Computing system, computerized method, computer-readable medium and program

Abstract

A computing system is provided, comprising one or more processors configured to provide a social media platform configured to provide a content feed to a user. The processor is further configured to generate user content interaction information by detecting user interactions with the content feed, and to provide a recommendation engine that selects content items for display in the content feed based on the generated user content interaction information. The processor is further configured to receive a refresh request to refresh the recommendation engine, and upon receiving the refresh request, refresh the recommendation engine, at least in part, by masking or resetting the user content interaction information. After the refresh, the processor is further configured to generate a refreshed content item via the recommendation engine based on the masked or reset user content interaction information, and to transmit the refreshed content item to the user device for display in the content feed.

Description

[Technical Field]

[0001] [CROSS-REFERENCE TO RELATED APPLICATIONS] This application claims priority to U.S. Application No. 17 / 935,902, filed September 27, 2022, entitled "Social Media Platform with Recommendation Engine Refresh," the entire contents of which are incorporated herein by reference in their entirety.

[0002] [Background technology] With the emergence of online social media platforms that allow users to share original media with the world within seconds, users upload, consume, and engage with a vast amount of content every day. Such content includes text posts, photos, long-form videos, and short-form videos. However, the sheer volume of content available on social media platforms makes it difficult for users to find uploaded content that matches their preferences. To solve this problem, recommendation systems have been developed that recommend content to appear in each user's content feed. Artificial intelligence (AI) model-based recommendation systems effectively create a content feed for each user on a social media platform by recommending content to the user based on the user's previous engagement with content on the social media platform. Summary of the Invention [Means for solving the problem]

[0003] To solve the problems discussed herein, computerized systems and methods are provided. In one aspect, a computerized system is provided, the computerized system comprising one or more processors configured to provide a social media platform configured to provide a content feed to a user computing device of a user by executing instructions stored in a memory. The processor is further configured to generate user content interaction information by detecting user interactions with the content feed, and to provide a recommendation engine that selects content items for display in the content feed based on the generated user content interaction information. The processor is further configured to receive a refresh request to refresh the recommendation engine via a graphical user interface (GUI) including a refresh selector, and to refresh the recommendation engine in response to receiving the refresh request, at least in part, by masking or resetting the user content interaction information. After the refresh, the processor is further configured to input the masked or reset user content interaction information into the recommendation engine. The processor is further configured to generate a refreshed content item via the recommendation engine based on the masked or reset user content interaction information and transmit the refreshed content item to the user equipment for display in the content feed.

[0004] This Summary is provided to introduce in a simplified form a selection of concepts that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to limit the scope of the claimed subject matter. Moreover, the claimed subject matter is not limited to implementations that solve any or all of the disadvantages noted in any part of this disclosure. [Brief explanation of the drawings]

[0005] [Figure 1] FIG. 1 is a schematic diagram of a computing system including a social media platform configured to provide a content feed to a user computing device, wherein content items in the content feed are selected based on user content interaction information, and the recommendation engine is refreshed by masking or resetting the user content interaction information. [Figure 2] 2 illustrates an exemplary graphical user interface (GUI) of a user computing device for a system according to FIG. 1 displaying a content feed. [Figure 3A] 2 illustrates an example GUI for the system of FIG. 1 displaying a content feed refresh selector on a user computing device. [Figure 3B] 2 illustrates an example GUI for the system of FIG. 1 displaying a content feed refresh selector on a user computing device. [Figure 4] 2 is a schematic diagram illustrating an exemplary embedding generated by an embedding generator and a masked embedding generated by a masking module of the system according to FIG. 1; FIG. [Figure 5] 2 shows an example of masking rules implemented by a masking module of the system according to FIG. 1 after a recommendation engine refresh request. [Figure 6] 2A and 2B show an example GUI of the system of FIG. 1 displaying a content feed on a user computing device before and after a refresh of the recommendation engine. [Figure 7] FIG. 1 illustrates a flowchart of a computerized method for refreshing a recommendation engine, according to one implementation of the present disclosure. [Figure 8] FIG. 1 illustrates an exemplary computing environment in which exemplary embodiments of the present disclosure may be implemented. DETAILED DESCRIPTION OF THE INVENTION

[0006] As described above, computer-based technologies have been developed to enhance users' experiences on social media platforms, in which artificial intelligence systems detect and utilize the user's interactions with content to determine the user's tastes and preferences. The systems then use this data to select content for the user, allowing the user to enjoy personalized media content selected in a personalized manner from among the plethora of media content available on the platform. However, if the user's interests change, the recommendation system may select content based on the user's previous content interactions that no longer reflect the user's preferences. As a result, the recommendation system may provide the user with content that the user is no longer interested in. This may reduce the user's enjoyment of the social media platform.

