Systems and methods for generating interactive displays of event sequences

By generating real-time event sequences and interactive displays, combined with machine learning models, the problem of low efficiency in real-time sports event data filtering and analysis in existing technologies has been solved, enabling fast, accurate, and customized data processing and user interaction.

CN123125035APending Publication Date: 2026-11-17STAT LLC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202580026688.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2024-04-04
Filing Date
2025-04-02
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing technologies cannot effectively integrate real-time event elements when generating interactive displays of sports event data, and the lack of user interactivity leads to low efficiency in data filtering and analysis.

Method used

By receiving multiple real-time event data, the system generates event sequences and formats interactive displays in real time, including real-time event elements and interactive elements. This allows users to update the displayed content based on user interaction and utilizes machine learning models for data processing and prediction.

Benefits of technology

It enables rapid, accurate, and customized analysis of real-time sports event data, reduces computing resource requirements, and improves user interactivity and data filtering efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123125035A_ABST
    Figure CN123125035A_ABST
Patent Text Reader

Abstract

According to the systems and techniques disclosed herein, a method for generating an interactive display may include receiving a plurality of real-time event data, the plurality of real-time event data including a plurality of real-time event actions associated with a match identifier. The method may further include generating an event sequence based on the plurality of real-time event actions. The method may further include generating the interactive display, the interactive display including at least a graphical representation of the event sequence. The graphical representation of the event sequence may include one or more real-time event elements, and one or more interactive elements. The one or more interactive elements may be configured to update the one or more real-time event elements in response to one or more user interactions. The method may further include transmitting the interactive display to a user interface.
Need to check novelty before this filing date? Find Prior Art

Description

Cross-references to related applications

[0001] This application claims priority to U.S. Provisional Application No. 63 / 574,420, filed April 4, 2024, which is incorporated herein by reference in its entirety. Technical Field

[0002] Various embodiments of this disclosure generally relate to computer-implemented techniques for generating interactive displays of event sequences, and more specifically to systems and methods for generating interactive displays of event sequences including real-time event elements. Background Technology

[0003] Filtering match-based data can typically be limited to set time constraints (e.g., the first and second halves of a football match) or based on specific timecodes (e.g., 10:09 in the first quarter of a match). This filtering does not include a unique visual timeline containing elements of real-time events, which allows users to manipulate match-based data in a fully customizable manner.

[0004] Unless otherwise stated herein, the materials described in this section are not prior art to the claims of this application and are not acknowledged as prior art or teachings of prior art by virtue of their inclusion in this section. Summary of the Invention

[0005] In one aspect, an exemplary embodiment of a method for generating an interactive display may include receiving a plurality of real-time event data, the plurality of real-time event data including a plurality of real-time event actions associated with a competition identifier. The method may further include generating an event sequence based on the plurality of real-time event actions. The event sequence may be generated in real time upon receiving the plurality of real-time event data. The method may further include generating the interactive display, the interactive display including at least a graphical representation of the event sequence. The graphical representation of the event sequence may include one or more real-time event elements and one or more interactive elements. The interactive display may be formatted in real time upon receiving the plurality of real-time event data. The one or more interactive elements may be configured to update the one or more real-time event elements in response to one or more user interactions. The method may further include transmitting the interactive display to a user interface.

[0006] In another embodiment, an exemplary model of a system for generating an interactive display includes: a memory storing instructions; and a processor operatively connected to the memory and configured to execute the instructions to perform operations. The operations may include receiving a plurality of real-time event data, the plurality of real-time event data including a plurality of real-time event actions associated with a competition identifier. The operations may further include generating an event sequence based on the plurality of real-time event actions. The event sequence may be generated in real time upon receiving the plurality of real-time event data. The operations may further include generating the interactive display, the interactive display including at least a graphical representation of the event sequence. The graphical representation of the event sequence may include one or more real-time event elements and one or more interactive elements. The interactive display may be formatted in real time upon receiving the plurality of real-time event data. The one or more interactive elements may be configured to update the one or more real-time event elements in response to one or more user interactions. The operations may further include transmitting the interactive display to a user interface.

[0007] In another embodiment, an exemplary embodiment of a non-transitory computer-readable medium storing instructions that perform operations when executed by one or more processors. The operations may include receiving a plurality of real-time event data, the plurality of real-time event data including a plurality of real-time event actions associated with a competition identifier. The operations may further include generating an event sequence based on the plurality of real-time event actions. The event sequence may be generated in real time upon receiving the plurality of real-time event data. The operations may further include generating the interactive display, the interactive display including at least a graphical representation of the event sequence. The graphical representation of the event sequence may include one or more real-time event elements and one or more interactive elements. The interactive display may be formatted in real time upon receiving the plurality of real-time event data. The one or more interactive elements may be configured to update the one or more real-time event elements in response to one or more user interactions. The operations may further include transmitting the interactive display to a user interface.

[0008] Additional objects and advantages of the disclosed aspects will be set forth in part in the following description, and in part will become apparent from the description, or may be learned by practice of the disclosed aspects. The objects and advantages of the disclosed aspects will be realized and obtained by means of the elements and combinations specifically pointed out in the appended claims.

[0009] It should be understood that the foregoing general description and the following detailed description are merely exemplary and illustrative, and do not limit the claimed aspects of the disclosure. Attached Figure Description

[0010] To gain a detailed understanding of the features described above, reference can be made to the embodiments for a more specific description of the briefly summarized disclosure, some of which are illustrated in the accompanying drawings. However, it should be noted that the drawings only show typical embodiments of the disclosure and should not be considered as limiting the scope of the disclosure, as other equivalent embodiments are permissible.

[0011] Figure 1 This is a block diagram illustrating a computing environment according to an example embodiment.

[0012] Figure 2 An exemplary environment for generating interactive displays according to one or more embodiments is described.

[0013] Figures 3A to 3E An exemplary graphical user interface comprising an interactive display for football matchups, according to one or more embodiments, is depicted.

[0014] Figures 4A to 4G An exemplary graphical user interface comprising an interactive display for a rugby game, according to one or more embodiments, is depicted.

[0015] Figures 5A to 5E An exemplary graphical user interface comprising an interactive display for a basketball game, according to one or more embodiments, is depicted.

[0016] Figure 6 A flowchart depicts an exemplary method for generating an interactive display according to one or more embodiments.

[0017] Figure 7 An exemplary data flow diagram for generating an interactive display is depicted according to one or more embodiments.

[0018] Figure 8 A flowchart for training a machine learning model is depicted according to one or more embodiments.

[0019] Figure 9 An example of a computing device according to one or more embodiments is depicted.

[0020] Figure 10A This is a block diagram illustrating a computing device according to an example embodiment.

[0021] Figure 10B This is a block diagram illustrating a computing device according to an example embodiment.

[0022] It is worth noting that, for the sake of simplicity and clarity, certain aspects of the accompanying drawings depict the overall configuration of various embodiments. Descriptions and details of well-known features and technologies may be omitted to avoid unnecessarily obscuring other features. Elements in the drawings are not necessarily drawn to scale; the dimensions of some features may be exaggerated relative to other features to improve understanding of the exemplary embodiments.

[0023] For ease of understanding, the same reference numerals have been used where possible to refer to the same elements common to the figures. It is contemplated that elements disclosed in one embodiment may be advantageously used in other embodiments without specific description. Detailed Implementation

[0024] The various aspects of this disclosure generally relate to computer-implemented techniques for generating interactive displays that include elements of real-time events. Interactive displays can allow users to filter team and athlete data based on real-time match events. Interactive displays can provide a visual timeline of sporting events (e.g., football matches) and thus allow users to make informed decisions using real-time match events, guided by data visualization, filtering of real-time match events, etc. One or more interactive elements can allow users to visually filter real-time match events, thereby generating substantially new interactive displays based on user interaction.

[0025] Additionally, shifts in match momentum (e.g., match events) can be measured by filtering real-time match events occurring during a live game. In the example, momentum can measure a shift in the game (e.g., at a specific point in time, a match event associated with a particular team becomes increasingly threatening to the opposing team). Therefore, momentum can measure the probability that a team (e.g., the team associated with a match event) will score one or more points within a predetermined amount of time. In various embodiments, momentum can be derived from ball possession value, which measures the impact that each individual match event might have on altering the probability of a team scoring one or more points within a predetermined amount of time. Therefore, data visualization can further guide users to make informed decisions using match-based data.

[0026] In exemplary use cases, multiple (e.g., three) mechanisms can be used to visualize and filter football matches. First, live match events can be filtered by a predetermined time range (e.g., first half, second half, extra time, etc.). Second, live match events can be filtered by event time. In the example, live event data can be filtered based on each individual event or set of events (e.g., goal scored, successful pass, etc.). Third, live match events can be filtered by a visual timeline. In the example, a visual timeline for football matches allows users to view match activities in different modes: event mode, momentum mode, and chance mode. In event mode, multiple unique football events can be mapped onto the visual timeline. In momentum mode, a visual representation of the probability of a team scoring can be shown in an interactive display. In chance mode, values ​​determined using data related to athletes, etc., can be mapped onto the visual timeline or interactive display. In various embodiments, live tabular data can be represented in an interactive display. In the example, as a user interacts with the interactive display, the tabular data can be updated to reflect the user's selections. Such selections can include team statistics, athlete statistics, and athlete statistics broken down by athlete position.

[0027] In another exemplary use case, several (e.g., four) mechanisms can be used to visualize and filter basketball games. First, live game events can be filtered by a predetermined time range (e.g., the entire game, first half, second half, first quarter, second quarter, etc.). Second, live game events can be filtered by scoring runs. In the example, such a running run could be a period of consecutive points scored by one team while the opposing team fails to score, or a period when one team outscores another. Such running runs can be highlighted in a visual timeline or represented as interactive elements in an interactive display. Third, live game events can be filtered by event time. In the example, live game events can be filtered by rebounds, missed shots, substitutions, turnovers, fouls, etc. Fourth, live game events can be filtered by a visual timeline. In the example, event patterns and lead trackers can be separate data visualizations that inform the user when high-intensity activity occurred during the game. In various embodiments, live tabular data can be represented in an interactive display. In the example, the tabular data can be updated to reflect the user's choices as the user interacts with the interactive display. This option can include team statistics, player statistics, shot assist matrix, shot consistency, and player comparison data.

[0028] In another exemplary use case, several (e.g., four) mechanisms can be used to visualize and filter rugby matches. First, live match events can be filtered by a predetermined time frame (e.g., the entire game, first half, second half, first quarter, second quarter, etc.). Second, live match events can be filtered by event time (e.g., first ten-yard shot). Third, live match events can be filtered by offensive possession. In the example, a user can group offensive possessions by whether they scored or not, or by user-selected offensive possession groups. Fourth, live match events can be filtered by a visual timeline. In the example, a visual timeline can be generated based on whether a team scored or not, or using winning probabilities. In various embodiments, live tabular data can be represented in an interactive display. In the example, as a user interacts with the interactive display, the tabular data can be updated to reflect the user's selections. Such selections can include team statistics, player statistics, and team efficiency data.

[0029] In various embodiments, the interactive display may include a zoom control that allows the user to interact with the visual timeline. This allows the user to zoom in or out on one or more data points, enabling precise segmentation of the data to view the granularity of time stamps for certain live match events by visually increasing the spacing between them.

[0030] Furthermore, the generated data (such as interactive elements and interactive displays) can be used as input to machine learning or artificial intelligence models. In the example, a user can interact with a language learning model (LLM) trained to respond to user questions or interactions using the generated data output. In this way, a user can interact with real-time match event data and live output (e.g., the generated interactive displays).

[0031] Therefore, this disclosure also provides performance prediction techniques based on machine learning. Additionally, using artificial intelligence-based techniques for natural language processing allows users to interact with the data. It can also reduce the logistical and financial challenges and / or undesirable outcomes associated with determining odds for expected athlete performance. More specifically, the techniques disclosed herein for generating displayed statistics for each athlete or team provide faster, real-time, more accurate, more efficient, and personalized processing of match event data compared to conventional techniques. The techniques disclosed herein further reduce the computational resources required for such processing by, for example, leveraging machine learning training to reduce the immediate processing load.

