presenting interface features, the disclosed system intelligently provides interface features with modified arrangements and / or combinations in a flexible and efficient manner.

[0007] Additional features and advantages of one or more embodiments of this disclosure are set forth in the following description and will be apparent in part from the description, or may be learned by practice of such exemplary embodiments. Attached Figure Description

[0008] With reference to the accompanying drawings, this disclosure describes one or more embodiments of the invention with additional specificity and detail. The following paragraphs briefly describe these drawings, in which:

[0009] Figure 1 An example system environment is shown that operates according to one or more embodiments of a smart interface feature system.

[0010] Figure 2 An overview of an intelligent interface feature system, according to one or more embodiments, generates updated graphical visualizations for a target time zone or target geographic region.

[0011] Figure 3 An example is shown of an intelligent interface feature system according to one or more embodiments using a feature visualization machine learning model to generate updated graphical visualizations;

[0012] Figure 4 An example is shown of an intelligent interface feature system according to one or more embodiments that uses a feature visualization machine learning model in the form of a feature visualization neural network to generate updated graphical visualizations.

[0013] Figure 5 An example process for training a feature visualization machine learning model according to an intelligent interface feature system based on one or more embodiments is illustrated;

[0014] Figures 6A-6B A candidate order of geographic regions for displaying interface features is shown according to one or more embodiments;

[0015] Figures 7A-7BAn example comparison of graphical visualizations according to one or more embodiments and updated graphical visualizations is shown;

[0016] Figure 8 An example digital content distribution settings interface according to one or more embodiments is shown;

[0017] Figure 9 A schematic diagram of a computing device for an intelligent interface feature system is shown according to one or more embodiments;

[0018] Figure 10 A flowchart is shown, according to one or more embodiments, of a series of actions for generating an updated graphical visualization for a target time zone using a feature visualization machine learning model based on client device interaction from a sample time zone;

[0019] Figure 11 A flowchart illustrating a series of actions, according to one or more embodiments, for generating an updated graphical visualization of a geographic region using a feature visualization machine learning model based on client device interaction; and

[0020] Figure 12 A block diagram of an example computing device according to one or more embodiments is shown. Detailed Implementation

[0021] This disclosure describes one or more embodiments of an intelligent interface feature system that utilizes machine learning methods to flexibly and efficiently provide interface features for display on client devices located in different time zones or geographical regions. Specifically, in some embodiments, compared to traditional digital content distribution systems, the intelligent interface feature system utilizes feature visualization machine learning models to generate additional graphical visualizations of graphic layouts, graphic combinations, or interface features to reduce client device interaction and improve performance metrics.

[0022] Specifically, in some cases, the intelligent interface feature system identifies client device interactions associated with interface features displayed within one or more graphical user interfaces of client devices in a sample time zone (or sample geographic region). In one or more embodiments, based on client device interactions in the sample time zone (or sample geographic region), the intelligent interface feature system utilizes a feature visualization machine learning model to generate updated graphical visualizations for a target time zone or target geographic region (e.g., to present interface features earlier, to present interface features in a different location, and / or to increase overall client device interactions). Furthermore, in some cases, the intelligent interface feature system compares multiple candidate orders or sequences of geographic regions to select an order or sequence of geographic regions for displaying interface features to achieve a specific performance metric.

[0023] As described above, in one or more embodiments, the intelligent interface feature system utilizes a feature visualization machine learning model to generate graphical visualizations of interface features. For example, the intelligent interface feature system generates graphical visualizations of a set of interface features, wherein the graphical visualizations include or span one or more graphical user interfaces that can be interactively navigated via a client device. In some cases, the intelligent interface feature system generates graphical visualizations to depict or visualize a set of interface features in a specific arrangement and / or combination within a graphical user interface displayed on a client device located within (or otherwise associated with) a sample time zone (or sample geographical region). For example, the intelligent interface feature system generates graphical visualizations for display via a client device located in a leading time zone preceding the target time zone.

[0024] As further mentioned, in one or more embodiments, the intelligent interface feature system updates or modifies the graphical visualization for a target time zone or target geographic region. For example, in some cases, the intelligent interface feature system updates the graphical visualization for the target time zone, as notified by client device interactions from sample (e.g., leading) time zones preceding the target time zone. For example, the intelligent interface feature system generates an updated graphical visualization based on client device interactions from sample time zones (or the same geographic region) for display on client devices located within the target time zone or target geographic region.

[0025] In some embodiments, the intelligent interface feature system determines client device interactions associated with a set of interface features included in a graphical visualization provided for a sample time zone (or sample geographic region). In these or other embodiments, the intelligent interface feature system also utilizes modified arrangements and / or combinations of interface features based on client device interactions to generate updated graphical visualizations for a target time zone (or target geographic region). For example, the intelligent interface feature system generates updated graphical visualizations to reposition, remove, and / or rearrange interface features based on client device interactions (e.g., to present highly selected interface features earlier or more prominently and / or to increase performance metrics such as clicks or conversions).

[0026] As also mentioned, in one or more embodiments, the intelligent interface feature system determines the order or sequence of geographic regions used to display interface features. For example, the intelligent interface feature system selects an initial geographic region and generates an order (e.g., a similarity chain) of geographic regions similar to the initial geographic region based on similarity scores between regions. In some cases, the intelligent interface feature system determines the similarity scores between geographic regions based on historical network user behavior.

[0027] In some embodiments, the intelligent interface feature system generates a specific similarity chain or other sequences of geographic regions for displaying the interface by testing candidate sequences of geographic regions. Specifically, in some embodiments, the intelligent interface feature system provides interface features for display on client devices in successive geographic regions based on multiple candidate sequences to determine which candidate sequence performs well relative to others (e.g., which candidate sequence produces at least a threshold performance metric associated with the interface features). In these or other embodiments, the intelligent interface feature system selects a sequence with at least a threshold performance and provides interface features for display sequentially or successively in the geographic regions according to the selected sequence.

[0028] As described above, in some embodiments, the intelligent interface feature system offers certain improvements or advantages over traditional digital content distribution systems. For example, in some embodiments, the intelligent interface feature system improves flexibility compared to traditional systems by dynamically adapting graphical visualizations to target time zones or target geographic regions. Specifically, compared to traditional systems that are rigidly fixed to uniformly distribute digital content across time zones or other geographic regions, the intelligent interface feature system utilizes feature visualization machine learning models to flexibly adapt graphical visualizations to time zones or other geographic regions. For example, the intelligent interface feature system generates graphical visualizations of interface features intelligently customized for specific time zones or geographic regions based on client device interactions related to interface features in a previous time zone or other geographic region.

[0029] At least in part, due to the increased flexibility of intelligent interface feature systems, some embodiments of intelligent interface feature systems also improve the efficiency of existing digital content distribution systems. Specifically, in some embodiments, intelligent interface feature systems intelligently rearrange, reorder, or otherwise modify interface features based on client device interactions. Therefore, compared to existing systems, intelligent interface feature systems reduce the number of client device interactions required to access the desired data and / or functionality. For example, intelligent interface feature systems generate updated or modified graphical visualizations with improved combinations and / or arrangements of interface features. In some cases, intelligent interface feature systems generate graphical visualizations that present more popular (e.g., highly selected) interface features in new locations and / or at earlier views within a series of graphical user interfaces.

[0030] While traditional systems typically require the same number of client device interactions to navigate through multiple graphical user interfaces to access highly selected interface features for each time zone or geographic region, intelligent interface feature systems adapt to reduce client device interactions by presenting highly selected interface features earlier in an updated graphical visualization. Therefore, compared to traditional systems, intelligent interface feature systems reduce the computational demands on handling client device interactions, thereby saving processing power and memory.

[0031] This at least partially contributes to improving the flexibility and efficiency of intelligent interface feature systems. In some embodiments, the intelligent interface feature system utilizes a novel machine learning model not found in existing systems: the feature visualization machine learning model. In practice, the intelligent interface feature system trains and utilizes the feature visualization machine learning model to generate updated graphical visualizations to improve performance metrics and / or reduce client device interactions by presenting interface features in modified combinations and / or arrangements.

[0032] As suggested in the foregoing discussion, this disclosure uses various terms to describe the features and benefits of intelligent interface feature systems. Additional details regarding the meaning of these terms as used in this disclosure are provided below. Specifically, the term "machine learning model" refers to a computer algorithm or set of computer algorithms that automatically improves a specific task through experience based on data usage. For example, a machine learning model may utilize one or more learning techniques to improve accuracy and / or effectiveness. Example machine learning models include various types of decision trees, support vector machines, Bayesian networks, linear regression, logistic regression, random forest models, or neural networks.

[0033] As an example of a specific machine learning model, the intelligent interface feature system utilizes a feature visualization machine learning model. As used herein, the term "feature visualization machine learning model" refers to a machine learning model that generates predictions of interface features and / or places interface features at appropriate locations within one or more graphical user interfaces (e.g., a sequence or set of graphical user interfaces). For example, in some cases, a feature visualization machine learning model generates predicted arrangements, combinations, designs, or organizations of one or more interface features or their corresponding locations. Such predictions may take the form of numerical indicators of interface features and / or locations (e.g., numbers in a matrix), a series of labels representing interface features and / or locations, binary classifications or classification series (e.g., a matrix indicating a "yes" or "no" term selected by the user), graphical representations of interface features and / or locations, or other appropriate outputs. For example, a feature visualization machine learning model generates predictions (or locations) of graphical visualizations of interface features based on client device interactions (e.g., reducing client device interactions to targeted popular interface features) and / or target performance metrics to, for example, increase clicks or increase conversions.

[0034] In some cases, feature visualization machine learning models include feature visualization neural networks. As used herein, the term "feature visualization neural network" refers to a neural network trained or tuned to generate or predict graphical visualizations based on client device interactions. Relatedly, the term "neural network" refers to a machine learning model that can be trained and / or tuned based on inputs to determine a classification or approximate an unknown function. For example, a neural network includes a model of interconnected artificial neurons (e.g., organized in layers) that communicate and learn to approximate complex functions and generate outputs (e.g., generated digital images) based on multiple inputs provided to the neural network. In some cases, a neural network refers to an algorithm (or set of algorithms) that implements deep learning techniques to model high-level abstractions in data. For example, a neural network can include convolutional neural networks, recurrent neural networks (e.g., LSTM), graph neural networks, or generative adversarial neural networks.

[0035] In some embodiments, the feature visualization machine learning model includes a collaborative filtering recommendation model. As used herein, the term "collaborative filtering recommendation model" refers to a machine learning model that generates predictions of graphical visualizations for interface features based on attributing client device interactions from one time zone (or geographic region) to another. For example, a collaborative filtering recommendation model determines predictions of interface features based on the determined order of time zones or geographic regions. In some cases, a collaborative filtering recommendation model determines the probability that one time zone (or region) will have the same (or similar within a threshold range) client device interactions as another time zone (or region) based on similarity scores between time zones (or regions). Therefore, based on the determination that two time zones (or regions) will have similar client device interactions, a collaborative filtering recommendation system can predict the graphical visualizations to be distributed to one time zone (or region) based on client device interactions in another time zone (or region).

[0036] As used herein, the term "interface feature" refers to a displayable digital content feature, object, or element within a graphical user interface. For example, interface features include elements depicted within a graphical user interface, such as titles, banners, digital images, blocks of digital text (e.g., text descriptions), icons (optional or non-optional), numbers, graphics, or certain other displayable objects of digital content. In some cases, interface features are related to a product or service. For example, interface features may include an advertising banner displayed within a website targeting a specific product. As another example, a website targeting a specific product may include a set of interface features associated with that product, such as a product name, product description, a digital image depicting the product, and one or more optional elements for adding the product to a digital shopping cart, scrolling through other digital images, viewing related products, or displaying reviews of the product.

