DCT

1:22-cv-01545

Recentive Analytics Inc v. Fox Corp

Log In

I. Executive Summary and Procedural Information

  • Parties & Counsel:
  • Case Identification: 1:22-cv-01545, D. Del., 02/24/2023
  • Venue Allegations: Venue is based on Defendants being organized under the laws of Delaware, and thus residing in the judicial district.
  • Core Dispute: Plaintiff alleges that Defendant’s internal predictive analytics tool, used to optimize the regional broadcast schedules for live sporting events, infringes four patents related to the dynamic, machine-learning-based generation of network maps and event schedules.
  • Technical Context: The technology at issue involves using computational and machine learning techniques to automate and optimize the complex process of scheduling television broadcasts to maximize viewership or other business metrics.
  • Key Procedural History: The complaint alleges that Plaintiff notified Defendant of its patents and invited licensing discussions beginning in July 2022, but Defendant terminated these discussions in October 2022, which may form the basis for a willfulness claim. This First Amended Complaint was filed to moot a previously-filed motion to dismiss.

Case Timeline

Date Event
2018 Fox was allegedly aware of Recentive's software platform
2019-10-10 Priority Date for ’811 and ’957 Patents
2021-02-02 Issue Date for U.S. Patent No. 10,911,811
2021-03-23 Issue Date for U.S. Patent No. 10,958,957
2021-05-27 Priority Date for ’367 and ’960 Patents
2022-01-01 (Approx.) Fox allegedly began building its internal predictive analytics platform
2022-07-12 Issue Date for U.S. Patent No. 11,386,367
2022-07-26 Recentive allegedly notified Fox of the ’811 Patent family
2022-08-24 Recentive allegedly provided additional information to Fox regarding the ’957 Patent
2022-10-18 Fox allegedly terminated licensing discussions with Recentive
2022-12-27 Issue Date for U.S. Patent No. 11,537,960
2023-02-24 Complaint Filing Date

II. Technology and Patent(s)-in-Suit Analysis

U.S. Patent No. 10,911,811 - “Systems and Methods for Automatically and Dynamically Generating a Network Map”

The Invention Explained

  • Problem Addressed: The patent describes the conventional process for creating broadcast schedules for events like NFL games as being entirely manual, static, and "incapable of responding to changing conditions" (’811 Patent, col. 1:25-26). This prior art process was allegedly unable to consider multiple schedule permutations or forecast the impact of a proposed change. (’811 Patent, col. 1:22-31).
  • The Patented Solution: The invention is a computer-implemented method that automates and improves this process by receiving an event schedule and using a machine learning technique to generate a "network map" that optimizes a parameter, such as overall television ratings (’811 Patent, col. 3:10-15; Abstract). A key feature is the ability to "automatically [and] dynamically" update the map in real time based on changing data, such as weather, news, or gambling information. (’811 Patent, col. 1:35-48, 3:22-34).
  • Technical Importance: The technology purports to replace a "crude and suboptimal" manual process with a dynamic, data-driven system capable of optimizing outcomes in the high-stakes environment of live television broadcasting. (Compl. ¶23).

Key Claims at a Glance

  • The complaint asserts independent claim 1 and dependent claims 6-9. (Compl. ¶¶25, 43).
  • The essential elements of independent claim 1 include:
    • Receiving a schedule for events at a first time and a second time.
    • Generating a network map that maps the events to television stations in various cities.
    • Using a machine learning technique to optimize an overall television rating.
    • Automatically updating the network map "on demand and in real time" based on changes to the schedule or "underlying criteria."
    • Using the map to determine which event will be displayed for each station at each time.