[0032] As used herein, a “machine learning model” generally encompasses instructions, data, and / or a model configured to receive input and apply one or more of weights, biases, classifications, or analyses to the input to generate an output. The output may include, for example, a classification of the input, an input-based analysis, a design, process, prediction, or recommendation associated with the input, or any other suitable type of output. Machine learning models are typically trained using training data (e.g., empirical data and / or samples of input data) that are fed into the model to build, tune, or modify one or more aspects of the model, such as weights, biases, criteria used to form classifications or clusters, etc. Aspects of a machine learning model may operate on the input in a linear, parallel, network-based (e.g., neural network), or via any suitable configuration.

[0033] As discussed in this paper, one or more machine learning models can be trained to understand the language of sports. Therefore, the machine learning models disclosed herein are sports machine learning models. Such sports machine learning models can be trained using sports-related data (e.g., tracking data, event data, etc., as discussed herein). A sports machine learning model trained to understand the language of sports based on sports-related data can be trained to adjust one or more weights, layers, nodes, biases, and / or synapses based on sports-related data. A sports machine learning model may include components that collectively relate one or more of the following (e.g., weights, layers, nodes, biases, and / or synapses): athletes and teams or leagues; teams and athletes or leagues; scores and teams; scoring events and athletes; sporting events and athletes or teams; victories and athletes or teams; losses and athletes or teams; and so on. Sports machine learning models can establish correlations between sports information and statistics within a competitive context. A sports machine learning model can be trained to adjust one or more weights, layers, nodes, biases, and / or synapses to correlate certain sports statistics based on the competitive context. For example, a team's victory metric can be automatically correlated with the opposing team's loss metric. As another example, score statistics can be viewed as a positive attribution for scoring teams and a negative attribution for losing teams. As yet another example, a given score can be ranked relative to one or more other scores based on its relative position compared to those scores.

[0034] As discussed in this paper, sports machine learning models can be trained on sports tracking and / or event data. This data can include athlete and / or object location information, movement information, trends, and changes. For example, a sports machine learning model can be trained by modifying one or more weights, layers, nodes, biases, and / or synapses to associate a given location with respect to the playing field of a venue and / or with one or more active subjects. As another example, a sports machine learning model can be trained by modifying one or more weights, layers, nodes, biases, and / or synapses to associate a given movement or trend with respect to the playing field of a venue and / or with one or more active subjects. As yet another example, a sports machine learning model can be trained by modifying one or more weights, layers, nodes, biases, and / or synapses to associate sporting events with corresponding time boundaries, teams, athletes, coaches, referee teams, and environmental data associated with the location of the sporting event.

[0035] Sports machine learning models can be trained by modifying one or more weights, layers, nodes, biases, and / or synapses to correlate position, movement, and / or trend information based on sports objectives. Sports objectives can be scoring-related objectives (e.g., points, goals, shots, number of shots, score value, etc.), game round results (e.g., passes, object (e.g., ball) movement, player positions, etc.), player positions, etc. Sports machine learning models can be trained based on sports objectives, game round results, player positions, etc., associated with a given sport (e.g., football, American football, basketball, baseball, tennis, golf, rugby, hockey, team sports, individual sports, etc.). For example, a football-based sports machine learning model can be trained to establish correlations or other associations of player position information relative to the football field. A football-based sports machine learning model can be further trained to establish correlations or other associations of sports data relative to the specific number of players and sports objectives in football.

[0036] Depending on various factors, one or more types of sports machine learning models (e.g., generative learning, linear regression, logistic regression, random forest, gradient boosting machine (GBM), deep learning, graph neural network (GNN), and / or deep neural network) can be determined based on the attributes of a given sport to which one or more machine learning models are applied. Attributes may include, for example, sport type (e.g., individual versus team sport), sport boundaries (e.g., time factors, number of athletes, object factors, possession periods (e.g., overlapping or independent), playing field type (e.g., restricted, unrestricted, virtual, real, etc.), athlete position, etc.

[0037] Depending on the context, a sports machine learning model can receive input including sports data for a given sport and can generate a matrix representation based on the features of that sport. The model can be trained to determine latent features for a given sport. For example, the matrix may include fields and / or subfields related to athlete information, team information, object information, sports boundary information, sports venue information, etc. Attributes associated with each field or subfield can be populated into the matrix based on received or extracted data. The model can then perform operations based on the generated matrix. Features can be updated based on the input data or updated training data, which is based on sports data, for example, features that the model was not previously trained to associate with the given sport. Therefore, the model can be trained iteratively based on sports data or simulated data.

[0038] The execution of a machine learning model can include deploying one or more machine learning techniques, such as transformer models, graph neural networks (GNNs), linear regression, logistic regression, random forests, gradient boosting machines (GBM), deep learning, and / or deep neural networks. Supervised and / or unsupervised training can be employed. For example, supervised learning can include providing training data and corresponding labels, such as real data. Unsupervised methods can include clustering, classification, etc. K-means clustering or K-nearest neighbors can also be used, and these can be supervised or unsupervised. A combination of K-nearest neighbors and unsupervised clustering techniques can also be used. Any suitable type of training can be used, such as randomized training, gradient boosting training, random seed training, recursive training, epoch-based training, or batch-based training.

[0039] While some examples in this document relate to certain types of machine learning and artificial intelligence, it should be understood that the techniques based on this disclosure are applicable to any suitable type of machine learning and / or artificial intelligence. It should also be understood that the examples above are illustrative only. The techniques and solutions disclosed herein are applicable to any suitable activity.

[0040] Although sporting events and aspects related to sporting events (e.g., events during a sporting event) are described as illustrative examples in various aspects of this document, the aspects of this document are not limited to such examples. For example, the aspects of this document can be implemented for other types of events or actions, such as financial activities, activities involving strategies and predictions of actions to be taken, or other implementations in which the performance of individuals, groups of individuals, or objects is evaluated or predicted.

[0041] While sporting events and aspects related to sporting events can be described with respect to a given sport, it should be understood that these aspects can be applied to any applicable sport, such as, but not limited to, team sports, individual sports, football, basketball, American football, rugby, golf, tennis, hockey, cricket, etc.

[0042] According to the embodiments disclosed herein, a transformer neural network can receive inputs (e.g., tensor layers), where each input corresponds to a given athlete, team, or match. The transformer neural network can output a generated prediction for one or more given athletes or teams based on such inputs. More specifically, the transformer neural network can output this generated prediction for a given athlete or team based on inputs associated with that athlete or team and further based on the influence of one or more other athletes or teams. Thus, as discussed herein, the predictions provided by the transformer neural network can take into account the influence of multiple athletes and / or teams when outputting predictions for a given athlete and / or team.

[0043] The systems described herein may include machine learning systems configured to generate one or more predictions. In some examples, the system may comprise transformer neural networks, graph neural networks, recurrent neural networks, convolutional neural networks, and / or feedforward neural networks. The system may implement a series of neural network instances (e.g., feedforward network (FFN) models) connected via a transformer neural network (e.g., a graph neural network (GNN) model). Although transformer neural networks are discussed in general herein, it should be understood that any applicable GNN or other neural network that can be graphically interpreted can be used to perform the techniques discussed herein with reference to transformer neural networks.

[0044] A transformer-based neural network can include an array of linear embedding layers, a transformer encoder, and an array of fully connected layers. The linear embedding layers map the component tensors of the received input to tensors with a common feature dimension. The transformer encoder performs attention along the time and agent dimensions. The fully connected layers embed the output from the last transformer layer of the transformer encoder into a tensor with the requested feature dimension for each target metric.

[0045] Transformer-based neural networks can be configured to receive input features through a set of linear embedding layers. These input features can be received at different resolutions and in time-series format. The input features can relate to athlete features, team features, and / or match features. The input features can be fed into the linear embedding layers as tuples of input tensors. For example, a tuple of three tensors can be provided, where the first tensor corresponds to all athletes in a match, the second tensor corresponds to the two teams in the match, and the third tensor corresponds to the match state.

[0046] Regarding the linear embedding layer set, the linear embedding layer can contain linear blocks for each input tensor of the tuple, and each block can map the input tensor to tensors with a common feature dimension D. The output of the linear embedding layer can be a tuple of tensors with a common feature dimension, which can be concatenated along the time dimension and the activity subject dimension to form a single tensor.

[0047] A Transformer encoder can be configured to receive a single tensor from a linear embedding layer. The Transformer encoder can be configured to learn embeddings configured to generate predictions for multiple actions per agent (e.g., each athlete and / or team). The Transformer encoder can include a series of axial Transformer encoder layers, each layer alternately applying attention along the temporal and agent dimensions. The Transformer encoder can include layers that alternately apply attention temporally to sequences of action events and spatially to sets of athletes and teams at each event time step. The Transformer encoder can include axial encoder layers configured to receive tensors from linear layers and apply attention along the temporal dimension and then along the agent dimension.

[0048] The attention mechanism implemented by the transformer encoder layer can be graphically interpreted on a dense graph, where each element is a node, and the attention mask is the inverse of the adjacency matrix defining the edges between nodes (therefore, when no attention mask is present, it means the graph is fully connected). In the case of axial attention used here, when the attention mask is applied to the time (row) dimension, the nodes in the graph can be arranged in a grid, and each node can be connected to all nodes in the same column and to all previous nodes in the same row. In this case, attention can be message-passing, where each node can receive messages describing the state of nodes in its neighborhood and then update its own state based on those messages. This attention scheme might mean that when making predictions for a particular athlete, the model can consider (i.e., pay attention to) nodes containing the athlete's previous states along the time series; and state nodes of other athletes, teams, and the current match state at the current time step. Nodes may not need to be homogeneous except for having the same feature dimensions; thus, a node representing an athlete can receive messages from nodes representing teams or from nodes representing athlete strength. Therefore, the model can learn the interactions between actors and ensure consistent predictions for each actor along the time series. The output of the Transformer encoder layer can be a tensor (e.g., an output embedding).

[0049] The final layer of a transformer-based neural network can be a fully connected layer. These layers can embed the output of the final transformer layer of the transformer encoder and map it to the feature dimension of each target metric. The final layer can output a target tuple containing a tensor for each modeled action in the modeled action set for each athlete and / or team. For example, the modeled actions can be empirical estimates of the distribution of sports statistics such as shots, goals, passes, etc.

[0050] Training a transformer-based neural network can involve selecting a corresponding loss function for each output target's distribution assumption. For example, for a Poisson distribution, the loss function could be the negative Poisson log-likelihood, and for a Bernoulli distribution, it could be the binary cross-entropy. During training, the loss can be calculated based on the true data values ​​of each target in the training set, and the loss values ​​can be summed. The optimizer can then update the model weights based on the total loss. The learning rate can be adjusted using a scheduling strategy with cosine annealing and no hot restart.

[0051] As discussed above, conventional user interfaces have many drawbacks related to the efficient operation of computers, thus offering users only limited options (e.g., filtering based on set time limits). Various aspects of this disclosure can address this problem of conventional user interfaces by providing improved user interfaces that allow users to more quickly access desired data processed / filtered / generated based on real-time or near real-time data stored in electronic systems / devices / applications.

[0052] Various aspects of this disclosure can improve the efficiency of electronic systems / devices / applications. For example, by providing various options to filter diverse sports data in real-time or near real-time and display the selected data in a single screen / window / view, users can see the most relevant data or functions without having to review all real-time data or calculate / process it separately. These aspects can also improve the speed at which users browse relevant data by allowing them to focus on relevant data using various filtering options, thereby saving computer resources and increasing computer processing speed.

[0053] Figure 1 This is a block diagram illustrating a computing environment 100 according to an example embodiment. The computing environment 100 may include a tracking system 102 (e.g., located at or communicating with one or more components located at the venue) communicating via a network 105, an organizational computing system 104, and one or more client devices 108.