[0037] In some embodiments, the intelligent interface feature system provides or presents interface features within a graphical visualization. As used herein, the term "graphical visualization" refers to a graphical arrangement, layout, design, or organization of one or more interface features within one or more graphical user interfaces. In some cases, a graphical visualization includes a graphical representation of an image, menu or optional option, product, tool, video, or other interface feature in a particular arrangement, layout, design, or organization. For example, a graphical visualization can refer to the layout of interface elements at a specific location within a webpage. As another example, a graphical visualization can refer to the arrangement of multiple webpages or multiple graphical user interfaces, each webpage or graphical user interface including corresponding interface features, and these features are presented in response to client device interactions to navigate between or across different webpages or graphical user interfaces. In this case, the graphical visualization can include highly selected interface features via client device interactions nested within several layers of a graphical user interface.

[0038] As described above, graphical visualization can include the arrangement and / or combination of interface features. As used herein, the term "arrangement" refers to the layout, placement, or formation of interface features. For example, an arrangement can include or indicate the pixel coordinate positions of individual interface elements, including indications of which interface features are included in which graphical user interfaces and where they are located. Relatedly, the term "combination" refers to the selection or set of interface features. For example, a combination can include the names or identifiers of interface features included within a graphical user interface, and may also (or alternatively) include the names or identifiers of interface features excluded from (or not included in) the graphical user interface.

[0039] In one or more embodiments, the intelligent interface feature system determines similarity scores between time zones or geographic regions based on web user behavior. As used herein, the term "web user behavior" refers to client device interactions received from a client device in relation to interface features or other digital content. For example, web user behavior includes information indicating the time and location of clicks, conversions, views, scrolls, and other behavioral data. In some cases, web user behavior indicates the time zone or geographic region in which an action such as a click, conversion, or view was initiated, and further indicates the timing of when the action occurred and / or the identity of the interface feature(s) clicked, viewed, or otherwise acted upon.

[0040] As described above, in some embodiments, the intelligent interface feature system generates updated graphical visualizations to improve performance metrics associated with previous versions of the graphical visualizations. As used herein, the term "performance metric" refers to a measure or metric that indicates a client device interaction in response to a particular interface feature (or combination of interface features). For example, performance metrics include selection counts, click counts, conversion counts, view counts, scroll counts, or certain other counts of client device interactions attributable to one or more interface features. In some cases, a performance metric indicates the total number of client device interactions of a certain type, while in others, it indicates the frequency (e.g., number of times per unit time) and / or recentity of client device interactions. In some embodiments, a performance metric is a combination (e.g., a weighted combination) of two or more client device interactions, such as a combination of clicks, views, and conversions attributable to an interface feature (or combination of interface features).

[0041] Additional details regarding the intelligent interface feature system will now be provided with reference to the accompanying drawings. For example, Figure 1 A schematic diagram of an example system environment for implementing the intelligent interface feature system 102 according to one or more embodiments is shown. (Reference) Figure 1 An overview of the intelligent interface feature system 102 is described below. A more detailed description of the components and processes of the intelligent interface feature system 102 is then provided with reference to the following figures.

[0042] As shown in the figure, the environment includes server 104, client devices 108a-108n, administrator device 110, database 112, and network 114. Each component of the environment communicates via network 114, and network 114 is any suitable network on which computing devices communicate. The following will combine... Figure 12 The example network will be discussed in more detail.

[0043] As described above, this environment includes client devices 108a-108n. Client devices 108a-108n are among a variety of computing devices, including smartphones, tablets, smart TVs, desktop computers, laptops, virtual reality devices, augmented reality devices, or similar devices. Figure 12 Another computing device described. Although Figure 1Multiple client devices 108a-108n are shown, but in some embodiments, the environment includes more or fewer client devices, each associated with a different user (e.g., a digital content viewer). In one or more embodiments, client devices 108a-108n are located in different time zones or geographic regions, wherein a first set of client devices (e.g., client devices 108a-108g) is located in a first (e.g., sample) time zone or a first (e.g., sample) geographic region, while a second set of client devices (e.g., client devices 108h-108n) is located in a second (e.g., target) time zone or a second (e.g., target) geographic region. Client devices 108a-108n communicate with server 104 via network 114. For example, client devices 108a-108n provide server 104 with information instructing client device interactions (e.g., clicks, views, and scrolling regarding interface features) and receive information from server 104, such as graphical visualizations including interface features. Therefore, in some cases, the intelligent interface feature system 102 on server 104 provides and receives information based on client device interaction via client devices 108a-108n.

[0044] like Figure 1 As shown, client devices 108a-108n include corresponding client applications 118a-118n. Specifically, client applications 118a-118n are web applications, local applications installed on client devices 108a-108n (e.g., mobile applications, desktop applications, etc.), or cloud-based applications in which all or part of their functionality is executed by server 104. Based on instructions from client applications 118a-118n, client devices 108a-108n present or display information to users, including graphical visualizations depicting one or more interface features.

[0045] like Figure 1 As shown, the environment includes server 104. Server 104 generates, tracks, stores, processes, receives, and sends electronic data, such as graphical visualizations and client device interactions. For example, server 104 receives data from client devices 108a-108n in the form of clicking, scrolling, viewing, or other client device interactions. In response, server 104 sends data to client devices 108a-108n to cause client devices 108a-108n to display or present graphical visualizations including interface features arranged or matched according to the client device interactions.

[0046] As described above, in some embodiments, server 104 communicates with client devices 108a-108n to send and / or receive data via network 114. In some embodiments, server 104 includes a distributed server, wherein server 104 includes multiple server devices distributed across network 114 and located in different physical locations. Server 104 may include a content server, application server, communication server, web hosting server, multidimensional server, or machine learning server. Server 104 may also access and utilize database 112 to store and retrieve information, such as feature visualization machine learning models, one or more graphical visualizations describing interface features, and client device interactions related to the displayed interface features.

[0047] like Figure 1 As further shown, the environment also includes an administrator device 110. Specifically, the administrator device 110 can be one of various computing devices, including smartphones, tablets, smart TVs, desktop computers, laptops, virtual reality devices, augmented reality devices, or similar devices. Figure 12 The other computing device mentioned above. In some cases, administrator device 110 communicates with server 104 to receive and / or provide information, such as digital content distribution parameters or settings.

[0048] For example, administrator device 110 includes administrator application 116. Specifically, the administrator application is a web application, a local application installed on administrator device 110 (e.g., a mobile application, desktop application, etc.), or a cloud-based application in which all or part of its functionality is performed by server 104. Based on instructions from administrator application 116, administrator device 110 presents or displays information to users, including a digital content distribution settings interface. Based on interactions with various interface elements within the digital content distribution settings interface, administrator device 110 can provide instructions to server 104 to modify the distribution of digital content within a graphical visualization provided to client devices located in different time zones or geographical regions.

[0049] like Figure 1As further shown, server 104 also includes an intelligent interface feature system 102 as part of digital content management system 106. For example, in one or more implementations, digital content management system 106 can store, generate, modify, edit, enhance, provide, distribute, and / or share digital content, such as interface features within graphical visualizations. For instance, digital content management system 106 provides graphical visualizations for display on client devices 108a-108n via client applications 118a-118n, for viewing and interacting with interface features within the graphical visualizations. In some implementations, digital content management system 106 provides tools to administrator device 110 via administrator application 116 to define settings or parameters for distributing interface features to different time zones or geographic regions.

[0050] In one or more embodiments, server 104 includes all or part of intelligent interface feature system 102. For example, intelligent interface feature system 102 operates on server 104 to generate updated graphical visualizations for client devices in a target time zone or target geographic region. In some cases, intelligent interface feature system 102 utilizes a feature visualization machine learning model, either locally on server 104 or from another network location (e.g., database 112), to generate updated graphical visualizations based on client device interactions from sample time zones or sample geographic regions.

[0051] In some cases, client devices 108a-108n include all or part of the intelligent interface feature system 102. For example, client devices 108a-108n may generate, obtain (e.g., download), or utilize one or more aspects of the intelligent interface feature system 102, such as a feature visualization neural network from server 104. In practice, in some implementations, such as... Figure 1 As shown, the intelligent interface feature system 102 is located in all or part of the client devices 108a-108n and / or the administrator device 110. For example, the intelligent interface feature system 102 includes a web-hosted application that allows the client devices 108a-108n and / or the administrator device 110 to interact with the server 104. For illustration, in one or more implementations, the client devices 108a-108n and / or the administrator device 110 access web pages supported and / or hosted by the server 104.

[0052] although Figure 1A specific arrangement of the environment is shown, but in some embodiments, the environment has a different component arrangement and / or may have a completely different number of components or sets of components. For example, as described above, the intelligent interface feature system 102 is implemented by client devices 108a-108n (e.g., wholly or partially located on client devices 108a-108n). Furthermore, in one or more embodiments, client devices 108a-108n and / or administrator device 110 communicate directly with the intelligent interface feature system 102, bypassing network 114. Additionally, in some embodiments, the environment includes a feature visualization machine learning model stored in database 112, which is maintained by server 104, client devices 108a-108n, administrator device 110, or a third-party device.

[0053] As described above, in one or more embodiments, the intelligent interface feature system 102 generates updated graphical visualizations, including modified interface features for display on client devices located in a target time zone or target geographic region. Specifically, the intelligent interface feature system 102 utilizes a feature visualization machine learning model to generate updated graphical visualizations from client device interactions in a sample time zone or sample geographic region. Figure 2 Example sequences of actions 202-218 for generating updated graphical visualizations according to one or more embodiments are shown. Figure 2 It provides an overview of generating updated graphical visualizations, and about Figure 2 Additional details regarding the specific actions are provided in the accompanying diagrams.

[0054] like Figure 2 As shown, the intelligent interface feature system 102 performs action 202 to determine the geographic region or time zone for the interface feature. Specifically, the intelligent interface feature system 102 determines or identifies a specific geographic region or time zone where the interface feature is presented for display on one or more client devices. For example, the intelligent interface feature system 102 determines a sample time zone or sample geographic region. In some cases, the intelligent interface feature system 102 determines the sample time zone as a leading time zone preceding the target time zone. In one or more embodiments, the intelligent interface feature system 102 distinguishes or identifies geographic regions based on their boundaries or limits and / or based on content distribution regions. In some embodiments, the intelligent interface feature system 102 determines geographic regions of arbitrary size, divided at any level of granularity, and varying to any degree between regions.

[0055] like Figure 2As further illustrated, the intelligent interface feature system 102 performs action 204 to generate a graphical visualization for a geographic region or time zone. Specifically, the intelligent interface feature system 102 generates a graphical visualization that includes or depicts one or more interface features (e.g., a set of interface features). For example, the intelligent interface feature system 102 generates a graphical visualization that is a series of graphical user interfaces, each including interface features displayed based on navigation interacted with via a client device. In some cases, the intelligent interface feature system 102 generates a graphical visualization for display on a client device located within a sample time zone or sample geographic region (as determined via action 202).

[0056] Furthermore, the intelligent interface feature system 102 performs action 206 to determine client device interactions for a sample geographic region or sample time zone. More specifically, the intelligent interface feature system 102 monitors or detects client device interactions related to interface features displayed within a graphical visualization, such as viewing, clicking, scrolling, and conversion. In some embodiments, the intelligent interface feature system 102 determines client device interactions for each individual client device and for each time zone or geographic region. As described above, the intelligent interface feature system 102 determines the number, frequency, and / or relevance (e.g., regarding individual interface features) of different types of client device interactions. In some cases, the intelligent interface feature system 102 determines or detects that some interface features are selected, viewed, or otherwise interacted with more frequently (or more often) than other interface features.

[0057] like Figure 2 As further illustrated, in some embodiments, the intelligent interface feature system 102 performs action 208 to determine a similarity score for a geographic region or time zone. Specifically, the intelligent interface feature system 102 determines a similarity score between two geographic regions (or two time zones) based on historical network user behavior. For example, the intelligent interface feature system 102 determines historical network user behavior, such as interactions between client devices and interface features determined via action 206. In some cases, the intelligent interface feature system 102 determines additional or alternative network user behavior related to other digital content that is not part of the graphical visualization. For example, based on specific device licensing and regulatory requirements, the intelligent interface feature system 102 monitors network traffic for different websites or applications to determine the type of interface features that receive client device interactions within a specific geographic region or time zone.

