DETAILED ACTION
1. The present application, filed on or after March 13, 2013, is being examined under the first inventor to file provisions of the AIA .
This is a regular application with a claim of priority to the parent Application, Application No. 17/389,532, now U.S. Patent No. 12,229,780.
The IDS filed January 9, 2025 has been considered in this Application.
Claims 1 - 20 are pending and examined as follows:
NOTE: interviews are welcome at any stage of prosecution. Please use the AIR form, the link for which can be found at the end of this action, to schedule the interview.
Claim Rejections - 35 USC § 112
2. The following is a quotation of 35 U.S.C. 112(b):
(B) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
Claims 1 - 20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor, or for pre-AIA the applicant regards as the invention.
Claim 1 is illustrative of the rejection. Claim 1 recites the following limitation:
replacing the first unstructured data in the untransformed transaction with the first cluster ID to obtain a transformed transaction;
generating a first query including the first cluster ID and based on the transformed transaction;
Broadest Reasonable Interpretation:
The Claim is constructed in the context or functionality of receiving an “unstructured” transaction for the purpose of determining possible fraud. See 0004. The quoted limitation appears to require that “that” transaction first be “transformed” by replacing unstructured data with a cluster ID to obtain “a transformed transaction.” This appears to conflict with the specification at least at the sections quoted below.
While the specification DOES clearly describe the transformation of transactions from an unstructured state to a structured state – and it does so in the context of replacing unstructured data with a cluster ID – it DOES NOT appear to perform this function in the context of examining any single transaction for potential fraud. It appears to perform this function of transformation of transactions for the purpose of providing such transactions to the feature generator. This feature generator then derives features – based on the transformed transactions and in the context of a cluster ID – and stores the features in the “feature store” 126, as illustrated in Fig. 1C below:
PNG
media_image1.png
668
978
media_image1.png
Greyscale
It seems clear from the specification that this “storage” is for the purpose of using the features – indexed with cluster ID’s - for matching purposes with the cluster ID of a single, future untransformed transaction which is being examined for potential fraud.
Thus, the relevant paragraphs of the specification read as follows:
“[0038] Turning to Figure 1C, the query generator (115) includes functionality to generate a query (166) from an untransformed transaction (140) and/or cluster IDs (162, 163), as described in Step 256 of Figure 2 below. The query (166) may include expressions that access the untransformed transaction (140) and/or cluster IDs (162, 163). An expression may include one or more operators, such as Boolean operators, aggregation operators, arithmetic operators, etc. Examples of aggregation operators may include: average, sum, count, maximum, etc. For example, an expression may calculate the total number of transactions over the past 60 days whose unstructured data (e.g., invoice memo) matches the cluster ID assigned to the untransformed transaction.
[0040] The feature generator (116) includes functionality to derive features from transformed transactions (124). The feature generator (116) includes functionality to store the derived features in the feature store (126). Features may be derived by executing one or more queries that access the transformed transactions (124). For example, the queries may access the cluster IDs of the transformed transactions (124). The features may be derived from multiple entities in the transformed transactions (124). For example, the multiple entities may include: a merchant, a customer, a bank account, and/or a payment.
[0041] Features that are derived using queries that access one or more cluster IDs of the transformed transactions are referred to as cluster-derived features (168). In contrast, features that are derived using queries that exclude any cluster IDs are referred to as raw features (170). For example, the raw features (170) may be derived using queries that access the structured data of the transformed transactions without referring to any cluster IDs.” (Emphasis Added)
Also see 0038; 0054; and 0062. Fig. 1B illustrates the process of generating transformed transactions which are then used – as illustrated in Fig. 1C above – to derive features for the fraud determination, which receives as input a “query result.” See e.g. 0017 and 0054.
Thus, features may be derived from transformed transactions – previously executed and based on transactions entered into by “multiple entities” – and then “stored” in the feature store 126. These features are then used – subsequently – to determine potential fraud with respect to a single untransformed transaction. Therefore, there does not appear to be any support in the specification for the claimed limitation that the single untransformed transaction – under examination for potential fraud - must first be “transformed” prior to generating a fraud score. It appears only necessary that the cluster ID be known while the transaction remains in an untransformed state, until the query result is obtained.
Clarification and/or correction is required.
Claim Rejections – 35 USC § 101
2. 35 USC § 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture and composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
A. Rejection Based on Abstract Idea
Claims 1 - 20 are rejected under 35 U.S.C. § 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more. Furthermore, this rejection is based on the 2019 Revised Patent Subject Matter Eligibility Guidance (2019 PEG).
B. Statutory Categories
Claim 1 is a method Claim and therefore falls into the category of a process. Claim 9 is a system Claim and also recites a processor and therefore falls into the category of machine/manufacture. Claim 17 is also a method Claim and therefore falls into the category of a process.