[0054] Network 105 can be of any suitable type, including a standalone connection via the Internet, such as a cellular network or a Wi-Fi network. In some embodiments, network 105 can use direct connections to connect terminals, services, and mobile devices, such as radio frequency identification (RFID), near field communication (NFC), Bluetooth™, Bluetooth Low Energy (BLE), Wi-Fi™, ZigBee™, ambient backscatter communication (ABC) protocol, USB, WAN, or LAN. Because the information transmitted may be personal or confidential, it may be necessary to encrypt or otherwise protect one or more of these connection types for security reasons. However, in some embodiments, the information transmitted may be less personal, and therefore, the choice of network connection may prioritize convenience over security.

[0055] Network 105 may include any type of computer networking arrangement for exchanging data or information. For example, network 105 may be the Internet, a private data network, a virtual private network using a public network, and / or other suitable connections that enable components in computing environment 100 to send and receive information between components of environment 100.

[0056] Tracking system 102 may be located within venue 106 and / or may communicate with components located within venue 106 (e.g., electronic communication, wireless communication, wired communication, etc.). For example, venue 106 may be configured to host a sporting event including one or more active subjects 112. Tracking system 102 may be configured to capture the movement of one or more active subjects (e.g., athletes) and one or more other associated active subjects (e.g., objects) (e.g., a ball, hockey, referees, etc.) on the playing field. In some embodiments, tracking system 102 may be an optical-based system using, for example, multiple fixed cameras, movable cameras, one or more panoramic cameras, etc. For example, a system consisting of six calibrated cameras (e.g., fixed cameras) may be used to project the three-dimensional positions of the athletes and the ball onto a two-dimensional top view of the playing field. In another example, a combination of fixed and non-fixed cameras may be used to capture the movement of all active subjects on the playing field as well as one or more associated objects. Using such a tracking system (e.g., tracking system 102), many different camera views of the playing field can be generated (e.g., high sideline view, free throw line view, player encirclement view, face-off view, end zone view, etc.).

[0057] In some embodiments, the tracking system 102 can be used for the broadcast signal stream of a given match. For example, the tracking system 102 can be used to generate a match file 110 to facilitate the broadcast signal stream of a given match. In this embodiment, each frame of the broadcast signal stream can be stored in the match file 110. The broadcast signal stream can be a signal stream formatted for broadcast through one or more channels (e.g., broadcast channels, internet-based channels, etc.). The match file 110 can be converted from a first format (e.g., a format output by one or more cameras or a format different from that output by one or more cameras) and can be converted to a second format (e.g., for broadcast transmission).

[0058] In some embodiments, the game file 110 may be further augmented with other event information corresponding to the event data, such as, but not limited to, game event information (pass, shot made, turnover, etc.) and contextual information (current score, remaining time, etc.). According to embodiments, the event data may be generated manually or by a computing system in real time (e.g., within approximately 30 seconds of the event occurring), as discussed herein. The computing system may generate the event data by, for example, analyzing tracking data (e.g., from tracking system 102) and / or one or more other data types (e.g., video streams, excitement data, etc.). The computing system may utilize machine learning models to determine when a given set of tracking data or a change in tracking data (e.g., a given athlete movement, object movement, its change, etc.) corresponds to an event (e.g., a scoring event, a penalty event, a ball possession-based event, a game play type event, etc.). Machine learning may be used to automatically identify the event data, trained to receive the game file 110 or a subset thereof as input and output game information and / or contextual information based on that input. According to the techniques disclosed herein, the machine learning model may be trained using supervised, semi-supervised, or unsupervised learning. Machine learning models can be trained by analyzing training data using one or more machine learning algorithms, as disclosed in this paper. Training data may include match files or simulated match files from historical matches, simulated matches, etc., and may include labeled and / or unlabeled data.

[0059] According to the embodiments disclosed herein, event data can be generated based on tracking data and / or content signal streams (e.g., in-venue video signal streams, broadcast signal streams, etc.). For example, tracking data can be generated by providing the content signal stream to one or more machine learning models. One or more machine learning models can identify athletes and / or objects in the content signal stream and convert them into digital representations. The digital representations of athletes and / or objects and their corresponding positions can be tracked to identify tracking data, such as movement data (e.g., position changes), movement variations, trends, etc. A prediction module can use this information to make predictions. The machine learning models can analyze the tracking data to determine the correlation between the tracking data and event types (e.g., goal score, pass, match round type, etc.). For example, tracking data can be used to determine when the digital representation of an object (e.g., the ball) crosses a scoring object (e.g., the goalpost). Based on this determination, the event type of a goal score can be identified. Furthermore, the digital representations of athletes who touched the object (e.g., the ball) before the goal score event can be identified as athletes who contributed to or otherwise caused the event (e.g., the goal). Therefore, the content signal stream can be used to generate tracking data, which can then be used to determine event data corresponding to certain sports events.

[0060] Tracking system 102 can be configured to communicate with organizational computing system 104 via network 105. For example, tracking system 102 can be configured to provide broadcast streams of matches or events to organizational computing system 104 in real-time or near real-time via network 105. As an example, tracking system 102 can provide one or more match files 110 in a first format (e.g., a format corresponding to a component based on tracking system 102). Alternatively or additionally, tracking system 102 or organizational computing system 104 can convert the broadcast stream (e.g., match file 110) from the first format to a second format. The second format can be based on organizational computing system 104. For example, the second format can be a format associated with data storage device 118, as discussed further herein.

[0061] The organization computing system 104 can be configured to process broadcast streams of matches. The organization computing system 104 may include at least a web client application server 114, a tracking data system 116, a data storage device 118, a round-by-round module 120, a filling module 122, and / or an interactive display module 124. Each of the tracking data system 116, the round-by-round module 120, the filling module 122, and the interactive display module 124 may consist of one or more software modules. One or more software modules may be a collection of code or instructions stored on a medium (e.g., the memory of the organization computing system 104) representing a series of machine instructions (e.g., program code) that implement one or more algorithmic steps. Such machine instructions may be actual computer code interpreted by the processor of the organization computing system 104 to implement the instructions, or alternatively, may be a higher-level encoding of the instructions interpreted to obtain actual computer code. One or more software modules may also include one or more hardware components. One or more aspects of the example algorithm may be executed by the hardware component (e.g., circuitry) itself, rather than as a result of instructions.

[0062] The tracking data system 116 can be configured to receive broadcast data from the tracking system 102 and generate tracking data based on the broadcast data. In some embodiments, the tracking data system 116 may employ an artificial intelligence and / or computer vision system configured to obtain athlete tracking data from the broadcast video signal stream.

[0063] To generate tracking data based on broadcast data, the tracking data system 116 may, for example, map pixels corresponding to each athlete and ball to points, and transform these points into a semantically meaningful event layer that can be used to describe athlete attributes. For example, the tracking data system 116 may be configured to ingest broadcast video received from the tracking system 102. In some embodiments, the tracking data system 116 may further classify each frame of the broadcast video into trackable and untrackable segments. In some embodiments, the tracking data system 116 may further calibrate the moving camera based on trackable and untrackable segments. In some embodiments, the tracking data system 116 may further use skeleton tracking to detect athletes within each frame. In some embodiments, the tracking data system 116 may further track and re-identify athletes over time. For example, the tracking data system 116 may re-identify athletes who are not in the camera's line of sight during a given frame. In some embodiments, the tracking data system 116 may further detect and track objects across multiple frames. In some embodiments, the tracking data system 116 may further utilize optical character recognition technology. For example, the tracking data system 116 can use optical character recognition technology to extract score information and remaining time information from the digital scoreboard of each frame.

[0064] This technology helps the tracking data system 116 generate tracking data based on the broadcast signal stream (e.g., broadcast video data). For example, the tracking data system 116 can perform this process to generate tracking data covering thousands of ball possession and / or broadcast frames. In addition to this process, the organization computing system 104 is not limited to generating tracking data based on broadcast video data. Instead, to provide descriptive analysis and useful feature representations for the interactive display module 124, the organization computing system 104 can be configured to map tracking data to a semantic layer (e.g., events).

[0065] The tracking data system 116 can be implemented using a machine learning model. According to the techniques disclosed herein, the machine learning model can be trained using supervised, semi-supervised, or unsupervised learning. The machine learning model can be trained by analyzing training data using one or more machine learning algorithms, as disclosed herein. Training data may include match files or simulated match files from historical matches, simulated matches, historical or simulated feature representations, etc., and may include labeled and / or unlabeled data. Labeled data may include location information, movement information, object information, trends, active entity identifiers, active entity re-identifiers, etc.

[0066] The turn-by-turn module 120 can be configured to receive turn-by-turn data from one or more third-party systems. For example, the turn-by-turn module 120 can receive a turn-by-turn signal stream corresponding to broadcast video data. In some embodiments, the turn-by-turn data can represent artificially generated data based on events that occur during the game. Although the goal of computer vision technology is to capture all data directly from the broadcast video stream, in some cases, the referee is still the final decision-maker regarding whether an event's outcome is valid. For example, in basketball, whether a shot is a two-pointer or a three-pointer (or whether it is valid, whether it is a traveling violation, whether there is a defensive / offensive foul, etc.) is decided by the referee. Therefore, in order to capture these data points, the turn-by-turn module 120 can utilize machine learning outputs and / or artificially labeled data that can reflect the referee's final decision. This data can be referred to as a turn-by-turn signal stream.

[0067] To aid in identifying events within the generated tracking data, the tracking data system 116 can merge or align the turn-by-turn data with the original generated tracking data (which may include match and time fields). The tracking data system 116 can utilize a fuzzy matching algorithm that combines turn-by-turn data, optical character recognition data (e.g., attack timer, score, remaining time, etc.), and match / ball position (e.g., the original tracking data) to generate aligned tracking data.

[0068] After alignment is complete, the tracking data system 116 can be configured to perform various operations on the aligned tracking system. For example, the tracking data system 116 can use per-round data to refine the positions of the player and the ball, as well as the precise frame of the end-of-control event (e.g., shot / rebound position). In some embodiments, the tracking data system 116 can be further configured to automatically detect events from the tracking data. In some embodiments, the tracking data system 116 can be further configured to enhance events using contextual information.

[0069] For automated event detection, the tracking data system 116 may include a neural network system trained to detect / refine various events in a sequential manner. For example, the tracking data system 116 may include an actor-action attention neural network system for detecting / refining one or more of the following: shooting, scoring, points, rebounding, passing, dribbling, calls, fouls, and / or ball possession. The tracking data system 116 may further include a large number of specialized event detectors trained to identify higher-level events. Exemplary higher-level events may include, but are not limited to, game possession, offensive / defensive transition, pressing, crossing, fast break, post-up, dribbling, isolation, screen, offside, handoff, off-ball screen, etc. In some embodiments, each specialized event detector may represent a neural network specifically trained to identify a particular event type. More generally, such event detectors can utilize any type of detection method. For example, specialized event detectors may use neural network methods or another machine learning classifier (e.g., random decision forest, SVM, logistic regression, etc.).

[0070] While mapping tracking data to events enables the capture of athlete representation, the tracking data system 116 can generate contextual information to enhance detected events in order to further construct the most comprehensive possible athlete representation. Exemplary contextual information may include defensive matchup information (e.g., who is guarding whom in each frame, defensive formation) and other defensive information, such as defensive strategies against screens or pressure.

[0071] In some embodiments, to measure influence, the tracking data system 116 may use a metric called an "influence score." An influence score captures the influence an athlete can have on each other athlete on the opposing team, ranging from 0 to 100. In some embodiments, the influence score value may be based on sports principles such as, but not limited to, proximity to an athlete, distance to a scoring target (e.g., the basket, goal, boundary, etc.), gap convergence rate, passing routes, routes to the scoring target, etc.

[0072] The imputation module 122 can be configured to use mean regression to create new athlete representations to reduce random noise in the features. For example, one of the major challenges of modeling with data that may only represent a limited number of matches per athlete (e.g., 20 to 30 matches) could be the high variance of low-frequency events in the tracking data. Therefore, the imputation module 122 can be configured to utilize an imputation method that could be a weighted average between the observations and the sample mean.