[0007] In view of the above-mentioned problems, a social media platform utilizing recommendation engine refresh is provided. FIG. 1 is a schematic diagram of a computing system 2 including a social media platform 16 configured to provide a content feed 10 to a user computing device 38 of a user, in which content items 12 in a content feed 10 have been selected via a recommendation engine 8 based on content interaction information 24, and the recommendation engine 8 has been refreshed by masking or resetting the user content interaction information 24. The computing system 2 may include one or more processors 4 configured to perform the functions and processes of the computing system 2 described herein by executing instructions using an associated memory 6. For example, the computing system 2 may comprise a cloud server platform including multiple server devices, and the one or more processors 4 may be one processor on a single server device or multiple processors on multiple server devices. Below, functions of the computing system 2 performed by the processor 4 are described by way of example, and this description should be understood to include execution on one or more processors 4 distributed across one or more of the aforementioned devices.

[0008] The social media platform generates a personalized content feed 10 for each user based on user and content data 18, including user information 20, including user identification (ID) 22, user content interaction information 24, content information 26, device information 28, search history 32, and recommendation history information 34, and provides the content feed 10 to the user's user computing device 38. The user and content data 18 is also used to personalize the user's experience with other services 60, such as advertising. The user information 20 includes, for example, a unique user identification 22 selected by the user upon account creation, a password, the user's language, country, gender, and interest categories. The content information 26 includes content characteristics, such as keywords in subtitles, hashtags, and audio content identification information. For example, the content information 26 may identify videos as being about cars, travel, cooking, or various other topics based on keywords present in the subtitles or based on hashtags. Additionally, the content information 26 may include the number of views or likes the content has received. Additionally, content information 26 may indicate whether the content uses a particular image or video filter, template, etc. Content information does not include user-specific information, but instead includes information about content such as videos, images, and text uploaded by the user and stored on the social media platform. Device information 28 includes information about the user's computing device 38, such as the location, device type, operating system, and time zone of the user's computing device 38. This information may be updated each time the user's computing device communicates with the social media platform. Computing device 38 may be any type of computing device, such as a smartphone, tablet computing device, head-mounted display device, laptop computer, desktop computer, smartwatch, etc.The search history 32 includes keywords that a user has entered as a search query into a search tool of a social media platform, and the recommendation history 34 includes content items 12 that have appeared in the content feed 10 and that have been previously recommended by the computing system 2.

[0009] User content interaction information 24 is generated by detecting user interactions with content feed 10 via processor 4 of system 2. Processor 4 is configured to detect user interactions with content items 12 in content feed 10, record the user's interactions as user content interaction information 24, and update the information as users view and interact with content items 12 by transmitting data indicative of updates 47 from user computing device 38 to computing system 2. Content items 12 may be any type of digital content type, such as video, audio, or images. For example, user content interaction information 24 may include other accounts the user follows, content the user has liked or shared, content the user has commented on, content the user has added to favorites, and content the user has flagged as "not interested."

[0010] The processor 4 is further configured to provide a recommendation engine 8 and input the generated user content interaction information 24 to the recommendation engine 8. The recommendation engine 8 may include a trained machine learning model 8A and a masking module 48 configured to selectively mask input to the trained machine learning model 8A based on instructions from the refresh request handler 45. The trained machine learning model 8A may be trained to predict content with which a user is likely to interact based on previous user interactions with social media platforms. The recommendation engine 8 selects content items 12 for display in the content feed 10 based on the generated user content interaction information 24 via the trained machine learning model 8A. The trained machine learning model 8A of the recommendation engine 8 can be built on a trained neural network, such as a Transformer model, which is a deep learning model that employs a self-attention mechanism to differentially weight the importance of each portion of input data. In addition to user content interaction information 24, user information 20, content information 26, and device information 28 may be input into a trained machine learning model 8A of recommendation engine 8 and used by the engine to select content items 12 for display in content feed 10. For example, a user looking to buy a car may follow a car dealer's account, watch videos uploaded by the car dealer featuring cars for sale, and leave comments on the videos. In this scenario, the trained machine learning model 8A of recommendation engine 8 is likely to select content items 12 featuring cars for sale that appear in content feed 10 for the user. Furthermore, if user information 20 indicates that the user is located in the United States, the trained machine learning model 8A of recommendation engine 8 may be trained to select content items 12 that depict cars for sale in the United States.