C. The Claim Recites an Abstract Idea
Claim 1 is illustrative of the rejection of all claims.
Claim 1 recites the limitation:
“5receiving a transaction record comprising an untransformed transaction including a first unstructured data; generating, by a first embedding model corresponding to the first unstructured data, a first vector from the first unstructured data; matching, by a first cluster model corresponding to the first unstructured data, the first vector to a first vector cluster;;”
This limitation, as drafted, is a process that, under its broadest reasonable interpretation, constitutes a method of organizing human activity, specifically, fundamental economic principles or practices. That is, analyzing this limitation in the context of the claim as a whole, it recites a process that falls within the grouping of abstract ideas comprising certain methods of organizing human activity. Fundamental economic principles or practices are examples of such methods. In this case, the fundamental economic principle or practice is the common practice of using machine learning to detect potential fraud. One of the most common aspects of reducing the dimensionality of data is to transform unstructured data into structured data using vector embeddings. This practice occurs literally billions of times every day.
Furthermore, the mere nominal recitation of terms - such as “processor” - does not remove the claim from the category of common or abstract methods of organizing human activity.
Thus, Claim 1 recites a judicial exception, namely, an abstract idea.
D. The Claim Does Not Integrate the Abstract Idea into a Practical Application
Moreover, this judicial exception is not integrated into a practical application. The possible “additional limitations” recited in the Claim that must be considered are as follows:
responsive to matching the first vector to the first vector cluster, assigning a first cluster ID corresponding to the first vector cluster to the first vector,
wherein the first cluster ID identifies the first vector cluster;
replacing the first unstructured data in the untransformed transaction with the first cluster ID to obtain a transformed transaction;
generating a first query including the first cluster ID and based on the transformed transaction;
processing the first query to generate a query result from a plurality of features of a plurality of prior transformed transactions;
processing, by a fraud determination model, the query result to generate a fraud score for the transformed transaction;
presenting the fraud score and the first cluster ID to a user of a software application;
updating the first cluster model to add or delete or modify vector clusters to generate a set of cluster IDs,
whereby generating the set of cluster IDs does not affect an input or output of the fraud determination model.
No additional computer components are mentioned in these limitations, and those quoted above are recited at a high level of generality. In fact, most of these are common mathematical concepts as well, at least in terms of the use of algorithms to generate vectors and then clustering those vectors to assign a cluster ID to the transaction.
No other particular computer functions or computer component interactions within this system are recited. Receiving a transaction, generating a vector, clustering, assigning a cluster ID, matching the cluster ID to derived features of previous transactions, and then using a ML model to generate a fraud score are all highly common computerized functions. These steps and functions are recited at a high level of generality. These functions are among the most common and generic computer functions, and they also are directed to abstract ideas and concepts associated with processing transactions and documents having both structured and unstructured data. This is what computers do. There is no specificity nor special functionality assigned to these steps in the claim.
Analyzing these additional limitations individually, and taking the claim as a whole and as an ordered combination, it is clear that these additional limitations do not serve to integrate the abstract idea into a practical application. They do not recite a technological solution to a technological problem. They do not improve the functioning of the computer system itself. In fact, there are very few computerized system components or functions recited. Thus, these limitations fail to recite with specificity any technical function or any improvement to the functioning of the computer system itself – if any. Therefore, the claim lacks the specificity required to transform the claim from one claiming only an outcome or a result – a fraud score - to one claiming a specific way of achieving that outcome or result.
Accordingly, the recitation of these generic components amounts to no more than mere instructions “to apply” the abstract idea exception using generic computer components. That is, the additional elements recited in the claim beyond the judicial exception(s) have been evaluated to determine whether those additional elements, considered individually and in combination, integrate the judicial exception(s) into a practical application. They do not.
E. Step 2B: The Claim Does Not Recite Significantly More than the Abstract Idea
This step involves the search for an “inventive concept.” However, it is clear from the case law and the MPEP that the considerations at issue are the same as those considered above with respect to the analysis of a practical application. See MPEP 2106.05(a) – (c) and (e). In other words, these analyses sharply overlap.
Therefore, based on the above analysis, the identified additional limitations do not provide “significantly more” than the abstract idea. The claim is therefore ineligible under §101. The other independent claims are, likewise, ineligible for the same reasons as they are virtually identical to Claim 10.
F. The Dependent Claims Do Not Recite Meaningful Additional Limitations
Similarly, Claim 2 recites the same abstract idea as Claim 1 by virtue of its dependency on Claim 1. Like Claim 1, this claim does not recite sufficient additional elements to integrate the abstract idea into a practical application. Claim 2 merely recites the abstract concept of training the fraud determination model.
Claim 3 merely recites the abstract concept of the use of a second query.
Claim 4 merely recites the abstract concept of generating vectors.
Claim 5 merely recites the abstract concept of a filtering function.