[0073] Therefore, for each athlete, the tracking data system 116, the round-by-round module 120, and the padding module 122 can work together to generate the raw dataset and the padding dataset for each athlete.

[0074] The interactive display module 124 can be configured or trained to generate an interactive display of event sequences. (See the description below.) Figure 2 Further discussion of interactive display module 124.

[0075] Data storage device 118 can be configured to store one or more match files 126. Each match file 126 may include video data for a given match. For example, the video data may correspond to multiple video frames captured by tracking system 102, tracking data generated from broadcast video by tracking data system 116, round-by-round data, enriched data, and / or padded training data. Match files 126 may be based, for example, on match file 110 as discussed herein. Match files 126 may be in a different format than match file 110. For example, a first-format match file 110 or a subset thereof may be transformed into a second-format match file 126. This transformation may be performed automatically based on the type and / or content of the first format and the type and / or content of the second format.

[0076] Client device 108 can communicate with organizational computing system 104 via network 105. Client device 108 can be operated by a user. For example, client device 108 can be a mobile device, tablet computer, desktop computer, or any computing system with the capabilities described herein. Users can include, but are not limited to, individuals such as subscribers, customers, potential customers, or clients of entities associated with organizational computing system 104, such as individuals who have received, will receive, or may receive products, services, or advice from entities associated with organizational computing system 104.

[0077] Client device 108 may include at least application 130. Application 130 may represent a web browser that allows access to a website, or it may be a standalone application. Client device 108 may access application 130 to access one or more functions of organizational computing system 104. Client device 108 may communicate via network 105 to request web pages, for example, from web client application server 114 of organizational computing system 104. Content displayed to client device 108 may be transmitted from web client application server 114 to client device 108, and then processed by application 130 for display via graphical user interface (GUI) of client device 108.

[0078] A prediction engine (e.g., which could be part of a data tracking system) can be configured to predict the basic formation of a team. Mathematically, the goal of the role alignment process could be to find a transformation... This transformation can Unstructured set of athlete trajectories Mapped to individual character trajectories An ordered set (e.g., a vector). Each athlete's trajectory itself can be an ordered set of positions. ,in As the main body of the event, A frame. In some embodiments, This can represent the optimal permutation matrix that achieves this sorting. The goal of the prediction engine could be to find the most probable set of two-dimensional (2D) probability density functions. :

[0079] In some embodiments, the equation can be transformed into an entropy minimization problem, where the objective is to reduce (e.g., minimize) the overlap (e.g., KL divergence) between each role. Therefore, in some embodiments, the total entropy is used as the basis for the problem. The final optimization equation can be expressed as:

[0080] The prediction engine can include a formation discovery module, a role assignment module, and a template module, each corresponding to a different stage of the prediction process. The formation discovery module can be configured to learn a distribution that maximizes the likelihood of the data. The role assignment module can be configured to map each athlete's position to a "role" distribution in each frame. Once the data is aligned, the template module can be configured to map each learned formation to a formation clustering template.

[0081] The organizational computing system can receive tracking data and / or event data for multiple events across multiple seasons or across a single match. For each event, a preprocessing agent can divide the match into multiple segments based on event information. In some embodiments, the preprocessing agent can divide the match into multiple segments based on various events that may occur throughout the match. For example, the preprocessing agent can divide the match into multiple segments based on one or more events, including but not limited to red cards, ejections, technical fouls, flagrant fouls, player disqualifications, substitutions, halftime, quarters, periods, overtime, etc. Generally, each of the multiple segments associated with the match can include an interval of necessary duration (e.g., at least one minute of playing time, at least two minutes of playing time, etc.). This necessary duration allows the organizational computing system to detect team formations.

[0082] Each segment may include its associated tracking dataset. Athlete tracking data can be captured by a tracking system configured to record the athlete's movements at a high frame rate (e.g., 10 Hz). Location. In some embodiments, athlete tracking data may further include single-frame event labels (e.g., pass, shot, cross) in each frame of the athlete tracking data. These frames may be referred to as “event frames”. As shown in the figure, the initial athlete tracking data can be represented as A collection of athlete trajectories Each athlete's trajectory can itself be an ordered set of positions. ,in As the main body of the event, For frames.

[0083] In some embodiments, the preprocessing agent can normalize the raw position data of the athletes. For example, the preprocessing agent can normalize the raw position data of the athletes in each segment so that all teams in the athlete tracking data are attacking from left to right and have a zero mean in each frame. This normalization can eliminate translation effects in the data. This can produce a set .

[0084] In some embodiments, the preprocessing agent can initialize the cluster centers of a normalized dataset using the average athlete position for pattern discovery. For example, the average athlete position can be determined by a set of... = { This indicates that the preprocessing agent can take the average position of each athlete in the normalized data and initialize the normalized data based on the average athlete position. This initialization of the normalized data based on the average athlete position can act as the initial position for each athlete to minimize data variance.

[0085] The organizational computing system can learn a pattern template from the tracking data of each segment. For example, the pattern discovery module can learn a distribution that maximizes the likelihood of the data. The pattern discovery module can structure the initial data into a single... vector, where, It can represent the total number of frames. It can represent the total number of participants in an activity (e.g., ten field players in football, five players in basketball, fifteen players in rugby, etc.), and It can represent the dimensions of the data (e.g., ).

[0086] Then, the formation discovery module can initiate the formation discovery algorithm. For example, the formation discovery module can initialize the K-means algorithm using the average position of the athletes and execute it until convergence. Executing the K-means algorithm until convergence produces better results than the conventional method of running a fixed number of iterations.

[0087] Then, the formation discovery module can initialize a Gaussian mixture model (GMM) using the cluster centers from the last iteration of the K-means algorithm. By parameterizing the distribution as a mixture of K Gaussians (where K equals the number of "roles," typically also N, i.e., the number of athletes), the formation discovery module can identify patterns in the data. The optimal matrix for likelihood maximization. In other words, the GMM can be configured to recognize... ,in, It can represent making data The optimal formation is obtained by maximizing the likelihood. Therefore, the formation discovery module does not stop the process after the last iteration of the K-means algorithm, but can use GMM clustering because elliptic clustering captures the shape of each athlete's role better than K-means clustering, which captures only the spherical nature of the data cloud for each role.

[0088] Furthermore, it is known that GMMs suffer from component collapse and may get trapped in ill-conditioned solutions. This collapse can lead to unreasonable clustering, such as meaningless outputs that may become unusable. To address this issue, the pattern discovery module can be configured to monitor the eigenvalues ​​of each component or parameter of the GMM throughout the expectation-maximization process. If the pattern discovery module determines that the eigenvalue ratio of any component becomes too large or too small, the next iteration can run a soft K-means update (e.g., a Gaussian mixture with spherical covariance) instead of a full covariance update. This process can be performed to ensure that the final clustering output is reasonable. For example, the pattern discovery module can monitor how the parameters of the GMM converge; if the parameters of the GMM are unstable (e.g., “out of control”), the pattern discovery module can identify this unstable behavior and then use a soft K-means update to slowly constrain the parameters back into the solution space.

[0089] For retrieval tasks involving large amounts of data, the implementation of the system requires the use of hash tables. By grouping similar match rounds together, only the "most likely" candidates are retrieved when a query is made. Comparisons can then be made locally between the candidates, and each match round within these groups is sorted by the degree of similarity. Previous systems attempted to cluster match rounds into similar groups using only one attribute (such as the trajectory of the ball). However, the semantics of match rounds can be captured more accurately using additional information, such as information about players (e.g., identity, trajectory, etc.) and events (passing, dribbling, shooting, etc.) as well as contextual information (e.g., whether the team is leading or trailing, how much time remains, etc.). Therefore, the implementation of this system utilizes information about the trajectory of the ball and players, as well as match events and context, to create hash tables, thereby effectively learning a "match round handbook" of representative match rounds of team or player behavior. This match round handbook is learned by selecting classification metrics that indicate interesting or discriminative match rounds. Suitable classification metrics may include the probability of predicting a score in football or basketball (e.g., expected score value ("EPV") or expected goal value ("EGV")). Other predictors can be chosen for performance variables, such as the probability of passing, the probability of shooting, the probability of moving along a specific direction / trajectory, or the probability of athlete fatigue / injury.

[0090] Categorical metrics are used to learn decision trees, a coarse-to-fine hierarchical approach where a question is posed at each node to group the data. The advantage of this approach is that it can be interpretable and multi-layered, acting as a "latent factor."

[0091] In this system implementation, a bottom-up approach is used to learn the decision tree. Various features are used sequentially to distinguish game rounds (e.g., first the ball, then the player closest to the ball, then the defensive player, etc.). By aligning trajectories, a reference point exists for each trajectory relative to its current position. This allows for asking more specific questions while maintaining generality (e.g., if a player is playing the "point guard" role, what are the distances between him / her and his / her teammates playing the "shooting guard" role, and between him / her and the defensive player playing the "point guard" role). Using this method avoids the need to exhaustively check all distances, a computationally intensive process extremely demanding in both basketball and soccer.

[0092] In another embodiment of the system, a top-down approach is used to learn the decision tree. In the first step, all match rounds are aligned with a template set. From this initial template set, using all ball and player information, match rounds are assigned to a set of K groups (clusters), thus forming the first layer of the decision tree. Then, backpropagation is used to prune unimportant players and divide each cluster into sub-clusters (second layer). The method continues until the leaf nodes of the tree represent a dictionary of match rounds that have predictive power for a specific task (e.g., scoring a goal) (third layer).

[0093] In addition to the raw trajectory information, in this embodiment of the system, the game rounds in the database are also associated with game event information and context information. The game events and context corresponding to a particular game round in the database can be inferred directly from the raw position tracking data (e.g., shot made or missed) or manually entered. Athlete role information (also inferred from position tracking data or entered separately) can then be used to train a model for the database by constructing features that encode game-specific information based on position and game data, and then calculating predicted values ​​(between 0 and 1) relative to a classification metric (e.g., expected score).

[0094] If a sufficient number of examples are available, these examples can be used to personalize the database model for specific athletes or match situations. However, in practice, a specific athlete or match situation may not be adequately represented by the match rounds in the database. Therefore, embodiments of this system can find examples similar to the situation of interest—whether it's finding athletes with similar characteristics or teams with similar playing styles. Using a more general representation of athletes and / or teams, each athlete or team is thus represented as a distribution of specific attributes, rather than using explicit team identities. Embodiments of this system use match rounds from a hash table / match round handbook learned through the distributed clustering process described above.

[0095] Furthermore, while various aspects have been discussed in relation to a particular sport, these aspects described are merely illustrative examples. The disclosed techniques are by no means limited to any specific sport. For example, these aspects can be implemented in other sports or activities, such as football, rugby, basketball, baseball, hockey, cricket, rugby union, tennis, etc.

[0096] Figure 2 Another exemplary environment 150 is depicted that can be used with the techniques presented herein. One or more user devices 162 can communicate via electronic network 160. One or more user devices 162 can be associated with a user (e.g., a user viewing and / or interacting with a generated interactive display, an administrator of one or more components of environment 150, etc.).

[0097] As will be discussed in further detail below, one or more interactive display computing systems 152 can communicate with one or more other components of the environment 150 via electronic network 160.

[0098] User device 162 may be configured to enable a user to access and / or interact with other systems in environment 150. For example, user device 162 may each be a computer system, such as a desktop computer, mobile device, tablet computer, augmented / virtual / extended reality device, etc. In some embodiments, user device 162 may include one or more electronic applications, such as programs, plugins, browser extensions, etc., installed on the memory of user device 162. In some embodiments, electronic applications may be associated with one or more components of other components in environment 150. For example, electronic applications may include one or more of system control software, system monitoring software, software development tools, etc. In some examples, user device 162 may be the same as or part of client device 108.