[0058] Based on client device interactions and / or other network user behaviors, the intelligent interface feature system 102 generates a similarity score. Specifically, the intelligent interface feature system 102 generates a similarity score between two geographic regions, which indicates a measure of similarity or resemblance between collective client device interactions and / or other network user behaviors associated with (e.g., occurring within) the corresponding geographic region. In some cases, the intelligent interface feature system 102 generates the similarity score based on additional or alternative factors, such as demographic similarities between geographic regions or time zones, such as age, gender, race, ethnicity, income, education, and language.

[0059] like Figure 2 As shown, in one or more embodiments, the intelligent interface feature system 102 performs action 210 to determine an order for geographic regions or time zones. Specifically, in some embodiments, the intelligent interface feature system 102 determines the order or sequence of geographic regions used to display interface features based on similarity scores between geographic regions. For example, the intelligent interface feature system 102 selects an initial geographic region (or receives its indication) and determines the next geographic region with the highest similarity score to that initial geographic region. In a similar manner, the intelligent interface feature system 102 determines each successive geographic region within the order by linking newly ordered geographic regions to the next geographic region, which has the highest similarity score relative to the newly ordered geographic region among the remaining geographic regions. Thus, based on the different initial geographic regions selected, the intelligent interface feature system 102 generates a completely new order of geographic regions based on similarity scores, linking regions together based on the highest available similarity score.

[0060] like Figure 2 As further illustrated, the intelligent interface feature system 102 performs action 212 to generate an updated graphical visualization. More specifically, the intelligent interface feature system 102 utilizes a feature visualization machine learning model to generate an updated arrangement and / or combination of interface features. In some cases, the intelligent interface feature system 102 utilizes a feature visualization machine learning model to generate an updated graphical visualization. For example, the intelligent interface feature system 102 utilizes a collaborative filtering recommendation model or a feature visualization machine learning model in the form of a feature visualization neural network to generate an updated graphical visualization for a target time zone or target geographic region.

[0061] In practice, the intelligent interface feature system 102 generates updated graphical visualizations to modify the arrangement and / or combination of interface features compared to the initial graphical visualizations provided to client devices within a sample time zone or sample geographic region. In some cases, the intelligent interface feature system 102 generates updated graphical visualizations earlier than the initial graphical visualizations to present popular interface features (e.g., associated with at least a threshold number of client device interactions such as viewing, clicking, scrolling, and / or converting). In some embodiments, the intelligent interface feature system 102 generates updated graphical visualizations to improve or increase specific performance metrics within a target time zone or target geographic region. For example, the intelligent interface feature system 102 receives indications of target performance metrics (e.g., increases, views, clicks, and / or conversions) from an administrator device 110. The intelligent interface feature system 102 also generates updated graphical visualizations to include interface features predicted to achieve the target performance metrics.

[0062] like Figure 2 As shown, the intelligent interface feature system 102 performs action 214 to provide an updated graphical visualization for display. Specifically, the intelligent interface feature system 102 provides an updated graphical visualization for display on client devices located in a target time zone or target geographic area. In practice, the intelligent interface feature system 102 provides an updated graphical visualization to display or present interface features with a modified layout and / or modified arrangement, wherein the interface features are located in different positions (e.g., at different coordinate locations and / or within different graphical user interfaces) compared to the initial graphical visualization.

[0063] As described above, the intelligent interface feature system 102 sometimes targets a time zone. Based on client device interactions from a sample (e.g., leading) time zone, the intelligent interface feature system 102 generates and provides an updated graphical visualization for display on client devices in the target time zone. In some cases, the updated graphical visualization includes popular interface features earlier or in more easily locating locations (e.g., with fewer navigation client device interactions). Therefore, for the target time zone, the intelligent interface feature system 102 reduces the number of client device interactions required to access certain interface features within the graphical visualization compared to the initial graphical visualization.

[0064] As described above, the intelligent interface feature system 102 sometimes targets geographic regions. In some embodiments, the intelligent interface feature system 102 provides updated graphical visualizations to an initial geographic region within a defined geographic region sequence (e.g., as determined via action 210). For example, the intelligent interface feature system 102 provides updated graphical visualizations for display on client devices located within the initial geographic region, while excluding client devices in other geographic regions (e.g., by limiting the provision of updated graphical visualizations or by providing the initial graphical visualizations instead). The intelligent interface feature system 102 also sequentially provides updated graphical visualizations to each successive geographic region.

[0065] like Figure 2 As further illustrated, in one or more embodiments, the intelligent interface feature system 102 performs action 216 to determine if there are additional geographic region sequences. Specifically, the intelligent interface feature system 102 determines whether every possible sequence of geographic regions has been tested to provide the updated graphical visualization. Once it is determined that more sequences remain, the intelligent interface feature system 102 returns to action 206 to determine client device interactions associated with the updated graphical visualization. In practice, the intelligent interface feature system 102 determines client device interactions that indicate one or more performance metrics associated with the graphical visualization provided to the geographic regions within the first sequence. For example, the intelligent interface feature system 102 determines clicks, views, scrolling, and / or conversions associated with specific interface features of the updated graphical visualization.

[0066] In some cases, when repeating actions for different orders, the intelligent interface feature system 102 repeats action 208 to determine new similarity scores between geographic regions. In practice, based on new client device interactions related to the updated graphical visualization (from each geographic region in the initial order), the intelligent interface feature system 102 determines new similarity scores, where regions that initially had low similarity scores may now have high similarity scores (and vice versa). Furthermore, the intelligent interface feature system 102 repeats action 210 to determine a new order of geographic regions based on the updated similarity scores. For example, the intelligent interface feature system 102 selects a new initial geographic region (or receives a new indication of it) and sorts the remaining geographic regions in a similarity chain as before, where the most similar available geographic region is linked to the last geographic region in that order.

[0067] In such a repetitive cycle, in some embodiments, the intelligent interface feature system 102 repeats actions 212 and 214 to generate another updated graphical visualization and provide the newly updated graphical visualization for display. Specifically, the intelligent interface feature system 102 generates the newly updated graphical visualization based on client device interactions with the previously updated graphical visualization. However, in some embodiments, the intelligent interface feature system 102 does not generate the newly updated graphical visualization, but instead skips action 212 and only performs action 214 to provide the previously updated graphical visualization for display. Specifically, the intelligent interface feature system 102 sequentially provides updated graphical visualizations (e.g., newly updated or previously updated graphical visualizations) for a new initial geographic region and for each subsequent geographic region.

[0068] The intelligent interface feature system 102 further repeats action 216 to determine if there are any additional sequences of remaining geographic regions. In effect, the intelligent interface feature system 102 determines whether all possible permutations of the geographic region sequences have been tested for interaction with the client device. However, in some embodiments, the intelligent interface feature system 102 determines whether a threshold number of sequences have been tested, rather than determining whether all possible candidate sequences have been tested. Once it is determined that more sequences remain, the intelligent interface feature system 102 repeats actions 206-216 as described above.

[0069] However, once it is determined that there are no remaining sequences (or sequences that have already been tested to the threshold number), the intelligent interface feature system 102 performs action 218 to select an order for presenting one or more interface features. Specifically, the intelligent interface feature system 102 selects an order from candidate sequences generated via action 210 and tested as described above, based on performance metrics. For example, the intelligent interface feature system 102 selects the order from the candidate sequences that produces the highest or best performance metric. In some cases, the intelligent interface feature system 102 selects the order that produces (or is predicted to produce) the highest number (or frequency) of one or more target performance metrics.

[0070] In practice, in one or more embodiments, the intelligent interface feature system 102 receives indications of one or more target performance metrics (e.g., from administrator device 110). For example, the intelligent interface feature system 102 receives indications of a specific number (or frequency) of clicks, views, scrolls, conversions (or combinations of two or more thereof) as target performance metrics. As another example, instead of receiving a specific number of indications, the intelligent interface feature system 102 receives indications to increase or maximize one type of client device interaction (e.g., clicks, views, scrolls, or conversions) or another type of client device interaction. In some cases, the intelligent interface feature system 102 also compares the target performance metric with performance metrics of various candidate sequences to select an order that satisfies the target performance metric.

[0071] As described above, in some of the described embodiments, the intelligent interface feature system 102 generates an updated graphical visualization for a target time zone or target geographic region (e.g., a geographic region following an initial geographic region in a geographic region sequence). Specifically, the intelligent interface feature system 102 generates the updated graphical visualization based on client device interactions from sample time zones or sample geographic regions. Figure 3 An updated graphical visualization generated according to one or more embodiments is shown.

[0072] like Figure 3 As shown, the intelligent interface feature system 102 identifies the sample time zone 302 and the target time zone 304. Specifically, the intelligent interface feature system 102 identifies the target time zone 304 based on an instruction received from the administrator device 110. For example, as part of a digital content activity for distributing digital content, the intelligent interface feature system 102 receives an instruction from the administrator to select a specific time zone as the target. In some cases, the intelligent interface feature system 102 receives an instruction for a target geographic area (not necessarily defined by a time zone) in a similar manner.

[0073] Furthermore, the intelligent interface feature system 102 identifies the sample time zone 302 as a time zone with threshold similarity relative to the target time zone 304. Specifically, the intelligent interface feature system 102 determines a similarity score between the target time zone 304 and multiple candidate time zones (e.g., one or more time zones preceding the target time zone 304). For example, the intelligent interface feature system 102 determines the similarity score based on historical network user behavior. Specifically, the intelligent interface feature system 102 monitors the behavior of network users over time in the target time zone 304 and other (e.g., preceding) time zones (including the sample time zone 302) to determine client device interactions, such as clicks, views, scrolling, and conversions, that are related to one or more interface features of the graphical visualization and / or to other digital content distributed via network 114 or some other network.

[0074] In response to the detection of similar types, quantities, and / or frequencies of client device interactions from client devices in different time zones (e.g., from client devices associated with user accounts sharing attributes such as demographic information), the intelligent interface feature system 102 determines a similarity score. Furthermore, the intelligent interface feature system 102 compares the similarity score to a threshold similarity score. Although the foregoing discussion primarily concerns time zones, the intelligent interface feature system 102 can also determine similarity scores between geographic regions (e.g., between a target geographic region and sample geographic regions) in a similar manner based on historical network user behavior indicating client device interactions over time.

[0075] like Figure 3 As shown, the intelligent interface feature system 102 determines a similarity score 306 between the sample time zone 302 and the target time zone 304. Furthermore, the intelligent interface feature system 102 determines that the similarity score 306 meets a threshold similarity score. However, in some embodiments, the intelligent interface feature system 102 does not necessarily determine the similarity score as the basis for selecting the sample time zone (e.g., sample time zone 302) relative to the target time zone (e.g., target time zone 304). Instead, the intelligent interface feature system 102 identifies the sample time zone as the time zone immediately preceding the target time zone 304 (e.g., with no other time zones in between).

[0076] like Figure 3 As further illustrated, the intelligent interface feature system 102 determines client device interactions 308 from a sample time zone 302 (or a sample geographic region). In fact, as described above, the intelligent interface feature system 102 determines client device interactions 308 associated with an initial graphical visualization provided for display via a client device located within the sample time zone 302 (or sample geographic region). For example, the intelligent interface feature system 102 detects or determines clicks, views, scrolling, and / or conversions associated with one or more interface features of the graphical visualization.

[0077] In practice, in some cases, the intelligent interface feature system 102 tracks client device interactions based on interface features to associate or categorize these interactions with individual interface features. For example, the intelligent interface feature system 102 detects clicks on an interface feature as client device interactions associated with that feature. In some cases, the intelligent interface feature system 102 compares client device interactions associated with each interface feature to determine which interface features are interacted with more than others. In some embodiments, the intelligent interface feature system 102 also (or alternatively) compares the number (or frequency) of client device interactions associated with an interface feature to a client device interaction threshold to, for example, identify interface features associated with at least a threshold number (or frequency) of client device interactions.