Claim 6 merely recites the abstract concept of a centroid for defining a cluster ID.
Claim 7 merely recites the abstract concept of the use of embedding models.
Claim 8 merely recites the abstract concept of updating a cluster model.
Claims 9 - 20 are virtually identical to various of the aforementioned claims and are ineligible for the same reasons as set forth above.
None of these claims provide any additional meaningful limitations, non-generic computer components, or specific assignments of functionality among those components. Likewise, if at all, these claims recite only generic, computer-related limitations which are recited at such a high level of generality as to be devoid of any meaningful limitations. These limitations do not recite improvements in the functioning of the computer or to any other technology or technical field.
Therefore, these claims do not include additional elements that are sufficient to integrate the abstract idea into a practical application, nor do they amount to significantly more than the recited abstract idea because the additional elements, when considered both individually and as an ordered combination, constitute only a mere instruction to “apply” the abstract idea.
Thus, Claims 1 - 20 constitute ineligible subject matter under 35 USC § 101 as being directed to an abstract idea without more.
Claim Rejections - 35 USC § 103
4. The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1 - 30 are rejected under 35 U.S.C. §103 as being unpatentable over U.S. Patent Publication No. 2022/0382784 to Osuala et al. (hereinafter “Osuala”) in view of U.S. Patent Publication No. 2020/0142989 to Bordawekar et al. (hereinafter “Bordawekar”)
Osuala is directly on point with the claimed invention and in the same field of endeavor – determining anomalies and fraud. (See at least [0037].)
The title of Osuala is: Determining an association rule
The Abstract reads as follows:
“A method and related system, comprising receiving a dataset comprising records, where each record of the records comprises information descriptive of an event corresponding to an entity. The records may be clustered resulting in clusters having categories respectively, each cluster category being indicative of an event category of the events. One or more event attributes descriptive of the events may be determined. Records having values of the determined event attributes may be selected from the dataset. The selected records may be grouped according to a grouping criterion, the grouping criterion being based on the values of the determined event attributes. At least one association rule may be determined using the groups and the cluster identifiers, where each association rule indicates a relationship between the event categories of a respective group.” (Emphasis Added)
While using slightly different terminology, Osuala is directly on point with the claimed invention – processing unstructured data in received documents or records (e.g. natural language text in the record) to generate vectors and clustering the vectors into similarity groups to facilitate ML processing. The vectors that represent the unstructured data in the document can be replaced with a cluster ID. Furthermore, it is clear that the records are purchase or retail transactions as explained in 0002 and 0004.
Thus, it is respectfully submitted that a person of ordinary skill in the art would readily understand that an “association rule” is considered to constitute the recited term “fraud score.”
Thus, Osuala teaches as follows in Fig. 1:
PNG
media_image2.png
829
1391
media_image2.png
Greyscale
Furthermore, Osuala teaches as follows:
“[0004] In one aspect, the disclosure relates to a computer-implemented method comprising receiving a dataset comprising records, where each record of the records comprises information descriptive of an event corresponding to an entity, and clustering the records, resulting in clusters having categories respectively, each cluster category being indicative of an event category of the events, where each record of the records is associated with a cluster identifier indicating the cluster to which the record belongs. The computer-implemented method may further comprise determining one or more event attributes descriptive of the events, selecting from the dataset records having values of the determined event attributes, and grouping the selected records according to a grouping criterion, the grouping criterion being based on the values of the determined event attributes, the grouping resulting in groups, where each group comprises a set of records representing respective ones of the event categories. The computer-implemented method may further comprise determining at least one association rule using the groups and the cluster identifiers, where each association rule indicates a relationship between the event categories of a respective group.” (Emphasis Added)
Therefore, Osuala in view of Bordawekar teaches:
1. A method, comprising: receiving a transaction record comprising an untransformed transaction including a first unstructured data; (See at least Osuala: 0027 and Fig. 1 at 103 “input data.”)
generating, by a first embedding model corresponding to the first unstructured data, a first vector from the first unstructured data (See at least Fig. 1 at 101 and 105, as well as 0042 and 0084);
matching, by a first cluster model corresponding to the first unstructured data, the first vector to a first vector cluster; (See at least 0056 wherein the “data store component 102” performs this matching function.)
responsive to matching the first vector to the first vector cluster, assigning a first cluster ID corresponding to the first vector cluster to the first vector, wherein the first cluster ID identifies the first vector cluster; (See at least 0056, as mentioned above, as well as 0066 – 0069 and Fig. 4.)
replacing the first unstructured data in the untransformed transaction with the first cluster ID to obtain a transformed transaction; (See at least 0056)
generating a first query including the first cluster ID and based on the transformed transaction; (See at least 0042)
processing the first query to generate a query result from a plurality of features of a plurality of prior transformed transactions; (See at least 0027)
processing, by a fraud determination model, the query result to generate a fraud score for the transformed transaction; (See at least 0026 – 0030)
presenting the fraud score and the first cluster ID to a user of a software application; and (See at least 0066)
updating the first cluster model to add or delete or modify vector clusters to generate a set of cluster IDs, whereby generating the set of cluster IDs does not affect an input or output of the fraud determination model. (See at least 0035, wherein a person of ordinary skill in the art would readily understand that the use of cluster ID’s would obviate the need to update the fraud model when the embedding or clustering models are being updated, or vice versa.