[0099] In various embodiments, environment 150 may include data storage device 164 (e.g., a database). Data storage device 164 may include server systems and / or data storage systems, such as computer-readable storage devices, such as hard disks, flash drives, disks, etc. In some embodiments, data storage device 164 includes and / or interacts with an application programming interface to exchange data with other systems (e.g., one or more other components of the environment). Data storage device 164 may include and / or act as a repository or source for storing event data, event actions, generated event sequences, output data, etc. (e.g., for transfer to user device 162 or any other component of environment 150). In some examples, data storage device 164 may be the same as or a part of data storage device 118.

[0100] In some embodiments, components of environment 150 are associated with public entities (e.g., service providers, account providers, etc.). For example, in some embodiments, interactive display computing system 152 and data storage device 164 may be associated with public entities. In some embodiments, one or more components of the environment may be associated with entities different from those associated with another component. For example, interactive display computing system 152 may be associated with a first entity (e.g., a service provider), while data storage device 164 may be associated with a second entity (e.g., a storage entity providing storage services to the first entity). The systems and devices of environment 150 may communicate in any arrangement. As will be discussed herein, the systems and / or devices of environment 150 may communicate to perform one or more operations, such as generating, training, or using machine learning models to process natural language data and other activities. In some examples, interactive display computing system 152 may be part of organizational computing system 104.

[0101] As discussed in further detail below, the interactive display computing system 152 can perform one or more of the following operations on a machine learning model configured to process natural language data: (i) generating, storing, training, communicating with, or using the model. The interactive display computing system 152 may include a machine learning model and / or instructions associated with the machine learning model, such as instructions for generating the machine learning model, training the machine learning model, using the machine learning model, etc. The interactive display computing system 152 may include instructions for retrieving data, adjusting data (e.g., based on the output of the machine learning model), and / or operating the display of the user device 162 to output a response generated in response to user input (e.g., data adjusted based on the machine learning model). The interactive display computing system 152 may include training data (e.g., language data) and may include real data (e.g., (i) training language data and (ii) training athlete performance data) to generate natural language responses.

[0102] like Figure 2The depicted interactive display computing system 152 may include a capture module 154. In various embodiments, the capture module 154 is configured to receive multiple real-time event data. The real-time event data may be collected and / or compiled by the interactive display computing system 152 or using components separate from the environment 150. The multiple real-time event data may include multiple real-time event actions associated with a match identifier. In examples, real-time event actions may include tabular and statistically significant actions (e.g., scoring a goal, successful pass, etc.) associated with one or more athletes, teams, etc., during a match / game. In various embodiments, the capture module 154 may receive event data nearly simultaneously with, or at least substantially simultaneously with, the occurrence of an event action. The interactive display computing system 152 may receive multiple real-time event data via network 160. Further, the multiple real-time event actions may be used (e.g., by the capture module 154 or the interactive display computing system 152) to update one or more event actions associated with a unique match identifier. In some examples, the capture module 154 may be part of a tracking data system 116 and / or an interactive display module 124.

[0103] According to some embodiments, the capture module 154 may receive in-venue or broadcast data associated with a sporting event. This in-venue or broadcast data can be used to generate event data discussed herein. For example, this in-venue or broadcast data may be provided to one or more event machine learning models. One or more event machine learning models may be trained based on training data, which includes historical or simulated in-venue or broadcast data, historical or simulated event data (e.g., labeled data), historical or simulated event actions (e.g., labeled data), etc. As discussed herein, the training data may be used to train the event machine learning model by modifying one or more weights, layers, synapses, biases, etc., of the event machine learning model according to a machine learning algorithm. Alternatively or additionally, this in-venue or broadcast data may supplement the received event data for validation and / or for generating event sequences.

[0104] The interactive display computing system 152 may also include an event sequence generation module 156. In various embodiments, the event sequence generation module 156 may be configured to generate an event sequence based on multiple real-time event actions. The event sequence may be generated in real time as multiple real-time event data are received. In this way, the event sequence can be updated to reflect real-time event actions. In some examples, the event sequence generation module 156 may be part of the interactive display module 124.

[0105] like Figure 2The depicted interactive display computing system 152 may further include a display generation module 157. In various embodiments, the display generation module 157 may be configured to generate an interactive display. The interactive display may include a graphical representation of an event sequence (including one or more real-time event elements) and at least one of one or more interactive elements. The interactive display may be formatted in real time upon receiving multiple real-time event data. In various embodiments, one or more interactive elements may be configured to update one or more real-time event elements in response to one or more user interactions. In some examples, the display generation module 157 may be part of the interactive display module 124.

[0106] The following will be about Figures 3A to 5E Exemplary embodiments of the interactive display are described in more detail below. In various embodiments, the interactive display is generated in real time as multiple real-time event data are received (e.g., by the capture module 154). The interactive display computing system 152 may also include a transmission module 158. In various embodiments, the transmission module 158 may be configured to transmit the interactive display to a user interface. In some examples, the transmission module 158 may be part of the interactive display module 124.

[0107] In various embodiments, the interactive display and / or graphical representation of the event sequence and / or other items discussed herein can be displayed in real time via applications (e.g., sports media applications, software applications, browser extensions, mobile applications, smart TV applications, etc.), widgets (e.g., sports information cards, sports media reports, interactive elements within software applications, etc.), various graphics, etc.

[0108] like Figure 2The depicted environment 150 may also include an electronic network 160. In various embodiments, the electronic network 160 may be a wide area network (“WAN”), a local area network (“LAN”), a personal area network (“PAN”), etc. In some embodiments, the electronic network 160 includes the Internet, and information and data provided between various systems are conducted online. “Online” can mean connecting to or accessing source data or information from a location remote from other devices or networks coupled to the Internet. Alternatively, “online” can mean connecting to or accessing an electronic network (wired or wireless) via a mobile communication network or device. The Internet is a global computer network system—a network of numerous networks in which a party on a computer or other device connected to the network can obtain information from any other computer and communicate with parties on other computers or devices. The most widely used part of the Internet is the World Wide Web (often abbreviated as “WWW” or referred to as the “Web”). A “website page” typically encompasses, for example, a location hosted and / or run by a computer system for online access, a data storage device, etc., and this location or data storage device may include data configured to cause programs such as web browsers to perform operations such as sending, receiving, or processing data, generating visual displays, and / or interactive interfaces. In some examples, electronic network 160 may be the same as or part of network 105.

[0109] Despite Figure 2 While depicted as separate components, it should be understood that in some embodiments, a component or part of a component in environment 150 may be integrated with or incorporated into one or more other components. In another example, interactive display computing system 152 may be integrated into a data storage system. The data storage system may be configured to communicate with and / or receive / transmit data with other components of environment 150 via electronic network 160. In some embodiments, the operation or aspects of one or more of the components discussed above may be distributed among one or more other components. Any suitable arrangement and / or integration of various systems and devices in environment 150 may be used.

[0110] Further aspects of the interactive display computing system 152 and how to generate interactive displays are discussed in more detail in the following methods. In the following methods and systems, various actions can be described as being generated by... Figure 2 The actions are performed by components of the environment 150 (such as the interactive display computing system 152, user equipment 162, or components thereof). However, it should be understood that in various embodiments, the various components of the environment 150 discussed above can execute instructions or perform actions, including those discussed below. Actions performed by the device can be considered as being performed by processors, actuators, etc., associated with the device. Furthermore, it should be understood that in various embodiments, various steps can be added, omitted, and / or rearranged in any suitable manner.

[0111] Figures 3A to 3E An exemplary graphical user interface is depicted, which includes an interactive display for football matchups. For example... Figure 3A The exemplary graphical user interface shown includes an interactive display 200. As described above, the interactive display 200 can be generated by a display generation module, such as the one mentioned above. Figure 2 The described display generation module. The interactive display 200 can be dynamically generated and updated in real time based on event data associated with sporting events. Content provided via the interactive display 200 can be sorted based on this sporting event data and filtered based on this data and / or user input. For example... Figure 3A The interactive display 200 shown may include a graphical representation 202 of an event sequence. The graphical representation 202 of the event sequence may include real-time event elements 209. In various embodiments, the real-time event elements 209 may graphically represent real-time event actions (e.g., those received by the capture module 154, such as those related to...). Figure 2 (The described capture module). For example... Figure 3A The real-time event actions shown (such as scoring a goal) can be graphically represented as a football. In the example, other real-time event actions (such as a successful pass, interception, conceding the ball, or no action) can be graphically represented as real-time event element 209.

[0112] like Figure 3A As shown, interactive elements 204 and 206 can be configured to update one or more real-time event elements 209 in response to one or more user interactions (e.g., mouse clicks, etc.). As shown, the graphical representation 202 of event sequences associated with a match identifier for a specific match (e.g., a football match) can be filtered according to a predetermined time range. As shown, when a user interacts with the full match time range button 204, the graphical representation 202 of the event sequences can be updated to display the event sequence of the entire match associated with the match identifier. As shown, the graphical representation 202 of the event sequences can be filtered according to different modes of match activity (e.g., event mode, momentum mode, and chance mode). As shown, when a user interacts with the event mode button 206, real-time event actions (graphically represented as real-time event elements 209) can be mapped to the graphical representation 202 of the event sequences according to that mode.

[0113] In addition, such as Figure 3AThe interactive display 200 shown may include a zoom control 208 that allows a user to interact with a graphical representation 202 of the event sequence. Thus, the user can zoom in or out on one or more real-time event elements 209, which represent real-time event data (e.g., data received by the capture module 154, such as information about...). Figure 2 The data points in the described capture module. The graphical representation 202 of the event sequence is thus updated to display precise slices of the real-time event data, thereby allowing the granularity of the time stamps of certain real-time match events to be viewed within the interactive display 200 by increasing the graphically represented spacing between the real-time event elements 209 in a visually interactive manner.

[0114] like Figure 3B As shown, a real-time event element data readout 218 can be generated in response to user interaction. When a user interacts with a real-time event element 219 graphically represented within a graphical representation 212 of the event sequence (e.g., mouse click), data readout 218 can be generated and displayed within an interactive display 210. Real-time event element data readout 218 can include a description of the sporting event corresponding to the real-time event element 219. This description can be generated by inputting event data corresponding to the real-time event element 219 into a sports language model. The sports language model can be trained to output this description based on sports-related training, as discussed herein. (The above is in conjunction with...) Figure 3A The user's interaction with the time range button 214 and / or the event mode button 216 can cause the graphical representation 212 of the event sequence to be updated according to the filter criteria selected by the user.

[0115] like Figure 3C As shown, one or more interactive elements 229, 231 of the interactive display 225 can update the graphical representation 227 of the event sequence according to the corresponding filtering mechanism. For example, selecting the full match time range button 229 and the momentum mode button 231 can update the graphical representation 227 of the event sequence. Further, as... Figure 3D As shown, one or more interactive elements 244, 246 of the interactive display 240 can update the graphical representation 242 of the event sequence according to a corresponding filtering mechanism. As shown, selecting the full match time range button 244 and the chance mode button 246 can update the graphical representation 242 of the event sequence. In various embodiments, and as... Figure 3EAs shown, when interacting with one or more real-time event elements of a graphical representation 252 of an event sequence, data readouts 254 can be generated and displayed within an interactive display 250. In various implementations, data readouts can be generated using data points of received real-time event data (including real-time event actions). As shown, data readout 254 may include a timestamp (e.g., the actual time the real-time event action occurred, its time relative to other real-time event actions, etc.), a descriptive identifier for the real-time event action (e.g., "penalty missed"), and one or more text elements describing the real-time event action (e.g., the athlete's name and a textual description of the real-time event action, etc.). As discussed herein, such text elements can be output by a sports large language model.

[0116] Figures 4A to 4E An exemplary graphical user interface is depicted, which includes an interactive display for a rugby match. For example... Figure 4A The exemplary graphical user interface shown includes an interactive display 300. As described above, the interactive display 300 can be generated by a display generation module, such as the one mentioned above. Figure 2 The description shows the generation module. For example... Figure 4A The interactive display 300 shown may include a graphical representation 302 of an event sequence. The graphical representation 302 of the event sequence may include real-time event elements 309. In various embodiments, the real-time event elements 309 may graphically represent real-time event actions (e.g., those received by the capture module 154, such as those related to...). Figure 2 (The described capture module).

