System, device, and method for generating and operating a gaming interface with a user-tailored scrollable feed of interactive games

Abstract

Generating and operating a gaming interface with a user-tailored and user-specific scrollable feed of interactive games, that is dynamically and uniquely tailored for each user based on computer-estimated user-preferences that are deduced based on deterministic rules or Machine Learning models or Artificial Intelligence engines. A computerized method stores data representing a plurality of online games; wherein each online game is associated with (i) a code-portion that runs the online game, and (ii) a screenshot or a preview video that visually demonstrate the online game. The method dynamically constructs and updates different, user- specific, user-tailored, scrollable feeds of online games for differentially displaying such game feeds on different end-user devices of different users, based on computer-generated estimations or predictions of which genre of online games or which graphical content theme is preferred by each user; taking into account data from monitored user interactions and user engagement with previously-displayed games and game-previews.

Description

System, Device, and Method for Generating and Operating a Gaming Interface with a User-Tailored Scrollable Feed of Interactive GamesCross-Reference to Related Applications

[0001] This patent application claims priority and benefit from US 63 / 627,178, filed on January 31, 2024, which is hereby incorporated by reference in its entirety.Field

[0002] Some embodiments are related to the field of computerized games.Background

[0003] Millions of people use electronic devices on a daily basis. Such electronic devices include, for example, desktop computers, laptop computers, smartphones, tablets, smartwatches, gaming consoles, and other devices.

[0004] Electronic devices are utilized for a variety of tasks and purposes; for example, browsing the Internet, watching videos, listening to music, reading online content, sending and receiving electronic mail (email) messages, engaging in Instant Messaging (IM), engaging in a video conference, performing online shopping, playing games, and other activities.Summary

[0005] Some embodiments provide systems, devices, and methods for generating and operating a gaming interface with a user-tailored scrollable feed of interactive games.

[0006] Some embodiments provide system and methods for generating and operating a gaming interface with a user-tailored and user-specific scrollable feed of interactive games, that is dynamically and uniquely tailored for each user based on computer-estimated userpreferences that are deduced based on deterministic rules or Machine Learning models or Artificial Intelligence engines. For example, a computerized method stores data representing a plurality of online games; wherein each online game is associated with (i) a code-portion that runs the online game, and (ii) a screenshot or a preview video that visually demonstrate the online game. The method dynamically constructs and updates different, user-specific, user- tailored, scrollable feeds of online games for differentially displaying such game feeds on different end-user devices of different users, based on computer-generated estimations or predictions of which genre of online games or which graphical content theme is preferred byeach user; taking into account data from monitored user interactions and user engagement with previously-displayed games and game -previews.

[0007] Some embodiments may provide other and / or additional benefits and / or advantages.Brief Description of the Drawings

[0008] Fig. 1 is a schematic block-diagram illustration of an interactive gaming system, in accordance with some demonstrative embodiments.

[0009] Figs. 2A to 2D are illustrations of screenshots taken from an electronic device, such as a smartphone or a tablet, demonstrating a continuous and user-tailored feed of on-screen games that can be swiped and scrolled by the user, in accordance with some demonstrative embodiments.

[0010] Figs. 3A and 3B are schematic illustrations of differential user-specific game feeds that are generated and / or displayed in parallel to two different users that operate two different end-user devices, in accordance with some demonstrative embodiments.Detailed Description of Some Demonstrative Embodiments

[0011] The Applicant has realized that some users of electronic devices may wish to play games, of a particular type or of various types; yet it may be difficult to discover such games and to start playing them rapidly.

[0012] For example, realized the Applicant, a typical user of a smartphone may need to access an “App Store”; may search for games using a keyword (e.g., “Blackjack”); may then be shown some irrelevant or promoted results; may then be able to see, at most, still screenshots of the game and not an actual stream and not an actual video; may then be required to download the game (which can take time, particularly on a slow connection and / or for games that are large in size), wait for the game to install itself, and then manually run or launch the game; and then, discover that the user does not like this game, and restart the process from the beginning, with an additional task of later uninstalling the previous game from this smartphone.

[0013] The Applicant has realized that this effort-consuming and time-consuming process is not enjoyable for many users, and prevents them from discovering and / or enjoying many games that would otherwise be interesting for them. The Applicant has also realized that many users do not bother to review (and to install, and run, and try) more than 10 or 20 of the numerous search results on an “app store”, and therefore many games remain entirely undiscovered by many users who would otherwise play them. The Applicant has also realizedthat the entire process of finding a new game and starting to engage with it, is typically a laborious task that many users skip or postpone or avoid.

[0014] The Applicant has realized that it may be advantageous and / or beneficial to provide an efficient and enjoyable and user-friendly system and user interface, that enables users of electronic devices (and particularly, smartphones and tablets) to rapidly and efficiently review and consider a large number of computerized games; and to rapidly and efficiently and almost instantly view a video clip or a real-time video stream of each such game; and to rapidly and efficiently and almost instantly commence playing a selected game without the need to download and install and launch a new “mobile app” for it; and to rapidly and efficiently and almost instantly switch from one game to another game, in an enjoyable and efficient manner.

[0015] Some embodiments of the present invention may provide a computerized system and a computerized method, that are configured to present on an electronic device of a user, a virtually infinite scrollable feed of online games, of various type or of a particular user-selected type (or, of a plurality of such user-selected types of games). The user may efficiently scroll through such feed, such as by swiping with his finger upwardly on a touch-screen of his smartphone or tablet, or by scrolling downwardly using a computer mouse scroll-wheel or using the “arrow down” key on a keyboard. The user may efficiently select a game from such scrollable feed, by hovering over it or by tapping it or clicking it. Optionally, a real-time video stream of the game is immediately streamed and shown to the user, or a pre-recorded video clip of the game is immediately streamed and shown to the user. The user may immediately join the game via an additional tap or click or gesture, or by tapping or selecting a “Play Right Now” on-screen button or GUI element; and the end-user device immediately enables that user to play that game, by himself (e.g., against a game server) or against one or more other players (who may be other human players and / or computerized players).

[0016] In accordance with some embodiments, the user may “subscribe” to, or may “follow”, a particular type of games (e.g., gambling games, arcade games, memory games); or a particular sub-type of games (e.g., Blackjack games, Slot Machine games, Poker games); and the system provides a user-tailored feed to different users based on such preferences. For example, the scrollable feed of games that is shown to User Adam on his smartphone may include only (or mostly) Blackjack and Slot Machine games; whereas, the scrollable feed of games that is shown to User Eve on her tablet may include only (or mostly) arcade games.

[0017] In accordance with some embodiments, the user may “subscribe” to, or may “follow”, a particular producer / publisher of games; and the system provides a user-tailored feed to different users based on such preferences. For example, the scrollable feed of gamesthat is shown to User Carla on her smartphone may include only (or mostly) games released by Publisher C; whereas, the scrollable feed of games that is shown to User David on his tablet may include only (or mostly) games released by Publisher D.

[0018] In some embodiments, the system may further tailor the feed of games to each user, based on two or more tailoring conditions or preferences; for example, showing to User Jane only “Blackjack” games that were released only by “Publisher E”.

[0019] In some embodiments, the system tracks and logs the actual engagement of each user with each game that is shown to him on his game-feed; and may analyze such engagement data, using a set of pre-defined rules and conditions, and / or using an Artificial Intelligence (Al) engine / Machine Learning (ML) engine / Deep Learning (DL) engine / Neural Network (NN) engine, to detect or to deduce additional preferences of each user, without such user explicitly selecting or explicitly indicating such preferences to the system. For example, the system may track and log that User Adam has scrolled rapidly and skipped several Blackjack games, and therefore he is not interested or is less interested in Blackjack games; and that User Adam then stopped at a Slot Machine game featuring fantasy characters and watched its video but did not play it; and then User Adam continued to scroll and stopped at a Slot Machine game featuring animals and played it for ten seconds; and then User Adam continued to scroll and stopped at a Slot Machine game featuring fruits and played it for seven minutes. The system may deduce that User Adam is more interested in Slot Machine games, relative to Blackjack games; and that User Adam may prefer games with content of fruit, relative to fantasy character. The system may then modify or adjust the scrollable feed of User Adam, to show more games of the type that was deduced to be more interesting to him, and to show less (or no) games of the type that was deduced to be less interesting to him. These are only non-limiting examples; and the system may generate other determinations or estimations by analyzing the swiping / browsing / tapping / watching / playing activity of each user, and its timing, its time-length, time-gaps, time intervals, or other parameters. For example, the system may further deduce that User Jane prefers games from Publisher A, as she typically spends more time in viewing their videos and / or playing them; whereas User Jane does not like games from Publisher B, as User Jane typically swipes away from them. Other estimations or deducing operations may be performed by the system.

[0020] In some embodiments, optionally, the system may be configured to generate a virtually infinite feed of games, based on a closed set of games that can be modified to create millions of permutations or variants or modifications. In a demonstrative embodiment, optionally, the system includes 100 game programs or game code units that contain the codefor operating each game; for example, program code for a Blackjack game, program code for a Poker game, program code for a Slot Machine game, and so forth. For each such program code, there are optionally provided a set of graphics / animations / sounds / video-clips / textual items / multimedia items that can be customized or replaced. For example, in such optional embodiments, the Slot Machine program game can use: (a) a first set of graphics that depict fruits; or (b) a second set of graphics that depict fantasy characters; or (c) a third set of graphics that depict animals; and so forth. Similarly, in such optional embodiments, the Title of each game (each program code) can be one of several alternate titles, such as: (a) Amazing Slot Machine, or (b) Superb Slot Machine, or (c) Colorful Slot Machine, and so forth. Similarly, in such optional embodiments, the graphic shape of the slot machine itself can be selected from, for example, (a) a square box, (b) a rectangular box, (c) a trapezoid box, and so forth. The system in such optional configurations can select - randomly or pseudo-randomly or based on pre-defined selection rules - a combination of those selectable parameters, thereby producing automatically millions of game variants from a single program code, and thus populating automatically a virtually infinite feed of games through which the user can scroll. In some embodiments, the optional modification or configuration of games may be user-specific or user-tailored; for example, the system may optionally monitor the engagement activities and / or the gaming interactions of User Carla, and may observe or deduce that User Carla spends more time with Slot Machine games that have animal graphics and not fruit graphics; and the system may optionally then tailor the game-feed for User Carla to show permutations of Slot Machine games that maintain the preferred animal graphics (and that avoid the non-preferred fruit graphics), in combination with other modifiers (e.g., modified title; modified look-and-feel of the slot machine; or the like). It is noted that the above on-the-fly customization of games may be an optional feature in some embodiments; as other embodiments may use, instead, a large dataset of pre -provided games having fixed or non-modifiable program code and / or fixed or non-changeable sets of graphics / audio effects / animations / soundtrack / textual components.

[0021] Some embodiments provide a dynamic social media-style gaming experience, and particularly a dynamic social media-style slot machine and / or live casino experience. The Applicant has realized that traditional slot machine games and traditional online casinos lack the engaging and interactive nature that characterizes other / modern digital platforms. Some embodiments innovatively combine or integrate (i) dynamic content delivery mechanisms similar to those implemented in some social media platforms (e.g., TikTok, Instagram) with (ii) an improved online gaming experience (e.g., online casino games / slot machines / card games experience), thereby creating a visually-captivating and interactive environment forusers; which further provides to users a unique ability to rapidly discover - and rapidly engage in - particular games to their likings.

[0022] Some embodiments provide a novel and innovative online gaming / casino platform, that generates presents and operates online casino games content in a manner that is generally inspired by the user interface and user experience (UI / UX) of social media platforms such as TikTok (which provides a scrollable feed for videos only) and Instagram (which provides a scrollable feed for videos and / or photos). The innovative platform utilizes Al / ML / DL / NN algorithms, such as those employing or leveraging Contrastive Language-Image PreTraining (CLIP) or other training or modeling techniques, to dynamically organize and update and tailor the content feed to each end-user, based on semantic analysis extracted from game footage and / or based on analysis of past engagement of that end-user and / or based on prediction or estimation of the user’s preferences or likings, and utilizing Al-driven suggestions and / or Al-driven tailoring of the scrollable game-feed based on machine-estimated / machine- deduced users' preferences, time of use, engagement history, and play history.

[0023] With regard to User Interface and Video Previews, the platform seamlessly blends or combines or generates a continuous or virtually-infinite scrolling feed of offered games, with an online casino environment. Users access a curated and user-specific feed of concise video previews (which may include live or real-time video streams of actual ongoing games, and / or pre-recorded video clips that are streamed instantly), showcasing various casino games and enabling the end-user to efficiently decide whether or not to join the game or to engage with the game, or conversely to skip it and move on (rapidly and efficiently) to another game in his tailored feed of games. These video previews, dynamically organized and personalized through Al-driven recommendations / suggestions / predictions / estimations, enable effortless and efficient browsing through thousands of games and quick access to games of interest.

[0024] With regard to Interactive Game Exploration, some embodiments provide and operate an interactive swipe / scroll mechanism, enabling users to effortlessly browse video previews showcasing an array of online games (particularly casino games), including (for example) roulette, blackjack, slot machines, and / or other games. Tapping (or clicking) on a desired video preview triggers the system to promptly and instantly launch the selected game, delivered on demand from the server, without a need to download or install any local program, directly or indirectly via an “app store”. After gameplay, the user can seamlessly return to his scrollable feed of games that is tailored to his estimated / predicted preferences, continuing exploration using the intuitive scrolling mechanics provided by the system.

[0025] The platform further enriches the gaming experience with social elements and with social interaction features. For example, a user can engage in (or create) live streaming sessions, sharing or up-streaming or broadcasting his real-time gaming experience within the platform with other end-users; and such other users may passively watch his streamed gameplay, and / or may actively interact with the stream gameplay via one or more interaction mechanism; for example, Observer A may passively view the streamed gameplay; Observer B may write comments in a comments section related to the gameplay; Observer C may compose and send a private message to the player using a chat interface; Observer D may compose or send or a public message to one or to some or to all of the other observers; Observer E may share the streamed gameplay with other friends; Observer F may efficiently select a shortlength response or reaction, such as a “thumb-up” icon or a “smiley face” emoji or other icon, which is conveyed by the system to the player and / or to other observers; or the like, thereby creating and fostering a vibrant community atmosphere.

[0026] Some embodiments enable users the capability to “follow” (or, to “subscribe to”) particular game providers or game publishers or game creators, or to “follow” (or, to “subscribe to”) particular types of games (e.g., Slot Machine games and not Blackjack games; or, games that feature animals graphics and not fruit graphics), and / or to otherwise prioritize content, including based on game previews, live gameplay sessions, or special events. This enables personalized content discovery based on individual preferences, enhancing user engagement and providing an efficient mechanism for users to find and enjoy particular games to their likings, out of a repertoire that may include hundreds of thousands of games which may otherwise be nearly impossible to manually browse by manual labor.

[0027] Some embodiments may provide one or more of the following technological:

[0028] (a) AI-Driven Personalization: Integration of a CLIP model, to extract semantics from game footage, enabling the curation of a personalized feed for users.

[0029] (b) Social Media-Style UX / UI of infinite or continuous scrollable feed of games, featuring an online casino experience; implementing a continuous scrolling feed of games that is tailored to user-declared preferences and / or machine-predicted / machine-estimated / machine-deduced user-preferences; enabling the user to seamlessly explore and rapidly discover, instantly preview and rapidly access a diverse range of games through an engaging and efficient interface. Such integration is facilitated by a server computer that is configured to capture or record or generate online game content, and configured to present it to users in the form of streaming videos (e.g., real-time streaming videos of real-time actual gameplay, and / or instant streaming of previously-recorded video clips of such gameplay). The server may utilizeAl / ML / LL / NN units, including one-shot Convolutional neural network (CNN) detectors, Large-Language Model (LLM) engines, Contrastive Language-Image Pre-training (CLIP), ensuring automatic and efficient capture of gaming content and / or selective tailoring of the gaming feed to each user. This convergence results in a unique and immersive user experience that combines the best elements of both social media and online casino environments.