[0078] As another example, the intelligent interface feature system 102 identifies or segments portions or percentages of attributes across multiple interface features in a client device interaction. For instance, the intelligent interface feature system 102 determines that a click, view, or conversion is triggered or caused by a combination of two or more interface features, and accordingly segments the attributes of the client device interaction. In some cases, the intelligent interface feature system 102 determines the proportion of a specific attribute for each interface feature based on a comparison of the time spent viewing an interface feature, the number of clicks on the interface feature, the number of views of the interface feature, and / or the number of times scrolling through the interface feature.

[0079] like Figure 3 As further illustrated, the intelligent interface feature system 102 utilizes the feature visualization machine learning model 310 to generate updated combinations and / or arrangements of interface features. For example, the intelligent interface feature system 102 generates an updated graphical visualization 314 from client device interaction 308. In practice, the intelligent interface feature system 102 generates the updated graphical visualization 314 for a target time zone 304 (or target geographic region). For example, the intelligent interface feature system 102 generates an updated graphical visualization to include one or more interface features arranged and / or combined differently from the interface features initially provided to the sample time zone 302.

[0080] In some cases, the intelligent interface feature system 102 generates an updated graphical visualization 314 to include one or more interface features that are the same as the initial graphical visualization and / or different from the initial graphical visualization, wherein the interface features are placed in modified positions within the same or different graphical user interfaces. For example, the intelligent interface feature system 102 generates an updated graphical visualization 314 to modify the placement of interface features that are associated with at least a threshold number (or frequency) of client device interactions based on client device interactions 308 and / or with more client device interactions than one or more other interface features. In practice, the intelligent interface feature system 102 generates the updated graphical visualization 314 to include one or more interface features based on selection counts from clicks (and / or counts of other client device interactions). Thus, the intelligent interface feature system 102 presents more popular or effective interface features in more prominent positions, such as higher positions on a webpage, more central positions on a webpage, or earlier positions in a series or sequence of webpages.

[0081] In some embodiments, the feature visualization machine learning model 310 is a collaborative filtering recommendation model. In these or other embodiments, the intelligent interface feature system 102 utilizes the feature visualization machine learning model 310 to generate updated arrangements and / or combinations of interface features. Specifically, the intelligent interface feature system 102 generates updated graphical visualizations 314 by performing collaborative filtering on client device interactions between sample time zones 302 (e.g., client device interactions 308) and target time zones 304. For example, based on a similarity score 306, the feature visualization machine learning model 310 generates predictions that a portion or percentage of client device interactions 308 from sample time zones 302 may also occur within the target time zone 304. In effect, the feature visualization machine learning model 310 generates predictions for different arrangements and / or combinations of interface features based on collaborative filtering.

[0082] For example, the feature visualization machine learning model 310 generates predictions for graphical visualizations of a target time zone 304. The feature visualization machine learning model 310 generates predictions based on the assumption that if the sample time zone 302 produces a set of client device interactions for an initial graphical visualization (and the sample time zone 302 and the target time zone 304 are within a threshold similarity), then the target time zone 304 will produce a specific set of client device interactions for different graphical visualizations with modified arrangements and / or combinations of interface features. In practice, based on monitoring client device interactions, the feature visualization machine learning model 310 generates or determines relationships between nodes and client devices representing interface features within the sample time zone 302 and / or the target time zone 304. Based on the determined relationships, the intelligent interface feature system 102 populates or predicts client device interactions between nodes that do not indicate actual observed or received data about client device interactions, such as predicted client device interactions for different arrangements and / or combinations of interface features from client devices in the target time zone 304.

[0083] Based on the established relationships or other factors, the feature visualization machine learning model 310 accordingly predicts the arrangement and / or combination of interface features for generating an updated graphical visualization 314 for the target time zone 304. For example, the intelligent interface feature system 102 generates an updated graphical visualization 314 to present one or more interface features (e.g., more popular or more effective interface features) in more effective locations compared to the initial graphical visualization. Thus, the intelligent interface feature system 102 handles fewer client device interactions from client device navigation to locate specific interface features.

[0084] like Figure 3As further illustrated, in some embodiments, the intelligent interface feature system 102 generates an updated graphical visualization 314 based on the target performance metric 312. Specifically, the intelligent interface feature system 102 receives an indication of the target performance metric 312 from the administrator device 110. For example, the intelligent interface feature system 102 receives an indication to add views (and / or clicks and / or conversions) for the updated graphical visualization 314 (or for a specific interface feature).

[0085] Based on the target performance metric 312 in conjunction with the client device interaction 308, the intelligent interface feature system 102 generates an updated graphical visualization 314. Specifically, the intelligent interface feature system 102 generates the updated graphical visualization 314 to include interface features predicted to achieve the target performance metric 312. For example, the intelligent interface feature system 102 utilizes a feature visualization machine learning model 310 to generate the updated graphical visualization 314, which depicts a combination of one or more interface features arranged and / or paired in a manner predicted to achieve the target performance metric 312.

[0086] As described above, in some of the described embodiments, the intelligent interface feature system 102 utilizes a feature visualization machine learning model to predict the arrangement and / or combination of interface features. The intelligent interface feature system 102 also generates updated graphical visualizations for a target time zone or target geographic region based on the predictions from the feature visualization machine learning model. Specifically, the intelligent interface feature system 102 utilizes a feature visualization machine learning model in the form of a feature visualization neural network. Figure 4 An updated graphical visualization is generated using a feature visualization machine learning model in the form of a feature visualization neural network, according to one or more embodiments.

[0087] like Figure 4 As shown, the intelligent interface feature system 102 utilizes a feature visualization machine learning model 406 to generate an updated graphical visualization 408 for a target time zone or target geographic region (e.g., target time zone 304). Specifically, the feature visualization machine learning model 406 predicts the updated arrangement and / or combination of interface features based on client device interactions 402. From the updated arrangement and / or combination, the intelligent interface feature system 102 generates the updated graphical visualization 408 based on client device interactions 402 from sample time zones or sample geographic regions (e.g., sample time zone 302). For example, the feature visualization machine learning model 406 predicts the arrangement and / or combination of interface features for the updated graphical visualization 408 by processing or analyzing client device interactions 402 to generate, encode, or extract hidden time zone vectors (or hidden geographic region vectors) from the client device interactions 402.

[0088] Specifically, the feature visualization machine learning model 406 utilizes one or more neurons within one or more network layers to generate a hidden vector representation of the client device interaction. In practice, the feature visualization machine learning model 406 generates a hidden vector representation that includes a hidden timezone vector, which represents the client device interaction 402 as a whole from the sample timezone (or the same geographical region). In some cases, the hidden timezone vector includes a numerical or mathematical representation of the client device interaction 402 that may not be interpretable by a human observer but can still be interpreted by the feature visualization machine learning model 406 and represent the client device interaction 402.

[0089] The intelligent interface feature system 102 also generates an updated graphical visualization 408 from the hidden time zone vector. More specifically, the intelligent interface feature system 102 utilizes one or more additional layers of the feature visualization machine learning model 406 to further process the hidden time zone vector and generate the updated graphical visualization 408. In effect, the intelligent interface feature system 102 generates the updated graphical visualization 408 based on the predicted arrangement and / or combination of interface features, reflecting the arrangement and / or combination of interface features to present more popular and / or more effective interface features in different (e.g., more prominent, more effective) locations.

[0090] In some cases, the intelligent interface feature system 102 utilizes placement and / or pairing predictions from the feature visualization machine learning model 406 to generate an updated graphical visualization 408 from the client device interaction 402 and a target performance metric 404 (e.g., target performance metric 312). For example, the feature visualization machine learning model 406 uses the target performance metric 404 to generate one or more weights or biases associated with the various layers or neurons within the feature visualization machine learning model 406. Based on the modified weights and biases according to the target performance metric 404, the intelligent interface feature system 102 adjusts how the feature visualization machine learning model 406 processes and delivers data, thereby modifying the output updated graphical visualization 408. In effect, the intelligent interface feature system 102 uses the target performance metric 404 to instruct the feature visualization machine learning model 406 (through modified weights and biases) to generate an updated graphical visualization 408 that is predicted to achieve the target performance metric 404 (as notified by the client device interaction 402).

[0091] although Figure 3-4The descriptions cover client device interactions from a single sample time zone or sample geographic region. However, in some embodiments, the intelligent interface feature system 102 generates updated graphical visualizations based on client device interactions from multiple sample time zones or sample geographic regions. For example, the intelligent interface feature system 102 identifies two or more leading time zones preceding the target time zone 304 and determines client device interactions from two or more leading time zones used as input to the feature visualization machine learning model 310 or 406. As another example, the intelligent interface feature system 102 determines client device interactions from two or more sample geographic regions for input to the feature visualization machine learning model 310 or 406, these two or more sample geographic regions having at least a threshold similarity score relative to the target geographic region. In some cases, the intelligent interface feature system 102 generates a combined set of client device interactions by generating a weighted combination of client device interactions from each of the two or more leading time zones (or two or more geographic regions).

[0092] For example, the intelligent interface feature system 102 gives greater weight to client device interactions from leading time zones that are closer to the target time zone in time (e.g., immediately preceding the target time zone). Additionally (or alternatively), the intelligent interface feature system 102 weights client device interactions from sample time zones or sample geographic regions based on similarity scores associated with the target time zone or target region (e.g., those with higher similarity scores are given greater weight). Furthermore, the feature visualization machine learning model 310 or 406 generates an updated graphical visualization based on client device interactions from combinations of two or more leading time zones.

[0093] In some embodiments, the intelligent interface feature system 102 performs the step of generating an updated graphical visualization that includes one or more interface features from a set of interface features for a target time zone. Figure 2 The above description of actions 206-212, including those concerning Figure 3-4 A more detailed description of the supporting actions 206-212 is provided, along with various embodiments and supporting actions and algorithms for performing steps to generate an updated graphical visualization that includes one or more interface features from a set of interface features for a target time zone.

[0094] For example, in some embodiments, performing the step of generating an updated graphical visualization that includes one or more interface features from a set of interface features for a target time zone includes detecting or determining client device interactions from client devices located within the sample time zone (e.g., as referenced). Figure 3(As described). In some embodiments, the step of generating an updated graphical visualization including one or more interface features from a set of interface features for a target time zone further includes generating an updated graphical visualization for the target time zone from client device interaction using a feature visualization machine learning model (e.g., as described in reference). Figure 3-4 (As described).

[0095] In one or more embodiments, the intelligent interface feature system 102 trains a feature visualization machine learning model (e.g., feature visualization machine learning model 310 or 406) to predict the arrangement and / or combination of interface features (e.g., for updated graphical visualizations). Specifically, the intelligent interface feature system 102 trains the feature visualization neural network based on sample client device interactions and corresponding ground truth graphical visualizations. Figure 5 A feature visualization machine learning model in the form of a trained feature visualization neural network is shown according to one or more embodiments.

[0096] like Figure 5 As shown, the intelligent interface feature system 102 performs an iterative training process to improve the accuracy of the feature visualization machine learning model 506. For example, the intelligent interface feature system 102 retrieves or accesses a set of sample client device interactions 504 from a database 502 (e.g., database 112). Furthermore, the intelligent interface feature system 102 inputs the sample client device interactions into the feature visualization machine learning model 506. Subsequently, the feature visualization machine learning model 506 generates a predicted graphical visualization 508 from the sample client device interactions 504. For example, the feature visualization machine learning model 506 generates sample feature vectors from the sample client device interactions 504 and further utilizes additional layers and neurons to process the sample feature vectors to generate an output in the form of a predicted graphical visualization 508.

[0097] like Figure 5 As further shown, the intelligent interface feature system 102 also performs a comparison 512 between the predicted graphical visualization 508 and the actual graphical visualization 510. Specifically, the intelligent interface feature system 102 accesses or retrieves the actual graphical visualization 510 from the database 502, wherein the actual graphical visualization 510 corresponds to or depicts the actual combination and arrangement of interface features that cause or trigger the sample client device interaction 504. Furthermore, the intelligent interface feature system 102 compares the actual graphical visualization 510 with the predicted graphical visualization 508.