U.S. Patent No. 11,386,367 - “Systems and Methods for Determining Event Schedules”

The Invention Explained

  • Problem Addressed: This patent addresses the difficulty of scheduling live events, such as concert tours, which involves a wide variety of factors like competing events, ticket prices, and venue availability. (’367 Patent, col. 1:26-30). The patent states that failing to properly consider these factors can lead to "sub-optimal schedules that generate little interest among consumers and/or result in low attendance or ticket sales." (’367 Patent, col. 1:30-33).
  • The Patented Solution: The invention is a computer-implemented method for generating an optimized event schedule for a series of live events. It involves providing event parameters (e.g., venue availability, performer fees) and target features (e.g., attendance, profit) to a machine learning model. (’367 Patent, Abstract). The model is "iteratively train[ed]" using historical data to improve its accuracy and can be updated in response to real-time changes to event parameters. (’367 Patent, col. 13:13-19).
  • Technical Importance: The technology aims to provide a sophisticated, automated tool for complex event logistics, moving beyond simple booking to predictive optimization of financial and attendance outcomes for tours or series of events. (Compl. ¶35).

Key Claims at a Glance

  • The complaint asserts independent claim 1 and dependent claims 4 and 8. (Compl. ¶¶31, 71).
  • The essential elements of independent claim 1 include:
    • Receiving event parameters (e.g., venue availability, ticket prices, performer fees) and event target features (e.g., attendance, profit).
    • Providing these to a machine learning (ML) model (neural network or support vector model).
    • "Iteratively training the ML model" using historical data to improve its accuracy.
    • Receiving user-specific parameters and weights for a future series of events.
    • Generating a schedule optimized relative to prioritized target features.
    • Detecting and providing a "real-time change" to the trained ML model to improve its accuracy.
    • Updating the schedule to keep it optimized in view of the real-time change.

U.S. Patent No. 10,958,957 - “Systems and Methods for Automatically and Dynamically Generating a Network Map”

  • Technology Synopsis: As a continuation of the ’811 Patent, this patent shares the same specification and is directed to the same technology of dynamically generating and optimizing network maps for broadcast events using machine learning. (Compl. ¶¶24, 26). The claimed method involves "obtaining" rather than "receiving" a schedule but is otherwise highly similar to that of the ’811 Patent. (’957 Patent, col. 10:9).
  • Asserted Claims: Independent claim 1 and dependent claims 7-9 and 12 are asserted. (Compl. ¶26).
  • Accused Features: The accused features are the same as those for the ’811 Patent: the "Fox Mapping Tool" used for scheduling NFL broadcasts. (Compl. ¶¶56-62).

U.S. Patent No. 11,537,960 - “Systems and Methods for Determining Event Schedules”

  • Technology Synopsis: As a continuation of the ’367 Patent, this patent shares the same specification and is directed to the same technology of using an iteratively trained ML model to generate optimized schedules for a series of live events. (Compl. ¶31). The claims are directed to a scenario with a first performer and at least one second performer, adding a competitive or cooperative element to the scheduling problem. (’960 Patent, col. 14:12-16:8).
  • Asserted Claims: Independent claim 1 and dependent claims 4 and 8 are asserted. (Compl. ¶31).
  • Accused Features: The accused features are the same as those for the ’367 Patent: the "Fox Mapping Tool" is alleged to generate event schedules based on various parameters and target features. (Compl. ¶¶88-99).

III. The Accused Instrumentality

Product Identification

  • The "Fox Mapping Tool," an internal networking and mapping analytics tool allegedly developed and used by Fox for the scheduling and regionalization of events. (Compl. ¶42).

Functionality and Market Context

  • The complaint alleges that the Fox Mapping Tool is used to create schedules for content in discrete time slots, such as NFL Sunday games at 1:00 p.m. and 4:00 p.m. ET, for broadcast across Fox's numerous affiliate stations in a plurality of U.S. cities. (Compl. ¶¶42, 44-45). The complaint alleges, on information and belief, that the tool uses machine learning to optimize for overall television ratings and that it updates network maps in real time based on changing conditions. (Compl. ¶¶46, 47). The complaint supports its allegations by referencing the public broadcast schedule for NFL games, which shows different games being assigned to different regions. A screenshot from a third-party website shows the NFL TV schedule for a specific date, listing games, their broadcast network (FOX or CBS), and the percentage of the U.S. covered by the broadcast. (Compl. p. 20, ¶45).