[0030] (c) Some embodiments provide Live Streaming and User Interaction, or enable live streaming of gaming sessions integrated with interactive user reactions. This feature allows users to follow preferred game providers, engage in live sessions, and react to such sessions within the platform. In some embodiments, User A may “follow” (or, may “subscribe to”) gameplays of User B, such that User A may be alerted or notified once User B engages in a new game, and such that the actual gameplay of User B is streamed in real time to User A.

[0031] (d) Some embodiments provide rapid and seamless Game Loading; introducing an intuitive swipe / scroll mechanism for effortless browsing of video previews, and providing a smooth and instant transition between game video previews and actual on-demand gameplay engagement. For example, games or game program codes are loaded in a streaming form from the server, and begin to play instantly or almost immediately on user devices, requiring virtually no pre-downloading / pre-installation on the user's device; thereby ensuring a seamless and instant gaming experience for users.

[0032] These features collectively redefine the online gaming landscape, integrating advanced technologies and social elements to offer a personalized, engaging, efficient, instantly responsive, and interactive gaming platform for users.

[0033] Some embodiments may utilize a Backend Infrastructure that utilizes WebRTC and a customized version of the Chromium engine for backend operations; and may implements VP8 / H.264 / H.265 video codecs for efficient video compression and for rapid video transmission, ensuring smooth and virtually instant content delivery. With regard to Client- Side Functionality, some embodiments utilize a tailored client interface and module, for example, implemented as a native mobile application or “mobile app”; or as a web-friendly web-site or web-page or web-based application (e.g., using JavaScript / HTML5 / CSS), or as a browser add-on / extension / plug-in, or as a stand-alone application, or even as a stand-alone browser); allowing efficient and rapid user interactions such as game selection, game preview, game engagement, bet placement / adjustments, spin initiation, and other gaming operations; wherein the end-user device transmits user actions to the backend streaming service for processing. The Backend Service Unit orchestrates the reception, storage, and transmission of the game data; the enforcement of game rules and constraints; the generation of feedback / gaming results in response to user actions and / or in response to actions of other participating players and / or in response to random or pseudo-random parameter values; as well as continuous delivery of video content in a scrollable format for the user’s gaming feed; optionally utilizing Al / ML / DL / NN based suggestions that are extracted from game preview semantics and / or that are based on previous user activity or previous user engagement, using a CLIP (or other suitable) Al / ML / DL / NN model.

[0034] Reference is made to Fig. 1 , which is a schematic block-diagram illustration of an interactive gaming system 100, in accordance with some demonstrative embodiments. For example, end-user device 101 (e.g., a smartphone) is operated by User A, and runs a gaming application 111 (e.g., implemented as a native “mobile app”); and similarly, end-user device 102 (e.g., a tablet) is operated by User B, and runs the same gaming application 112 (which may be implemented as a native “mobile app”, or may be implemented otherwise, such as a mobile -friendly website or webpage, or as a browser, or as a browser extension / plug-in / addon).

[0035] Each of devices 101-102 communicates with a Gaming Server 120, over one or more wired and / or wireless communication networks and / or over one or more wired or wireless communication links; for example, over Wi-Fi, over Bluetooth, over a Cellular connection, over a chain of communications links, or the like.

[0036] Gaming Server 120 includes, or is operable associated with, a Games Repository 121; which stores, for each game, both (i) game program code 122 and (ii) modifiable / interchangeable game elements 123. For example, a game of Slot Machine may include a fixed program code; and also, multiple sets of graphic elements (e.g., Animal graphics; Fruit graphics) and / or audio elements (e.g., pop music themed sound effects and soundtrack; rock- n-roll themed sound effects and soundtrack) and / or textual elements (e.g., game title of “Fantasy Slot Machine”, game title of “Animals Slot Machine”) and / or GUI elements (e.g., square buttons to Hold a slot machine reel; circular buttons to Hold a slot machine reel). The game program code may use placeholders or pointers, that enable dynamic automatic and efficient modification of the gameplay based on a set of elements selected from a variety of options.

[0037] For example, a Game Constructor Unit 124 operates to take a particular game program code, and to add to it a particular set of graphics / sounds / texts / GUI elements from a particular set of options, to thus generate a particular variant of that game. The selection may be performed randomly or pseudo-randomly, and / or by taking into account previous userengagement patterns that were exhibited by each user.

[0038] For example, a User Engagement Tracking & Analysis Unit 125 operates to monitor and log and track all the interactions by each user, and their timing and their time-length; for example, observing and tracking that User A has scrolled past Game 1 without watching its video preview; or that User A has stopped scrolling at Game 4 and watched its video preview for 5 seconds and then proceeded to scroll away; or that User B stopped scrolling at Game 8 and watched its video preview for 20 seconds and then proceeded to join that game and played it for 6 minutes; or the like. User Engagement Tracking & Analysis Unit 125 may analyze those engagement data-points, optionally by using an Al / ML / DL / NN Engine 126, to deduce or determine or estimate or predict user-specific patterns that are then used for user-specific tailoring of the game feed by a User-Specific Scrollable Game Feed Generator 127.

[0039] For example, the User-Specific Scrollable Game Feed Generator 127 may select particular games from the Games Repository 121 based on the deduced user preferences ; and / or may skip, or may avoid selecting, one or more games based on such deduced user preferences. Furthermore, in some embodiments, as an optional feature, the User-Specific Scrollable Game Feed Generator 127 may optionally activate an On-The-Fly Game Modification Unit 128 (which is an optional component, in some embodiments), to optionally tailor a particular game to the deduced (or declared) preferences of a particular user; such as, by deducing that User B prefers (based on his past or recent user engagement data) Slot Machine games (and not Blackjack games) featuring Animals graphics (and not Food graphics); and therefore, specifically populating the game feed for User B with Slot Machine game(s) (and not Blackjack games) that feature Animals graphics (and not Food graphics). It is noted that the on-the-fly modification / customization / tailoring of a specific game to a specific user, by modifying or replacing graphics / sound effects / theme / textual elements / title / other components, is an optional feature that need not necessarily be used or may not necessarily be included; for example, other embodiments may use a closed list or a closed dataset of hundreds or thousands of fixed, non-modifiable, games that are pre -provided to the system with their fixed program code and with their fixed / non-replaceable set of graphics / animations / sound effects / soundtrack / textual element / title.

[0040] In some embodiments, the Al / ML / DL / NN Engine 126 may deduce and utilize insights generated from a group of users, in order to tailor the game feed of a particular user. For example, the Al / ML / DL / NN Engine 126 may deduce that within a group of 500 users, there were 300 users who preferred to engage with Slot Machine games with Animals graphics; and that 280 of those 300 users has also preferred to subsequently engage with Slot Machine games with Fantasy Characters graphics; and therefore, the Al / ML / DL / NN Engine 126may provide an insight or a command to the User- Specific Scrollable Game Feed Generator 127 and / or to the On-The-Fly Game Modification Unit 128, that once a new user is estimated to also like Slot Machine games with Animals graphics, then the game feed for such new user would also include at least one game of Slot Machine with Fantasy Characters as this is an estimated preference of such type of users.

[0041] The user-tailored feed of games is served to each end-user device separately; for example, by serving to each end-user device a Title of each game, followed by a static photo or image of the game, and / or followed by an instantly-playing video clip of that game (e.g., real-time gameplay, or pre-recorded gameplay); and / or optionally followed by, or having, a button indicating “Learn More” to obtain further details, and a button of “Play Now” or “Join This Game Now” to enable the user to engage with this game; followed by a similar set of elements for the next game of that user-specific game-feed, and so forth. Optionally, a Streaming Engine 129 may be responsible for the streaming operations; and a Game Previews Video Repository 130 may store pre-recorded video clips of game previous; and a Game Stills Repository 131 may store still images / photos of each game for rapid access and rapid serving of such images / photos into the gaming feed of each user.

[0042] Optionally, a Predictive Pre-Loading / Caching Unit 132 may operate to pre-load or to pre-serve or to cache one or more video previous / images / photos / text elements, while the user is browsing or scrolling or swiping his game feed. For example, while User A is looking at the preview image / video of Game 6 in his game feed, the Predictive Pre-Loading / Caching Unit 132 may already serve or send to the end-user device of User A the video preview and / or the images and / or the text elements that would enable him to rapidly view Game 7 and even Game 8 on his game feed.

[0043] Gaming Server 120 further includes a User-Actions Input / Output Controller 133, which is responsible for receiving and processing inputs that the end-user provides via gestures on his device (e.g., tap operations, touch-screen operations or gestures, on-screen selection or tapping or dragging operations, keyboard / keypad operations performed via a physical keyboard / keypad or via an on-screen keyboard / keypad); and to transfer such incoming commands or incoming gestures or user-input to a Game Control Engine 134 that runs the actual game code and that responds to the user-input based on the game program code. The User- Actions Input / Output Controller 133 then sends back to the end-user device, directly or via the Streaming Engine 129, data indicating the results of the user’s actions; for example, graphics or animation elements that are responsive to the user’s action, sound effects, gaming results, or the like.

[0044] In some embodiments, an entirety of the gameplay processing of each game is performed exclusively on Gaming Server 120; and is not performed on end-user devices 101 or 102, which have a limited set of roles of (a) showing the game feed to the user, (b) capturing user-gestures and sending them to the Gaming Server 120 for processing there, and (c) receiving back from the Gaming Server 120 the processing results, particularly in the format of video content that is instantly streamed and displayed on the end-user device. This architecture enables rapid and instantaneous gameplay and engagement, without the need for an end-user to download / install / launch a program code or a separate “mobile app” for each game that he selects in his interactive game feed. In other embodiments, optionally, at least some of the processing operations may be performed locally on the end-user device; for example, by serving from the Gaming Server 120 to the end-user device a relevant code portion (e.g., in JavaScript / HTML / CSS) that is then run locally on the end-user device and that causes a “spinning reel” animation (of a slot machine) or a “spinning roulette” animation or a “cards being dealt” animation to be displayed for three seconds on the end-user device as part of the actual gameplay.

[0045] Gaming Server 120 may further include, or may be operable associated with, a Broadcasting / Observing Engine 135; which may take an actual real-time gameplay of a particular game that is being played by User A (and / or by User B); and by broadcasting or streaming that gameplay to User C and User D who can view and observe the gameplay on their devices; and by enabling User C and User D to send their reactions / comments to the observed gameplay, which in turn would be broadcast or shown to other users who observe the same game and / or to the player(s) that are currently engaging with the game. Optionally, a Chat & Reaction Module 136 may be responsible for providing to each such user an on-screen interface to convey its reaction out of several depicted representations (e.g., emoji graphics, icons, thumb-up / thumb-down, textual labels such as “Well Done!”, or the like) and / or by providing a chat panel in which an observer / a user may type free-style text that other / s can see on their devices.

[0046] In some embodiments, a Follow & Subscribe Module 137 may be responsible for: receiving an indication from User B, that he would like to subscribe to games that are published / created by Publisher X, and creating a programmed indication to populate the game -feed of User B (partially, or mostly, or exclusively) with games published by Publisher X, and notifying User B whenever Publisher X releases a new game.

[0047] Additionally or alternatively, in some embodiments, the Follow & Subscribe Module 137 may be responsible for: receiving an indication from User B, that he would like tosubscribe to games that are played by User A, and creating a programmed indication to populate the game-feed of User B (partially, or mostly, or exclusively) with games that were played in the past and / or that are played now by User A, and / or notifying User B whenever User A engages with a new game.

[0048] Additionally or alternatively, in some embodiments, the Follow & Subscribe Module 137 may be responsible for: receiving an indication from User B, that he prefers or likes games of a particular type (e.g., Slot Machines) and / or sub-type (e.g., Slot Machine games with Animal graphics); and creating a programmed indication to populate the game-feed of User B (partially, or mostly, or exclusively) with games that belong to this type or sub-type; and / or notifying User B whenever a new game of that type or sub-type is released or published and / or is being played by other user / s.

[0049] In some embodiments, optionally, a Real / Virtual Money Clearing Unit 138 may be responsible for: (a) collecting real-world funds from users (e.g., via credit card, debit card, electronic funds transfer, wire transfer); (b) paying out real-world funds to users; (c) collecting or receiving or sending or paying out or converting crypto-currency coins or tokens; (d) providing / receiving / sending virtual assets or virtual representations of “funds” of users or of “rewards” of users, such as, virtual credit points or virtual credit balance, virtual “stars”, or the like; providing or receiving other types of rewards or awards, virtual “medals”, or coupons or vouchers that can be redeemed or exchanged for real-life goods and services and / or for online goods and services and / or for virtual goods, or the like.

[0050] In some embodiments, optionally, a Leaderboard Module 139 may be responsible for maintaining, updating, and presenting a Leaderboard or an achievements board or scoreboard for each game, or for each game-type or game sub-type; keeping track of the top players in each such game or category, and displaying a list of the highest achievers and their respective achievers, to those players and / or to other observers.

[0051] Reference is made to Figs. 2A to 2D, which are illustrations of screen-shots (201 to 204, respectively) from an end-user device (e.g., smartphone or tablet) that is part of an interactive gaming system, in accordance with some demonstrative embodiments. They demonstrate the smooth and continuous scrolling of games or game -preview elements (e.g., screenshot / thumbnail image; video preview; animated preview).

[0052] For example, Fig. 2A shows a Roulette online game filling up most of the screen as a first or initial game in a user-specific games feed. Also shown, for demonstrative purposes, is a partial view of the top-portion of the next game in the feed, which s a Slot Machine game having a classic casino theme.

[0053] As the user swipes upwardly with his finger, or scrolls the screen-view downwardly, the first game (Roulette) is gradually disappearing upwardly; Fig. 2B shows only the bottompart of that Roulette online game, followed below it by a full view of the Slot Machine game that follows it in the feed.

[0054] A further swiping upwardly of the game feed brings up the arrangement demonstrated in Fig. 2C: the slot machine game is now occupying the top part of the screen, and the next game - an online Checkers game - is starting to partially appear at the bottom part of the screen.

[0055] A further swiping upwardly of the game feed brings up the arrangement demonstrated in Fig. 2D: the Checkers game is now occupying the top part of the screen, and the next game - an animal-themed Slot Machine game - is starting to partially appear at the bottom part of the screen.

[0056] As demonstrated, the screen of the end-user device is responsive to the swiping or scrolling motions or gestures of the user, and shows portions of game preview elements as a user-specific / user-tailored continuous scrollable feed.

[0057] Fig. 2C and Fig. 2D further demonstrate that a slot machine game can be dynamically modified by the gaming server (e.g., as an optional feature of some embodiments), and / or that game program code can be dynamically re-used (e.g., as an optional feature in some embodiments), and / or that game layout and content may be adjusted or configured or tailored or modified to meet machine-deduced player preferences (e.g., as an optional features in some embodiments); such that, for example, the slot machine game of Fig. 2C features a first layout of elements and a first set of graphics corresponding to a Classic Casino Theme, whereas the slot machine game of Fig. 2D features a second, different, layout of elements and a second, different, set of graphics corresponding to an Animals Theme.

[0058] Reference is made to Figs. 3A and 3B, which are schematic illustrations of differential user-specific game feeds (301 and 302, respectively) that are generated and / or displayed at the same time or in parallel or concurrently to two different users that operate two different end-user devices, in accordance with some demonstrative embodiments.

