PTAB

IPR2022-00439

Ebates Performance Marketing Inc v. IBM Corp

Log In
Key Events
Petition

1. Case Identification

2. Patent Overview

  • Title: Customer Self Service System Resource Results Annotator
  • Brief Description: The ’676 patent discloses a customer self-service system for resource search and selection. The system applies an "ordering and annotation function" that maps a "user context vector" with a query response set to generate and present a personalized, "annotated response set" to the user.

3. Grounds for Unpatentability

Ground 1: Claims 1-2 are obvious over Uchiyama alone or in combination with Wasfi

  • Prior Art Relied Upon: Uchiyama (Application # 2002/0065802) and Wasfi (a 1998 publication titled "Collecting User Access Patterns for Building User Profiles and Collaborative Filtering").
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Uchiyama, which discloses a system for personalizing internet search results, teaches nearly all limitations of claim 1. Uchiyama’s system collects user data to create user profiles and uses that information to re-order search results to display URLs most compatible with a user’s profile higher in the list. This meets the limitations of receiving a resource response set and applying an ordering and annotation function. For the "user context vector" limitation, Petitioner contended that while Uchiyama discloses user profiles, it does not explicitly label them as vectors.
    • Motivation to Combine: A person of ordinary skill in the art (POSITA) would have been motivated to represent Uchiyama's user profiles as vectors, a common practice for handling large data sets in information retrieval. Wasfi explicitly taught using vector space representation to compare user profile vectors with web page vectors via a similarity function to rank results. A POSITA would combine Wasfi's vector-based comparison method with Uchiyama’s system to efficiently implement the comparison between user profiles and search results, which was Uchiyama’s stated goal.
    • Expectation of Success: A POSITA would have had a reasonable expectation of success because using vectors and mathematical functions for comparison was a well-understood and routine technique in the field of information retrieval.

Ground 2: Claim 3 is obvious over the Ground 1 references in view of Lee

  • Prior Art Relied Upon: Uchiyama, Wasfi, and Lee (Patent 6,463,428).
  • Core Argument for this Ground:
    • Prior Art Mapping: Claim 3 adds the limitation that annotations include elements for "bolding" one or more resource results. Petitioner asserted that the primary references teach annotating results by re-ordering them or using a "compatibility gauge."
    • Motivation to Combine: Lee explicitly teaches displaying more relevant database search results using bold text and bold lines to distinguish them from less relevant results. A POSITA would combine Lee's teaching with the Uchiyama/Wasfi system because bolding is a simple, common, and space-efficient method for visually indicating relevance to a user, improving the interface's usability.
    • Expectation of Success: Modifying Uchiyama’s system to bold text based on relevance scores would have been a straightforward programming task.

Ground 3: Claim 4 is obvious over the Ground 1 references in view of Littlefield

  • Prior Art Relied Upon: Uchiyama, Wasfi, and Littlefield (Patent 6,564,208).
  • Core Argument for this Ground:
    • Prior Art Mapping: Claim 4 adds limitations for determining primary results to be displayed on a first screen and secondary results for presentation on a secondary screen.
    • Motivation to Combine: Littlefield addresses the known problem of search queries returning too many results to display at once. It teaches the common solution of paginating results, displaying a first page with the top N hits and providing a control to access subsequent pages. A POSITA would apply this conventional display technique to Uchiyama’s system to manage large result sets and avoid overwhelming the user, thereby improving search efficiency.
    • Expectation of Success: Implementing result pagination was a standard feature in search engines and would have been a routine modification.

Ground 4: Claims 5-13 are obvious over the Ground 1 references in view of Boyan

  • Prior Art Relied Upon: Uchiyama, Wasfi, and Boyan (a 1997 publication titled "A Machine Learning Architecture for Optimizing Web Search Engines").
  • Core Argument for this Ground:
    • Prior Art Mapping: Claims 5-13 introduce concepts of an adaptive system that improves over time based on user feedback, including a database of user interactions, an "annotation scoring metric," a supervised learning algorithm, and an adaptable annotation function. Petitioner argued Uchiyama discloses a goal of improving results over time by collecting user data but fails to teach the specific mechanism for doing so.
    • Motivation to Combine: Boyan discloses a machine learning architecture that optimizes a search engine’s ranking function based on historical user interaction data (i.e., the rank of links a user selects). It teaches using this implicit feedback to calculate a performance metric and employing a supervised learning algorithm ("simulated annealing") to tune the ranking function to improve performance. A POSITA would combine Boyan's machine learning methods with Uchiyama’s system to provide the specific back-end mechanism needed to achieve Uchiyama's stated goal of dynamically improving search relevance.
    • Expectation of Success: A POSITA would have reasonably expected success in implementing Boyan’s optimization techniques, as Uchiyama already provided the necessary infrastructure for collecting the user interaction data required by Boyan’s method.

4. Key Claim Construction Positions

  • Petitioner argued that the terms "mechanism for receiving a resource response set," "mechanism for receiving a user context vector," and "processing mechanism for receiving user interaction data" are means-plus-function terms that are indefinite under 35 U.S.C. §112 because the specification fails to disclose adequate corresponding structure.
  • However, Petitioner contended that under Intel Corp. v. Qualcomm Inc., the Board can and should still conduct an obviousness analysis because the claimed functions are performable by a general-purpose computer or processor. For the purpose of the inter partes review (IPR), Petitioner proposed that the Board construe the structure for these terms as "a general-purpose computer" or "a general-purpose computer processor."

5. Arguments Regarding Discretionary Denial

  • Petitioner argued against discretionary denial under Fintiv, asserting that the parallel district court litigation was in its earliest stages. At the time of filing, a scheduling order had only recently been entered, no significant discovery had occurred, and trial was scheduled nearly two years in the future.
  • Furthermore, Petitioner stipulated that it would not pursue in the litigation the same prior art or arguments asserted in the IPR petition, which weighs heavily against denial.

6. Relief Requested

  • Petitioner requested the institution of an IPR and the cancellation of claims 1-13 of the ’676 patent as unpatentable.