PTAB: Ex parte Scott Michael Zoldi

Certain methods of organizing human activity, Transformation of an abstract idea into a patent-eligible application, Step 2b


PTAB Appeal 2021-005276

Abstract from Application on Appeal

A system and method is disclosed as using archetype-based n-grams based on an event sequence of the real-time transactions, the n-grams providing a probability based on a specific sequence of behavioral events and their likelihood, and in which high probability n-grams represent typical behaviors of customers in a same peer group, and low probability n-grams represent rare event sequences and increased risk.

Representative Claim on Appeal

  • 1. A method comprising:
  • (a) receiving, by one or more data processors from a client system, a scoring request and event data associated with a structured, ordered sequence of events, the event data comprising a concatenated string comprising one or more event characteristics, wherein the event data further comprises one or more merchant categories associated with a transaction;
  • (b) generating, by the one or more processors, one or more event vectors from the event data, at least one of the one or more event vectors representing a unique temporal trait associated with the one or more event characteristics;
  • (c) generating, by the one or more processors, a clustering of customer, account, device, or channel based on archetypes derived from event history associated with the customer, account, device, or channel;
  • (d) generating, by the one or more data processors, an n-gram for the structured, ordered sequence of events within at least one of the one or more event vectors, an n-gram representing historical occurrence of at least one event, within an associated event vector, to improve fraud detection by providing an indication of a possibility of fraudulent activity for the at least one event;
  • (e) generating, by the one or more data processors, a probability of occurrence of an event based on the n-gram within the associated event vector and the clustering of the customer, account, device, or channel;
  • (f) retrieving, by the one or more data processors, real-time transaction profiles with recursive fraud features associated with a payment card of the transaction, the real-time transaction profiles including past transaction behavior of one or more customers;
  • (g) extracting, by the one or more data processors, a probability table of the n-gram;
  • (h) generating, by the one or more data processors and based on the transaction profiles and the n-gram, a score for the event, the score representing the probability of occurrence of the event in a context of the associated clustering of the customer, account, device, or channel; and
  • (i) transmitting, by the one or more data processors and responsive to the scoring request, the score to the client system, the client system making, based on the score, approve/decline decisions for the transaction associated with the event.

Rejection of Claims on Appeal

Claims 1-5 and 7-21 are rejected under 35 U.S.C. § 101 as being directed to patent-ineligible subject matter.

Summary of Appeal Analysis

The appeal concerns an invention related to a method for detecting fraudulent activities in a structured, ordered sequence of events using event data, event vectors, and clustering. The method involves receiving a scoring request and event data associated with a transaction, generating event vectors from the event data, generating a clustering of customer, account, device, or channel based on archetypes derived from event history, generating an n-gram for the structured sequence of events, generating a probability of occurrence of an event based on the n-gram and the clustering, retrieving real-time transaction profiles with recursive fraud features associated with a payment card, extracting a probability table of the n-gram, generating a score for the event based on the transaction profiles and the n-gram, and transmitting the score to the client system. The client system makes approve/decline decisions for the transaction associated with the event based on the score.

The rejection is based on 35 U.S.C. § 101, which states that an invention is patent-eligible if it claims a “new and useful process, machine, manufacture, or composition of matter.” However, the Supreme Court has interpreted this section to include implicit exceptions, such as laws of nature, natural phenomena, and abstract ideas, which are not patentable. The patent eligibility determination uses a two-step framework described in Mayo and Alice. The first step involves determining whether the claim is directed to one of the patent-ineligible concepts, such as abstract ideas. If the claim is directed to an abstract idea, the second step involves examining the elements of the claim to determine whether it contains an “inventive concept” sufficient to “transform” the claimed abstract idea into a patent-eligible application.

The Examiner rejected claims 1-5 and 7-21 under 35 U.S.C. § 101 as being directed to patent-ineligible subject matter. The Appellant appeals this decision, arguing that the claimed method is not an abstract idea and contains an inventive concept that transforms the claimed subject matter into a patent-eligible application. The Administrative Patent Judges (Daniel S. Song, Edward A. Brown, and Brett C. Martin) have jurisdiction over the appeal under 35 U.S.C. § 6(b). The summary of the appeal demonstrates that the invention in question involves a method for detecting fraudulent activities in a sequence of events using event data, event vectors, and clustering. The method is claimed to improve fraud detection by providing an indication of the possibility of fraudulent activity for a particular event, and to generate a score for the event representing the probability of occurrence in the context of the associated clustering of the customer, account, device, or channel. The Appellant argues that the claimed method is not an abstract idea and contains an inventive concept that transforms the claimed subject matter into a patent-eligible application. The Examiner rejected the claims under 35 U.S.C. § 101, and the Appellant appeals this decision. The outcome of the appeal will be determined by the Administrative Patent Judges based on their analysis of the claimed subject matter and the arguments presented by the Appellant.