[0059] Fig. 3A demonstrates that the games feed 301 that is generated for User Adam, and is gradually displayed on his end-user device, begins with a Slot Machine game having a Christmas theme; then has a Roulette game having a fantasy theme; and then has a Jigsaw Puzzle having an animal theme. The system monitors the user interactions and engagement of User Adam, and detects that User Adam spent increased time with the Slot Machine game having a Christmas theme. Therefore, the system deduces or estimates or predicts, that UserAdam prefers Slot Machine games, and / or that User Adam prefers Christmas themed games. Therefore, the system decides to dynamically populate accordingly the subsequent items in the game feed for User Adam: the next game in Adam’s user-tailored feed is a Christmas themed game (of Blackjack), based on the system’s estimate that Christmas themed games are preferred by User Adam; and the next game in Adam’s user-tailored feed is a Slot Machine game (having an animal theme), based on the system’s estimate that Slot Machine games are preferred by User Adam. The system continues to monitor the interactions and engagement of User Adam with his particular games feed, and with particular games that he selects to preview and / or to play, and the system continues to dynamically modify the feed for User Adam according to system-deduced insights regarding his preferences. In some embodiments, the feed can be updated or modified or constructed dynamically while Adam is scrolling / swiping / playing / interacting; whereas, in other embodiments, the feed is updated only between usage sessions, such that (for example) the games feed that will be displayed to User Adam on Tuesday is constructed by the system on Monday night based on a cumulative analysis of Adam’s interactions with the feed and with games during the entirety of that Monday.

[0060] Fig. 3B demonstrates that the games feed 302 that is generated for User Janet, and is gradually displayed on her end-user device, begins with Blackjack game having an animal theme; and then has a Roulette game having a Fantasy theme; and then has a Checkers game having a Science Fiction theme. The system monitors the user interactions and engagement of User Janet, and detects that User Janet spent increased time with the Roulette game having a Fantasy theme. Therefore, the system deduces or estimates or predicts, that User Janet prefers Roulette games, and / or that User Janet prefers Fantasy themed games. Therefore, the system decides to dynamically populate accordingly the subsequent items in the game feed for User Janet: the next game in Janet’s user-tailored feed is a Fantasy themed game (of Poker), based on the system’s estimate that Fantasy themed games are preferred by User Janet; and the next game in Janet’s user-tailored feed is Roulette game (having a flowers theme), based on the system’s estimate that Roulette games are preferred by User Janet. The system continues to monitor the interactions and engagement of User Janet with her particular games feed, and with particular games that she selects to preview and / or to play, and the system continues to dynamically modify the feed for User Janet according to system-deduced insights regarding her preferences. In some embodiments, the feed can be updated or modified or constructed dynamically while Janet is scrolling / swiping / playing / interacting; whereas, in other embodiments, the feed is updated only between usage sessions, such that (for example) the games feed that will be displayed to User Janet on Tuesday is constructed by the system onMonday night based on a cumulative analysis of Janet’s interactions with the feed and with games during the entirety of that Monday.

[0061] As demonstrated, different games need not have the same screen-size dimensions or resolutions; and the feed can dynamically accommodate games having different sizes or dimensions. Some embodiments provide a computerized method, comprising: (a) storing on a server data representing a plurality of online games; wherein each online game is associated with (i) a code-portion that runs the online game, and (ii) a screenshot or a thumbnail video that visually demonstrate the online game; and then, (b) causing a first end-user device of a first end-user to display a first scrollable feed of online games, wherein the first scrollable feed of online games is tailored to the first end-user based on analysis of past interactions of said first end-user, and in parallel or concurrently or in a time-slot that is at least partially overlapping, causing a second end-user device of a second end-user to display a second, different, scrollable feed of online games, wherein the second scrollable feed of online games is tailored to the second end-user based on analysis of past interactions of said second end-user.

[0062] In some embodiments, the tailoring of the scrollable feed is further based on explicit preferences selected by the end-user, including preferred game genres, themes, or publishers.

[0063] In some embodiments, the tailoring of the scrollable feed is enhanced by tracking the time spent by the end-user on previewing or playing individual games.

[0064] In some embodiments, the server is configured to analyze end-user interactions using machine learning models to predict game preferences.

[0065] In some embodiments, the thumbnail video associated with each game is streamed in real time from the server as the end-user scrolls through the feed.

[0066] The method of claim 1 , wherein the scrollable feed includes an option for the enduser to save games to a personal favorites list for later access.

[0067] In some embodiments, the scrollable feed includes recommendations for games that are currently popular among a group of users with similar preferences.

[0068] In some embodiments, the end-user can filter the scrollable feed by game type, theme, or level of complexity.

[0069] In some embodiments, the server adjusts the feed content based on the time of day or duration of end-user activity sessions.

[0070] In some embodiments, the server provides real-time notifications within the feed when a new game matching the end-user's preferences becomes available.

[0071] In some embodiments, the scrollable feed incorporates social features, including the ability to view and join games currently being played by the end-user's friends.

[0072] In some embodiments, each game in the scrollable feed includes a visible rating or review score based on aggregated user feedback.

[0073] In some embodiments, the server provides an interactive tutorial for each game available in the scrollable feed.

[0074] In some embodiments, the end-user can interact with a game preview by performing gestures or selecting buttons to simulate gameplay without fully launching the game.

[0075] In some embodiments, the scrollable feed dynamically integrates promotional content, including limited-time events or bonus features for specific games.

[0076] In some embodiments, the server provides an option to toggle between different feeds tailored for single-player games and multiplayer games.

[0077] In some embodiments, the tailoring of the feed includes prioritizing games that the end-user has previously marked as enjoyable or replayed multiple times.

[0078] In some embodiments, the scrollable feed includes games categorized into themed collections, such as seasonal games or games featuring specific visual styles.

[0079] In some embodiments, the server ensures that games displayed in the feed do not repeat within a single browsing session unless explicitly selected by the end-user.

[0080] In some embodiments, the tailoring of the scrollable feed takes into account the device specifications of the end-user device, optimizing games for the screen size or processing power.

[0081] In some embodiments, the server supports interactive user feedback mechanisms, allowing the end-user to rate or report games directly within the scrollable feed.

[0082] In some embodiments, the scrollable feed includes a feature that allows the enduser to preview gameplay statistics or leaderboards associated with each game.

[0083] In some embodiments, the tailoring of the feed is enhanced through integration with third-party user profiles or social media accounts to identify user interests.

[0084] In some embodiments, the scrollable feed incorporates voice command functionality, enabling the end-user to search for games or navigate the feed using spoken queries.

[0085] In some embodiments, the server dynamically adjusts the scrollable feed to reduce the bandwidth usage based on the network conditions of the end-user device.

[0086] In some embodiments, the tailoring of the feed includes excluding games that the end-user has explicitly indicated as uninteresting or undesirable.

[0087] In some embodiments, the scrollable feed includes a “Try Now” button for each game, enabling a brief demo or trial session before fully engaging with the game.

[0088] In some embodiments, the server monitors engagement patterns across multiple end-user devices associated with the same user account to optimize the tailoring of the feed.

[0089] In some embodiments, the scrollable feed includes games featuring different difficulty levels, with visual indicators for easy, medium, or hard modes.

[0090] In some embodiments, the server integrates with parental controls to filter out games based on age restrictions or content ratings.

[0091] In some embodiments, the server enables the end-user to share specific games or game previews from the feed directly with other users via messaging or social media platforms.

[0092] In some embodiments, the tailoring of the feed incorporates preferences for specific gameplay durations, displaying shorter or longer games based on end-user habits.

[0093] In some embodiments, the scrollable feed displays real-time user counts for multiplayer games, showing how many players are currently engaged in each game.

[0094] In some embodiments, the server includes a feature for integrating promotional incentives, such as bonus points or in-game rewards, for engaging with specific games in the feed.

[0095] In some embodiments, the scrollable feed highlights games that are part of tournaments or competitive events currently available for participation.

[0096] In some embodiments, the server enables the end-user to customize the visual layout of the scrollable feed, including the size or arrangement of game previews.

[0097] In some embodiments, the scrollable feed is configured to display a virtually infinite sequence of games by dynamically loading additional game previews from the server as the end-user scrolls.

[0098] In some embodiments, the virtually infinite scrollable feed is generated by continuously retrieving and displaying game content based on the end-user's browsing behavior and preferences.

[0099] In some embodiments, the virtually infinite scrollable feed ensures seamless user experience by preloading game previews before they become visible on the end-user device.

[0100] In some embodiments, the virtually infinite scrollable feed allows the end-user to navigate backward and forward without interruptions by caching previously loaded game content.

[0101] In some embodiments, the virtually infinite scrollable feed dynamically prioritizes the order of games displayed based on real-time analytics of end-user engagement patterns.

[0102] In some embodiments, the server dynamically modifies game elements, including graphics, sound effects, or themes, in real time to align with the end-user’s deduced preferences.

[0103] In some embodiments, the server generates on-the-fly game variations by combining predefined program code with selectable modifiable components, such as visual styles, soundtracks, or user interface layouts.

[0104] In some embodiments, the on-the-fly modification of games includes adapting the difficulty level or game mechanics based on real-time analysis of the end-user's engagement history.

[0105] In some embodiments, the server modifies game previews in real time by altering graphical elements to create tailored visual experiences for the end-user.

[0106] In some embodiments, the on-the-fly modification of games includes replacing or customizing thematic elements, such as replacing fruit-themed slot machine graphics with animal-themed graphics, based on the end-user's preferences.

[0107] Some embodiments provide a computerized method, comprising: (a) storing game data and associated media, including video previews, on a server; (b) analyzing past interactions of an end-user with displayed games; (c) generating a tailored scrollable feed of games for the end-user based on said analysis; (d) streaming video previews of games from the server to the end-user device in real time; and (e) enabling seamless selection and immediate gameplay initiation from the tailored feed without local game installation.

[0108] Some embodiments provide a computerized method, comprising: (a) monitoring end-user engagement with game previews, including scrolling speed and dwell time; (b) utilizing machine learning algorithms to deduce user-specific gaming preferences; (c) dynamically updating a personalized game feed in real time based on deduced preferences; (d) preloading game previews to minimize latency during user browsing; and (e) providing realtime feedback on game interactions to further refine user preference models.

[0109] Some embodiments provide a computerized method, comprising: (a) presenting a virtually infinite scrollable feed of games on an end-user device; (b) tailoring the feed based on explicit and implicit user preferences; (c) enabling rapid game previewing through real-time video streaming; (d) providing a selectable option to immediately join gameplay without downloading; and (e) tracking user actions for adaptive feed customization.

[0110] A computerized method, comprising: (a) storing a plurality of games, each with modifiable elements, on a gaming server; (b) dynamically modifying game previews using predefined parameters, such as themes and graphics; (c) streaming tailored previews based onuser preferences to the end-user device; (d) providing an interactive swipe mechanism for efficient feed navigation; and (e) initiating selected games directly from the feed without requiring additional user installations.

[0111] Some embodiments provide a computerized method, comprising: (a) generating a personalized, scrollable feed of games based on user interactions and preferences; (b) integrating live gameplay previews into the feed for rapid discovery; (c) dynamically adjusting the game order based on real-time engagement patterns; (d) enabling game selection and instantaneous server-based gameplay; and (e) incorporating user feedback to further refine the presented content.

[0112] Some embodiments provide a computerized method, comprising: (a) analyzing user interaction data, including preferences for specific game types or themes; (b) organizing a tailored feed of interactive games accessible on a touch-enabled device; (c) dynamically streaming video clips associated with each game as the user scrolls; (d) allowing instant transitions from preview to active gameplay; and (e) modifying the feed continuously using Al-driven predictions.

[0113] Some embodiments provide a computerized method, comprising: (a) creating a scrollable feed of games dynamically populated with server-stored content; (b) tailoring feed content for each user using machine-learning-derived insights; (c) providing real-time previews, including live or recorded gameplay footage; (d) enabling interactive game selection via user gestures or clicks; and (e) continuously optimizing the feed using historical and session-specific engagement data.

[0114] Some embodiments provide a computerized method, comprising: (a) receiving a request from an end-user device to display a scrollable feed of games; (b) determining user preferences based on declared and deduced engagement patterns; (c) presenting video previews of games sequentially within the feed; (d) facilitating immediate gameplay initiation directly from the feed; and (e) analyzing user interactions to refine subsequent feed presentations.

[0115] Some embodiments provide a computerized method, comprising: (a) storing multiple games with customizable components on a central server; (b) generating game previews tailored dynamically to user preferences; (c) streaming previews continuously as users scroll through the feed; (d) allowing immediate user interaction and gameplay activation; and (e) adapting feed content and sequence based on user interaction metrics.

[0116] Some embodiments provide a computerized method, comprising: (a) identifying user preferences from historical engagement and explicit settings; (b) generating a personalized scrollable game feed displayed on an end-user device; (c) providing dynamic previews forinstant visual representation of each game; (d) enabling seamless gameplay activation without downloading game-specific apps; and (e) monitoring user activity to improve future feed personalization.

[0117] In some embodiments, the tailoring of the first and second scrollable feeds is based on different sets of user interaction data specific to the first and second end-users.

[0118] In some embodiments, the server applies distinct machine-learning algorithms for the first and second end-users to generate personalized feed recommendations.

[0119] In some embodiments, the first scrollable feed is populated with games of a particular genre preferred by the first end-user, while the second scrollable feed contains games of a different genre preferred by the second end-user.

[0120] In some embodiments, the tailoring of the first and second feeds considers distinct time-of-use patterns for the first and second end-users.

[0121] In some embodiments, the first end-user's feed emphasizes single-player games, while the second end-user's feed highlights multiplayer games based on respective preferences.

[0122] In some embodiments, the first and second feeds are dynamically updated in real time based on ongoing interactions of the first and second end-users.

[0123] In some embodiments, the server excludes games previously skipped by the first end-user from the first feed, while retaining them in the second feed for the second end-user.

[0124] In some embodiments, the first scrollable feed includes promotional content targeted at the first end-user, while the second scrollable feed presents different promotions tailored to the second end-user.

[0125] In some embodiments, the tailoring of the feeds is enhanced by analyzing the respective device specifications of the first and second end-users to optimize game compatibility.

[0126] In some embodiments, the server adjusts the visual layout or design of the first feed differently from the second feed, based on the end-users’ individual aesthetic preferences.

[0127] In some embodiments, the first end-user’s feed is filtered to show games from a specific publisher, while the second end-user’s feed includes games from a different publisher.

[0128] In some embodiments, the first feed incorporates games that are frequently played by the first end-user's social circle, while the second feed includes games preferred by the second end-user’s social circle; such as, by deducing the social circle of the user from his social network account and from users who are “friends” of each such user on a social network.

[0129] In some embodiments, the first scrollable feed contains newly released games prioritized for the first end-user, while the second scrollable feed focuses on classic games prioritized for the second end-user.

[0130] In some embodiments, the server provides the first and second end-users with respective yet different options to customize their feeds further, creating distinctly tailored experiences.

[0131] In some embodiments, the first end-user’ s feed emphasizes games with shorter play durations, while the second end-user’s feed includes games with longer gameplay sessions, based on respective user behaviors.

[0132] In some embodiments, the server tracks the time spent by each end-user on each game preview, using this data to prioritize similar games in the respective tailored feed.

[0133] In some embodiments, the server monitors scrolling speed and pauses during feed navigation to identify games that capture each end-user’s attention, adjusting the feed accordingly.

[0134] In some embodiments, the server analyzes the frequency of game selections by each end-user to emphasize frequently selected game types in the respective tailored feed.

[0135] In some embodiments, the server logs the number of interactions, including taps and clicks, for each game preview to refine user-specific preferences for future feed construction.

[0136] In some embodiments, the server tracks and analyzes the sequence of games skipped by each end-user, using this data to deprioritize skipped game types in the tailored feed.

[0137] In some embodiments, the server monitors engagement duration with gameplay sessions after game selection, prioritizing similar games for the respective end-user’s feed.

[0138] In some embodiments, the server uses Al / ML / DL algorithms to detect patterns in user interactions, such as genre preference, and dynamically updates the tailored feed for each end-user.

[0139] In some embodiments, the server captures and processes gestures, including swiping directions and speeds, to infer and construct end-user preferences for feed content.

[0140] In some embodiments, the server tracks the end-user’s reaction to promotional content in the feed, adjusting the feed content to increase relevance based on interaction data.