[0098] To perform comparison 512, in some embodiments, the intelligent interface feature system 102 utilizes a loss function such as a mean squared error loss function or a cross-entropy loss function to determine a loss metric between the predicted graphical visualization 508 and the real graphical visualization 510. Based on comparison 512, the intelligent interface feature system 102 also performs backpropagation 514. Specifically, the intelligent interface feature system 102 backpropagates to modify the internal parameters of the feature visualization machine learning model 506, such as weights and biases. By modifying the weights and biases, the intelligent interface feature system 102 adjusts how the feature visualization machine learning model 506 processes and transmits information to reduce the loss metric determined via comparison 512.

[0099] In addition, the intelligent interface feature system 102 is designed for multiple iterations or epoch repetitions. Figure 5 The process shown continues until the feature visualization machine learning model 506 generates a predicted graphical visualization that satisfies a threshold loss metric (or threshold accuracy). For example, for each iteration, the intelligent interface feature system 102: i) accesses a set of sample client device interactions, ii) generates a predicted graphical visualization from the sample client device interactions using the feature visualization machine learning model 506, iii) compares the predicted graphical visualization with the true graphical visualization of the sample client device interactions corresponding to the respective iteration (via a loss function), and iv) backpropagates to reduce the loss metric by modifying the parameters of the feature visualization machine learning model 506. By utilizing the iterative training process, the intelligent interface feature system 102 generates accurate updated graphical visualizations for display on client devices within a target time zone or target geographic region.

[0100] As described above, in one or more embodiments, the intelligent interface feature system 102 determines a sequence of geographic regions for displaying interface features. Specifically, the intelligent interface feature system 102 determines the order in which geographic regions are provided to offer one or more interface features for display within an updated graphical visualization, to achieve, for example, a target performance metric. Figures 6A-6B The order in which geographic regions are determined for displaying one or more interface features is shown according to one or more embodiments.

[0101] like Figure 6AAs shown, the intelligent interface feature system 102 identifies or identifies multiple geographic regions. For example, the intelligent interface feature system 102 identifies geographic regions defined by geographic coordinates, geographic hashes, demographic information, digital content interests, and / or time zones. In some cases, the intelligent interface feature system 102 identifies geographic regions in the form of digital content distribution regions, where, for example, the digital content received by a region in the northeastern United States differs from that received by a region in the southeastern United States due to differences in geography, demographics, and content interests. As shown in the figure, the intelligent interface feature system 102 identifies three distinct geographic regions: Region 1, Region 2, and Region 3.

[0102] Based on identified geographic regions, the intelligent interface feature system 102 generates a candidate order of multiple geographic regions. Specifically, the intelligent interface feature system 102 generates the candidate order by selecting an initial geographic region and generating a similarity chain originating from that initial geographic region. For example, the intelligent interface feature system 102 determines the similarity score between the initial geographic region and every other geographic region.

[0103] In practice, as described above, the intelligent interface feature system 102 determines the similarity score based on historical network user behavior indicating that the client device interacts with specific interface features and / or other digital content. The intelligent interface feature system 102 also determines the similarity score based on other or additional factors such as user demographics within the region, time zone similarity between regions, and / or language similarity between regions. Furthermore, the intelligent interface feature system 102 selects the geographic region with the highest similarity score relative to the initial geographic region as the subsequent geographic region following the initial geographic region in this sequence.

[0104] The intelligent interface feature system 102 also links additional subsequent geographic regions to an ordered similarity chain. For example, the intelligent interface feature system 102 determines the similarity score of each geographic region relative to subsequent geographic regions following the initial geographic region. Based on the similarity score, the intelligent interface feature system 102 selects an additional subsequent geographic region immediately following the initial geographic region. In effect, the intelligent interface feature system 102 selects the additional subsequent geographic region as the geographic region with the next highest (or highest available) similarity score relative to the subsequent geographic region (e.g., the highest similarity score after the initial geographic region's similarity score). The intelligent interface feature system 102 continues linking geographic regions in this order by determining similarity scores (relative to each subsequently added geographic region) and selecting those regions with the highest available similarity scores to append to the ordered similarity chain.

[0105] like Figure 6AAs shown, the intelligent interface feature system 102 determines a similarity score of 0.85 between region 1 and region 2 (e.g., in the range of 0 to 1). To determine the similarity score, the intelligent interface feature system 102 compares historical network user behavior in region 1 with historical network user behavior in region 2, including client device interaction types, the quantity (or frequency) of each type, and / or a comparison of the timing of client device interactions relative to various interface features or other digital content. Based on the comparison of historical network user behavior (attached to other information such as demographics as described above), the intelligent interface feature system 102 determines a similarity score of 0.85 between region 1 and region 2.

[0106] Furthermore, the intelligent interface feature system 102 determines a similarity score between region 1 and region 3, where the similarity score is below 0.85. Therefore, the intelligent interface feature system 102 selects region 2 as the successor region to the initial region 1 from the depicted candidates. Additionally, the intelligent interface feature system 102 determines a similarity score of 0.70 between region 2 and region 3 based on a comparison of client device interaction information for each region. Since the similarity score of 0.70 is lower than the highest available similarity score for the remaining regions relative to region 2, the intelligent interface feature system 102 appends region 3 to the candidate order, after region 2.

[0107] In one or more embodiments, the intelligent interface feature system 102 sequentially distributes or provides interface features for display on client devices located in each successive geographic region of the candidate order. Specifically, the intelligent interface feature system 102 generates an initial graphical visualization including the interface features for display on client devices in region 1 (e.g., the initial region). Subsequently, the intelligent interface feature system 102 utilizes a feature visualization machine learning model to generate an updated graphical visualization from client device interactions in region 1 (related to the interface features of the graphical visualization), as described above, to include a modified presentation of the interface features for display on client devices located in region 2. Furthermore, the intelligent interface feature system 102 generates another updated graphical visualization from client device interactions in region 2 (and / or region 1) for display on client devices in region 3.

[0108] like Figure 6AAs shown, the intelligent interface feature system 102 tracks or receives data indicating user device network behavior related to interface features. For region 1, the intelligent interface feature system 102 identifies 8 clicks and 10 views (related to interface features) from user A and 15 clicks and 20 views from user B. Furthermore, for region 2, the intelligent interface feature system 102 identifies 7 clicks and 10 views for user C and 14 clicks and 18 views for user D. Additionally, for region 3, the intelligent interface feature system 102 identifies 5 clicks and 9 views for user E and 13 clicks and 30 views for user F.

[0109] like Figure 6B As shown, the intelligent interface feature system 102 generates another candidate order of geographic regions. More specifically, the intelligent interface feature system 102 selects or identifies different initial regions for the new candidate order. For example, the intelligent interface feature system 102 selects region 2 as the initial geographic region. Furthermore, the intelligent interface feature system 102 selects region 1 as the subsequent geographic region following region 2 based on a comparison of the similarity scores of the geographic region relative to region 2. In fact, the similarity score of region 1 relative to region 2 is 0.85. In addition, the intelligent interface feature system 102 also compares the similarity scores of geographic regions to select the geographic region with the highest available similarity score to follow region 1. As shown, the intelligent interface feature system 102 selects region 3 to follow region 1 based on a similarity score of 0.6.

[0110] Similar to the above about Figure 6A The intelligent interface feature system 102 is described according to... Figure 6B The intelligent interface feature system 102 assigns interface features based on a candidate order. Specifically, it generates a graphical visualization including interface features for display on client devices in region 2 (e.g., the initial region). The intelligent interface feature system 102 also determines client device interactions associated with the interface features of region 2 (e.g., 7 clicks and 10 views from user C, and 14 clicks and 18 views from user D), and uses a feature visualization machine learning model to generate an updated graphical visualization from client device interactions for region 1 (e.g., a subsequent geographic region). Furthermore, the intelligent interface feature system 102 determines client device interactions from region 1 (e.g., 10 clicks and 10 views from user A, and 15 clicks and 22 views from user B), and uses a feature visualization machine learning model to generate another updated graphical visualization including interface features for region 3 from the client device interactions of region 1.

[0111] like Figures 6A-6BAs shown, the intelligent interface feature system 102 determines performance metrics for candidate order. For example, the intelligent interface feature system 102 determines multiple client device interactions, such as clicks, views, scrolling, and / or conversions, reflecting how interface features are executed when distributed sequentially according to each candidate order. As shown in the figure, the intelligent interface feature system 102 determines performance metrics for... Figure 6A The candidate order was determined by 62 total clicks and 97 total views (associated with interface features). Also shown in the figure, the intelligent interface feature system 102 determined the target... Figure 6B The candidate order had a total of 66 clicks and 102 views (associated with interface features).

[0112] Based on performance metrics of different candidate orders, the intelligent interface feature system 102 learns how to select initial regions. Specifically, based on generating and testing different candidate orders over time, the intelligent interface feature system 102 learns which geographical regions will produce orders or similarity chains that will achieve specific performance metrics. In practice, in some cases, the intelligent interface feature system 102 selects from multiple candidate orders (e.g., Figures 6A-6B Choose an order from the candidate order shown to use for displaying or distributing interface features based on performance metrics.

[0113] For example, the intelligent interface feature system 102 selects an order based on comparing a performance metric with a target performance metric. As described above, the intelligent interface feature system 102 receives an indication of the target performance metric from the administrator device 110. In some cases, the intelligent interface feature system 102 receives the target performance metric to increase client device interaction, such as clicks, views, scrolling, and / or conversions. As described above, the intelligent interface feature system 102 determines the total click count for the first candidate order to be 62 and the total view count to be 97. Figure 6B As shown, the intelligent interface feature system 102 determines the total click count for the second candidate order to be 66 and the total view count to be 102. Therefore, based on the target performance metric of increasing client device interaction (e.g., clicks and / or views), the intelligent interface feature system 102 selects... Figure 6B The candidate order for distributing interface features is shown in the diagram. In practice, the intelligent interface feature system 102 selects... Figure 6B The candidate order is adjusted to increase clicks and views.

[0114] As described above, in some of the described embodiments, the intelligent interface feature system 102 generates an updated graphical visualization to modify the arrangement and / or combination of interface features. Specifically, the intelligent interface feature system 102 generates an updated graphical visualization to select one or more interface features and place them in different positions compared to the initial graphical visualization. Figures 7A-7BA comparison between an initial graphical visualization and an updated graphical visualization according to one or more embodiments is shown.

[0115] like Figure 7A As shown, the intelligent interface feature system 102 generates and provides a graphical visualization 702a for display on the graphical user interface of the client device 108a. As illustrated, the graphical visualization 702a includes interface features such as a title 704, selectable menu options 706, and selectable digital images 708a. In some cases, the graphical visualization 702a includes... Figure 7A These are additional interface features not shown in the diagram, but rather nested within other graphical user interfaces. For example, based on the selection of selectable menu options 706, the intelligent interface feature system 102 provides different graphical user interfaces to be displayed as part of a graphical visualization 702a, which includes products targeted at "men".

[0116] In fact, in some embodiments, the graphical visualization 702a is displayed on multiple graphical user interfaces, such as... Figure 7A The nested version of client device 108a is shown in this example. In these or other embodiments, to access a specific interface feature, the user of client device 108a navigates through multiple interface layers across multiple graphical user interfaces. In response to determining that a specific interface feature is desirable (e.g., based on a threshold number of client device interactions), intelligent interface feature system 102 generates an updated graphical visualization to present the interface feature in a more prominent position, eliminating the need to navigate across multiple graphical user interfaces to access it.

[0117] Along these lines, Figure 7B An updated graphical visualization 702b is shown displayed on the graphical user interface of client device 108h. This updated graphical visualization 702b includes one or more interface features with a modified arrangement and / or combination from graphical visualization 702a. In fact, the updated graphical visualization 702b includes combinations of interface features different from those depicted in graphical visualization 702a. Specifically, the updated graphical visualization 702b includes a digital image 708b and selectable menu options 710 (etc.). As described above, based on a threshold number of detected client device interactions with selectable menu options 710 for “shoes,” the intelligent interface feature system 102 generates the updated graphical visualization 702b to pre-deploy the selectable menu options 710 (e.g., eliminating the need for navigation through other graphical user interfaces to access it). Furthermore, the intelligent interface feature system 102 provides the digital image 708b in a more prominent position than the digital image 708a in the initial graphical visualization 702a.