IV. Analysis of Infringement Allegations

’811 Patent Infringement Allegations

Claim Element (from Independent Claim 1) Alleged Infringing Functionality Complaint Citation Patent Citation
A computer-implemented method for dynamically generating a network map, the method comprising: receiving a schedule for a first plurality of live events scheduled to start at a first time and a second plurality of live events scheduled to start at a second time; The Fox Mapping Tool allegedly receives the schedule of NFL Sunday games, which occur at discrete time slots (e.g., 1 p.m. and 4 p.m. ET). ¶44 col. 9:67-10:3
generating, based on the schedule, a network map mapping the first plurality of live events and the second plurality of live events...to a plurality of television stations for a plurality of cities... The Fox Mapping Tool maps the scheduled NFL games to Fox's numerous affiliate stations in different U.S. cities and markets. ¶45 col. 10:4-16
wherein generating the network map comprises using a machine learning technique to optimize an overall television rating across the first plurality of live events and the second plurality of live events; The complaint alleges Fox employs a "Vice President of Data Products and ML Strategy" and uses the tool to optimize overall television ratings, citing an article about Fox achieving high ratings. ¶46 col. 10:17-20
automatically updating the network map on demand and in real time based on a change to at least one of (i) the schedule and (ii) underlying criteria... The complaint alleges "on information and belief" that the Fox Mapping Tool updates network maps in real time based on changing conditions. ¶47 col. 10:21-27
using the network map to determine for each station (i) the first live event...and (ii) the second live event...that will be displayed... On information and belief, the tool schedules and Fox broadcasts distinct NFL games on its affiliate stations in the 1 p.m. and 4 p.m. timeslots. ¶48 col. 10:28-32

’367 Patent Infringement Allegations

Claim Element (from Independent Claim 1) Alleged Infringing Functionality Complaint Citation Patent Citation
receiving one or more event parameters for series of live events, wherein the one or more event parameters comprise at least one of venue availability, venue locations, proposed ticket prices, performer fees... The Fox Mapping Tool is alleged to generate an event schedule based on event parameters, demonstrated by the fact that Fox distributes content at discrete time slots. ¶72 col. 14:2-9
providing the one or more event parameters and the one or more event target features to a machine learning (ML) model... The complaint alleges, based on Fox employing a "Vice President of Data Products and ML Strategy," that the Fox Mapping Tool uses a machine learning technique. ¶73 col. 14:10-14
iteratively training the ML model to identify relationships between different event parameters and the one or more event target features using historical data... The complaint alleges "on information and belief" that the Fox Mapping Tool has this capability and is iteratively trained using historical data. ¶74 col. 14:15-19
detecting a real-time change to the one or more user specific event parameters; The complaint alleges "on information and belief" that the Fox Mapping Tool detects real-time changes to event parameters. ¶79 col. 14:36-38
updating, via the trained ML model, the schedule for the future series of live events such that the schedule remains optimized... The complaint alleges "on information and belief" that the tool generates a schedule that remains optimized in view of real-time changes. ¶82 col. 14:43-48

Identified Points of Contention

  • Evidentiary Questions: For both patent families, the complaint relies heavily on allegations made "on information and belief" to meet key claim limitations related to the inner workings of the accused tool (e.g., "automatically updating in real time," "iteratively training," "detecting a real-time change"). A central issue will be whether Plaintiff can uncover, through discovery, direct evidence that the Fox Mapping Tool actually performs these specific, claimed functions, as the currently cited evidence (e.g., executive job titles, high ratings, public schedules) is circumstantial.
  • Scope Questions (’367 and ’960 Patents): A significant legal and technical question is whether the claims of the ’367 family, which are framed in the context of creating a schedule for a "series of live events" (e.g., a concert tour) using parameters like "venue fees" and "performer fees," can be read to cover the accused activity. Fox’s alleged activity is selecting which pre-scheduled NFL games to broadcast in which designated market regions, which may involve a different set of parameters and a different technical problem than planning a tour from scratch.