[0141] In some embodiments, the server monitors and records the time intervals between game selections to deduce user-specific decision-making patterns for feed optimization.

[0142] In some embodiments, the server evaluates user interactions, including hovering over specific game previews, to identify tentative interests and reflect them in the tailored feed.

[0143] In some embodiments, the server analyzes user engagement with thematic elements, such as graphical styles or soundtrack preferences, to construct a more personalized feed for each end-user.

[0144] In some embodiments, the server tracks the order of game previews clicked by each end-user and uses this data to present preferred game categories earlier in the feed.

[0145] In some embodiments, the server evaluates the repetition of end-user interactions with similar games to prioritize or exclude games based on recurring behaviors.

[0146] In some embodiments, the server logs engagement with multiplayer or single-player modes to prioritize respective game types in the tailored feed for each end-user.

[0147] In some embodiments, the server integrates user browsing history, including skipped games, into a predictive model that dynamically modifies feed content for future sessions.

[0148] In some embodiments, the server uses real-time analytics to assess end-user preferences during active feed navigation, immediately adjusting feed content to align with observed behaviors.

[0149] In some embodiments, the server tracks and analyzes geographic location and time zone of each end-user, tailoring the feed to optimize game availability and relevance.

[0150] In some embodiments, the server evaluates user interactions during specific time periods to identify habitual gaming patterns, constructing a tailored feed that reflects these behavioral trends.

[0151] In some embodiments, the server utilizes supervised machine learning models to analyze historical user interactions and predict user preferences for constructing tailored game feeds.

[0152] In some embodiments, the server implements reinforcement learning algorithms to continuously refine feed content based on feedback from user interactions, such as game selections or skips.

[0153] In some embodiments, the server applies neural networks to analyze patterns in user engagement data, including viewing time and click-through rates, to optimize feed personalization.

[0154] In some embodiments, the server leverages unsupervised machine learning models to cluster similar user behaviors and tailor feeds accordingly for end-users exhibiting comparable interaction patterns.

[0155] In some embodiments, the server utilizes natural language processing (NLP) models to interpret user-provided feedback or preferences, incorporating them into the construction of personalized feeds.

[0156] In some embodiments, the server employs deep learning models to identify complex user behavior patterns and predict preferences for specific game genres or styles.

[0157] In some embodiments, the server integrates decision-tree-based machine learning models to evaluate user interactions and prioritize game categories in the tailored feed.

[0158] In some embodiments, the server uses convolutional neural networks (CNNs) to analyze visual elements of games preferred by users and tailor the feed to highlight similar games.

[0159] In some embodiments, the server implements ensemble learning methods to combine predictions from multiple machine learning models, ensuring enhanced accuracy in feed personalization.

[0160] In some embodiments, the server uses predictive analytics powered by Al to forecast future user preferences based on aggregated historical interaction data.

[0161] In some embodiments, the server employs a Contrastive Language-Image Pretraining (CLIP) model to extract semantic features from game previews and align them with user preferences.

[0162] In some embodiments, the server utilizes recurrent neural networks (RNNs) or long short-term memory (LSTM) models to analyze sequential user interactions and predict trends in gaming preferences.

[0163] In some embodiments, the server applies probabilistic ML / DL models to assign confidence scores to predicted user preferences, dynamically adjusting feed content based on accuracy thresholds.

[0164] In some embodiments, the server uses transfer learning techniques to enhance feed personalization by applying insights from general gaming datasets to specific end-user behaviors.

[0165] In some embodiments, the server employs Al-based anomaly detection algorithms to identify and respond to unusual interaction patterns, such as sudden changes in user preferences.

[0166] In some embodiments, the server integrates federated learning models to analyze interaction data locally on user devices, preserving privacy while tailoring feeds effectively.

[0167] In some embodiments, the server combines collaborative filtering with machine learning techniques to recommend games based on preferences of similar users within a shared dataset.

[0168] In some embodiments, the server utilizes hybrid Al models combining contentbased filtering and ML / DL models to recommend games in a user-tailored manner by matching user preferences with game metadata.

[0169] In some embodiments, a system can be implemented by using some, or most, or all, of the following units; which may be implemented using hardware components and / or software components:

[0170] End-User Device, which can be a smartphone, tablet, laptop computer, desktop computer, gaming device, gaming console, smart television, Augmented Reality (AR) or Virtual Reality (VR) or Mixed Reality (MR or XR) device or gear or helmet or glasses or wearable device, a portable electronic device, a mobile electronic device, a handheld electronic device, or other device operated by the end-user, capable of displaying the scrollable game feed and receiving user inputs via touch, touch-screen, clicks, taps, gestures, and / or other input methods (e.g., voice commands).

[0171] Gaming Server hosting game data, user interaction logs, and machine learning algorithms responsible for tailoring and delivering user-specific feeds.

[0172] Games Repository, which is a database storing game program codes, media elements, and associated metadata for generating game feeds and previews.

[0173] Feed Generator Unit, which is a component (e.g., on the server) responsible for assembling and / or populating user-specific scrollable feeds based on real-time interaction data and user preferences.

[0174] Interaction Tracking Unit, which can be a client-side module and / or a server-side module, that logs user activities, such as scrolling, tapping, swiping, clicking, engagement times, engagement start, engagement end, engagement frequency, zoom in / zoom out gestures, touch-screen gestures, and / or other user interactions for further analysis.

[0175] Machine Learning / Deep Learning Engine, such as an Al-based component that analyzes user interaction data using supervised, unsupervised, or deep learning techniques to deduce user preferences.

[0176] User Preference Database, or storage unit or dataset that stores explicit user preferences, such as preferred genre selections, preferred game publishers, preferred themes.

[0177] Streaming Engine, such as a server component enabling real-time delivery of video previews and live gameplay streams to user devices, and / or delivering the code -portions that are required to run on the client-side end-user device in order to run a particular game.

[0178] Game Previews Repository, such as a collection or database or dataset of prerecorded video clips and thumbnails used for presenting game previews in the scrollable feed.

[0179] Dynamic GUI Module, which can dynamically adjust the visual layout of the feed based on device specifications and / or user preferences; such as, by re-sizing a game to it a particular layout of device specification or screen resolution, or by selecting a particular gameversion (out of several pre-defined / hard-coded game versions) that matches a particular device specification or screen resolution or user preference.

[0180] Gesture Recognition Unit, that can interpret user gestures, such as swipes or pinches, to control feed navigation or game selection.

[0181] Recommendation Engine, such as an Al-driven module that predicts additional games of interest based on user interaction patterns and preferences.

[0182] Real-Time Feedback Sub-system, that adapts feed content dynamically during a user session based on real-time interaction data; for example, dynamically replacing Game 17 that was planned to be displayed to User Adam in his feed if he scrolls five more games ahead, with another game based on User Adam’s interactions so far with Games 1 to 12 in his userspecific feed. For example, the original planned feed was to show Blackjack games as Games 4 and 9 and 17; but the system detects that the user swipes rapidly (e.g., twice as rapidly relative to other places) over Game 4 and Game 9, and therefore, the system dynamically determines or estimates that this user is not interested in Blackjack games, and dynamically replaces Game 17 from being the planned Blackjack game to being a Slot Machine game. In a contrarian example, the system may detect that the user lingered on Games 5 and 8 and 12, which are all Holiday Themed games; and therefore, the system may dynamically adjust Games 17 and onward, or only Games 17 and 20 and 26, to also be Holiday Themed games for this user.

[0183] Social Interaction Unit, enabling users to share game previews, live gameplay sessions, and reactions with other users or social networks.

[0184] Predictive Caching Unit, that can pre-load game previews or other feed elements to minimize or reduce latency during browsing.

[0185] Game Selection Interface, such as an on-screen interface allowing users to select games directly from the scrollable feed for instant engagement.

[0186] Game Control Engine, such as a backend system that processes user inputs during gameplay and sends responses to the user device.

[0187] On-The-Fly Modification Module, which is an optional unit that customizes or modifies game graphics, sounds, or themes in real-time based on user preferences and / or based on recent user interactions and / or historic user interactions.

[0188] User Engagement Analysis Unit, configured to perform monitoring and analyzing user interactions, such as skipped games or engagement duration, to refine feed construction.

[0189] Leaderboard Management Unit, configured for tracking and displaying rankings or achievements within individual games or game categories for competitive users.

[0190] Notification Unit, configured for sending real-time alerts about new game releases, promotions, or activity from followed users or publishers.

[0191] Live Streaming Unit, enabling users to broadcast their gameplay sessions or view live streams from other users.

[0192] Follow & Subscribe Unit, enabling users to follow specific games, publishers, or other players, influencing the content of their game feeds.

[0193] Backend Infrastructure, or other server architecture supporting game processing, feed generation, and interaction tracking, ensuring system reliability and scalability.

[0194] Video Compression / Encoding Engine, that utilizes codecs such as H.264 / H.265 / HEVC / AVC / VP8 / VP9 to encode and / or optimize video transmission for game previews and / or live streams.

[0195] User Profile Management Unit, managing user accounts, preferences, and interaction history, enabling personalized experiences across multiple devices.

[0196] Al-Powered Content Filtering Unit, configured for filtering inappropriate or unwanted games from the feed based on user-defined or machine-deduced preferences.

[0197] Multiplayer Interaction Unit, enabling users to join or interact with multiplayer games directly from the scrollable feed.

[0198] Virtual Currency Handling Unit, configured for managing virtual assets, including game credits, rewards, or in-app purchases, for seamless transactions.

[0199] Analytics Dashboard Unit, or other backend interface providing insights into user behavior, system performance, and engagement metrics for operators or administrators.

[0200] Some embodiments use automatic detailed construction of a User-Specific Feed of Games. The process of constructing a user-specific feed of games involves multiple layers of data acquisition, analysis, modeling, and real-time adaptation. The goal is to deliver a highly personalized, dynamic feed that enhances user engagement and satisfaction. Below, we describe this process in detail, introducing novel techniques and methodologies to achieve a technically robust and scalable solution.

[0201] Some embodiments use acquisition of Comprehensive User Data. For example, the construction of a personalized feed begins with collecting a wide array of data from the user. This data includes, for example: (a) Explicit Preferences, collected through onboarding questionnaires or settings where users manually specify preferred genres, preferred themes, preferred gameplay durations, or preferred publishers; (b) Behavioral Data, captured during user interactions, including swipe patterns, dwell times on previews, selection frequency, and gameplay duration; (c) Contextual Data, including real-time factors such as the user’s device type, time of day, geographic location, and network speed; (d) Social Connections, such as data on the user’ s in-platform connections or external social media links to analyze peer preferences and interactions; (e) Historical Data, past interaction logs, such as games played, skipped, or marked as favorites, stored across sessions. This diverse dataset forms the foundation for building a deeply personalized experience that tailors the user’s games feed to the specific user’ s characteristics.

[0202] Some embodiments may use Feature Engineering for User Interaction Data. For example, raw data collected from user interactions is processed into structured features that represent actionable insights. Some techniques may include: (a) Interaction Encoding, in which the system performs encoding of each interaction, such as swiping or selecting, into numeric representations, capturing metrics like speed, direction, and intensity; (b) Temporal Features, in which the system analyzes how interaction patterns change over time, such as increased preference for casual games during specific hours; (c) Engagement Scores, in which the system assigns weighted scores to actions (e.g., spending 30 seconds on a preview versus playing a game for 5 minutes), indicating levels of interest; (d) Preference Vectors, as the system can use vectorization techniques to create multidimensional representations of user preferences, summarizing behavioral trends. These features can be stored in a centralized database for further analysis.

[0203] Some embodiments may use Hierarchical Clustering for Content Categorization. For example, in order to tailor the games feed, games are first categorized into clusters based on their characteristics. This ensures the system can match user preferences to a curated subset of content. Some steps may include, for example: (a) Content Feature Extraction, in which the system extract metadata, such as genre, theme, difficulty, visual style, and soundtrack, for each game; and additional features, like developer reputation or peer ratings, can also be included; (b) Hierarchical Clustering, in which the system uses clustering algorithms (e.g., hierarchical agglomerative clustering) to group games into hierarchical categories; for instance, action games might be subdivided into subgenres like platformers, shooters, or role-playing games;(c) Dynamic Category Expansion, in which the system can be configured for continuously adding new games to clusters using similarity metrics derived from semantic analysis, ensuring updated recommendations. This hierarchical structure enables efficient mapping of user preferences to relevant content categories.

[0204] Some embodiments may use a Hybrid Recommendation System, as the games feed can leverage a hybrid recommendation engine, combining multiple approaches to optimize personalization; for example: (a) Collaborative Filtering, or identifying similar users based on historical engagement data and recommending games preferred by peers; (b) Content-Based Filtering, or matching user preferences to game metadata, such as genres or themes, extracted during feature engineering; (c) Context-Aware Filtering, or incorporating contextual factors, such as device type or session duration, to adjust recommendations dynamically; (d) Knowledge Graph Integration, or building and updating a knowledge graph that connects users, games, genres, publishers, and interaction data; this graph enables reasoning over indirect relationships, such as recommending a new game by the same publisher as a frequently played title. These approaches can be weighted adaptively based on user behavior.

[0205] Some embodiments may use Real-Time Feed Construction Pipeline; such that the games feed can be constructed in real-time through a multi-stage pipeline, ensuring low latency and high responsiveness. Some operations may include, for example: (a) Query Processing, such that when a user accesses the feed, the system queries the database for the latest user interaction data and preference vectors, (b) Dynamic Prioritization, by applying priority rules to determine the order of games in the feed; for instance, games from favorite genres are prioritized higher; or, newly released games are shown early in the feed, (c) Contextual Adjustments, by adjusting the feed to align with contextual factors, such as avoiding high- bandwidth previews for users on slower connections, (d) Batch Processing, by pre-loading a batch of top-ranked games to ensure seamless scrolling, while asynchronously evaluating subsequent games.

[0206] Some embodiments may use Reinforcement Eearning (RE) for Continuous Improvement and to optimize games feed generation. Some features may include: (a) Reward Function, defining success metrics, such as increased game selections, longer gameplay durations, or higher user satisfaction ratings; (b) Action Space, allowing the model to experiment with various feed arrangements, such as varying the position of game previews or introducing new categories; (c) Exploration-Exploitation Balance, that uses RL algorithms such as Deep Q-Networks (DQN), to balance the introduction of novel content withreinforcement of proven preferences. Over time, the RL model adapts to maximize long-term user engagement.

[0207] Some embodiments may optionally use Multi-modal Al integration or MultiModalities Model integration, using a Large Multi-Modalities Model (LMM or LMMM) such as OpenAI ChatGPT 4o, to analyze diverse data sources and / or behavioral data signals and to further refine recommendations predictions and game feed construction and tailoring. The multi-modalities model may perform, for example: (a) Visual Processing, by applying computer vision techniques to analyze game preview thumbnails, extracting features like color schemes or character types that align with user preferences; (b) Audio Analysis, by using audio processing models to evaluate game soundtracks, matching users to games with preferred audio styles; (c) Textual Analysis, by leveraging natural language processing (NLP) to analyze game descriptions, peer reviews, or user-provided feedback for additional insights. This multimodal approach ensures holistic personalization to the specific user based on Al-deduced features.

[0208] Some embodiments may optionally use define Personalization Ethics and Fairness Rules, to ensure ethical use of Al in feed personalization. For example, the system can be configured to incorporate safeguards such as: (a) Bias Detection, by monitoring Al outputs to detect and mitigate biases, ensuring fair representation of diverse game categories; (b) Transparency, by allowing users to view and modify the factors influencing their feed, such as enabling or disabling specific genres or publishers; (c) Privacy Protections, by storing sensitive user data securely and anonymizing data before processing to prevent misuse.