[0117] like Figure 4A As shown, interactive elements 304 and 306 can be configured to update one or more real-time event elements 309 in response to one or more user interactions (e.g., mouse clicks, etc.). As shown, the graphical representation 302 of event sequences associated with a game identifier for a specific game (e.g., a rugby game) can be filtered according to a predetermined time range. As shown, when a user interacts with the full game time range button 304, the graphical representation 302 of the event sequences can be updated to display the event sequence of the entire game associated with the game identifier. As shown, the graphical representation 302 of the event sequences can be filtered according to different modes of game activity (e.g., event mode and leader tracker mode). As shown, when a user interacts with the event mode button 306, real-time event actions (graphically represented as real-time event elements 309) can be mapped to the graphical representation 302 of the event sequences according to the selected mode.

[0118] In addition, such as Figure 4AThe interactive display 300 shown may include a zoom control 308 that allows a user to interact with a graphical representation 302 of an event sequence. Thus, the user can zoom in or out on one or more real-time event elements 309, which represent real-time event data (e.g., data received by the capture module 154, such as information about...). Figure 2 The data points in the described capture module. The graphical representation 302 of the event sequence is thus updated to display precise slices of the real-time event data, thereby allowing the granularity of the time stamps of certain real-time match events to be viewed within the interactive display 300 by increasing the graphically represented spacing between the real-time event elements 309 in a visually interactive manner.

[0119] like Figure 4A The interactive display 300 shown may include one or more graphical representations 310 of a game statistics tab. As shown, when the "Team Statistics" tab is selected via user interaction, the tab may display one or more graphical representations of real-time event data. In the example, such graphical representations of real-time event data may include side-by-side comparisons of team statistics such as points, field goals made, field goal attempts, etc.

[0120] like Figure 4B As shown, the event element slider 317 of the interactive display 315 can update the graphical representation of the event sequence in response to one or more user interactions. When the user interacts with the event element slider 317 (e.g., dragging the slider from one side of the graphical representation of the event sequence to the other), real-time event elements can be filtered, selected, displayed in an updated manner, etc. In various implementations, selecting one or more real-time event elements by interacting with the event element slider 317 can cause the graphical representation 323 of the match statistics tab to be updated based on the selected event elements. For example, the graphical representation of score points can be updated to include only those scores included within the selected one or more real-time event elements.

[0121] As mentioned above Figure 4A The user's interaction with the time range button 319 and / or the event mode button 321 can cause the graphical representation of the event sequence to be updated according to the filter criteria selected by the user.

[0122] like Figure 4CAs shown, one or more interactive elements 334, 336 in the interactive display 330 can update the graphical representation 332 of the event sequence according to the corresponding filtering mechanism. For example, selecting the full match time range button 334 and the event mode button 336 can update the graphical representation 332 of the event sequence. Additionally, one or more interactive elements 334, 336 can update the graphical representation 338 of the match statistics tab according to the corresponding filtering mechanism. Figure 4C As shown, when a user interacts with the graphical representation 332 of the event sequence, the real-time event action readout 340 can be filled with a text representation of data points included in the real-time event data and graphically represented by one or more real-time event elements.

[0123] Furthermore, such as Figure 4D As shown, the interactive display 350 includes one or more interactive elements 354, 358 that can update the graphical representation of the event sequence 352 and / or one or more match statistics tabs 360 according to the corresponding filtering mechanism. Additionally, as... Figure 4D As shown, the interactive display 350 may include a zoom control 356 that allows the user to interact with a graphical representation 352 of the event sequence. Thus, the user can zoom in or out on one or more real-time event elements, which represent real-time event data (e.g., data received by the capture module 154, such as information about...). Figure 2 The data points in the described capture module. The graphical representation 352 of the event sequence is thus updated to display precise slices of the real-time event data, thereby allowing the granularity of the time stamps for certain real-time match events to be viewed within the interactive display 350 by increasing the spacing between the graphically represented real-time event elements in a visually interactive manner.

[0124] In various embodiments, and as Figure 3E The graphical representation of the text machine learning output 370 shown can be displayed within the interactive display 365 in the following ways: based on user interaction, selection of one or more filter criteria and / or interactive elements; displayed in response to selection of one or more real-time event elements; displayed automatically based on the generation of event sequences, etc. In various implementations, the generative machine learning model trained to process natural language (e.g., a sports language learning model) can thus be used with natural language processing and computational systems (e.g., such as...) Figure 2The depicted interactive display computing system 152 captures, generates, or stores data to process real-time event data to generate text machine learning output 370 that can be displayed on the interactive display 365. The text machine learning output 370 can be formatted in generally understandable natural language and can utilize data from the interactive display computing system 152 or other components separate from the environment 150 (such as...). Figure 2 The captured event data is described. In various embodiments, the text machine learning output 370 may be generated by a machine learning model based on one or more predictions, conclusions, patterns, etc., identified or determined using real-time event data.

[0125] Generative machine learning models can receive real-time event data as input. They can be iteratively trained based on training data (such as event data received at or generated by the capture module 154). Based on this real-time updated training data, the generative machine learning model can output natural language output. This natural language output can be generated by first generating a numerical value in response to the input, and then converting the numerical value into natural language output. Therefore, a natural language model can be trained on sports event data to produce outputs specific to that data.

[0126] Generally, the artificial intelligence or machine learning models disclosed in this paper include a set of variables, such as nodes, neurons, and filters, which are tuned (e.g., weighted or biased) to different values ​​through the application of training data. In supervised learning, for example, when the real data provided as training data is known, training can be performed by feeding training data samples into the model, whose variables are set to initial values ​​(e.g., randomized, Gaussian noise-based, or based on pre-trained model settings). The output can be compared with the real data to determine the error, which can then be backpropagated through the model to adjust the values ​​of the variables.

[0127] Training can be performed in any suitable manner, such as in batches, and can include any suitable training method, such as stochastic or non-stochastic gradient descent, gradient boosting, random forests, etc. In some embodiments, a portion of the training data can be retained during training and / or used to validate the trained machine learning model, for example, by comparing the output of the trained model with real data from that portion of the training data to evaluate the accuracy of the trained model. The training of the machine learning model can be configured to enable the machine learning model to learn the association between natural language data and real-time event data, such that the trained machine learning model is configured to generate output in response to received input.

[0128] In various embodiments, the variables of the machine learning model can be correlated with each other in any suitable arrangement to generate output. For example, in some embodiments, the machine learning model may include a natural language processing architecture configured to identify, separate, and / or extract linguistic features from input text data. For example, the machine learning model may include one or more convolutional neural networks (“CNNs”) configured to identify features in text, and may include additional architectures (e.g., connection layers, neural networks, etc.) configured to determine relationships between the identified features in order to determine and generate natural language responses for inputs, interactions, etc.

[0129] In some embodiments, the machine learning or artificial intelligence model may include a recurrent neural network (“RNN”). Generally, RNNs are a class of feedforward neural networks well-suited for processing input sequences. In some embodiments, the machine learning model may include a long short-term memory (“LSTM”) model and / or a sequence-to-sequence (“Seq2Seq”) model. An LSTM model may be configured to generate output based on samples, a process that takes into account at least some previous samples and / or outputs. A Seq2Seq model may be configured to, for example, receive text as input and generate output in real time. In some embodiments, the machine learning model may include a transformer model and / or a graph neural network (GNN) model. Such models may be configured to generate output based on input data.

[0130] like Figures 4F to 4G The exemplary graphical user interface shown includes an interactive display 380. As described above, the interactive display 380 can be generated by a display generation module, such as the one mentioned above. Figure 2 The description shows the generation module. For example... Figures 4F to 4G The interactive display 380 shown may include a graphical representation 383 of an event sequence. The graphical representation 383 of the event sequence may include real-time event elements. In various embodiments, the real-time event elements may graphically represent real-time event actions (e.g., as received by the capture module 154, such as regarding...). Figure 2 (The described capture module).

[0131] like Figures 4F to 4G As shown, interactive elements 381, 382, ​​and 384 can be configured to update one or more real-time event elements in response to one or more user interactions (e.g., mouse clicks, etc.). As shown, a graphical representation 383 of a sequence of events associated with a match identifier for a specific match (e.g., a rugby match) can be filtered according to a predetermined time range.

[0132] like Figure 4FAs shown, when the user interacts with the time button 381 and the full match time range button 384, the graphical representation 383 of the event sequence can be updated to display the event sequence of the full match associated with the match identifier. Figure 4G As shown, the graphical representation of the event sequence 383 can be updated to display the event sequence related to scoring / non-scoring offensive rounds when the user interacts with the offensive round button. Live game events can be filtered using the offensive round button. In the example, the user can group offensive rounds by scoring or non-scoring offensive rounds, or by a user-selected group of offensive rounds.

[0133] like Figures 4F to 4G The interactive display 380 shown may include one or more graphical representations 386 of the game statistics tab. As shown, when the "Team Statistics" tab is selected via user interaction, the tab may display one or more graphical representations of real-time event data. In the example, such graphical representations of real-time event data may include side-by-side comparisons of team statistics such as points, total play, total yards, yards per play, passing yards, rushing yards, sacks, penalties, and penalty kicks.

[0134] Figures 5A to 5E An exemplary graphical user interface is depicted, which includes an interactive display for basketball games. For example... Figure 5A The exemplary graphical user interface shown includes an interactive display 400. As described above, the interactive display 400 can be generated by a display generation module, such as the one mentioned above. Figure 2 The description shows the generation module. For example... Figure 5A The interactive display 400 shown may include a graphical representation 404 of an event sequence. The graphical representation 404 of the event sequence may include real-time event elements. In various embodiments, the real-time event elements may graphically represent real-time event actions (e.g., as received by the capture module 154, such as regarding...). Figure 2 (The described capture module).

[0135] like Figure 5A As shown, the interactive elements can be configured to update one or more real-time event elements in response to one or more user interactions (e.g., mouse clicks, etc.). As shown, a graphical representation 404 of the event sequence associated with a game identifier for a specific game (e.g., a basketball game) can be filtered according to a predetermined time range. As shown, when a user interacts with the full game time range button 402, the graphical representation 404 of the event sequence can be updated to display the event sequence for the entire game associated with the game identifier.

[0136] like Figure 5A The interactive display 400 shown may include one or more graphical representations 406 of a match statistics tab. As shown, when the "Team Statistics" tab is selected via user interaction, the tab may display one or more graphical representations of real-time event data. In the example, such graphical representations of real-time event data may include side-by-side comparisons of team statistics (such as scores).

[0137] like Figure 5B As shown, the event element slider 414 of the interactive display 410 can update the graphical representation of the event sequence in response to one or more user interactions. When a user interacts with the event element slider 414 (e.g., dragging the slider from one side of the graphical representation of the event sequence to the other), real-time event elements can be filtered, selected, displayed in an updated manner, etc. In various implementations, selecting one or more real-time event elements by interacting with the event element slider 414 can cause the graphical representation 416 of the match statistics tab to be updated based on the selected event elements. For example, the graphical representation of score points can be updated to include only those scores included within the selected one or more real-time event elements.

[0138] As mentioned above Figure 5A The user's interaction with the time range button 412 can cause the graphical representation of the event sequence to be updated according to the filter criteria selected by the user.

[0139] like Figure 5C The exemplary graphical user interface shown includes an interactive display 420. As described above, the interactive display 420 can be generated by a display generation module, such as the one mentioned above. Figure 2 The description shows the generation module. For example... Figure 5C The interactive display 420 shown may include a graphical representation 423 of an event sequence. The graphical representation 423 of the event sequence may include real-time event elements 424. In various embodiments, the real-time event elements 424 may graphically represent real-time event actions (e.g., those received by the capture module 154, such as those related to...). Figure 2 (The described capture module).