[0011] Briefly, attention is directed to FIG. 2 , which illustrates an example GUI 40 for displaying a content feed 10 on a user computing device 38. The processor 4 is further configured to provide the user computing device 38 with a graphical user interface (GUI) 40 configured to display the content feed 10. As shown in FIG. 2 , the GUI 40 for the content feed 10 includes content items 12, a “LIKE” icon 80, a “COMMENTS” icon 82, a “FAVORITES” icon 84, and a settings icon 86 for a user to view and enter user account settings. In this example, content items 12 including automobiles selected by a trained machine learning model 8A of the recommendation engine 8 are displayed on the content feed 10. The “LIKE” icon 80 takes the form of a heart symbol, and a user may use the “LIKE” button by tapping on the “LIKE” icon 80 for content items 12 that the user enjoys. The “COMMENTS” icon 82 allows a user to add comments to the content items 12. A "FAVORITES" icon 84 allows the user to add content items 12 to the user's favorites folder for easy later access. Tapping these icons serves as an interaction between the user and the content 22, which is detected and stored as user-content interaction information 24 as described above. GUI 40 also includes a settings icon 86 that allows the user to manage the user's account settings for purposes described herein.

[0012] The GUI 40 further includes a refresh selector 42. When the refresh selector 42 is selected, the user computing device 38 is configured to send a refresh request to the computing system 2. A refresh request handler 45 executed by the processor 4 of the computing system 2 is configured to receive a refresh request 44 to refresh the recommendation engine 8 from the user computing device 38 in response to a user selection of the refresh selector 42. The refresh request handler 45 may be configured to implement the requested refresh operation by communicating with a masking module 48 of the recommendation engine 8 and instructing the masking module to perform masking according to masking rules 50. Briefly, attention is directed to FIGS. 3A and 3B , which show an exemplary GUI 40 displaying the recommendation engine refresh selector 42 on the user computing device 38. When the user taps the settings icon 86 of FIG. 2 , the GUI 40 displays account settings 90 that allow the user to manage settings such as security, privacy, and content feed preferences 92, as shown in FIG. 3A . When a user selects a content feed preference 92, the GUI 40 displays a recommendation engine refresh selector 42, as shown in FIG. 3B , that allows the user to make a refresh request to refresh the recommendation engine 8. When a refresh request is made, the GUI 40 may display text that interprets the intent of the recommendation engine refresh and the impact that activating such a recommendation engine refresh will have on the user's experience. The text may also prompt the user to "cancel" or "confirm" the refresh. Selecting the "confirm" option activates the recommendation engine refresh. Additionally, the processor 4 may be configured to receive a request from the user to cancel the recommendation engine refresh and, in response, execute the request to cancel the recommendation engine refresh. This request can be made if the user is dissatisfied with the content after the recommendation engine refresh.

[0013] Returning to FIG. 1 , in response to receiving the refresh request 44 as described above, the processor 4 is configured to refresh the recommendation engine 8, at least in part, by masking or resetting the user content interaction information 24. Upon receiving the request, the processor 4, via the masking module 48, masks a portion of the user and content data 18, including the user information 20 and the user content interaction information 24, in accordance with the masking rules 50. The processor 4 is further configured to input the masked or reset user content interaction information 24 and the user information 20 to the trained machine learning model 8A of the recommendation engine 8. Masking the user content interaction information 24 and the user information 20 may include masking raw data of the user content interaction information 24 and the user information 20, and masking embeddings corresponding to the user content interaction information 24 and the user information 20.

[0014] Briefly, attention is drawn to FIG. 4 , which shows a schematic diagram of the process of generating embeddings via embedding generator 112 and masking the generated embeddings via masking module 48. Initially, feature identifications (IDs) 102 are extracted from raw user and content data 18, including user-content interaction information 24, via feature extraction module 110. Each feature ID 102 uniquely identifies a respective feature within the dataset. For example, a feature ID 102 extracted from user information 20 may include the United States as the user's country and English as the user's language. The feature IDs extracted from user-content interaction information 24 may include content that the user liked and commented on. After the feature IDs 102 are extracted, embeddings 104 for the feature IDs 102 are generated via embedding generator 112. The embeddings 104 are numeric vectors (e.g., [1, 0.5, 2.1...]) that represent the feature IDs 102. The embeddings 104 are provided to the masking module 48 to generate masked embeddings 106 for the particular target information by setting the vector values ​​to a predetermined masking value, e.g., 0. Alternatively, masking may be performed by setting the vector values ​​to other values ​​that cause the masked embeddings 106 to be effectively ignored or filtered out by the recommendation engine 8. In the illustrated example, embeddings 1 (EMB 1) through 100 (EMB 100), which represent user information 20 and user-content interaction information 24, are masked. These embeddings are referred to as masked embeddings 106. Embeddings 101 (EMB 101) through 100 (EMB 100), which represent other information, e.g., content information 26 and device information 28, are unmasked. These embeddings are referred to as unmasked embeddings 107. The masked embeddings 106, along with the unmasked embeddings 107, are input to a trained machine learning model 8A of the recommendation engine 8. To select a content item, masking embeddings 106 having a predetermined masking value, for example 0, are filtered out by the trained machine learning model 8A of the recommendation engine 8.That is, in the illustrated example, user information 20 and user-content interaction information 24 are not considered by the trained machine learning model 8A of recommendation engine 8 when selecting refreshed content items 12. Furthermore, because the recommendation engine refresh does not delete all user-specific data associated with the user account but only masks some of the user and content data 18, such as user information 20 and user-content interaction information 24, the user can continue to use the platform's services without setting up a new account. By not masking device information 28, as an example, the user can continue to browse content targeted to the user's geographic location. Furthermore, because user search history 32 is not deleted, the social media platform's features, such as a search tool that provides personalized search results, function as the user is accustomed to after refreshing recommendation engine 8. Additionally, because user information 20 and user content interaction information 24 are masked rather than deleted, and search history 32 and recommendation history 34 are not deleted, sponsored content such as advertisements can be delivered to users based on those data sources after the recommendation engine is refreshed, thereby displaying more relevant sponsored content to each user than would be the case if such content were deleted.