[0118] In some embodiments, the intelligent interface feature system 102 provides a graphical visualization 702a for display on a client device 108a located in a sample time zone or an initial geographic region, and provides an updated graphical visualization 702b for display on a client device 108h located in a target time zone or a subsequent geographic region. In practice, as described herein, the intelligent interface feature system 102 determines client device interactions associated with interface features of the graphical visualization 702a, and then uses a feature visualization machine learning model (e.g., feature visualization machine learning model 310 or 406) to generate the updated graphical visualization 702b from the client device interactions.

[0119] Apart from Figures 7A-7B In addition to the modifications shown, the intelligent interface feature system 102 can also modify interface features in other ways. For example, in some embodiments, the intelligent interface feature system 102 generates an updated graphical visualization 702b that depicts the digital image 708a in colors different from those of the digital image 708a, based on detected client device interactions with various color versions of the digital image 708a (e.g., to select the most popular color). In some cases, the intelligent interface feature system 102 modifies other interface features, such as product descriptions. For example, based on client device interactions within the graphical visualization 702a, the intelligent interface feature system 102 determines that a shorter (or longer) product description is more effective in evoking client device interactions. Therefore, the intelligent interface feature system 102 generates a shorter (or longer) product description for the product within the updated graphical visualization 702b.

[0120] In some cases, the intelligent interface feature system 102 modifies product descriptions, titles, or other interface features based on an adapted language. For example, the intelligent interface feature system 102 determines more effective words for triggering client device interactions regarding product descriptions, titles, banners, or other interface elements based on interactions with the client device of the graphical visualization 702a. Therefore, the intelligent interface feature system 102 generates modified interface elements for inclusion in the updated graphical visualization 702b by changing the language of the interface elements to use more effective words.

[0121] As described above, in some embodiments, the intelligent interface feature system 102 receives digital content distribution settings from an administrator device (e.g., administrator device 110). Specifically, the intelligent interface feature system 102 provides a graphical user interface, such as a digital content distribution settings interface, whereby the administrator sets parameters for distributing updated graphical visualizations to various target time zones or target geographic regions. Figure 8 An example digital content distribution settings interface according to one or more embodiments is shown.

[0122] like Figure 8As shown, administrator device 110 displays or presents a digital content distribution settings interface 802. Within the digital content distribution settings interface 802, intelligent interface feature system 102 provides various interactive or selectable elements for modifying how graphical visualizations are distributed to client devices. For example, intelligent interface feature system 102 provides selectable options (e.g., radio buttons) for distributing updated graphical visualizations including one or more interface features by time zone sequence or by regional similarity. Based on the selection of distribution according to time zone sequence, intelligent interface feature system 102 performs the methods described herein to determine client device interactions from sample time zones and generate updated graphical visualizations for target time zones based on client device interactions.

[0123] Alternatively, based on the selection of an option to distribute graphical visualizations via regional similarity, the intelligent interface feature system 102 performs the methods described herein to identify initial geographic regions, generate an order of geographic regions based on similarity scores between regions, and generate one or more updated graphical visualizations for sequential distribution to geographic regions in that order based on client device interactions from regions (e.g., actions from one or more previous regions affect the generation of the next updated graphical visualization for subsequent regions).

[0124] like Figure 8 As further shown, the digital content distribution settings interface 802 includes options for modifying one or more target performance metrics. For example, the digital content distribution settings interface 802 includes sliders for adjusting the impact metric or weight for each of multiple client device interaction types. As illustrated, the digital content distribution settings interface 802 includes options for adjusting the emphasis for clicks, options for adjusting the emphasis for views, and options for adjusting the emphasis for conversions (e.g., where sliding left decreases emphasis and sliding right increases emphasis).

[0125] In practice, in some embodiments, the intelligent interface feature system 102 generates a combined target performance metric that includes a weighted combination of different types of client device interactions based on settings configured via administrator device 110. For example... Figure 8 As shown, the digital content distribution settings interface 802 indicates low emphasis for clicks, slightly higher emphasis for views, and strong emphasis for conversions. Based on the target performance metric, the intelligent interface feature system 102 generates a combined target performance metric by generating a weighted combination according to the emphasis level (e.g., low weight for clicks, slightly higher weight for views, and heavy weight for conversions). Furthermore, the intelligent interface feature system 102 generates an updated graphical visualization as described herein to present interface features with a combination and / or arrangement predicted to achieve the target performance metric (e.g., the combined target performance metric).

[0126] Now for reference Figure 9 This will provide additional details regarding the components and capabilities of the intelligent interface feature system 102. Specifically, Figure 9 A schematic diagram of an intelligent interface feature system 102 on an example computing device 900 (e.g., one or more of client devices 108a-108n, administrator device 110, and / or server 104) is shown. Figure 9 As shown, the intelligent interface feature system 102 includes a client device interaction manager 902, a sequence manager 904, a graphical visualization manager 906, a user interface manager 908, and a storage manager 910.

[0127] As described above, the intelligent interface feature system 102 includes a client device interaction manager 902. Specifically, the client device interaction manager 902 manages, maintains, identifies, tracks, monitors, detects, or identifies client device interactions. For example, the client device interaction manager 902 identifies client device interactions from client devices located in one or more time zones or geographic areas, such as a sample time zone, a target time zone, a sample geographic area, or a target geographic area. The client device interaction manager 902 also identifies client device interactions associated with specific interface features displayed within a graphical visualization, attributing (partial) client device interactions to interface features to determine which are more popular or effective (e.g., which elicit more interactions).

[0128] Furthermore, the intelligent interface feature system 102 includes a sequence manager 904. Specifically, the sequence manager 904 manages, maintains, determines, arranges, organizes, generates, or identifies the order or sequence of geographic regions. For example, as described herein, the sequence manager 904 determines the order of geographic regions by generating a similarity chain based on similarity scores. In practice, the sequence manager 904 determines or selects the order in which geographic regions are displayed within a graphical visualization to improve or enhance specific target metrics. For example, the sequence manager 904 determines similarity scores between geographic regions (or time zones). Furthermore, the sequence manager 904 links geographic regions based on similarity scores and sequentially presents interface features to client devices located in each successive geographic region.

[0129] As shown in the figure, the intelligent interface feature system 102 also includes a graphical visualization manager 906. Specifically, the graphical visualization manager 906 manages, maintains, deploys, generates, or creates graphical visualizations. For example, as described herein, the graphical visualization manager 906 utilizes a feature visualization machine learning model based on client device interaction to generate updated graphical visualizations. In practice, the graphical visualization manager 906 generates updated graphical visualizations for a target time zone or target geographic region based on client device interactions from sample time zones or sample geographic regions.

[0130] Furthermore, the intelligent interface feature system 102 includes a user interface manager 908. Specifically, the user interface manager 908 renders, displays, or provides graphical visualizations for display. For example, the user interface manager 908 provides an initial graphical visualization (e.g., in the form of one or more graphical user interfaces) for display on client devices in a sample time zone or sample geographic region. Additionally, the user interface manager 908 provides an updated graphical visualization for display on client devices in a target time zone or target geographic region. In some cases, the user interface manager 908 provides an updated graphical visualization that describes one or more interface features with a modified layout and / or arrangement (e.g., to reduce client device interaction and / or achieve target performance metrics).

[0131] The intelligent interface feature system 102 also includes a storage manager 910. The storage manager 910 operates in conjunction with or includes one or more storage devices such as a database 912 (e.g., database 112), which store various types of data, such as feature visualization machine learning models, client device interaction data, interface feature data, historical network user behavior data, time zone data defining different time zones, and geographic data defining different geographic regions.

[0132] In one or more embodiments, each component of the intelligent interface feature system 102 communicates with each other using any suitable communication technology. Furthermore, components of the intelligent interface feature system 102 communicate with one or more other devices, including the aforementioned client devices. It should be recognized that, although in Figure 9 The components of the intelligent interface feature system 102 are shown to be separate, but any sub-component can be combined into fewer components, such as a single component, or divided into more components that can serve a particular implementation. Furthermore, although the intelligent interface feature system 102 is described in conjunction with... Figure 9 The components, however, at least some of the components used to perform operations in conjunction with the intelligent interface feature system 102 described herein can be implemented on other devices within the environment.

[0133] Components of the intelligent interface feature system 102 may include software, hardware, or both. For example, components of the intelligent interface feature system 102 may include one or more instructions stored on a computer-readable storage medium and executable by a processor of one or more computing devices (e.g., computing device 900). When executed by one or more processors, the computer-executable instructions of the intelligent interface feature system 102 may cause the computing device 900 to perform the methods described herein. Alternatively, components of the intelligent interface feature system 102 may include hardware, such as a dedicated processing device that performs a particular function or group of functions. Additionally or alternatively, components of the intelligent interface feature system 102 may include a combination of computer-executable instructions and hardware.

[0134] Furthermore, components of the intelligent interface feature system 102 that perform the functions described herein can be implemented, for example, as part of a standalone application, a module of an application, an application plugin including a content management application, one or more library functions that can be called by other applications, and / or a cloud computing model. Therefore, components of the intelligent interface feature system 102 can be implemented as part of a standalone application on a personal computing device or mobile device. Alternatively or additionally, components of the intelligent interface feature system 102 can be implemented in any application that allows the creation and delivery of marketing content to users, including but not limited to… MARKETING Applications in, such as Adobe and ADOBE "ADOBE", "ADOBE MARKETING CLOUD", "ADOBE CAMPAIGN" and "ADOBE ANALYTICS" are registered trademarks or trademarks of Adobe Systems Incorporated in the U.S. and / or other countries.

[0135] Figures 1-9 The corresponding text and examples provide several different systems, methods, and non-transient computer-readable media for generating updated graphical visualizations for a target time zone or target geographic region using a feature visualization machine learning model based on client device interactions from a sample time zone or sample geographic region. In addition to the foregoing, embodiments may also be described in terms of flowcharts including actions for achieving specific results. For example, Figure 10-11 A flowchart illustrating an example sequence or series of actions according to one or more embodiments is shown.

[0136] Although Figures 10-11 Actions according to a particular embodiment are shown, but alternative embodiments may omit, add, reorder, and / or modify them. Figures 10-11 Any action shown. Figure 10-11The action can be performed as part of a method. Alternatively, a non-transient computer-readable medium can include instructions that, when executed by one or more processors, cause a computing device to perform... Figure 10-11 The action. In other embodiments, a system can perform... Figure 10-11 The actions described herein can be repeated or performed in parallel with each other, or in parallel with different instances of the same or other similar actions.

[0137] Figure 10 An example series of actions 1000 is illustrated for generating updated graphical visualizations for a target time zone using a feature visualization machine learning model based on client device interactions from a sample time zone. Specifically, the series of actions 1000 includes an action 1002 for generating a graphical visualization for a sample time zone. For example, action 1002 relates to generating a graphical visualization of a set of interface features for display within one or more graphical user interfaces of multiple client devices located in the sample time zone. In some embodiments, action 1002 relates to generating a graphical visualization in response to client device interactions navigating across multiple graphical user interfaces to include a set of interface features across the multiple graphical user interfaces. In some cases, action 1002 relates to generating one or more arrangements or combinations of interface feature sets for a sample time zone prior to the target time zone.

[0138] As shown in the figure, the action series 1000 also includes action 1004 for determining client device interactions. Specifically, action 1004 involves determining client device interactions related to interface features from a set of interface features within one or more graphical user interfaces, for a sample time zone.

[0139] Furthermore, Action Series 1000 includes Action 1006, which generates an updated graphical visualization from client device interactions. Specifically, Action 1006 involves generating an updated graphical visualization of one or more interface features from a set of interface features, based on client device interactions associated with interface features, targeting a target time zone, and utilizing a feature visualization machine learning model. For example, Action 1006 involves using a feature visualization machine learning model that includes a collaborative filtering recommendation model to determine the arrangement or combination of one or more interface features from client device interactions associated with a sample time zone, and generating an updated graphical visualization for the target time zone to reflect the arrangement or combination of those interface features.