[0140] like Figure 5CAs shown, interactive element 421 can be configured to update one or more real-time event elements 424 in response to one or more user interactions (e.g., mouse clicks, etc.). As shown, a graphical representation 423 of event sequences associated with a game identifier for a specific game (e.g., a basketball game) can be filtered according to a predetermined time range. As shown, when a user interacts with the full game time range button 421, the graphical representation 423 of the event sequences can be updated to display the event sequence for the entire game associated with the game identifier.

[0141] like Figure 5C The interactive display 420 shown may include one or more graphical representations 426 of a match statistics tab. As shown, when the "Team Statistics" tab is selected via user interaction, the tab may display one or more graphical representations of real-time event data. In the example, such graphical representations of real-time event data may include side-by-side comparisons of team statistics (such as scores).

[0142] In various embodiments, real-time data can be represented in an interactive display. In one example, this data can be updated to reflect the user's selections as the user interacts with the interactive display. Such selections may include team statistics, player statistics, shot assist matrix, shot streaks, and player comparison data. When the shot streak button is selected, the user can identify players or teams that have made multiple consecutive successful shots.

[0143] like Figure 5D and Figure 5E The exemplary graphical user interface shown includes an interactive display 430. As described above, the interactive display 430 can be generated by a display generation module, such as the one mentioned above. Figure 2 The description shows the generation module. For example... Figure 5D and Figure 5E As shown, the interactive display 430 may include a rotary button 432. When the user selects the rotary button 432 via interaction, the interactive display 430 may show a graphical representation of athlete rotation data. In the example, real-time game events may be filtered based on athlete participation (e.g., who is on or off the basketball court). In the example, system 152 may collect, process, and / or generate athlete on / off data. Athlete on / off data may include real-time event data for each athlete. In the example, athlete on / off data may further include performance data for each athlete when one or more other athletes are on or off the court.

[0144] In the example, such as Figure 5DAs shown, when the substitute button is unchecked, only players on the court can be displayed on the interactive display 430. Figure 5E As shown, when the substitute button is checked, all athletes can be displayed on the interactive display 430.

[0145] Figure 6 A flowchart illustrating an exemplary method for generating an interactive display is provided. Exemplary method 500 begins at step 505, in which multiple real-time event data (e.g., data generated by a competition identifier) ​​are received, including multiple real-time event actions associated with a competition identifier. Figure 2 The captured module 154 described receives the data. Multiple real-time event actions may include one or more event actions associated with a match identifier, such as scoring a goal, successful pass, interception, conceding a goal, etc. At step 510, an event sequence is generated based on the multiple real-time event actions. The event sequence may be generated in real-time upon receiving multiple real-time event data.

[0146] At step 515, an interactive display is generated, which includes at least a graphical representation of the event sequence and one or more interactive elements (e.g., by means of...). Figure 2 The interactive display is generated by the depicted display generation module 157. The interactive display can be generated in real time upon receiving multiple real-time event data. One or more interactive elements can be configured to update the interactive display with one or more real-time event elements in response to one or more user interactions. In an example, the interactive display can be generated based on a set of filtering rules activated by one or more user interactions. In such an example, the filtering rules can include filtering by event time, by one or more modes (as described above), etc. In various embodiments, filtering by event time can represent a first filtering rule, and filtering by one or more modes can represent a second filtering rule. In an example, the first filtering rule can correspond to displaying a first set of real-time event elements associated with a first event sequence period in response to a first user interaction. In an example, the second filtering rule can correspond to displaying a first set of real-time event statistics associated with a first event sequence period in response to a first user interaction.

[0147] In various implementations, scrolling interactive elements (e.g., sliders) can be configured to cause a graphical representation of an event sequence to move forward or backward in time in response to one or more user interactions, as described above. In the example, the interactive display can be regenerated in real time as multiple real-time event data are received. At step 520, the interactive display is transmitted to the user interface (e.g., transmitted by transmission module 158 via network 160 to the display of user device 162, such as...). Figure 2 (As depicted).

[0148] Figure 7 An exemplary data flow diagram 600 for generating an interactive display is depicted. This can be achieved (e.g., via network 160, as per [reference]). Figure 2 (As described) receiving one or more sets of real-time event data from a data source in the form of a first data subset 602 and a second data subset 604 (e.g., received by the capture module 154, such as regarding...). Figure 2 (As described). As shown, the first data subset 602 and the second data subset 604 can be placed in the buffer 606. In various embodiments, the buffer 606 can be operatively connected to the interactive display computing system 152, or can be integrated into the interactive display computing system 152 or one or more components of the interactive display computing system 152.

[0149] The first data subset 602 can be received at a first time and can be provided to the buffer 606. The buffer 606 can verify the first data subset 602 (e.g., via the verification module at 608) and can determine that the first data subset 602 is incomplete. For example, the buffer 606 can determine that the first data subset 602 has a missing value in the sequence counter, can determine that the first data subset 602 is missing event data, can determine that the first data subset 602 includes the later part of the event data in chronological order but does not include the earlier part of the corresponding data in chronological order, and so on. Based on the determination that the first data subset 602 is incomplete, the verification module can instruct the buffer 606 to retain the first data subset without transmitting it to the interface generator. As shown, the second data subset 604 can be received at a second time after the first time. The buffer 606 can verify the first data subset 602 and the second data subset 604 (e.g., via the verification module at 610) and can determine that the first data subset 602 plus the second data subset 604 is complete. For example, buffer 606 can determine that the first data subset 602 plus the second data subset 604 includes a complete counter, can determine that the first data subset 602 plus the second data subset 604 has complete event data, and can determine that the first data subset 602 plus the second data subset 604 includes all consecutive parts of a given event, etc. Based on the determination that the first data subset 602 plus the second data subset 604 is complete, the verification module can instruct buffer 606 to transmit the first data subset 602 and the second data subset 604 at 612 to the interface generator (e.g., to the display generation module 157, as per [reference]). Figure 2 (As described). Therefore, data flow diagram 600 depicts techniques for ensuring that a complete dataset is received (e.g., from multiple sources) before event data is provided via display generation module 157 for visualization and filtering. This technique ensures that incomplete data is not presented via display generation module 157.

[0150] Figure 8 A flowchart for training a machine learning model is depicted. For example... Figure 8 As shown in flowchart 700, training data 712 may include one or more of stage inputs 714 and known outputs 718 associated with the machine learning model to be trained. Stage inputs 714 may come from any applicable source, including the components or sets shown in the accompanying figures provided herein. For machine learning models generated based on supervised or semi-supervised training, known outputs 718 may be included. Unsupervised machine learning models may be trained without using known outputs 718. Known outputs 718 may include known or expected outputs for future inputs that are similar to or belong to the same category as stage inputs 714 but do not have corresponding known outputs.

[0151] Training data 712 and training algorithm 720 can be provided to training unit 730, which can apply training data 712 to training algorithm 720 to generate a trained machine learning model 750. According to an embodiment, comparison results 716 can be provided to training unit 730, which compare the previous outputs of the corresponding machine learning model to apply the previous results to retrain the machine learning model. Comparison results 716 can be used by training unit 730 to update the corresponding machine learning model. Training algorithm 720 can utilize machine learning networks and / or models, including but not limited to deep learning networks (such as graph neural networks (GNN), deep neural networks (DNN), convolutional neural networks (CNN), fully convolutional networks (FCN), and recurrent neural networks (RCN)), probabilistic models (such as Bayesian networks and graph models), and / or discriminative models (such as decision forests and maximum margin methods). The output of flowchart 700 can be the trained machine learning model 750.

[0152] The machine learning models disclosed herein can be trained by adjusting one or more weights, layers, and / or biases during the training phase. During the training phase, historical or simulated data can be provided as input to the model. The model can adjust one or more of its weights, layers, and / or biases based on this historical or simulated information. Based on this training, the adjusted weights, layers, and / or biases can be configured in a production version of the machine learning model (e.g., a trained model). After training is complete, the machine learning model can output machine learning model outputs according to the topics disclosed herein. Depending on the implementation, one or more machine learning models disclosed herein can be continuously updated based on feedback associated with the use or implementation of the machine learning model outputs.

[0153] It should be understood that the aspects in this disclosure are merely exemplary, and other aspects may include various combinations of features from other aspects, as well as additional or fewer features.

[0154] Generally, any process or operation discussed in this disclosure that is understood to be computer-implementable (such as the process shown in the flowcharts disclosed herein) can be executed by one or more processors of a computer system, such as any system or device in the exemplary environment disclosed herein, as described above. A process or process step executed by one or more processors may also be referred to as an operation. One or more processors may be configured to execute such a process by accessing instructions (e.g., software or computer-readable code) that, when executed by the one or more processors, cause the one or more processors to perform the process. The instructions may be stored in the memory of the computer system. The processor may be a central processing unit (CPU), a graphics processing unit (GPU), or any suitable type of processing unit.

[0155] A computer system (such as a system or device that implements the processes or operations described in the examples above) may include one or more computing devices, such as one or more of the systems or devices disclosed herein. One or more processors of the computer system may be located in a single computing device or distributed among multiple computing devices. The memory of the computer system may include the respective memory of each of the multiple computing devices.

[0156] Figure 9 This is a simplified functional block diagram of a computer 800 according to an exemplary aspect of this disclosure, which may be configured as an apparatus for performing the methods disclosed herein. For example, computer 800 may be configured as a system according to an exemplary aspect of this disclosure. In various aspects, any system herein may be computer 800, which includes, for example, a data communication interface 820 for packet data communication. Computer 800 may also include a central processing unit (“CPU”) 802 in the form of one or more processors for executing program instructions. Computer 800 may include an internal communication bus 808 and storage units 806 (such as ROM, HDD, SDD, etc.) that can store data on a computer-readable medium 822, but computer 800 may also receive programming and data via network communication.

[0157] The computer 800 may also have a memory 804 (such as RAM) that stores information for performing the techniques presented herein (e.g., regarding...). Figures 1 to 8The system and methods described herein contain instructions 824, but instructions 824 may be temporarily or permanently stored within other modules of the computer 800 (e.g., processor 802 and / or computer-readable medium 822). The computer 800 may also include input and output ports 812 and / or a display 810 for connection to input and output devices such as a keyboard, mouse, touchscreen, monitor, display, etc. Various system functions may be implemented in a distributed manner on multiple similar platforms to distribute the processing load. Alternatively, these systems may be implemented by appropriately programming a single computer hardware platform.

[0158] The program aspect of this technology can be considered a "product" or "artifact," typically in the form of executable code and / or associated data, carried or embodied in some machine-readable medium. "Storage" media include any or all tangible memory or associated modules of computers, processors, etc., such as various semiconductor memories, tape drives, hard disk drives, etc., which can provide non-transitory storage for software programming at any time. All or part of the software can sometimes be communicated via the Internet or various other telecommunications networks. For example, such communication enables the loading of software from one computer or processor to another, such as from a management server or host computer of a mobile communication network to a server's computer platform and / or from a server to a mobile device. Therefore, another medium that can carry software elements includes light waves, radio waves, and electromagnetic waves, such as those used through wired and fiber optic terrestrial networks and physical interfaces between local devices on various air links. Physical elements carrying such waves (such as wired or wireless links, optical links, etc.) can also be considered as media carrying software. As used herein, unless limited to non-transitory, tangible "storage" media, terms such as "computer or machine readable medium" refer to any medium that participates in providing instructions to a processor for execution.

[0159] Figure 10AAn architecture of a computing system 900 according to an example embodiment is illustrated. System 900 may represent at least a portion of an organizational computing system 104. One or more components of system 900 may communicate electrically with each other using bus 905. System 900 may include a processing unit (CPU or processor) 910 and a system bus 905 that couples various system components, including system memory 915 (such as read-only memory (ROM) 920 and random access memory (RAM) 925), to processor 910. System 900 may include a cache of high-speed memory directly connected to, adjacent to, or integrated as part of processor 910. System 900 may copy data from memory 915 and / or storage device 930 to cache 912 for fast access by processor 910. In this way, cache 912 can provide performance improvements, thereby avoiding latency for processor 910 while waiting for data. These and other modules may control or be configured to control processor 910 to perform various actions. Other system memory 915 may also be available. Memory 915 may include various different types of memory with different performance characteristics. Processor 910 may include: any general-purpose processor and hardware or software modules, such as services 1 932, 2 934, and 3 936 stored in storage device 930, which are configured to control processor 910; and dedicated processors, in which software instructions are incorporated into the actual processor design. Processor 910 may essentially be a fully self-contained computing system, containing multiple cores or processors, buses, memory controllers, caches, etc. Multi-core processors may be symmetric or asymmetric.