[0015] As described above, after a recommendation engine refresh, the user information 20 and the user content interaction information 24 are masked by the masking module 48 according to the masking rules 50. FIG. 5 shows an example of the masking rules 50 implemented by the masking module 48 of the system 2 according to FIG. 1. Before a refresh request is made, the user and content data 18 information is not masked or reset, as shown at 200. According to the masking rules 50, the refresh of the recommendation engine 8 may include a first temporary refresh that masks or resets the user content interaction information 24 and the user information 20, including the user identification information 22, until a first predetermined threshold number of views 50A is reached. In the illustrated example, after a refresh request is made, the user information 20, including the user identification information 22, and the user content interaction information 24 are masked according to the masking rules 50, as shown at 202, until the first predetermined threshold number of views of the content item 12 after the refresh. The first predetermined number of views 50A may range from 25 to 100 views, and in one particular embodiment, is 50 views. After the first predetermined number of views occurs, the user content interaction information 24 and the user information 20, excluding the user identification information 22, are unmasked, so that only the user identification information 22 remains masked, as shown at 204. Thus, in accordance with the masking rule 50, a refresh of the recommendation engine may include a second temporary refresh after the first predetermined threshold number of views 50A occurs, unmasking the user content interaction information 24 and the user information 20, excluding the user identification information 22, and masking the user identification information 22, until a second predetermined threshold number of views 50B is reached. The second predetermined number of views may range from 150 to 300 views, and in one particular embodiment, is 200 views. The computing system 2 is further configured to terminate the second temporary refresh by unmasking the user identification information 22 after a second predetermined threshold number of views 50B has been reached.Therefore, after the second predetermined number of views 50B, the masking of the user identification information 22 is also removed, as shown at 206, and no other information is masked.

[0016] After the recommendation engine refresh, the processor 4 is configured to generate refreshed content items 12 via the recommendation engine 8 based on the masked or reset user content interaction information 24 and user information 20, and based on the unmasked or reset user device information 28, search history 32, and recommendation history 34, and to send the refreshed content items 12 to the user computing device 38 for display in the content feed 10. FIG. 6 shows example GUIs of the system 2 according to FIG. 1 displaying the content feed 10 on the user computing device 38 before and after a recommendation engine refresh. The illustrated example 250 shows the GUI 40 before the recommendation engine refresh, in which the recommendation engine 8 includes the content items 12 selected for the user in accordance with the present invention. In the illustrated situation, the content feed 10 before the refresh includes content items 12 featuring automobiles, according to the recommendation engine's selection. In contrast, the illustrated example 252 shows the GUI 40 after the recommendation engine refresh, in which the content feed 10 includes the content items 12 that would be selected for the new user. In the illustrated situation, after the content refresh, instead of the content items 12 featuring automobiles, the content feed 10 includes content items 12 that are preferred by a broad demographic of users of the social media platform 16. As the user continues to use the social media platform 16 of Figure 1, the computing system 2 repeats the process of collecting new data, determining the user's new preferences, and selecting appropriate content items 12 for inclusion on the user's content feed 10 so that the feed accurately reflects the user's most recent preferences.

[0017] In one configuration of computing system 2, the processor 4 may be configured to mask or reset portions of user and content data 18, such as user information 20 and user content interaction information 24, to allow users to refresh advertising content similar to the content feed 10 described above, and to provide the masked information to other services 60, such as advertisements provided by social media platforms 16.