[0209] Some embodiments may utilize a Feedback Loop for Iterative Refinement, in order to continuously refine the games feed; optionally using one or more of the following: (a) Implicit Feedback, by analyzing engagement metrics, such as skipped games or session durations, to adjust feed construction; (b) Explicit Feedback, by allowing users to provide direct input, such as rating games or flagging irrelevant content; (c) automatically performing A / B testing, and experimenting with different feed configurations across user groups to identify the most effective designs.

[0210] Some embodiments may be configured to predict User Preferences for Games in a Tailored Feed. Predicting which game a user would prefer to see in their feed, or which game they are more likely to select and play, can be implemented using a combination of data analysis, ML / DL, and behavioral modeling. This process relies on collecting user data, analyzing historical interactions, and implementing predictive algorithms. In a demonstrative implementation, the predictive process may include one or more of the following seven steps.

[0211] The first step is Comprehensive User Data Collection, or gathering data from various sources to build a detailed user profile. This data includes: (a) Interaction Data, or information on how users interact with the game feed, such as scrolling speed, preview dwell times, and click-through rates; (b) Gameplay Data, or metrics on games played, including frequency, duration, and level of engagement (e.g., completing levels or in-app purchases); (c) Content Preferences, such as explicitly selected preferences, favorite genres, themes, or publishers; (d) Demographic Information, such as age, gender, location, profession, and device type, which can influence gaming habits and preferences; (e) Temporal Patterns, or other timebased factors, such as the user’s preferred gaming hours or frequency of engagement during specific days of the week; (f) Social Influences, or fata on the user’s connections and their preferences within the same platform or across linked social networks. This dataset forms the foundation for predictive modeling.

[0212] The second step is Feature Engineering for Predictive Analysis: once data is collected, it is transformed into structured features to enable effective prediction. Some operations of this step may include: (a) identifying Behavioral Trends, or patterns such as genre preference shifts over time or increased engagement with multiplayer games; (b) generating Content Affinity Scores, or numerical scores representing the user’s affinity toward specific themes, mechanics, or visual styles; (c) identifying Interaction Sequences, or sequential data capturing the order in which games were browsed, previewed, or selected, providing insight into decision-making processes; (d) identifying and collecting Feedback Signals, such as implicit signals (e.g., skipping a preview) and explicit signals (e.g., user ratings or reviews) indicating game preferences. The feature engineering stage ensures that raw data is converted into meaningful inputs for predictive models.

[0213] The third step includes employing ML / DL models for Predicting Preferences, to analyze the features and predict user preferences. The models may include, for example: (a) Collaborative Filtering Models, such as a ML / DL model that identifies users with similar interaction patterns and recommends games preferred by those users; for example, if User A frequently plays puzzle games and shares behavioral similarities with User B, the system recommends puzzle games to User B. (b) Content-Based Filtering Models, for example, a ML / DL model that matches user preferences to game metadata, such as genre, theme, or visual style; for example, a user engaging heavily with fantasy-themed games is more likely to be presented with similar content, (c) Recurrent Neural Networks (RNNs) that can process sequential data, such as interaction order or gameplay timelines, to predict the next likely game selection; for example, after playing multiple fast-paced arcade games, the system predicts thatthe user is likely to select another arcade game, (d) Reinforcement Learning Models, that continuously learn from user feedback and adjusts predictions to optimize long-term engagement; for example, if a recommended game is selected and played extensively, similar games are prioritized in future predictions, (e) Hybrid Models, that combine collaborative filtering, content-based filtering, and reinforcement learning to enhance accuracy; for example, a hybrid model can recommend games by considering both user interaction history and broader trends among similar users.

[0214] The fourth step includes Real-Time Prediction and Feed Construction, as predictions can be made in real time as users interact with the system. The process involves, for example: (a) Dynamic Ranking, by assigning scores to available games based on their predicted likelihood of selection, ensuring the most relevant games are shown first; (b) Context- Aware Adjustments, by adapting predictions to account for real-time factors such as device type, network speed, or session length; (c) Experimentation and Exploration, by introducing new or less familiar games alongside predicted favorites to keep the feed fresh and engaging. The system balances personalized recommendations with opportunities for discovery, ensuring sustained user interest.

[0215] The fifth step may optionally include Advanced Predictive Techniques, to further refine predictions; such as: (a) Knowledge Graphs, or a graph-based representation connecting users, games, genres, publishers, and interaction data; for example, if a user has shown interest in games by Publisher A and similar users have engaged with Publisher B, games by Publisher B are recommended, (b) Contrastive Language-Image Pretraining (CLIP), using an Al model for analyzing visual and textual features of game previews to align them with user preferences; for example, the system identifies that a user prefers colorful graphics and recommends visually similar games, (c) Predictive Analytics for Similarity Metrics, wherein the system calculates similarity scores between games based on metadata and user engagement data, prioritizing games with higher scores higher scores for inclusion in the feed; for example, if a user enjoys puzzle games with time-based challenges, the system prioritizes games with similar mechanics, (d) Multi- Armed Bandit Algorithms, that can balance exploration (introducing new games) and exploitation (recommending known favorites) by testing different games and evaluating user responses; for example, the system shows a mix of familiar games and new releases, tracking the user’s reaction to optimize future predictions, (e) Temporal Context Modeling, wherein the system analyzes time-specific factors to refine predictions, such as recommending shorter games during weekdays and longer, more immersive games duringweekends; for example, a casual game might be suggested for a brief evening session, while a role-playing game is presented for longer weekend play.

[0216] A sixth step that can be used is Feedback Integration for Continuous Improvement, as the system can refine predictions by integrating user feedback into its predictive models. This includes, for example: (a) Implicit Feedback, by tracking engagement metrics such as time spent browsing previews, number of selections, or duration of gameplay, (b) Explicit Feedback, by allowing users to rate games, mark them as favorites, or provide comments, which are directly incorporated into future recommendations, (c) performing A / B Testing to test different prediction models on subsets of users to identify the most effective algorithms for specific user segments. This feedback loop ensures the system evolves alongside user preferences.

[0217] A seventh step may include operations for ensuring Scalability and Fairness, as the predictive system can be designed to scale efficiently and to maintain fairness. For Scalability, the system can utilize distributed computing and parallel processing to handle large user bases and diverse game libraries without latency. For Fairness, the system can prevent or reduce algorithmic biases by monitoring model outputs and ensuring equitable representation of game categories, genres, and publishers. By addressing these concerns, the system ensures that predictions remain relevant and balanced for all users.

[0218] Accordingly, predicting which game a user is likely to prefer or play involves a sophisticated interplay of data collection, feature engineering, machine learning models, and real-time adjustments. By leveraging advanced Al techniques like RNNs, reinforcement learning, and knowledge graphs, the system can dynamically construct personalized feeds that not only align with user preferences but also introduce engaging new content. The integration of feedback mechanisms and fairness considerations ensures continuous improvement and equitable access to diverse gaming options, creating an engaging and satisfying user experience.

[0219] Some embodiments may optionally implement Emotion-Based Feed Personalization:Integrate emotion recognition technology, such as facial expression analysis or sentiment analysis of user text inputs, to adjust the game feed dynamically. For example, the system can prioritize relaxing games when stress indicators are detected or fast-paced games during periods of high energy.

[0220] Some embodiments may optionally implement Cross-Platform Synchronization, to enable synchronization of user preferences, game progress, and tailored feeds across multiple devices. For instance, a user can seamlessly switch between their smartphone, tablet, anddesktop, maintaining a consistent and personalized experience without interruptions or data loss.

[0221] Some embodiments may optionally implement Gamified Discovery Mechanism, by introducing gamification elements in the feed, such as achievements for discovering new genres or earning rewards for trying unfamiliar games. This encourages exploration and increases engagement by creating a sense of progression and accomplishment through the feed itself.

[0222] Some embodiments may optionally implement Voice- Activated Navigation, to allow users to interact with the game feed through voice commands, such as searching for specific genres or asking for recommendations. This feature enhances accessibility, providing an intuitive way for users to explore and select games hands-free.

[0223] Some embodiments may optionally implement Real-Time Social Recommendations, or may incorporate a feature that shows games actively being played by friends or trending within a user’s social circle in real time. This fosters a sense of community and encourages multiplayer engagement, driving more interactive and shared experiences.

[0224] Some embodiments may optionally implement an Adaptive Feed Layout, as the system can dynamically modify the feed’s visual presentation based on the user’s device orientation, preferences, or accessibility needs. For example, a horizontal feed for tablet users or larger preview thumbnails for users with visual impairments.

[0225] Some embodiments may optionally take into account an Energy Optimization Mode, as the system can introduce a low-power mode that reduces resource-intensive animations and streaming quality on user devices when the battery level is low. This ensures users can continue exploring the feed and playing games without prematurely draining their device’s battery.

[0226] Some embodiments may optionally implement Interactive Storyline-Based Game Previews, as the system can be configured to provide storyline-driven interactive previews for games, allowing users to make simple choices or experience a mini-demo directly in the feed. This enhances user immersion and helps users decide on games without committing to full downloads or sessions.

[0227] Some embodiments may optionally implement Augmented Reality (AR) Integration, or can enable AR-enhanced game previews, where users can view game elements, characters, or environments overlaid on their real-world surroundings. This unique feature elevates the browsing experience and offers an innovative way to interact with potential game selections.

[0228] Some embodiments may solve, prevent, cure and / or mitigate one, or some, or most, or all, of the following problems or disadvantages, from conventional gaming platforms may suffer. (1) Slow and tedious and effort-consuming Game Discovery Process, as traditional systems require extensive searching in app stores, filtering through irrelevant results, and relying on static images, making game discovery time-consuming; whereas, some embodiments streamline discovery with tailored, scrollable feeds and instant previews, significantly reducing effort and improving user satisfaction. (2) High Storage and Download Requirements, as downloading and installing games consumes significant device storage and time, especially for large games; whereas, some embodiments enable immediate streaming and play without installation, conserving storage and minimizing wait times, ensuring instant access to games. (3) Lack of Personalization, as traditional platforms offer generic recommendations, failing to match diverse user preferences; whereas, some embodiments personalize feeds using advanced Al, ensuring users see games aligned with their tastes, improving engagement and reducing frustration from irrelevant suggestions. (4) Limited Game Previews, as static screenshots or limited trailers in traditional systems do not fully represent gameplay; whereas, some embodiments provide dynamic, interactive previews or live streams, enabling users to better evaluate games before committing to play or purchase. (5) Inflexible User Interfaces, as traditional game interfaces often lack adaptability for varying devices or user preferences; whereas, some embodiments feature dynamic layouts, adjustable feed presentation, and user-tailored navigation, enhancing accessibility and usability for diverse audiences. (6) Disconnected Gaming Ecosystem, as games in traditional systems often lack social integration, making it difficult to connect with friends or follow trends; whereas, some embodiments include real-time social recommendations, multiplayer engagement options, and live streaming, fostering a more connected gaming experience. (7) Energy Inefficiency, as some high-performance gaming apps can drain device batteries quickly, discouraging prolonged use; whereas, some embodiments may include energy optimization features, offering low-power modes for game browsing and previews, prolonging battery life without compromising the user experience. (8) Overwhelming Game Libraries, as traditional massive game catalogs make finding relevant content challenging, leading to decision fatigue; whereas, some embodiments mitigate this by curating user-specific feeds with precise recommendations, reducing overwhelm and enabling users to explore games more efficiently. (9) Infrequent Updates to User Interests, as traditional systems often fail to adapt to evolving user preferences in real-time; whereas, some embodiments can use deterministic rules and / or ML / DL modelsto continuously refine game feeds, ensuring the content stays relevant as user tastes and behaviors change.

[0229] Some embodiments may provide some, or most, or all, of the following benefits or advantages: (1) Instant Game Access, as users can seamlessly preview and play games without downloading or installation, reducing wait times and enabling immediate engagement; this eliminates storage constraints and ensures a smooth, frustration-free gaming experience. (2) Personalized Recommendations, as the system tailors the game feed based on user preferences, behaviors, and engagement patterns, ensuring highly relevant suggestions; this enhances user satisfaction by prioritizing games most likely to align with their interests. (3) Dynamic Game Previews, as interactive video previews, including live or pre-recorded streams, allow users to experience gameplay before committing; this transparent approach reduces uncertainty and improves decision-making for users. (4) Effortless Game Discovery, as the curated scrollable feed eliminates the need for tedious keyword searches, presenting a streamlined interface for users to explore and find new games effortlessly. (5) Real-Time Adaptation, as the system can use deterministic rules and / or ML / DL models to analyze real-time user interactions to dynamically update game feeds, ensuring continuously relevant recommendations that evolve with changing preferences. (6) Cross-Device Compatibility, as the system can synchronize user profiles and preferences across multiple devices, offering a consistent gaming experience whether users switch between smartphones, tablets, or computers. (7) Reduced Energy Consumption, as with energy optimization modes, the system can minimize power usage during game previews or feed browsing, allowing users to engage for longer periods without depleting battery life. (8) Enhanced Social Integration, as the platform incorporates social features, enabling users to view friends’ activity, join multiplayer games, and share game previews; this fosters community engagement and interactive experiences. (9) Exploration of Diverse Content, as the tailored feed balances user-specific preferences with new or trending games, encouraging discovery of genres or styles users might not have considered, broadening their gaming horizons. (10) Streamlined Navigation, as the intuitive scrollable feed and gesture-based interface simplify browsing, reducing the complexity of navigating large game libraries and providing a more enjoyable user experience. (11) Scalable Content Delivery, as the cloud-based infrastructure ensures scalable delivery of game feeds and previews, accommodating growing user bases and expanding game libraries without performance bottlenecks. (12) Adaptive Accessibility Features, as dynamic GUI adjustments can be configured to cater to different user needs, such as enlarged thumbnails or alternate layouts for users with visual or physical impairments, making the system accessible to a broader audience.

[0230] Some embodiments provide systems, devices, and methods for generating and operating a user-specific, scrollable feed of interactive games, designed to enhance the discovery and accessibility of games. The system addresses challenges in traditional gaming systems, such as the time-consuming and effort-intensive process of finding, downloading, and installing games. The system focuses on providing a streamlined, efficient, and user-friendly mechanism for game discovery and engagement by tailoring the content to individual user preferences and enabling near-instant access to games without the need for downloading or installing separate applications.

[0231] Some embodiments may include the following features or functionality, or some of them. (1) Scrollable Feed of Games, as the system presents a large or a virtually infinite feed of games that can be navigated easily using gestures, such as swiping on a touchscreen or scrolling with a mouse or keyboard. The feed is designed to display a sequence of games tailored to each user’s preferences and interactions. (2) Tailored User Experience, as the system provides personalized game recommendations by analyzing explicit preferences, such as user- declared game types or publishers, as well as implicit behavioral data, including engagement time, scrolling speed, and selection frequency. This tailored approach ensures that the feed is highly relevant to each individual user. (3) Game Previews, as each game in the feed is accompanied by visual representations, such as thumbnails, video previews, or real-time streams, allowing users to quickly assess gameplay before deciding to engage. These previews can include pre-recorded clips or live streams from ongoing games. (4) Seamless Gameplay Access, as the system eliminates the need for downloading and installing games by enabling users to instantly start playing selected games from the feed. Games are streamed directly to the user’s device, ensuring a smooth transition from discovery to engagement. (5) Adaptive Feed Content, as the feed can be continuously updated based on user interactions. For example, games that are skipped or ignored are deprioritized, while those that attract user interest are emphasized. This adaptability ensures that the content remains relevant over time. (6) Interaction Tracking and Analysis, as the system tracks various user actions, such as dwell times on previews, game selection, and gameplay duration. This data is analyzed to refine user preferences and optimize the feed. Advanced methods, such as artificial intelligence (Al) and ME / DL models, are employed to detect patterns in user behavior. (7) Social Integration, as the platform includes features that allow users to connect with others. This includes sharing game previews, joining multiplayer games, or interacting with live gameplay streams. Social features are integrated into the feed to foster community engagement. (8) Game Personalization, as in some embodiments, the system dynamically modifies games to alignwith user preferences. For example, graphics, themes, or audio elements can be adjusted based on past interactions. This personalization enhances the user experience and aligns the gameplay with individual tastes. (9) Content Categorization, as the system can categorize games into various genres, sub-genres, and themes, enabling efficient filtering and browsing. Users can follow specific categories or publishers, ensuring their feed focuses on preferred types of games. (10) Performance Optimization, as the system can provide high-quality game streaming with minimal latency. It leverages advanced compression techniques and predictive caching to ensure smooth delivery of game previews and gameplay, even under varying network conditions.