[0160] To enable users to interact with the computing system 900, the input device 945 can represent any number of input mechanisms, such as a microphone for voice, a touch-sensitive screen for gesture or graphical input, a keyboard, a mouse, motion input, voice, etc. The output device 935 (e.g., a display) can also be one or more of a variety of output mechanisms known to those skilled in the art. In some cases, a multimodal system allows users to provide multiple types of input to communicate with the computing system 900. The communication interface 940 typically manages and controls user input and system output. There are no limitations on operation on any particular hardware arrangement, and therefore, the basic features described herein can be easily replaced as improved hardware or firmware arrangements are developed.

[0161] Storage device 930 may be a non-volatile memory and may be a hard disk or other type of computer-readable medium that can store computer-accessible data, such as a magnetic tape cassette, flash memory card, solid-state memory device, digital multifunction disk, cartridge, random access memory (RAM) 925, read-only memory (ROM) 920 and combinations thereof.

[0162] Storage device 930 may include services 932, 934, and 936 for controlling processor 910. Other hardware or software modules are contemplated. Storage device 930 may be connected to system bus 905. In one aspect, a hardware module performing a specific function may include software components stored in a computer-readable medium that combine with necessary hardware components (such as processor 910, bus 905, output device 935, etc.) to perform the function.

[0163] Figure 10B A computer system 950 with a chipset architecture is shown, which may represent at least a portion of an organizational computing system 104. The computer system 950 may be an example of computer hardware, software, and firmware that can be used to implement the disclosed techniques. System 950 may include a processor 955, which represents any number of physically and / or logically distinct resources capable of executing software, firmware, and hardware configured to perform the identified computations. The processor 955 may communicate with a chipset 960, which may control the inputs and outputs of the processor 955. In this example, the chipset 960 outputs information to an output 965 (e.g., a display) and may read and write information to a storage device 970, which may include, for example, magnetic media and solid-state media. The chipset 960 may also read and write data to RAM 975. A bridge 980 may be provided for interfacing with various user interface components 985 to interface with the chipset 960. Such a user interface component 985 may include a keyboard, microphone, touch detection and processing circuitry, pointing devices (such as a mouse), etc. Typically, the input to the system 950 can come from any of a variety of sources, including machine-generated and / or human-generated input.

[0164] Chipset 960 can also interface with one or more communication interfaces 990 that may have different physical interfaces. Such communication interfaces may include interfaces for wired and wireless local area networks (LANs), broadband wireless networks, and personal area networks (PANs). Some applications of the methods disclosed herein for generating, displaying, and using a GUI may include receiving ordered datasets via a physical interface, or generating such ordered datasets by the machine itself through processor 955 analyzing data stored in storage device 970 or RAM 975. Furthermore, the machine may receive input from a user via user interface component 985 and perform appropriate functions, such as browsing functions, by interpreting this input using processor 955.

[0165] It is understood that the example systems 900 and 950 may have more than one processor 910, or be part of a group or cluster of computing devices networked together to provide greater processing power.

[0166] While the foregoing describes embodiments herein, other and further embodiments may be devised without departing from its basic scope. For example, aspects of this disclosure may be implemented in hardware or software or a combination of hardware and software. One embodiment described herein may be implemented as a program product for use with a computer system. The program of the program product defines the functionality of the embodiment (including the methods described herein) and may be contained on a variety of computer-readable storage media. Illustrative computer-readable storage media include, but are not limited to: (i) non-writable storage media on which information is permanently stored (e.g., read-only memory (ROM) devices within a computer, such as CD-ROM discs readable by a CD-ROM drive, flash memory, ROM chips, or any type of solid-state non-volatile memory); and (ii) writable storage media on which changeable information is stored (e.g., floppy disks within a floppy disk drive, or hard disk drives, or any type of solid-state random access memory). Such computer-readable storage media are embodiments of this disclosure when carrying computer-readable instructions directing the functionality of the disclosed embodiments.

[0167] Those skilled in the art will understand that the foregoing examples are exemplary and not restrictive. All arrangements, enhancements, equivalents, and modifications that will be apparent to those skilled in the art upon reading the specification and studying the accompanying drawings are intended to be included within the true spirit and scope of this disclosure. Therefore, the appended claims are intended to cover all such modifications, arrangements, and equivalents that fall within the true spirit and scope of these teachings.

[0168] While the disclosed methods, devices, and systems are described with reference to transmission data as an example, it should be understood that the disclosed aspects are applicable to any environment, such as desktop or laptop computers, car entertainment systems, home entertainment systems, etc. Furthermore, the disclosed aspects are applicable to any type of Internet protocol.

[0169] It should be understood that in the foregoing description of exemplary aspects of the invention, various features of the invention are sometimes combined together in a single aspect, figure, or description thereof in order to simplify this disclosure and aid in understanding one or more of the inventive aspects. However, this method of disclosure should not be construed as reflecting an intention that the claimed invention requires more features than expressly recited in each claim. Rather, as reflected in the appended claims, an inventive aspect lies in fewer than all features of a single foregoingly disclosed aspect. Therefore, the claims following the detailed description are hereby expressly incorporated into that detailed description, wherein each claim is independently a separate aspect of the invention.

[0170] Furthermore, while some aspects described herein include some features included in other aspects but not others, as those skilled in the art will understand, combinations of features from different aspects are intended to be within the scope of the invention and to form different aspects. For example, any aspect of the claimed aspects may be used in any combination in the appended claims.

[0171] Therefore, although certain aspects have been described, those skilled in the art will recognize that other and further modifications can be made thereto without departing from the spirit of the invention, and it is intended to claim protection for all such changes and modifications falling within the scope of the invention. For example, functions can be added or removed from the block diagrams, and operations can be interchanged between functional blocks. Operations can be added or removed to the methods described within the scope of the invention.

[0172] The subject matter disclosed above should be considered illustrative rather than restrictive, and the appended claims are intended to cover all such modifications, enhancements, and other embodiments falling within the true spirit and scope of this disclosure. Therefore, to maximize legal permissibility, the scope of this disclosure will be determined by the broadest permissible interpretation of the appended claims and their equivalents, and should not be limited to or confined to the specific embodiments described above. While various embodiments of this disclosure have been described, it will be apparent to those skilled in the art that further embodiments are possible within the scope of this disclosure. Accordingly, this disclosure is not limited except in accordance with the appended claims and their equivalents.

Claims

1. A computer-implemented method for generating an interactive display, the method comprising: Multiple real-time event data are received by one or more processors, the multiple real-time event data including multiple real-time event actions associated with a competition identifier; An event sequence is generated by one or more processors based on the plurality of real-time event actions, wherein the event sequence is generated in real time upon receiving the plurality of real-time event data; The interactive display is generated by the one or more processors, the interactive display including at least a graphical representation of the event sequence and one or more interactive elements, the graphical representation including one or more real-time event elements, wherein the interactive display is formatted in real-time upon receiving the plurality of real-time event data, and wherein the one or more interactive elements are configured to update the one or more real-time event elements in response to one or more user interactions; and The interactive display is transmitted to the user interface by the one or more processors.

2. The computer-implemented method as described in claim 1, wherein, The interactive display is generated based on a set of filtering rules.

3. The computer-implemented method as described in claim 2, wherein, The set of filtering rules includes a first filtering rule, which corresponds to displaying a first set of real-time event elements associated with a first event sequence period in response to a first user interaction.

4. The computer-implemented method as described in claim 2, wherein, The set of filtering rules includes a second filtering rule, which corresponds to displaying a first set of real-time event statistics associated with a first event sequence period in response to a first user interaction.

5. The computer-implemented method as described in claim 1, wherein, The multiple real-time event actions include at least one of scoring a goal, successful passing, interception, conceding the ball, or no action.

6. The computer-implemented method of claim 1, further comprising: The first subset of real-time event data from the plurality of real-time event data is received by the one or more processors; as well as The first subset of real-time event data is provided to the buffer by the one or more processors.

7. The computer-implemented method of claim 6, further comprising: The one or more processors receive a second subset of real-time event data from the plurality of real-time event data; The one or more processors provide the second subset of real-time event data to the buffer; as well as The one or more processors determine that the plurality of real-time event data has been completed based on the first subset of real-time event data and the second subset of real-time event data.

8. A system for generating interactive displays, the system comprising: A memory, wherein the memory stores instructions; and a processor operatively connected to the memory and configured to execute the instructions to perform operations including: Multiple real-time event data are received by one or more processors, the multiple real-time event data including multiple real-time event actions associated with a competition identifier; An event sequence is generated by one or more processors based on the plurality of real-time event actions, wherein the event sequence is generated in real time upon receiving the plurality of real-time event data; The interactive display is generated by the one or more processors, the interactive display including at least a graphical representation of the event sequence and one or more interactive elements, the graphical representation including one or more real-time event elements, wherein the interactive display is formatted in real-time upon receiving the plurality of real-time event data, and wherein the one or more interactive elements are configured to update the one or more real-time event elements in response to one or more user interactions; and The interactive display is transmitted to the user interface by the one or more processors.

9. The system of claim 8, wherein, The interactive display is generated based on a set of filtering rules.

10. The system of claim 9, wherein, The set of filtering rules includes a first filtering rule, which corresponds to displaying a first set of real-time event elements associated with a first event sequence period in response to a first user interaction.

11. The system of claim 9, wherein, The set of filtering rules includes a second filtering rule, which corresponds to displaying a first set of real-time event statistics associated with a first event sequence period in response to a first user interaction.

12. The system of claim 8, wherein, The multiple real-time event actions include at least one of scoring a goal, successful passing, interception, conceding the ball, or no action.

13. The system of claim 8, wherein, The one or more interactive elements are configured to update the graphical representation of the event sequence in response to the one or more user interactions.

14. The system of claim 8, wherein, The scrolling interactive element is configured to cause the graphical representation of the event sequence to move forward or backward in time in response to one or more user interactions.

15. A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, perform operations including: Multiple real-time event data are received by one or more processors, the multiple real-time event data including multiple real-time event actions associated with a competition identifier; The one or more processors generate an event sequence based on the plurality of real-time event actions, wherein, The event sequence is generated in real time upon receiving the multiple real-time event data; An interactive display is generated by the one or more processors, the interactive display including at least a graphical representation of the event sequence and one or more interactive elements, the graphical representation including one or more real-time event elements, wherein the interactive display is formatted in real time upon receiving the plurality of real-time event data, and wherein the one or more interactive elements are configured to update the one or more real-time event elements in response to one or more user interactions; as well as The interactive display is transmitted to the user interface by the one or more processors.

16. The non-transitory computer-readable medium of claim 15, wherein, The interactive display is generated based on a set of filtering rules.

17. The non-transitory computer-readable medium of claim 16, wherein, The set of filtering rules includes a first filtering rule, which corresponds to displaying a first set of real-time event elements associated with a first event sequence period in response to a first user interaction.

18. The non-transitory computer-readable medium of claim 16, wherein, The set of filtering rules includes a second filtering rule, which corresponds to displaying a first set of real-time event statistics associated with a first event sequence period in response to a first user interaction.

19. The non-transitory computer-readable medium of claim 15, wherein, The one or more interactive elements are configured to update the graphical representation of the event sequence in response to the one or more user interactions.

20. The non-transitory computer-readable medium of claim 15, wherein, The scrolling interactive element is configured to cause the graphical representation of the event sequence to move forward or backward in time in response to one or more user interactions.