[0018] 7 illustrates a flowchart of a computerized method 300 for refreshing a recommendation engine according to one implementation of the present disclosure. The method 300 may be implemented by the hardware and software of the computing system 2 described above, or by other suitable hardware and software. In step 304, the method 300 may include providing a social media platform configured to provide the content feed to a user's user computing device.

[0019] At step 306, the method may further include generating user content interaction information by detecting user interaction with the content feed. At step 308, the method may further include providing a recommendation engine that selects content items for display in the content feed based on user information including user identification information for the user and the generated user content interaction information. Additionally, content information, device information, and other information may be input to the recommendation engine.

[0020] In step 310, the method may further include providing a graphical user interface (GUI) configured to display the content feed to a user device, the GUI including a refresh selector, and receiving the refresh request from the user device in response to a user selection of the refresh selector.

[0021] In step 312, the method may further include, in response to receiving the refresh request, refreshing the recommendation engine at least in part by masking or resetting the user content interaction information. As shown in step 314, the user information and the user content interaction information may be masked by generating embedded information representing the user information and the user content interaction information and masking the embedded information.

[0022] At step 316, the method may further include inputting the masked or reset user information and user content interaction information into the recommendation engine. At step 318, the method may further include generating a refreshed content item via the recommendation engine based on the masked or reset user information and user content interaction information. At step 320, the method steps may further include transmitting the refreshed content item to the user computing device for display in the content feed.

[0023] The above-described systems and methods may be implemented to reset a user's content feed on a social media platform without requiring the recreation of a new user account or the deletion of data reflecting the user's engagement with content on the platform. The above-described systems and methods can create impactful changes in the content feed that are quickly perceived by the user without interrupting the user's continued use of the platform.

[0024] In some embodiments, the methods and processes described herein may be coupled to the computing system of one or more computing devices. In particular, such methods and processes may be implemented as computer application programs or services, application programming interfaces (APIs), libraries, and / or other computer program products.

[0025] 8 schematically illustrates a non-limiting embodiment of a computing system 600 capable of implementing one or more of the methods and processes described above. Computing system 600 is shown in simplified form. Computing system 600 may embody computing system 2, described above and shown in FIG. 1. Computing system 600 may take the form of one or more personal computers, server computers, tablet computers, home entertainment computers, network computing devices, gaming devices, mobile computing devices, mobile communication devices (e.g., smartphones) and / or other computing devices, and wearable computing devices, such as smart watches and head-mounted augmented reality devices.

[0026] Computing system 600 includes a logical processor 602, a volatile memory 604, and a non-volatile storage device 606. Computing system 600 may optionally include a display subsystem 608, an input subsystem 610, a communication subsystem 612, and / or other components not shown in FIG.

[0027] Logical processor 602 comprises one or more physical devices configured to execute instructions. For example, a logical processor may be configured to execute instructions that are part of one or more applications, programs, routines, libraries, objects, components, data structures, or other logical constructs. Such instructions may be implemented to perform a task, implement a data type, transform the state of one or more components, achieve a technical effect, or achieve a desired result.

[0028] A logical processor may include one or more physical processors (hardware) configured to execute software instructions. Additionally or alternatively, a logical processor may include one or more hardware logic circuits or firmware devices configured to execute hardware-implemented logic or firmware instructions. The processors of logical processor 602 may be single-core or multi-core, and the instructions executed thereon may be configured for sequential, parallel, and / or distributed processing. Individual components of a logical processor may optionally be distributed among two or more separate devices that may be remotely located and / or configured for coordinated processing. Aspects of a logical processor may be virtualized and executed by remotely accessible network-connected computing devices configured in a cloud computing configuration. In such cases, it should be understood that these virtualized aspects execute on different physical logical processors of various different machines.

[0029] Non-volatile storage device 606 includes one or more physical devices configured to hold instructions executable by a logical processor to implement the methods and processes described herein. When such methods and processes are implemented, the state of non-volatile storage device 606 may be transformed, for example, to hold different data.

[0030] The non-volatile storage device 606 may comprise removable and / or embedded physical devices. The non-volatile storage device 606 may include optical memory (e.g., CD, DVD, HD-DVD, Blu-Ray Disc, etc.), semiconductor memory (e.g., ROM, EPROM, EEPROM, flash memory, etc.), and / or magnetic memory (e.g., hard disk drives, floppy disk drives, tape drives, MRAM, etc.), or other mass storage technologies. The non-volatile storage device 606 may include non-volatile, dynamic, static, read / write, read-only, sequential access, position-addressable, file-addressable, and / or content-addressable devices. It will be understood that the non-volatile storage device 606 is configured to retain instructions even when power to the non-volatile storage device 606 is disconnected.