[0232] In some embodiments, the technical architecture of the system may include the following components or sub-systems. (1) Central Server System, to store game data, manage user profiles, and handle feed generation. The server hosts a repository of games, including program code, graphical assets, and video previews. (2) Feed Generation Unit, responsible for generating the scrollable feed for each user. This module evaluates user interaction data and applies machine learning algorithms to construct a feed tailored to individual preferences. (3) User Interaction Monitoring, a component that tracks and logs user behavior during feed navigation and gameplay. This data is analyzed to improve future recommendations and ensure the feed remains relevant to the user. (4) Streaming Engine, that enables instant or near-instant or low-latency delivery of game previews and gameplay to user devices. It supports real-time streaming of both pre-recorded and live game sessions. (5) Innovative User Interface, as the feed is presented through a dynamic and intuitive scrollable or swipe-capable interface that adapts to different devices and user preferences; users can efficiently and rapidly navigate the feed using touch gestures, mouse scrolls, or keyboard inputs. (6) Al and ML / DL integration, such as, employing Al / ML / DL models to analyze user data, detect preferences, and predict which games a user is likely to enjoy. These insights are used to refine feed content and ensure accurate recommendations.

[0233] The system may provide a variety of benefits, such as: (1) Efficiency, as users can rapidly discover and play games without the need for lengthy downloads or installations. (2) Personalization, as user-tailored feeds ensure that users are presented with games that align closely with their preferences and behaviors. (3) Accessibility, as the system’s interface and real-time streaming capabilities make it accessible across various devices and user scenarios. (4) Engagement, as features like game previews, social interactions, and dynamic content adaptation enhance overall user engagement and satisfaction.

[0234] Some embodiments are applicable across various gaming contexts, including mobile devices, desktop environments, and smart TVs. Some embodiments are particularly well-suited for gaming platforms that aim to offer large, diverse libraries of games while maintaining an efficient and user-centric discovery process. The technology and features of some embodiments can also be integrated with online casinos, multiplayer gaming platforms, or educational game repositories to enhance user interaction and retention. Some embodiments may thus provide a robust solution for personalized game discovery and engagement by leveraging advanced technologies, including Al, real-time streaming, and user behavior analysis. The system can address inefficiencies of traditional gaming systems, and offers a dynamic, adaptive platform tailored to individual user preferences. By integrating innovative features, the system improves the overall gaming experience, making it more accessible, efficient, and engaging.

[0235] Some embodiments may optionally use a Feedback Loop Mechanism for Tailoring the Games Feed. For example, a Feedback Loop Mechanism can further enhance the tailoring of the games feed by continuously incorporating user interactions and responses into the recommendation system. The mechanism works as an iterative process that collects real-time data, processes it to extract actionable insights, and adjusts the feed dynamically based on these insights.

[0236] Components of the Feedback Loop Mechanism may include: (a) Interaction Monitoring Unit (IMU), that tracks all user actions on the platform, including scrolling, pausing, clicking, selecting games, exiting previews, and gameplay metrics like duration and success rates. It also logs contextual data such as the time of day, device type, and network conditions, (b) Data Collection and Logging Module, as user interactions are continuously logged in a central repository. This data is timestamped and indexed, ensuring that the platform maintains a chronological record of user activities for longitudinal analysis, (c) Behavioral Analysis Engine (BAE), which processes the raw interaction data to identify patterns and extract features, such as preferences for specific genres, engagement levels with certain themes, or responsiveness to game difficulty levels. Techniques like clustering and dimensionality reduction can be applied to summarize user behavior effectively, (d) Real-Time Adjustment Layer, such as a middleware system that dynamically integrates insights from the analysis engine into the live feed. This layer ensures that new recommendations reflect the latest user preferences without requiring a system restart or manual input, (e) Reinforcement Learning Module, that can use a reward-based learning approach; for example, the system assigns scores to specific user actions, such as clicking on a game preview or playing a game for an extendedperiod. These scores are used to train a reinforcement learning model, optimizing future feed adjustments, (f) Content Recommendation Engine, which is responsible for presenting games in the feed based on updated user profiles. It integrates collaborative filtering, content-based filtering, and hybrid recommendation models, dynamically incorporating feedback signals.

[0237] For example, the operational workflow of the Feedback Eoop may be as follows, (a) Data Capture: As users interact with the games feed, the IMU captures all relevant actions and sends the data to the logging module. For example, if a user spends significant time previewing puzzle games but skips action games, these interactions are recorded, (b) Real- Time Processing:The captured data is forwarded to the Behavioral Analysis Engine, where algorithms like decision trees or neural networks process the data to update the user’ s preference profile. Key metrics, such as dwell time or the number of previews skipped, are converted into numerical weights representing user interest levels, (c) Feed Adjustment: Insights from the analysis engine are fed into the Real-Time Adjustment Eayer. For example, if the system detects increased interest in multiplayer games, the next scrollable section of the feed might prioritize games featuring multiplayer options or co-op modes, (d) Reward and Feedback Evaluation: the Reinforcement Eearning Module evaluates the effectiveness of recommendations based on user actions. Positive outcomes, such as selecting or playing a recommended game, increase the recommendation weight for similar games. Negative outcomes, such as skipping or exiting previews quickly, reduce the weight for similar content, (e) Model Refinement, as the recommendation system updates its underlying machine learning models periodically, incorporating aggregated feedback data to improve prediction accuracy. Batch processing ensures that historical trends and newly captured behaviors are balanced in the updated model, (f) Iterative Learning, as the feedback loop mechanism repeats continuously, ensuring that every new interaction refines the feed further. Over time, the system develops a robust understanding of each user’s preferences, even adapting to subtle or evolving behavioral changes.

[0238] Some embodiments may optionally use advanced techniques for Feedback Integration, such as from the following: (a) Multi-Armed Bandit Algorithms, that balance exploration and exploitation, occasionally introducing unfamiliar games into the feed to test for latent preferences while prioritizing proven favorites, (b) Predictive Modeling, by using sequential models like Recurrent Neural Networks (RNNs) or Long Short-Term Memory (LSTM) networks, to predict future preferences based on historical interaction patterns, (c) Federated Learning, in order to maintain user privacy; the system can process feedback locallyon user devices and share aggregated insights with the central server, ensuring personalization without compromising sensitive data, (d) Anomaly Detection, as some Al algorithms can identify unusual interaction patterns, such as a sudden disinterest in a preferred genre, and adjust the feed dynamically to account for temporary behavioral shifts.

[0239] In some embodiments, implementing a feedback loop mechanism can offer several advantages, such as: (a) Adaptability, as the feed evolves in real time, ensuring it aligns with current user preferences, (b) Improved Accuracy, as continuous learning reduces irrelevant recommendations, enhancing the user experience, (c) Engagement Optimization, as prioritizing relevant content can allow the system to encourage deeper interaction and longer session durations, (d) Scalability, as the modular design supports simultaneous feedback integration for a large user base without significant performance degradation.

[0240] Accordingly, some embodiments may use a feedback loop mechanism that enhances the personalization of the games feed by iteratively learning from user interactions. By leveraging advanced analytics, machine learning, and real-time adjustments, the system provides a continuously evolving, user-centric gaming experience while addressing scalability and privacy concerns. This approach ensures that users receive tailored, engaging, and relevant recommendations.

[0241] Some embodiments may monitor, track, measure, detect, sense, log, and / or analyze some or all of the following types of User Interactions, for the purpose of constructing a user- tailored / user-specific scrollable feed of games, and / or for predicting or estimating which game(s) are more likely to be played or engaged by a particular user. (1) Preview Dwell Time: The system can measures how long a user spends viewing a game preview, indicating their level of interest. Longer dwell times suggest potential engagement, while shorter times imply disinterest or irrelevance of the game to the user’s preferences. (2) Scrolling Speed: The system can analyze the speed at which users scroll through the feed. Faster scrolling may indicate disinterest in the current content, while slower scrolling or pauses suggest increased attention to specific games. (3) Game Selection Frequency: The system can track how often users select games from the feed. High selection rates for specific genres or themes indicate strong preferences that can guide future feed content adjustments. (4) Gameplay Duration: The system can record the length of time users spend playing a selected game. Longer durations suggest a deeper interest in that game type, while brief sessions may indicate dissatisfaction or a mismatch with preferences. (5) Skipping Patterns: The system can monitor games that are consistently skipped or ignored in the feed. Frequent skipping of a particular genre or theme suggests a lack of interest and helps refine future recommendations. (6) Preview EngagementActions: The system can capture specific actions during game previews, such as zooming into details, clicking “Learn More”, or adjusting preview settings. These actions reflect curiosity or intent to explore further. (7) Interaction Sequence: The system can analyze the order in which games are previewed, skipped, or selected. Patterns in sequential behavior can indicate the user’s decision-making process and preferences over time. (8) Revisit Frequency: The system can track how often users return to previously previewed or played games. Repeated visits suggest strong interest and can be used to prioritize similar games in the feed. (9) Preferred Time of Engagement: The system can monitor the times of day or week when users engage with the feed. Time-specific preferences can tailor recommendations to align with their gaming habits (e.g., casual games during short breaks). (10) Multiplayer Interaction Preference: The system can monitor whether users gravitate toward multiplayer games, co-op modes, or solo experiences. This helps the system prioritize games with similar interaction styles. (11) Genre Diversity: The system can analyze the variety of genres interacted with over time. Users showing consistent interest in specific genres can have their feeds concentrated on those, while diverse preferences are met with broader recommendations. (12) Reaction to Promotions: The system can track user responses to promotional content, such as discounts or featured games. Positive responses, like clicks or selections, suggest sensitivity to promotions and inform tailored marketing strategies. (13) Content Exploration Depth: The system can measure how far users scroll in the feed before making a selection. Deeper exploration suggests they enjoy browsing, while shallow engagement implies a need for higher-priority recommendations. (14) Engagement with Tutorials: The system can monitor interactions with game tutorials or instructions. Users engaging deeply with tutorials may prefer games with a learning curve, while skipping them might indicate a preference for intuitive, straightforward games. (15) Thematic Preferences: The system can analyze the themes users engage with most, such as fantasy, science fiction, pet-related / animal-related themes, holiday themes, or realistic settings. These insights help refine the visual and narrative elements of games prioritized in the feed. (16) Response to Updates: The system can monitor engagement with updated content or newly added games. Quick adoption of new releases indicates a preference for fresh experiences, while slow uptake suggests a need for familiar or established games. (17) Audio and Visual Preferences: The system can capture responses to specific audio styles (e.g., upbeat or atmospheric) and visual designs (e.g., minimalist or detailed). This helps in recommending games with matching sensory appeal. (18) In-Game Achievement Behavior: The system can track whether users aim for high scores, complete challenges, passing or completing levels, defeating a virtual enemy, or engaging in casual play or leisure play. Competitive orachievement-oriented players can be offered games with leaderboards, while others receive relaxing or story-driven options. (19) Social Engagement Levels: The system can monitor user interactions with social features, such as sharing previews, inviting friends, or watching live streams. High social activity suggests prioritizing multiplayer or community-based games in their feed. Other types of user interactions can be monitored and analyzed, to further adapt the scrollable games feed to the specific user.

[0242] Some embodiments can further perform tailoring a games feed using location data or geo-location data, introducing a dimension of personalization that accounts for geographical, cultural, and regulatory differences among users. By incorporating location-based insights, the system can refine the user experience, ensuring the feed aligns with regional preferences, legal requirements, and contextual needs. Below is a detailed explanation of how the feed of games can be tailored using such data.

[0243] The system begins by collecting location data through the user’s device, leveraging Global Positioning System (GPS) data, Internet Protocol (IP) address geo-location, Wi-Fi triangulation, beacon based geo-location, and / or cellular network signals. Once the user’s location is identified, it is processed in conjunction with other data points to inform feed adjustments. The location data may be mapped to various attributes such as the country, city, time zone, or even specific cultural or regional characteristics. For instance, if User 1 is located in France and User 2 is in Germany, the system applies localized adjustments tailored to each user’ s environment.

[0244] For example, one approach involves adjusting the feed based on language preferences associated with the user’s location. For User 1 in France, the feed might prioritize games with French-language options, ensuring all previews, descriptions, and gameplay instructions are presented in French. Similarly, for User 2 in Germany, the system would highlight games available in German, offering a seamless and accessible experience. This adjustment requires mapping the available languages for each game in the repository and dynamically filtering or ranking games to match the user’s linguistic context. If a game is available in multiple languages, the system could even display the preferred language by default in the game preview, reducing friction for users.

[0245] Another use of location data involves tailoring the feed to regional content preferences. Cultural factors significantly influence gaming habits, with certain genres, themes, or visual styles resonating more with specific regions. For example, strategy games may have a higher appeal in Germany, while casual or narrative-driven games may be more popular in France. The system can integrate regional gaming trend data, collected from global userbehavior or third-party market analyses, to adjust the prioritization of games. For User 1, the feed might highlight games inspired by French art, history, or culture, while User 2’ s feed could focus on strategy games with German historical themes or mechanics.

[0246] Regulatory compliance is another aspect of tailoring game feeds based on location. Gaming regulations vary widely between regions, affecting the availability of certain games or game features. For instance, France and Germany may have different restrictions on in-game purchases, loot boxes, or age -rated content. The system leverages a database of regulatory rules mapped to geographical locations, ensuring the feed adheres to local laws. For User 1 in France, the feed could exclude games with unregulated gambling elements, while User 2’s feed in Germany might adjust to highlight games adhering to German regulatory standards.

[0247] Location data can also inform feed adjustments by accounting for regional time zones and user activity patterns. By analyzing time zone data, the system aligns feed recommendations with expected user activity periods. For instance, if User 1 in France typically plays games in the evening, the system might promote relaxing or single -player games during those hours. Conversely, if User 2 in Germany engages with the platform during lunch breaks, the feed could prioritize quick-play or casual games suited to short sessions. These adjustments enhance the relevance of recommendations by aligning them with the user’s daily routines.

[0248] Geo-location data further enables event-based or season-based tailoring of the feed. For example, during national holidays or festivals, the system could highlight games with themes or promotions related to the event. For User 1 in France, the feed might feature games celebrating Bastille Day with French-themed content or discounts. For User 2 in Germany, Oktoberfest-themed games or seasonal multiplayer events could take precedence. These localized promotions create a sense of relevance and engagement by connecting gaming content with the user’s immediate environment.

[0249] Weather data, derived from geo-location information, offers another opportunity for user-specific geo-located feed customization. The system can incorporate weather conditions into its recommendation engine, providing content that aligns with the user’s physical context. For instance, on a rainy day in France, User l’s feed might emphasize cozy, narrative-driven games, while sunny weather in Germany might lead to the promotion of outdoor-themed or adventure games for User 2. In another example, if the user is located in a region that is currently experiencing snow or heavy snow, then snow-themed games can be prioritized or pushed to populate the games feed of that user. These subtle contextual adjustments can enhance the user experience by aligning the content with their surroundings.

[0250] Location data can also be used to optimize server connections and reduce latency in game streaming. For users in France, the system might route data through a nearby European server, ensuring smooth and responsive gameplay. Similarly, for users in Germany, server connections can be optimized based on proximity and network quality. By integrating geolocation with network optimization, the system not only improves user experience but also ensures reliable content delivery. Finally, location data enables tailored social features within the game feed. Users in specific regions may be more inclined to engage with games featuring multiplayer modes or local leaderboards. For User 1 in France, the feed might prioritize games where other French players are active, fostering a sense of community. Similarly, User 2 in Germany could be presented with games hosting regional tournaments or events, encouraging participation and competition.