[0140] In some embodiments, action 1006 involves utilizing a feature visualization machine learning model, including a feature visualization neural network, to extract a hidden timezone vector representing the sample timezone from client device interactions relative to one or more interface features associated with a sample timezone, and to generate an updated graphical visualization for a target timezone from the hidden timezone vector. In some cases, action 1006 involves generating an updated graphical visualization to include one or more interface features within a single graphical user interface, without requiring client device interactions navigating across multiple graphical user interfaces.

[0141] like Figure 10 As further shown, action series 1000 includes action 1008, which provides an updated graphical visualization for display. Specifically, action 1008 relates to providing an updated graphical visualization of one or more interface features for display within a graphical user interface of a client device located in a target time zone. For example, action 1008 relates to providing an updated graphical visualization of one or more interface features for display within a graphical user interface of a client device in a target time zone. In some cases, action 1008 relates to providing one or more arrangements or combinations of interface features that differ from the set represented in the graphical visualization of a sample time zone. In these or other cases, action 1008 relates to providing one or more interface features within a single graphical user interface without requiring client device interaction navigating across multiple graphical user interfaces. For example, action 1008 relates to providing a modified product description.

[0142] In some embodiments, action series 1000 includes actions that determine a similarity score between a sample time zone and a target time zone based on historical network user behavior within the sample time zone and the target time zone. In these or other embodiments, action series 1000 includes actions that utilize a feature visualization machine learning model to determine the arrangement or combination of one or more interface features for the target time zone based on the similarity score between the sample time zone and the target time zone.

[0143] In one or more implementations, action series 1000 includes an action that compares the similarity scores of multiple candidate sample time zones with respect to a target time zone. Furthermore, action series 1000 may include an action that selects a sample time zone from the multiple candidate sample time zones based on the comparison. In some cases, action series 1000 includes an action that determines a selection count corresponding to an interface feature from client device interactions associated with the interface feature, and selects one or more interface features for an updated graphical visualization based on the selection count.

[0144] In some cases, action series 1000 includes actions to generate additional graphical visualizations of an additional set of interface features for display via client devices located in additional sample time zones. Furthermore, action series 1000 includes actions to determine client device interactions related to the additional set of interface features for the additional sample time zones. Additionally, action series 1000 includes actions to generate updated graphical visualizations for target time zones based on client device interactions related to the additional set of interface features.

[0145] Figure 11 An example series of actions 1100 is shown for generating updated graphical visualizations sequentially for geographic regions using a feature visualization machine learning model based on client device interaction. Specifically, the series of actions 1100 includes an action 1102 that generates similarity scores for geographic regions. For example, action 1102 involves generating similarity scores for multiple geographic regions based on historical web user behavior to provide interface features for display on client devices.

[0146] As shown in the figure, action series 1100 also includes action 1104, which determines the order of geographic regions. Specifically, action 1104 involves determining the order of multiple geographic regions, including an initial geographic region and subsequent geographic regions, based on a similarity score. For example, action 1104 involves monitoring network user behavior within each of the multiple geographic regions and, based on the network user behavior and for a given geographic region among the multiple geographic regions, determining a similarity score relative to each other geographic region among the multiple geographic regions.

[0147] In some cases, action 1104 involves selecting an initial geographic region from multiple geographic regions, determining a subsequent geographic region that is the geographic region with the highest similarity score relative to the initial geographic region as immediately following the initial geographic region in the order, and determining an additional subsequent geographic region that is the next geographic region with the second highest similarity score relative to the subsequent geographic region as immediately following the subsequent geographic region in the order.

[0148] Furthermore, the action series 1100 includes an action 1106 to determine client device interactions from an initial geographic region. Specifically, action 1106 involves determining client device interactions associated with interface features displayed in a graphical visualization on a client device located within the initial geographic region for the initial geographic region. In some embodiments, the action series 1100 includes an action to select an initial geographic region. Selecting an initial geographic region may involve determining multiple candidate orders of multiple geographic regions, generating updated graphical visualizations of interface features for successive geographic regions within the multiple candidate orders, comparing performance metrics of the interface features associated with each candidate order within the multiple candidate orders, and selecting an initial geographic region based on the compared performance metrics.

[0149] Furthermore, action series 1100 includes action 1108, which generates an updated graphical visualization for subsequent geographic regions. Specifically, action 1108 involves interacting with client devices associated with interface features and utilizing a feature visualization machine learning model to generate an updated graphical visualization of interface features for subsequent geographic regions. For example, action 1108 involves using a feature visualization machine learning model to generate an updated graphical visualization of interface features to improve one or more performance metrics, including clicks or conversions.

[0150] like Figure 11 As further shown, action series 1100 includes action 1110, which provides an updated graphical visualization for display. Specifically, action 1110 relates to providing an updated graphical visualization of interface features for display within the graphical user interface of a client device in a subsequent geographic area.

[0151] In some embodiments, action series 1100 includes actions to determine client device interactions from subsequent geographic regions associated with interface features. Furthermore, action series 1100 includes actions to generate additional updated graphical visualizations of interface features interacting with the client device from subsequent geographic regions, for additional subsequent geographic regions following the sequence of subsequent geographic regions within a plurality of geographic regions. Additionally, action series 1100 includes actions to provide additional updated graphical visualizations of interface features for display on client devices within the additional subsequent geographic regions.

[0152] In some cases, Action Series 1100 includes the following actions: by monitoring web user behavior from multiple geographic regions associated with interface features, determining one or more performance metrics, including clicks or conversions associated with interface features, selecting a new initial geographic region based on the performance metrics of the interface features, and determining an updated order of multiple geographic regions, including the new initial geographic region and a new subsequent geographic region, based on similarity scores.

[0153] Embodiments of this disclosure may include or utilize a dedicated or general-purpose computer including computer hardware, such as one or more processors and system memory, as discussed in more detail below. Embodiments within the scope of this disclosure also include physical and other computer-readable media for carrying or storing computer-executable instructions and / or data structures. Specifically, one or more processes described herein may be implemented at least in part as instructions contained in a non-transitory computer-readable medium and executable by one or more computing devices (e.g., any media content access device described herein). Generally, a processor (e.g., a microprocessor) receives instructions from a non-transitory computer-readable medium (e.g., memory, etc.) and executes those instructions to perform one or more processes, including one or more processes described herein.

[0154] Computer-readable media can be any available medium accessible by a general-purpose or special-purpose computer system. A computer-readable medium storing computer-executable instructions is a non-transitory computer-readable storage medium (device). A computer-readable medium carrying computer-executable instructions is a transmission medium. Therefore, by way of example and not limitation, embodiments of this disclosure may include at least two distinct types of computer-readable media: non-transitory computer-readable storage media (devices) and transmission media.

[0155] Non-transient computer-readable storage media (devices) include RAM, ROM, EEPROM, CD-ROM, solid-state drives (“SSDs”) (e.g., RAM-based), flash memory, phase-change memory (“PCM”), other types of memory, other optical disc storage, disk storage or other magnetic storage devices, or any other medium that can be used to store desired program code in the form of computer-executable instructions or data structures and that can be accessed by a general-purpose or special-purpose computer.

[0156] A “network” is defined as one or more data links that enable the transmission of electronic data between computer systems and / or modules and / or other electronic devices. When information is transmitted or provided to a computer via a network or other communication connection (hardwired, wireless, or a combination of hardwired and wireless), the computer correctly regards that connection as a transmission medium. Transmission media may include networks and / or data links that can be used to carry desired program code in the form of computer-executable instructions or data structures and are accessible by general-purpose or special-purpose computers. Combinations of the foregoing should also be included within the scope of computer-readable media.

[0157] Furthermore, upon arrival at various computer system components, program code in the form of computer-executable instructions or data structures can be automatically transferred from the transmission medium to a non-transitory computer-readable storage medium (device) (and vice versa). For example, computer-executable instructions or data structures received via a network or data link can be cached in RAM within a network interface module (e.g., a "NIC") and then ultimately transferred to the computer system RAM and / or a less volatile computer storage medium (device) at the computer system. Therefore, it should be understood that non-transitory computer-readable storage media (devices) can be included in computer system components that also (or even primarily) utilize the transmission medium.

[0158] Computer-executable instructions include, for example, instructions and data that, when executed at a processor, cause a general-purpose computer, a special-purpose computer, or a special-purpose processing device to perform a particular function or group of functions. In some embodiments, the computer-executable instructions are executed on a general-purpose computer to transform the general-purpose computer into a special-purpose computer that implements the elements of this disclosure. The computer-executable instructions may be, for example, binary, intermediate format instructions such as assembly language, or even source code. Although the subject matter has been described in language specific to structural features and / or methodological actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the features or actions described above. Rather, the described features and actions are disclosed as exemplary forms for implementing the claims.

[0159] Those skilled in the art will understand that this invention can be implemented in network computing environments with many types of computer system configurations, including personal computers, desktop computers, laptop computers, message processors, handheld devices, multiprocessor systems, microprocessor-based or programmable consumer electronics, network PCs, minicomputers, mainframe computers, mobile phones, PDAs, tablet computers, pagers, routers, switches, etc. This disclosure can also be implemented in a distributed system environment, where local and remote computer systems perform tasks via network links (via hardwired data links, wireless data links, or a combination of hardwired and wireless data links). In a distributed system environment, program modules can reside in local and remote memory storage devices.

[0160] Embodiments of this disclosure can also be implemented in a cloud computing environment. In this specification, "cloud computing" is defined as a model for enabling on-demand network access to a shared pool of configurable computing resources. For example, cloud computing can be used in the market to provide ubiquitous and convenient on-demand access to a shared pool of configurable computing resources. The shared pool of configurable computing resources can be rapidly provisioned via virtualization, released with minimal management effort or service provider interaction, and then scaled accordingly.

[0161] Cloud computing models can be composed of various features, such as on-demand self-service, widespread network access, resource pooling, rapid elasticity, and measurement services. Cloud computing models can also expose various service models, such as Software as a Service (“SaaS”), Platform as a Service (“PaaS”), and Infrastructure as a Service (“IaaS”). Cloud computing models can also be deployed using different deployment models, such as private clouds, community clouds, public clouds, and hybrid clouds. In this specification and claims, a “cloud computing environment” refers to an environment that uses cloud computing.

[0162] Figure 12 An example computing device 1200 (e.g., computing device 900, client devices 108a-108n, and / or server 104) that can be configured to perform one or more of the processes described above is illustrated in block diagram form. It will be understood that the intelligent interface feature system 102 may include an implementation of the computing device 1200. For example... Figure 12 As shown, the computing device 1200 may include a processor 1202, a memory 1204, a storage device 1206, an I / O interface 1208, and a communication interface 1210. Furthermore, the computing device 1200 may include input devices such as a touchscreen, mouse, and keyboard. In some embodiments, the computing device 1200 may include a... Figure 12 The components shown are fewer or more components. A more detailed description will follow. Figure 12 The components of the computing device 1200 shown.

[0163] In a particular embodiment, processor(s) 1202 includes hardware for executing instructions, such as those that constitute a computer program. By way of example and not limitation, in order to execute instructions, processor(s) 1202 may retrieve (or fetch) instructions from internal registers, internal caches, memory 1204, or storage device 1206, and decode and execute them.

[0164] Computing device 1200 includes memory 1204 coupled to processor(s) 1202. Memory 1204 can be used to store data, metadata, and programs executed by processor(s). Memory 1204 may include one or more of volatile and non-volatile memory, such as random access memory (“RAM”), read-only memory (“ROM”), solid-state drive (“SSD”), flash memory, phase-change memory (“PCM”), or other types of data storage. Memory 1204 may be internal or distributed memory.

[0165] Computing device 1200 includes storage device 1206, which includes memory for storing data or instructions. By way of example and not limitation, storage device 1206 may include the aforementioned non-transient storage media. Storage device 1206 may include hard disk drive (HDD), flash memory, universal serial bus (USB) drive, or combinations of these or other storage devices.