[0031] Volatile memory 604 may include a physical device with random access memory. Volatile memory 604 is typically utilized by logical processor 602 to temporarily store information during the processing of software instructions. It will be appreciated that if power to volatile memory 604 is removed, volatile memory 604 typically will not continue to store instructions.

[0032] Aspects of the logic processor 602, volatile memory 604, and non-volatile storage device 606 may be integrated together into one or more hardware logic components, which may include, for example, field programmable gate arrays (FPGAs), program and application specific integrated circuits (PASICs / ASICs), program and application specific standard products (PSSPs / ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0033] The terms "module," "program," and "engine" may be used to describe aspects of computing system 600 that are typically implemented in software to perform specific functions, including transformations by a processor using portions of volatile memory to specifically configure a program to perform a certain function. Thus, a module, program, or engine may be instantiated using portions of volatile memory 604 via logic processor 602 executing instructions held in non-volatile storage 606. It will be understood that different modules, programs, and / or engines may be instantiated from the same application, service, code block, object, library, routine, API, function, etc. Similarly, the same module, program, and / or engine may be instantiated from different applications, services, code blocks, objects, routines, APIs, functions, etc. The terms "module," "program," and "engine" may include individuals or groups of executable files, data files, libraries, drivers, scripts, database records, etc.

[0034] If included, the display subsystem 608 may be used to present a visual representation of the data maintained by the non-volatile storage device 606. This visual representation may take the form of a graphical user interface (GUI). As the methods and processes described herein modify the data maintained by the non-volatile storage device and transform the state of the non-volatile storage device, the state of the display subsystem 608 may be similarly transformed to visually represent the changes in the underlying data. The display subsystem 608 may include one or more display devices utilizing virtually any type of technology. Such display devices may be combined with the logic processor 602, volatile memory 604, and / or non-volatile storage device 606 in a shared housing, or such display devices may be peripheral display devices.

[0035] If included, the input subsystem 610 may include and interact with one or more user input devices, such as a keyboard, mouse, touchscreen, or game controller. In some embodiments, the input subsystem may include and interact with selected natural user input (NUI) components. These components may be integral or peripheral, and input action transmission and / or processing may be handled on-board or off-board. Exemplary NUI components may include microphones for speech and / or voice recognition; infrared, color, stereo, and / or depth cameras for machine vision and / or gesture recognition; head trackers, eye trackers, accelerometers, and / or gyroscopes for motion detection and / or intent recognition; electric field sensing components for assessing brain activity; and / or any other suitable sensors.

[0036] If included, the communications subsystem 612 may be configured to communicatively couple the various computing devices described herein with each other or with other devices. The communications subsystem 612 may include wired and / or wireless communication devices compatible with one or more different communications protocols. As a non-limiting example, the communications subsystem may be configured to communicate via a wireless telephone network, or a wired or wireless local or wide area network, e.g., HDMI® over a Wi-Fi connection. In some embodiments, the communications subsystem may enable the computing system 600 to send and receive messages to other devices over a network, such as the Internet.

[0037] The following paragraphs provide additional support for the claims of the present application. One aspect provides a computing system. The computing system may include one or more processors configured to provide a social media platform configured to provide a content feed to a user computing device of a user by executing instructions stored in a memory. The processor may be further configured to generate user content interaction information by detecting user interactions with the content feed, and to provide a recommendation engine that selects content items for display in the content feed based on the generated user content interaction information. The processor may be further configured to receive a refresh request to refresh the recommendation engine, and in response to receiving the refresh request, refresh the recommendation engine at least in part by masking or resetting the user content interaction information. The processor may be further configured to input the masked or reset user content interaction information to the recommendation engine, generate a refreshed content item via the recommendation engine based on the masked or reset user content interaction information, and transmit the refreshed content item to the user computing device for display in the content feed. The processor may further be configured to provide a graphical user interface (GUI) including a refresh selector configured to display the content feed to the user computing device, and to receive the refresh request from the user computing device in response to a user selection of the refresh selector.

[0038] According to this aspect, the recommendation engine may select a content item for display based on the generated user-content interaction information and user information, the user information including at least user identification information for the user. The recommendation engine may further select the content item for display based on the generated user-content interaction information, the user information, and device information of the user.

[0039] According to this aspect, the processor may be further configured to, in response to receiving the refresh request, refresh the recommendation engine, at least in part, by additionally masking or resetting the user information, including the user identification information.