[0251] In some embodiments, leveraging location data to tailor game feeds offers a comprehensive approach to personalization. By considering language preferences, regional trends, regulatory compliance, time zones, seasonal events, weather, server optimization, and social features, the system can deliver a highly contextualized and engaging gaming experience. These location-based adjustments not only enhance the relevance of recommendations but also ensure that users feel understood and catered to, fostering loyalty and deeper engagement with the platform.

[0252] Additionally or alternatively, some embodiments can perform tailoring a games feed to the time-of-day, day-of-the-week, or day-of-the-month. This allows the system to align game recommendations with the user’s contextual environment, behavioral patterns, and psychological state during specific times. Incorporating temporal factors into the feed personalization process requires a robust system capable of collecting, analyzing, and acting upon time-sensitive data. This approach enhances user engagement by providing games that match the user’s likely mindset or availability during these periods.

[0253] For example, the system begins by collecting interaction data from users across different times of the day and days of the week or month. For example, it tracks when users are most active on the platform, how long they engage, and which types of games they prefer during these periods. This data is logged with timestamps and stored in a central database, creating a detailed timeline of user activity. Over time, this temporal interaction data builds a profile of user preferences correlated to specific time periods.

[0254] In order to tailor the feed based on time-of-day, the system identifies behavioral patterns associated with different periods, such as morning, afternoon, evening, or late night. Morning hours may correspond to shorter sessions, as users often have limited time beforestarting their day. In this case, the feed might prioritize quick-play games, such as casual puzzle games or arcade titles, that can be enjoyed in brief intervals. Conversely, evening hours may reflect a more relaxed environment where users have more leisure time. In this situation, the feed can be tailored to emphasize immersive genres like role-playing games, narrative-driven experiences, or multiplayer titles. Late-night gaming sessions, often linked to reduced cognitive energy, may favor games with simpler mechanics or soothing aesthetics to align with the user’s likely state of mind.

[0255] Similarly, day-of-week patterns provide another layer of possible games feed customization. The system analyzes trends indicating how user preferences shift between weekdays and weekends. Weekdays may be characterized by constrained schedules, so recommendations might focus on games that can be played in shorter bursts or paused conveniently. For example, a user’s feed could include turn-based strategy games or lightweight casual games that do not demand extended commitment. On weekends, when users often have more free time, the system might recommend games requiring longer engagement, such as open-world exploration games or competitive multiplayer experiences. Furthermore, weekends might also highlight games with social features, such as online co-op modes, as users are more likely to have time to interact with friends.

[0256] The system can also adapt to recurring monthly patterns, such as the beginning, middle, or end of the month. These periods may coincide with shifts in user priorities. For example, near the beginning of the month, users may be more inclined to try new games as they explore recently released titles. The system could prioritize games that have just launched or feature promotional content. Toward the middle of the month, users might revisit games they have already played or explore games within their familiar preferences. In this scenario, the feed can be adapted to focus on offering updates, expansions, or new challenges in games they are already engaged with. At the end of the month, users might seek games with shorter- term goals, as they may aim to complete certain objectives or try out specific content before transitioning to new experiences in the upcoming month.

[0257] Temporal factors also enable integration with seasonal or holiday-based gaming patterns. For example, a feed tailored for weekday evenings during December may prioritize holiday-themed games or limited-time events tied to the season. Similarly, specific days of the week, such as Fridays or Saturdays, could feature games that align with social gatherings, such as party games or competitive multiplayer titles. On the other hand, Sunday mornings might promote relaxing single -player games that match a quieter pace.

[0258] The system’s ability to incorporate time-sensitive data can be powered by algorithms that process temporal interaction data in real time. For instance, the system can use ML / DL models to detect correlations between time periods (or time -points, or day-of-week, or day-of-month, or time-of-day) and user engagement metrics, such as the frequency of game selections, session durations, and abandonment rates. Reinforcement learning techniques can further refine the system by optimizing feed performance based on outcomes. If a recommendation aligns with the user’s behavior during a specific time period, the system assigns a higher weight to similar recommendations in future iterations. Conversely, if users skip or ignore certain game types during specific periods, the system deprioritizes those recommendations.

[0259] In some embodiments, in order to enhance the effectiveness of time-based tailoring, predictive analytics can anticipate user needs. For example, the system may be configured to predict that a user typically engages in casual games during weekday mornings and prepare the feed accordingly before the user logs in. This proactive approach minimizes latency in delivering relevant recommendations and improves the overall user experience.

[0260] Additionally or alternatively, contextual factors such as the user’s local time zone can play a role that the system takes into account. The system aligns its recommendations with the user’s real-time environment, ensuring that global users receive personalized feeds based on their respective time zones. For instance, while one user in Europe may receive a feed optimized for late-night relaxation, another user in North America may simultaneously see a feed tailored to their afternoon activity.

[0261] The dynamic nature of temporal tailoring also allows integration with other data points, such as seasonal or time-specific promotions. For example, games with in-game events that are active during specific periods can be prominently featured in the feed, encouraging timely engagement. Similarly, promotional content such as weekend discounts or limited-time offers could be prioritized for users accessing the feed during those windows.

[0262] In some embodiments, tailoring a games feed based on time-of-day, day-of-week, or day-of-month or other temporal factors involves a comprehensive analysis of user behavior patterns and contextual data. By leveraging interaction logs, machine learning algorithms, and predictive analytics, the system dynamically adjusts recommendations to align with the user’s temporal context. This approach ensures that the games presented in the feed are not only personalized but also relevant to the user’s immediate environment, optimizing engagement and satisfaction. Through real-time adaptability and ongoing refinement, the system creates a deeply contextualized and responsive gaming experience.

[0263] Some embodiments provide a computerized method, comprising: (a) storing on a server data representing a plurality of online games; wherein each online game is associated with (i) a code -portion that runs the online game, and (ii) a screenshot image or a preview video that visually demonstrate the online game; (b) causing a first end-user device of a first end-user to display a first scrollable feed of online games, wherein the first scrollable feed of online games is tailored to the first end-user based on analysis of past interactions of said first enduser; and in parallel, causing a second end-user device of a second end-user to display a second, different, scrollable feed of online games, wherein the second scrollable feed of online games is tailored to the second end-user based on analysis of past interactions of said second end-user.

[0264] In some embodiments, the computerized method comprises: dynamically constructing a different, user-specific, user-tailored, scrollable feed of online games for displaying on different end-user devices of different users, based on computer-generated estimation of which genre of online games is preferred by each of said different users.

[0265] In some embodiments, the computerized method comprises: dynamically constructing a different, user-specific, user-tailored, scrollable feed of online games for displaying of different end-user devices of different users, based on computer-generated insights or Al-generated insights regarding user preferences, that are derived from analysis of past interactions of each user with online games.

[0266] In some embodiments, the method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; based on analysis of differential swiping speed that the first user exhibits when swiping along the first scrollable feed of online games, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games.

[0267] In some embodiments, the computerized method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; based on analysis of time-measured lingering of the first enduser over a preview of a particular online game that belongs to a particular genre, during scrolling along the first scrollable feed of online games, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games.

[0268] In some embodiments, the computerized method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; measuring engagement time of the first user, with different online games that belong to different genres, that were presented to the first end-user in the first scrollable feed of online games; based on analysis of differential engagement times thatthe first end-user exhibits at different online games in the first scrollable feed of online games, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games.

[0269] In some embodiments, the computerized method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; tracking geo-location data of the first end-user device while the first end-user is scrolling along the first scrollable feed of online games and while the first end-user is engaging with online games; determining a game-to-location correlation between (i) a particular type of online games that the first end-user plays or for which he exhibits lingering engagement, and (ii) geo-location data of the first end-user device; based on said game-to-location correlation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games.

[0270] In some embodiments, the computerized method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; tracking temporal data of the first end-user device while the first end-user is scrolling along the first scrollable feed of online games and while the first enduser is engaging with online games; determining a time-and-day to game correlation, between (i) a particular type of online games that the first end-user plays or for which he exhibits lingering engagement, and (ii) a temporal characteristic that includes at least one of: time-of- day, day-of-week, day-of-month, in-holiday time; based on said time-and-day to game correlation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games.

[0271] In some embodiments, the computerized method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; tracking device orientation data of said first end-user device, while the first end-user is scrolling along the first scrollable feed of online games and while the first end-user is engaging with online games, using at least one of: (i) an accelerometer of said first end-user device, (ii) a gyroscope unit of said first end-user device, (iii) a compass unit of said first end-user device, (iv) a spatial orientation sensor of said first end-user device, (v) a spatial angular tilt sensor of said first end-user device; determining a correlation between (i) particular online games for which monitored user interactions exhibit a longer engagement time by the first time -user, and (ii) device orientation data that was sensed during engagement with said particular online games; based on said correlation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games.

[0272] In some embodiments, the computerized method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; determining a correlation between (i) particular online games for which monitored user interactions exhibit a longer engagement time by the first time -user, and (ii) a particular graphical theme that is featured in said particular online games; based on said correlation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games.

[0273] In some embodiments, the computerized method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; detecting that the first end-user lingers for a first time -period T1 over a preview of a first particular online game that is displayed in the first scrollable feed of online games; detecting that the first end-user lingers for a second time -period T2 over a preview of a second particular online game that is displayed in the first scrollable feed of online games; wherein T2 is at least twice Tl; generating an estimation that the first end-user prefers online games that belong to a game genre of the second online game over online games that belong to a game genre of the first online game; based on said estimation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to increase a percentage of games that belong to the game genre of the second online game, and to decrease a percentage of games that belong to a game genre of the first online game.

[0274] In some embodiments, the computerized method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; detecting that the first end-user actively plays for a first timeperiod Tl with first particular online game that is displayed in the first scrollable feed of online games; detecting that the first end-user actively plays for a second time -period T2 with a second particular online game that is displayed in the first scrollable feed of online games; wherein T2 is at least twice Tl; generating an estimation that the first end-user prefers online games that belong to a game genre of the second online game over online games that belong to a game genre of the first online game; based on said estimation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to increase a percentage of games that belong to the game genre of the second online game, and to decrease a percentage of games that belong to a game genre of the first online game.

[0275] In some embodiments, the method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed ofonline games; determining which particular online game was played by the first end-user for the longest cumulative time during most-recent N usage sessions, wherein N is a pre-defined integer; generating an estimation that the first end-user prefers online games that belong to a game genre of said particular online game; based on said estimation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to increase a percentage of games that belong to the game genre of said particular online game, and to decrease a percentage of games that belong to other game genres.

[0276] In some embodiments, the computerized method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; determining which particular online game was played by the first end-user for the longest cumulative time during most-recent N usage sessions, wherein N is a pre-defined integer; and further determining that said particular online game is associated with a particular graphical theme of content; generating an estimation that the first end-user prefers online games that incorporate said particular graphical theme of content that is associated with said particular online game; based on said estimation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to increase a percentage of games that exhibit said particular graphical theme of content, and to decrease a percentage of games that exhibit other graphical themes of content.

[0277] In some embodiments, the computerized method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; extracting from monitored user interactions of the first enduser, an extracted temporal feature that is at least one of: (a) time-length of actively playing with each game, (b) time-length of watching a preview video or a preview animation of each game; feeding said extracted temporal feature that was extracted from monitored user interactions of the first end-user, into a computerized engine that uses at one of: (i) a pre-trained Machine Learning model, or (ii) a Large Multi-Modalities Model, or (iii) a set of pre-defined deterministic rules; and commanding said computerized engine to generate predictions indicating which particular online games are more likely to be played by the first end-user; based on said predictions, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to populate therein the particular online games that the computerized engine predicts as more likely to be played by the first end-user.

[0278] In some embodiments, the computerized method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; extracting from monitored user interactions of the first end-user, an extracted content-theme feature that is at least one of: (a) a graphical theme of characters that are depicted in online games played by the first end-user, (b) a graphical theme of objects that are depicted in online games played by the first end-user; feeding said extracted content-theme feature that was extracted from monitored user interactions of the first end-user, into a computerized engine that uses at one of: (i) a pre-trained Machine Learning model, or (ii) a Large Multi-Modalities Model, or (iii) a set of pre-defined deterministic rules; and commanding said computerized engine to generate predictions indicating which particular online games are more likely to be played by the first end-user; based on said predictions, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to populate therein the particular online games that the computerized engine predicts as more likely to be played by the first end-user.

[0279] In some embodiments, the method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; extracting from monitored user interactions of the first end-user, a userengagement feature that is at least one of: (a) a binary signal indicating whether the first enduser watch or did not watch a video preview of particular games, (b) a binary signal indicating whether the first end-user actively played or did not actively play particular games, (c) a binary signal indicating whether the first user played with particular games more than T seconds or not more than T seconds, wherein T is a pre-defined threshold value; (d) a binary signal indicating whether or not the first user shared with another user a recommendation to play a particular game; feeding said extracted user-engagement feature that was extracted from monitored user interactions of the first end-user, into a computerized engine that uses at one of: (i) a pre-trained ML model, or (ii) a Large Multi-Modalities Model, or (iii) a set of predefined deterministic rules; and commanding said computerized engine to generate predictions indicating which particular online games are more likely to be played by the first end-user; based on said predictions, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to populate therein the particular online games that the computerized engine predicts as more likely to be played by the first end-user.

[0280] In some embodiments, the method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; based on analysis of monitored user interactions of the first end-user, determining that the first end-user engages for a longer time, on average, with online games that were published by a particular game -publisher, relative to online games that were published by other game-publishers; generating an estimation that the first end-user prefersonline games that were published by said particular game -publisher; based on said estimation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to increase a percentage of games that were published by said particular game -publisher, and to decrease a percentage of games that were published by other game-publishers.

[0281] In some embodiments, the computerized method comprises: monitoring user interactions of a plurality of different users, that utilize a respective plurality of end-user devices to interact with online games that are presented via different user-specific scrollable feeds of online games; training a Machine Learning model on features extracted from said user interactions of said plurality of different users, and constructing a Machine Learning model that predicts which type of games are more likely to be engaged by a particular end-user that exhibits a particular feature of user interactions; dynamically constructing or modifying a subset of online games that populate a scrollable feed of online games that is displayed to said particular user, based on predictions generated by said Machine Learning model.

[0282] In some embodiments, the computerized method comprises: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; tracking at least one behavioral data-item that is selected from the group consisting of: (i) audio of the first end-user captured while he interacts with the first scrollable feed of online games, (ii) video of the first end-user captured while he interacts with the first scrollable feed of online games, (iii) one or more images of the first end-user captured while he interacts with the first scrollable feed of online games, (iv) one or more body-gestures of the first end-user captured while he interacts with the first scrollable feed of online games; feeding the user interactions, and the at least one behavioral data-item, into a Large MultiModalities Model (LMMM), and prompting the LMMM to generate predictions indicating which online games are more likely to be engaging to the first end-user and which other online games are less likely to be engaging to the first end-user; based on LMMM-generated predictions, dynamically constructing or modifying a subset of online games that populate a scrollable feed of online games that is displayed to said particular user.

[0283] Some embodiments may include one or more hardware processors, that are configured to execute code; and that are operably associated with one or more memory units, that are configured to store code and data; wherein the one or more hardware processors are configured to perform a computerized method as described above.

[0284] Some embodiments may include a non-transitory storage article or storage medium, having stored thereon instructions or machine -readable code that, when executed by a machineor by a hardware processor, cause the machine or the hardware processor to perform a method as described above.

[0285] Although portions of the discussion herein relate, for demonstrative purposes, to wired links and / or wired communications, some embodiments are not limited in this regard, but rather, may utilize wired communication and / or wireless communication; may include one or more wired and / or wireless links; may utilize one or more components of wired communication and / or wireless communication; and / or may utilize one or more methods or protocols or standards of wireless communication.