[0166] The computing device 1200 also includes one or more input or output (“I / O”) devices / interfaces 1208 provided to allow a user to provide input (such as user strokes) to the computing device 1200, receive output from the computing device 1200, and otherwise transmit data to the computing device 1200. These I / O devices / interfaces 1208 may include a mouse, keypad or keyboard, touchscreen, camera, optical scanner, network interface, modem, other known I / O devices, or combinations of these I / O devices / interfaces 1208. The touchscreen can be activated using a writing device or a finger.

[0167] I / O device / interface 1208 may include one or more devices for presenting output to a user, including but not limited to a graphics engine, a display (e.g., a screen), one or more output drivers (e.g., display drivers), one or more audio speakers, and one or more audio drivers. In some embodiments, device / interface 1208 is configured to provide graphical data to the display for presentation to the user. The graphical data may represent one or more graphical user interfaces and / or any other graphical content that may serve a particular implementation.

[0168] The computing device 1200 may also include a communication interface 1210. The communication interface 1210 may include hardware, software, or both. The communication interface 1210 provides one or more interfaces for communication (e.g., packet-based communication) between the computing device and one or more other computing devices 1200 or one or more networks. By way of example and not limitation, the communication interface 1210 may include a network interface controller (NIC) or network adapter for communicating with Ethernet or other wired networks, or a wireless NIC (WNIC) or wireless adapter for communicating with wireless networks such as Wi-Fi. The computing device 1200 may also include a bus 1212. The bus 1212 may include hardware, software, or both for coupling components of the computing device 1200 to each other.

[0169] In the foregoing description, the invention has been described with reference to specific exemplary embodiments thereof. Various embodiments and aspects of the invention have been described with reference to the details discussed herein, and the accompanying drawings illustrate various embodiments. The foregoing description and drawings are illustrative of the invention and should not be construed as limiting the invention. Numerous specific details have been described to provide a thorough understanding of various embodiments of the invention.

[0170] This invention may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are to be considered illustrative rather than restrictive in all respects. For example, the methods described herein may be performed with fewer or more steps / actions, or the steps / actions may be performed in a different order. Furthermore, the steps / actions described herein may be repeated or performed in parallel with each other, or in parallel with different instances of the same or similar steps / actions. Therefore, the scope of the invention is indicated by the appended claims, and not by the foregoing description. All modifications within the meaning and equivalent scope of the claims should fall within its scope.

Claims

1. A system for displaying interface features, comprising: One or more memory devices, This includes feature visualization machine learning models and data corresponding to the set of interface features; One or more computing devices are configured to enable the system to: Compare the similarity scores of multiple candidate sample time zones with respect to a target time zone, wherein the similarity scores are determined based on historical network user behavior and user characteristics within the respective candidate sample time zone and the target time zone, the historical network user behavior being indicated by at least one of the following: interaction type with interface features, number or frequency of interactions, and interaction timing, and the user characteristics including at least one of the following: age, gender, education, and language; Based on the comparison, a sample time zone is selected from the plurality of candidate sample time zones; A graphical visualization of the interface feature set of the first arrangement is generated for display within one or more graphical user interfaces of multiple client devices located in the sample time zone; For the sample time zone, determine the client device interactions within the one or more graphical user interfaces that are related to the interface features in the first set of interface features; Based on client device interactions associated with the interface features of the first arrangement and according to the similarity score between the sample time zone and the target time zone, an updated graphical visualization of the interface feature set for the target time zone is generated by repositioning, removing, and / or rearranging one or more interface features in the interface feature set using the feature visualization machine learning model: Weights and biases associated with neurons in the feature visualization machine learning model are generated based on a target performance metric, wherein the target performance metric includes one or more of the following: click counts or conversion counts for interface features; and Using the weights and the biases, an updated graphical visualization depicting the visual modifications to the set of interface features is generated; as well as The updated graphical visualization of one or more interface features of the second arrangement is provided for display within the graphical user interface of a client device located in the target time zone.

2. The system according to claim 1, wherein, The one or more computing devices are further configured to cause the system to generate the updated graphical visualization of the one or more interface features by utilizing the feature visualization machine learning model, including a collaborative filtering recommendation model, in the following manner: For the updated graphical visualization of the target time zone, one or more of the arrangement of one or more interface features or the combination of one or more interface features are determined from the client device interaction associated with the sample time zone; as well as Generate the updated graphical visualization for the target time zone to reflect one or more of the arrangement or combination of the one or more interface features.

3. The system according to claim 2, wherein, The one or more computing devices are further configured to enable the system to: Based on historical network user behavior and user characteristics within the sample time zone and the target time zone, the similarity score between the sample time zone and the target time zone is determined; as well as Based on the similarity score between the sample time zone and the target time zone, the feature visualization machine learning model is used to determine one or more of the arrangement or combination of one or more interface features for the target time zone.

4. The system according to claim 1, wherein, The one or more computing devices are further configured to cause the system to generate the updated graphical visualization of the one or more interface features by: utilizing the feature visualization machine learning model including a feature visualization neural network, in order to: Extract a hidden timezone vector representing the sample timezone from the client device interaction associated with one or more interface features related to the sample timezone; and The updated graphical visualization for the target time zone is generated from the hidden time zone vector.

5. The system according to claim 1, wherein, The one or more computing devices are further configured to enable the system to: In response to client device interactions navigating across multiple graphical user interfaces, the graphical visualization is generated to include the set of interface features across the multiple graphical user interfaces; as well as The updated graphical visualization is generated to include one or more interface features within a single graphical user interface, without requiring client device interaction across multiple graphical user interfaces.

6. The system according to claim 1, wherein, The one or more computing devices are further configured to enable the system to: From the client device interaction associated with the interface feature, determine the selection count corresponding to the interface feature; and Based on the selection count, select one or more interface features for the updated graphical visualization.

7. The system according to claim 1, wherein, The one or more computing devices are further configured to enable the system to: Generate additional graphical visualizations of additional interface feature sets for display via client devices located in additional sample time zones; For the additional sample time zone, determine the client device interaction related to the additional interface feature set; as well as Based on the client device interaction associated with the additional interface feature set, the updated graphical visualization for the target time zone is generated.

8. A non-transient computer-readable medium comprising instructions that, when executed by at least one processor, cause a computing device to: Based on historical network user behavior and user characteristics within multiple geographic regions, a similarity score is generated for the multiple geographic regions to provide interface features for display on client devices. The historical network user behavior is indicated by at least one of the following: interaction type with interface features, number or frequency of interactions, and interaction timing. The user characteristics include at least one of the following: age, gender, education, and language. The order for the plurality of geographic regions is determined based on the similarity score, and the order for the plurality of geographic regions includes an initial geographic region and subsequent geographic regions. For the initial geographic region, client device interactions related to interface features are determined, and the interface features are displayed in a graphical visualization of a first arrangement of interface feature sets on client devices located within the initial geographic region. Based on the client device interaction associated with the interface features and according to the similarity score between the initial geographic region and the subsequent geographic region, an updated graphical visualization of the interface features set for the subsequent geographic region is generated by repositioning, removing, and / or rearranging the interface features using a feature visualization machine learning model: Weights and biases associated with neurons in the feature visualization machine learning model are generated based on performance metrics, wherein the performance metrics include one or more of the following: click counts or conversion counts for the interface features; and Using the weights and the biases, an updated graphical visualization depicting the visual modifications to the set of interface features is generated; as well as The updated graphical visualization of the interface features of the second arrangement is provided for display within the graphical user interface of the client device in the subsequent geographic region.

9. The non-transient computer-readable medium of claim 8, further comprising instructions that, when executed by the at least one processor, cause the computing device to: Determine client device interactions from the subsequent geographic region that are related to the interface features; For additional subsequent geographic regions within the sequence of the plurality of geographic regions, following the subsequent geographic regions, additional updated graphical visualizations of the interface features are generated from client device interactions from the subsequent geographic regions; and Provide the additional updated graphical visualization of the interface features for display on client devices within the additional subsequent geographic region.

10. The non-transient computer-readable medium of claim 8, further comprising instructions that, when executed by the at least one processor, cause the computing device to: The performance metric is determined by monitoring network user behavior related to the interface features from the multiple geographic regions. A new initial geographic region is selected based on the performance metric of the interface features; as well as Based on the similarity score, an updated order for the plurality of geographic regions is determined, the updated order for the plurality of geographic regions including the new initial geographic region and the new subsequent geographic regions.

11. The non-transient computer-readable medium of claim 8, further comprising instructions, which, when executed by the at least one processor, cause the computing device to select the initial geographic region in such a manner as follows: Determine multiple candidate orders for the multiple geographical regions; For successive geographical regions within the multiple candidate sequences, an updated graphical visualization of the interface features is generated; Compare the performance metrics of the interface features associated with each candidate order within the plurality of candidate orders; as well as The initial geographic region is selected based on the comparison of the performance metrics.

12. The non-transient computer-readable medium of claim 8, further comprising instructions that, when executed by the at least one processor, cause the computing device to generate the updated graphical visualization of the interface features using the feature visualization machine learning model to improve one or more performance metrics including clicks or conversions.

13. The non-transient computer-readable medium of claim 8, further comprising instructions, which, when executed by the at least one processor, cause the computing device to generate the similarity scores for the plurality of geographic regions in such a way as: Monitor network user behavior within each of the multiple geographic regions; and Based on the network user behavior and for a given geographic region among the plurality of geographic regions, determine a similarity score for each other geographic region within the plurality of geographic regions.

14. The non-transient computer-readable medium of claim 8, further comprising instructions, which, when executed by the at least one processor, cause the computing device to determine the order for the plurality of geographic regions in such a way as follows: Select the initial geographic region from the plurality of geographic regions; The geographic region with the highest similarity score to the initial geographic region is identified as the subsequent geographic region, which is used to follow the initial geographic region in the order. as well as The next geographic region with the next highest similarity score for the subsequent geographic region is identified as an additional subsequent geographic region, which follows the subsequent geographic region in the order.

15. A computer-implemented method, comprising: Compare the similarity scores of multiple candidate sample time zones with respect to a target time zone, wherein the similarity scores are determined based on historical network user behavior and user characteristics within the respective candidate sample time zone and the target time zone, the historical network user behavior being indicated by at least one of the following: interaction type with interface features, number or frequency of interactions, and interaction timing, and the user characteristics including at least one of the following: age, gender, education, and language; Based on the comparison, a sample time zone is selected from the plurality of candidate sample time zones; A graphical visualization of the first set of interface features is generated for display within one or more graphical user interfaces of multiple client devices located in the sample time zone; Based on client device interactions associated with the interface features of the first arrangement and according to the similarity score between the sample time zone and the target time zone, an updated graphical visualization of the interface feature set for the target time zone is generated by repositioning, removing, and / or rearranging one or more interface features in the interface feature set using a feature visualization machine learning model: Weights and biases associated with neurons in the feature visualization machine learning model are generated based on a target performance metric, wherein the target performance metric includes one or more of the following: click counts or conversion counts for interface features; and Using the weights and the biases, an updated graphical visualization depicting the visual modifications to the set of interface features is generated; as well as An updated graphical visualization of one or more interface features of the second arrangement is provided for display within the graphical user interface of a client device in the target time zone.

16. The computer-implemented method according to claim 15, wherein, The graphical visualization of the interface feature set includes: generating one or more of the arrangements or combinations of the interface feature set for the sample time zones preceding the target time zone.

17. The computer-implemented method according to claim 15, wherein, Providing the updated graphical visualization for display within the graphical user interface of the client device in the target time zone includes providing one or more of an arrangement or combination that is different from the arrangement or combination of the set of interface features represented in the graphical visualization for the sample time zone.

18. The computer-implemented method according to claim 15, wherein: Generating the graphical visualization includes: generating the set of interface features for display across the multiple graphical user interfaces in response to client device interactions navigating across the multiple graphical user interfaces; and Providing the updated graphical visualization includes providing the one or more interface features within a single graphical user interface, without requiring client device interaction across multiple graphical user interfaces.

19. The computer-implemented method according to claim 15, wherein, Providing the updated graphical visualization for display within the graphical user interface of the client device in the target time zone includes: providing a modified product description.

Images