[0040] According to this aspect, refreshing the recommendation engine may include a first temporary refresh that masks or resets the user content interaction information and the user information including the user identification information until a first predetermined threshold number of views is reached. After the first predetermined threshold number of views occurs, refreshing the recommendation engine may further include a second temporary refresh that unmasks the user content interaction information and the user information excluding the user identification information and masks the user identification information until a second predetermined threshold number of views is reached.

[0041] According to this aspect, the processor may be further configured to terminate the second temporary refresh by unmasking the user identifying information after a second predetermined threshold number of views has been reached.

[0042] According to this aspect, the processor may be further configured to generate embedded information based on the user content interaction information, and refresh the recommendation engine by masking or resetting the user content interaction information by masking or resetting the embedded information representing the user content interaction information.

[0043] According to this aspect, the processor may be further configured to receive a request from a user to cancel the recommendation engine refresh, and to cancel the recommendation engine refresh in response to receiving the request.

[0044] According to another aspect of the present disclosure, a computerized method is provided. The computerized method may include providing a social media platform configured to provide the content feed to a user computing device of a user. The computerized method may further include generating user content interaction information by detecting user interaction with the content feed. The computerized method may further include providing a recommendation engine that selects content items for display in the content feed based on the generated user content interaction information. The computerized method may further include receiving a refresh request to refresh the recommendation engine and refreshing the recommendation engine, at least in part, by masking or resetting the user content interaction information in response to receiving the refresh request. The computerized method may further include inputting the masked or reset user content interaction information into the recommendation engine. The computerized method may further include generating refreshed content items via the recommendation engine based on the masked or reset user content interaction information and transmitting the refreshed content items to the user computing device for display in the content feed. The computerized method may further include providing a graphical user interface (GUI) configured to display the content feed to a user computing device, the GUI including a refresh selector, and receiving the refresh request from the user computing device in response to a user selection of the refresh selector.The computerized method may further include selecting, via the recommendation engine, a content item for display based on the generated user content interaction information and user information, the user information including at least user identification information for the user. The computerized method may further include, in response to receiving the refresh request, refreshing the recommendation engine at least in part by additionally masking or resetting the user information including the user identification information. The computerized method may further include generating embedded information based on the user content interaction information, and refreshing the recommendation engine by masking or resetting the user content interaction information by masking or resetting embedded information representing the user content interaction information.

[0045] According to this aspect, refreshing the recommendation engine may include a first temporary refresh that masks or resets the user content interaction information and the user information including the user identification information for a first predetermined threshold number of views. The refresh of the recommendation engine may further include a second temporary refresh that unmasks the user content interaction information and the user information excluding the user identification information and masks the user identification information for a second predetermined threshold number of views.

[0046] According to another aspect of the present disclosure, a computer-readable medium is provided. The computer-readable medium may include instructions that, when executed by one or more processors, cause the one or more processors to: provide a social media platform configured to provide a content feed to a user computing device of a user; and generate user content interaction information by detecting user interaction with the content feed. These steps may further include providing a recommendation engine that selects content items for display in the content feed based on the generated user content interaction information, user information including at least user identification information, and device information of the user. These steps may further include receiving a refresh request to refresh the recommendation engine; and refreshing the recommendation engine, at least in part, by masking or resetting the user content interaction information and the user information in response to receiving the refresh request. The steps may further include inputting the masked or reset user content interaction information, the masked or reset user information, and the device information of the user into the recommendation engine, and generating, via the recommendation engine, refreshed recommended content items based on the masked or reset user content interaction information and the masked or reset user information. The steps may further include transmitting the refreshed content items to the user computing device for display in the content feed.

[0047] It will be understood that the configurations and / or approaches described herein are exemplary in nature and are susceptible to numerous variations, and therefore, these specific embodiments or examples should not be considered in a limiting sense. The particular routines or methods described herein may represent one or more of any number of processing strategies. As such, various operations shown and / or described may be performed in the order shown and / or described, in other orders, in parallel, or may be omitted. Similarly, the order of the processes described above may be changed.

[0048] The subject matter of the present disclosure includes novel and non-obvious combinations and sub-combinations of the various process, system, configurations, and other features, functions, operations and / or properties disclosed herein, and all equivalents thereof.

Claims

1. 1. A computing system comprising: By executing instructions stored in memory, providing a social media platform configured to provide a content feed to a user's user computing device; generating user content interaction information by detecting user interactions with the content feed; providing a recommendation engine that selects content items for display in the content feed based on the generated user content interaction information; receiving a refresh request to refresh the recommendation engine; In response to receiving the refresh request, refreshing the recommendation engine, at least in part, by masking or resetting the user-content interaction information.