[0286] Some embodiments of the present invention may be implemented by using a special-purpose machine or a specific -purpose device that is not a generic computer, or by using a non-generic computer or a non-general computer or machine. Such system or device may utilize or may comprise one or more components or units or modules that are not part of a “generic computer” and that are not part of a “general purpose computer”, for example, cellular transceivers, cellular transmitter, cellular receiver, GPS unit, location-determining unit, accelerometer(s), gyroscope(s), device-orientation detectors or sensors, device-positioning detectors or sensors, or the like.

[0287] Some embodiments of the present invention may be implemented as, or by utilizing, an automated method or automated process or a computerized process, or a machine- implemented method or process, or as a semi-automated or partially-automated method or process, or as a set of steps or operations which may be executed or performed by a computer or machine or system or other device.

[0288] Some embodiments may be implemented by using code or program code or machine -readable instructions or machine -readable code, which may be stored on a non- transitory storage medium or non-transitory storage article (e.g., a CD-ROM, a DVD-ROM, a physical memory unit, a physical storage unit), such that the program or code or instructions, when executed by a processor or a machine or a computer, cause such processor or machine or computer to perform a method or process as described herein. Such code or instructions may be or may comprise, for example, one or more of: software, a software module, an application, a program, a subroutine, instructions, an instruction set, computing code, words, values, symbols, strings, variables, source code, compiled code, interpreted code, executable code, static code, dynamic code; including (but not limited to) code or instructions in high-level programming language, low-level programming language, object-oriented programming language, visual programming language, compiled programming language, interpreted programming language, C, C++, C#, Java, JavaScript, SQL, Ruby on Rails, Go, Cobol, Fortran,ActionScript, AJAX, XML, JSON, Lisp, Eiffel, Verilog, Hardware Description Language (HDL), BASIC, Visual BASIC, MATLAB, Pascal, HTML, HTML5, CSS, Perl, Python, PHP, Dart programming language, machine language, machine code, assembly language, or the like.

[0289] Discussions herein utilizing terms such as, for example, “processing”, “computing”, “calculating”, “determining”, “establishing”, “analyzing”, “checking”, “detecting”, “measuring”, or the like, may refer to operation(s) and / or process(es) of a processor, a computer, a computing platform, a computing system, or other electronic device or computing device, that may automatically and / or autonomously manipulate and / or transform data represented as physical (e.g., electronic) quantities within registers and / or accumulators and / or memory units and / or storage units into other data or that may perform other suitable operations.

[0290] The terms “plurality” and “a plurality”, as used herein, include, for example, “multiple” or “two or more”. For example, “a plurality of items” includes two or more items.

[0291] References to “one embodiment”, “an embodiment”, “demonstrative embodiment”, “various embodiments”, “some embodiments”, and / or similar terms, may indicate that the embodiment(s) so described may optionally include a particular feature, structure, or characteristic, but not every embodiment necessarily includes the particular feature, structure, or characteristic. Furthermore, repeated use of the phrase “in one embodiment” does not necessarily refer to the same embodiment, although it may. Similarly, repeated use of the phrase “in some embodiments” does not necessarily refer to the same set or group of embodiments, although it may.

[0292] As used herein, and unless otherwise specified, the utilization of ordinal adjectives such as “first”, “second”, “third”, “fourth”, and so forth, to describe an item or an object, merely indicates that different instances of such like items or objects are being referred to; and does not intend to imply as if the items or objects so described must be in a particular given sequence, either temporally, spatially, in ranking, or in any other ordering manner.

[0293] Some embodiments may be used in, or in conjunction with, various devices and systems, for example, a Personal Computer (PC), a desktop computer, a mobile computer, a laptop computer, a notebook computer, a tablet computer, a server computer, a handheld computer, a handheld device, a Personal Digital Assistant (PDA) device, a handheld PDA device, a tablet, an on-board device, an off-board device, a hybrid device, a vehicular device, a non-vehicular device, a mobile or portable device, a consumer device, a non-mobile or nonportable device, an appliance, a wireless communication station, a wireless communication device, a wireless Access Point (AP), a wired or wireless router or gateway or switch or hub, a wired or wireless modem, a video device, an audio device, an audio-video (A / V) device, awired or wireless network, a wireless area network, a Wireless Video Area Network (WVAN), a Local Area Network (LAN), a Wireless LAN (WLAN), a Personal Area Network (PAN), a Wireless PAN (WPAN), or the like.

[0294] Some embodiments of the present invention may be used in conjunction with one way and / or two-way radio communication systems or devices, cellular radio-telephone communication systems or devices, a mobile phone, a cellular telephone, a wireless telephone, a Personal Communication Systems (PCS) device, a PDA or handheld device which incorporates wireless communication capabilities, a mobile or portable Global Positioning System (GPS) device, a device which incorporates a GPS receiver or transceiver or chip, a device which incorporates an RFID element or chip, a Multiple Input Multiple Output (MIMO) transceiver or device, a Single Input Multiple Output (SIMO) transceiver or device, a Multiple Input Single Output (MISO) transceiver or device, a device having one or more internal antennas and / or external antennas, Digital Video Broadcast (DVB) devices or systems, multistandard radio devices or systems, a wired or wireless handheld device, e.g., a Smartphone, a Wireless Application Protocol (WAP) device, or the like.

[0295] Some embodiments may comprise, or may be implemented by using, an “app” or application which may be downloaded or obtained from an “app store” or “applications store”, for free or for a fee, or which may be pre-installed on a computing device or electronic device, or which may be otherwise transported to and / or installed on such computing device or electronic device.

[0296] Functions, operations, components and / or features described herein with reference to one or more embodiments of the present invention, may be combined with, or may be utilized in combination with, one or more other functions, operations, components and / or features described herein with reference to one or more other embodiments of the present invention. The present invention may thus comprise any possible or suitable combinations, rearrangements, assembly, re-assembly, or other utilization of some or all of the modules or functions or components that are described herein, even if they are discussed in different locations or different chapters of the above discussion, or even if they are shown across different drawings or multiple drawings.

[0297] While certain features of some demonstrative embodiments of the present invention have been illustrated and described herein, various modifications, substitutions, changes, and equivalents may occur to those skilled in the art. Accordingly, the claims are intended to cover all such modifications, substitutions, changes, and equivalents.

Claims

CLAIMS1. A computerized method, comprising:(a) storing on a server data representing a plurality of online games; wherein each online game is associated with (i) a code-portion that runs the online game, and (ii) a screenshot image or a preview video that visually demonstrate the online game;(b) causing a first end-user device of a first end-user to display a first scrollable feed of online games, wherein the first scrollable feed of online games is tailored to the first end-user based on analysis of past interactions of said first end-user; and in parallel, causing a second end-user device of a second end-user to display a second, different, scrollable feed of online games, wherein the second scrollable feed of online games is tailored to the second end-user based on analysis of past interactions of said second end-user.

2. The computerized method of claim 1, comprising: dynamically constructing a different, user-specific, user-tailored, scrollable feed of online games for displaying on different end-user devices of different users, based on computer-generated estimation of which genre of online games is preferred by each of said different users.

3. The computerized method of claim 1, comprising: dynamically constructing a different, user-specific, user-tailored, scrollable feed of online games for displaying of different end-user devices of different users, based on computer-generated insights regarding user preferences, that are derived from analysis of past interactions of each user with online games.

4. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; based on analysis of differential swiping speed that the first user exhibits when swiping along the first scrollable feed of online games, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games.

5. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; based on analysis of time-measured lingering of the first end-user over a preview of a particular online game that belongs to a particular genre, during scrolling along the first scrollable feed of online games, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games.

6. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; measuring engagement time of the first user, with different online games that belong to different genres, that were presented to the first end-user in the first scrollable feed of online games; based on analysis of differential engagement times that the first end-user exhibits at different online games in the first scrollable feed of online games, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games.

7. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; tracking geo-location data of the first end-user device while the first end-user is scrolling along the first scrollable feed of online games and while the first end-user is engaging with online games; determining a game-to-location correlation between (i) a particular type of online games that the first end-user plays or for which he exhibits lingering engagement, and (ii) geolocation data of the first end-user device; based on said game-to-location correlation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games.

8. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; tracking temporal data of the first end-user device while the first end-user is scrolling along the first scrollable feed of online games and while the first end-user is engaging with online games; determining a time-and-day to game correlation, between (i) a particular type of online games that the first end-user plays or for which he exhibits lingering engagement, and (ii) a temporal characteristic that includes at least one of: time-of-day, day-of-week, day-of-month, in-holiday time; based on said time-and-day to game correlation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games.

9. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; tracking device orientation data of said first end-user device, while the first end-user is scrolling along the first scrollable feed of online games and while the first end-user is engaging with online games, using at least one of: (i) an accelerometer of said first end-user device, (ii) a gyroscope unit of said first end-user device, (iii) a compass unit of said first end-user device, (iv) a spatial orientation sensor of said first end-user device, (v) a spatial angular tilt sensor of said first end-user device; determining a correlation between (i) particular online games for which monitored user interactions exhibit a longer engagement time by the first time-user, and (ii) device orientation data that was sensed during engagement with said particular online games; based on said correlation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games.

10. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; determining a correlation between (i) particular online games for which monitored user interactions exhibit a longer engagement time by the first time-user, and (ii) a particular graphical theme that is featured in said particular online games; based on said correlation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games.

11. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; detecting that the first end-user lingers for a first time-period T1 over a preview of a first particular online game that is displayed in the first scrollable feed of online games; detecting that the first end-user lingers for a second time-period T2 over a preview of a second particular online game that is displayed in the first scrollable feed of online games; wherein T2 is at least twice Tl; generating an estimation that the first end-user prefers online games that belong to a game genre of the second online game over online games that belong to a game genre of the first online game; based on said estimation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to increase a percentage of games that belong to the game genre of the second online game, and to decrease a percentage of games that belong to a game genre of the first online game.

12. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; detecting that the first end-user actively plays for a first time -period Tl with first particular online game that is displayed in the first scrollable feed of online games; detecting that the first end-user actively plays for a second time-period T2 with a second particular online game that is displayed in the first scrollable feed of online games; wherein T2 is at least twice Tl; generating an estimation that the first end-user prefers online games that belong to a game genre of the second online game over online games that belong to a game genre of the first online game; based on said estimation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to increase a percentage of games that belong to the game genre of the second online game, and to decrease a percentage of games that belong to a game genre of the first online game.

13. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; determining which particular online game was played by the first end-user for the longest cumulative time during most-recent N usage sessions, wherein N is a pre-defined integer; generating an estimation that the first end-user prefers online games that belong to a game genre of said particular online game; based on said estimation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to increase a percentage of games that belong to the game genre of said particular online game, and to decrease a percentage of games that belong to other game genres.

14. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; determining which particular online game was played by the first end-user for the longest cumulative time during most-recent N usage sessions, wherein N is a pre-defined integer; and further determining that said particular online game is associated with a particular graphical theme of content; generating an estimation that the first end-user prefers online games that incorporate said particular graphical theme of content that is associated with said particular online game; based on said estimation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to increase a percentage of games that exhibit said particular graphical theme of content, and to decrease a percentage of games that exhibit other graphical themes of content.

15. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; extracting from monitored user interactions of the first end-user, an extracted temporal feature that is at least one of: (a) time-length of actively playing with each game, (b) timelength of watching a preview video or a preview animation of each game;feeding said extracted temporal feature that was extracted from monitored user interactions of the first end-user, into a computerized engine that uses at one of: (i) a pre-trained Machine Learning model, or (ii) a Large Multi-Modalities Model, or (iii) a set of pre-defined deterministic rules; and commanding said computerized engine to generate predictions indicating which particular online games are more likely to be played by the first end-user; based on said predictions, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to populate therein the particular online games that the computerized engine predicts as more likely to be played by the first end-user.

16. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; extracting from monitored user interactions of the first end-user, an extracted contenttheme feature that is at least one of: (a) a graphical theme of characters that are depicted in online games played by the first end-user, (b) a graphical theme of objects that are depicted in online games played by the first end-user; feeding said extracted content-theme feature that was extracted from monitored user interactions of the first end-user, into a computerized engine that uses at one of: (i) a pre-trained Machine Learning model, or (ii) a Large Multi-Modalities Model, or (iii) a set of pre-defined deterministic rules; and commanding said computerized engine to generate predictions indicating which particular online games are more likely to be played by the first end-user; based on said predictions, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to populate therein the particular online games that the computerized engine predicts as more likely to be played by the first end-user.

17. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; extracting from monitored user interactions of the first end-user, a user-engagement feature that is at least one of:(I) a binary signal indicating whether the first end-user watch or did not watch a video preview of particular games,(II) a binary signal indicating whether the first end-user actively played or did not actively play particular games,(III) a binary signal indicating whether the first user played with particular games more than T seconds or not more than T seconds, wherein T is a pre-defined threshold value;(IV) a binary signal indicating whether or not the first user shared with another user a recommendation to play a particular game; feeding said extracted user-engagement feature that was extracted from monitored user interactions of the first end-user, into a computerized engine that uses at one of: (i) a pre-trained Machine Learning model, or (ii) a Large Multi-Modalities Model, or (iii) a set of pre-defined deterministic rules; and commanding said computerized engine to generate predictions indicating which particular online games are more likely to be played by the first end-user; based on said predictions, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to populate therein the particular online games that the computerized engine predicts as more likely to be played by the first enduser.

18. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; based on analysis of monitored user interactions of the first end-user, determining that the first end-user engages for a longer time, on average, with online games that were published by a particular game -publisher, relative to online games that were published by other game-publishers; generating an estimation that the first end-user prefers online games that were published by said particular game-publisher; based on said estimation, dynamically constructing or modifying a subset of online games that populate said first scrollable feed of online games, to increase a percentage of games that were published by said particular game -publisher, and to decrease a percentage of games that were published by other game-publishers.

19. The computerized method of claim 1, comprising: monitoring user interactions of a plurality of different users, that utilize a respective plurality of end-user devices to interact with online games that are presented via different userspecific scrollable feeds of online games;training a Machine Learning model on features extracted from said user interactions of said plurality of different users, and constructing a Machine Learning model that predicts which type of games are more likely to be engaged by a particular end-user that exhibits a particular feature of user interactions; dynamically constructing or modifying a subset of online games that populate a scrollable feed of online games that is displayed to said particular user, based on predictions generated by said Machine Learning model.

20. The computerized method of claim 1, comprising: monitoring user interactions of the first end-user, that utilizes the first end-user device to interact with the first scrollable feed of online games; tracking at least one behavioral data-item that is selected from the group consisting of: (i) audio of the first end-user captured while he interacts with the first scrollable feed of online games, (ii) video of the first end-user captured while he interacts with the first scrollable feed of online games, (iii) one or more images of the first end-user captured while he interacts with the first scrollable feed of online games, (iv) one or more body-gestures of the first end-user captured while he interacts with the first scrollable feed of online games; feeding the user interactions, and the at least one behavioral data-item, into a Large Multi-Modalities Model (LMMM), and prompting the LMMM to generate predictions indicating which online games are more likely to be engaging to the first end-user and which other online games are less likely to be engaging to the first end-user; based on LMMM-generated predictions, dynamically constructing or modifying a subset of online games that populate a scrollable feed of online games that is displayed to said particular user.

21. A system comprising : one or more hardware processors, that are configured to execute code, that are operably associated with one or more memory units that are configured to store code and data; wherein the one or more hardware processors are configured to perform a computerized process comprising:(a) storing on a server data representing a plurality of online games; wherein each online game is associated with (i) a code-portion that runs the online game, and (ii) a screenshot or a thumbnail video that visually demonstrate the online game;(b) causing a first end-user device of a first end-user to display a first scrollable feed of online games, wherein the first scrollable feed of online games is tailored to the first end-user based on analysis of past interactions of said first end-user; and in parallel, causing a second end-user device of a second end-user to display a second, different, scrollable feed of online games, wherein the second scrollable feed of online games is tailored to the second end-user based on analysis of past interactions of said second end-user.

Images