V. Key Claim Terms for Construction

  • Term: "automatically updating the network map on demand and in real time" (’811 Patent, Claim 1)

  • Context and Importance: This term is central to the alleged technical improvement over static, manual scheduling. The complaint's allegations for this element are made on "information and belief" without direct evidentiary support, making its definition critical. Practitioners may focus on this term because the level of automation and the meaning of "real time" will determine whether Fox's alleged conduct infringes.

  • Intrinsic Evidence for Interpretation:

    • Evidence for a Broader Interpretation: The specification suggests the update can be triggered by a wide range of events, including "a manual change performed by a user" or changes in underlying data like weather or news. (’811 Patent, col. 2:3-5, col. 3:22-34). This could support an interpretation where "automatic" refers to the system's response rather than the absence of any human trigger.
    • Evidence for a Narrower Interpretation: The patent contrasts the invention with the "entirely manual" prior art. (’811 Patent, col. 1:24-25). A defendant could argue that "automatically" requires a system that operates largely without human intervention or decision-making in the update process itself, beyond an initial trigger.

  • Term: "iteratively training the ML model" (’367 Patent, Claim 1)

  • Context and Importance: This limitation defines a specific process for improving the ML model's accuracy. Like the "real-time" updating element, the complaint alleges this capability "on information and belief." (Compl. ¶74). The dispute will likely center on what specific technical actions constitute "iteratively training" and whether Plaintiff can prove Fox performs them.

  • Intrinsic Evidence for Interpretation:

    • Evidence for a Broader Interpretation: The claim language itself is general, stating the training uses "historical data corresponding to one or more previous series of live events." (’367 Patent, col. 14:17-18). Plaintiff may argue any process that refines the model with new historical data meets this limitation.
    • Evidence for a Narrower Interpretation: The patent describes this as a step that "improves the accuracy of the ML model." (’367 Patent, col. 14:18-19). Defendant may argue this requires a specific, structured feedback loop where the model's performance is measured and used to retrain the model in a recurring, or "iterative," fashion, a process for which there is no direct evidence in the complaint.

VI. Other Allegations

  • Indirect Infringement: The complaint alleges that Fox proceeded with its infringing activity with "specific intent to cause (or willful blindness to causing) infringement." (Compl. ¶51). While not pleaded as a separate count, this language tracks the standard for induced infringement under 35 U.S.C. § 271(b). The factual basis is Fox's alleged development and internal use of the infringing tool.
  • Willful Infringement: Willfulness is alleged for all four patents. The complaint asserts that Fox had pre-suit knowledge of Recentive’s technology generally since at least 2018 and specific notice of the patents-in-suit via letters and licensing discussions beginning in July 2022. (Compl. ¶¶4, 36-38, 49, 52). The allegation is that Fox continued its infringement despite this knowledge, constituting objectively reckless conduct. (Compl. ¶52).

VII. Analyst’s Conclusion: Key Questions for the Case

  • A central issue will be one of evidentiary sufficiency: Can the Plaintiff, through discovery, substantiate its "information and belief" allegations that the accused "Fox Mapping Tool" performs the specific, internal machine learning functions required by the claims—namely, the "real-time updating" of the '811 patent family and the "iteratively training" of the '367 patent family?
  • A second key question will be one of claim scope: Can the claims of the '367 and '960 patents, which are described in the context of creating an optimized schedule for a series of events like a concert tour using parameters such as "venue availability" and "performer fees," be construed to cover the accused activity of selecting which pre-scheduled broadcast to air in various regional markets?
  • Finally, the case will raise a question of technical equivalence: Does the public data cited in the complaint—showing a finished broadcast schedule—provide a sufficient factual basis to plausibly allege infringement of claims directed to the underlying process of dynamically generating and optimizing that schedule using specific machine learning techniques?