1. A computing system comprising one or more processors configured to:

2. The one or more processors further comprise: inputting the masked or reset user content interaction information into the recommendation engine; generating, via the recommendation engine, refreshed content items based on the masked or reset user-content interaction information; transmitting the refreshed content item to the user computing device for display within the content feed; 10. The computing system of claim 1, configured to:

3. The one or more processors further comprise: providing a graphical user interface including a refresh selector configured to display the content feed on the user computing device; receiving the refresh request from the user computing device in response to a user selection of the refresh selector; 10. The computing system of claim 1, configured to:

4. The recommendation engine selects content items for display based on the generated user-content interaction information and user information, the user information including at least user identification information for the user. The computing system of claim 1 .

5. The recommendation engine selects the content items for display based on the generated user-content interaction information, the user information, and the user's device information. The computing system of claim 4 .

6. The one or more processors are further configured to, in response to receiving the refresh request, refresh the recommendation engine at least in part by additionally masking or resetting the user information, including the user identification information.

5. The computing system of claim 4, configured to:

7. Refreshing the recommendation engine includes a first temporary refresh that masks or resets the user content interaction information and the user information, including the user identification information, until a first predetermined threshold number of views is reached. The computing system of claim 6 .

8. The refresh of the recommendation engine includes a second temporary refresh that unmasks the user content interaction information and the user information, excluding the user identification information, and masks the user identification information after the first predetermined threshold number of views occurs until a second predetermined threshold number of views is reached. The computing system of claim 7 .

9. The one or more processors are further configured to terminate the second temporary refresh by unmasking the user identifying information after the second predetermined threshold number of views is reached. The computing system of claim 8 .

10. The one or more processors further comprise: generating embedded information based on the user content interaction information; Refreshing the recommendation engine by masking or resetting the user content interaction information by masking or resetting embedded information representing the user content interaction information.

10. The computing system of claim 1, configured to:

11. The one or more processors further comprise: receiving a request from a user to cancel a refresh of the recommendation engine; In response to receiving the request, canceling the refresh of the recommendation engine.

10. The computing system of claim 1, configured to:

12. 1. A computerized method for refreshing a recommendation engine, comprising: providing a social media platform configured to provide a content feed to a user's computing device; generating user content interaction information by detecting user interactions with the content feed; providing a recommendation engine that selects content items for display in the content feed based on the generated user content interaction information; receiving a refresh request to refresh the recommendation engine; In response to receiving the refresh request, refreshing the recommendation engine, at least in part, by masking or resetting the user-content interaction information; and A computerized method comprising:

13. inputting the masked or reset user content interaction information into the recommendation engine; generating, via the recommendation engine, refreshed content items based on the masked or reset user-content interaction information; transmitting the refreshed content item to the user computing device for display within the content feed; The computerized method of claim 12 further comprising:

14. providing a graphical user interface including a refresh selector configured to display the content feed on a user computing device; receiving the refresh request from the user computing device in response to a user selection of the refresh selector; The computerized method of claim 12 further comprising:

15. and selecting, via the recommendation engine, a content item for display based on the generated user-content interaction information and user information, the user information including at least user identification information for the user.

13. The computerized method of claim 12.

16. In response to receiving the refresh request, refreshing the recommendation engine at least in part by additionally masking or resetting the user information, including the user identification information.

16. The computerized method of claim 15, further comprising:

17. Refreshing the recommendation engine includes a first temporary refresh that masks or resets the user content interaction information and the user information, including the user identification information, for a first predetermined threshold number of views.

17. The computerized method of claim 16.

18. The refresh of the recommendation engine includes a second temporary refresh that unmasks the user content interaction information and the user information excluding the user identification information for a second predetermined threshold number of views, and masks the user identification information.

18. The computerized method of claim 17.

19. generating embedded information based on the user content interaction information; refreshing the recommendation engine by masking or resetting the user content interaction information by masking or resetting embedded information representing the user content interaction information; The computerized method of claim 12 further comprising:

20. A computer-readable medium storing instructions, comprising: The instructions, when executed by one or more processors: providing a social media platform configured to provide a content feed to a user computing device of the user; generating user content interaction information by detecting user interactions with the content feed; providing a recommendation engine that selects content items for display in the content feed based on the generated user-content interaction information, user information including at least user identification information, and device information of the user; receiving a refresh request to refresh the recommendation engine; In response to receiving the refresh request, refreshing the recommendation engine, at least in part, by masking or resetting the user-content interaction information and the user information; inputting the masked or reset user content interaction information, the masked or reset user information, and the device information of the user into the recommendation engine; generating, via the recommendation engine, refreshed recommended content items based on the masked or reset user-content interaction information and the masked or reset user information; transmitting the refreshed content item to the user computing device for display within the content feed; causing the one or more processors to execute Computer-readable medium.

Images