Therefore, subject to further consideration of the cited reference and subject to the broadest reasonable interpretation of the relevant limitations, Osuala appears to teach the essential elements of Claim 1; however, for the avoidance of doubt, Bordawekar is cited for its teachings related to the generation of a fraud score.
Bordawekar is in the exact same field of endeavor as Osuala and the claimed invention in that it uses ML models to vectorize unstructured data and group the vectors into similarity groups defined by cluster ID’s, and then perform queries based on the clusters.
The title is: Method and system for supporting inductive reasoning queries over multi-modal data from relational databases
Thus, Bordawekar teaches as follows in the Abstract:
“A system and a method for performing queries, including generating text representations of features of various types of data, building a multi-modal word embedding model to capture relationships between the various types of data, and based on the multi-modal word embedding model, performing an inductive reasoning query.” (Emphasis Added)
The scores generated by Bordawekar are described in at least 0063.
Therefore, it would have been obvious to one of ordinary skill in the relevant art at the time of filing the claimed invention to have modified the transaction analysis teachings of Osaula to add the query and similarity score teaches of Bordawekar. The motivation to do so comes from Osaula. As quoted above, Osaula teaches that machine learning techniques can be applied to tokenize (e.g. vectorize) unstructured data and the vectors can be clustered into similar groups. This data structure can then be used to generate association rules, such as fraud scores. It would greatly enhance the efficiency and accuracy of the system of Osaula to use the query and similarity score teachings of Bordawekar.
With regard to Claims 2 - 8, Osuala in view of Bordawekar teaches:
2. The method of claim 1, 2. The method of claim 1, further comprising:generating a plurality of features from a plurality of prior transformed transactions,the plurality of features comprising cluster-derived features including cluster IDs of the plurality of prior transformed transactions; andtraining the fraud determination model on the cluster-derived features and non-cluster derived features of the plurality of features of the plurality of prior transformed transactions, to generate the fraud score indicating a probability that the transformed transaction is fraudulent. (See at least 0027 and 0032 relating to the attributes (e.g. features) used to train the model of Osaula.)
3. The method of claim 1, further comprising:deriving, from the plurality of prior transformed transactions and using a second query, a cluster-derived feature, wherein the second query comprises at least one cluster ID; andderiving, from the plurality of prior transformed transactions and using a third query, a raw feature, wherein the third query excludes cluster IDs,wherein the plurality of features of the plurality of prior transformed transactions comprise the cluster-derived feature and the raw feature. (See at least 0030 wherein a “pattern” is defined in terms of cluster identifiers.)
4. The method of claim 1, wherein a plurality of untransformed transactions comprises a plurality of unstructured data, and wherein transforming the plurality of untransformed transactions to the plurality of prior transformed transactions comprises:generating a plurality of vectors from the plurality of unstructured data of the plurality of untransformed transactions;assigning, for the plurality of vectors, a plurality of matching cluster IDs by matching respective vectors of the plurality of vectors with respective matching vector clusters; andreplacing the plurality of unstructured data of the plurality of untransformed transactions with the plurality of matching cluster IDs. (See at least 0042)
5. The method of claim 1, further comprising:generating a second vector from unstructured data included in another untransformed transaction;obtaining a subset of a plurality of untransformed transactions satisfying a filter criterion;generating a plurality of vectors from a plurality of unstructured data of the subset of the plurality of untransformed transactions;generating a plurality of similarity scores between the second vector and the plurality of vectors;generating another fraud score using the plurality of similarity scores; and determining, using the another fraud score, that the another untransformed transaction is fraudulent. (See at least 0042 wherein a person of ordinary skill in the art would understand that a vector “space” would relate to a plurality of vectors.)
6. The method of claim 1, wherein:the first cluster ID is based on vectors within a threshold distance of a centroid of the first vector cluster,the centroid represents an average of the vectors in the first vector cluster,wherein the first cluster ID is expressed in a fixed format comprising an integer or alphanumeric string, andthe first cluster model is trained to cluster vectors from the first unstructured data. (See at least 0096)
7. The method of claim 1, wherein a plurality of embedding models are trained to convert untransformed transactions in training data to vectors corresponding to a plurality of n-grams. (See at least 0056, wherein it would be easily determined that a plurality of models could be used.)
8. The method of claim 1, wherein updating the first cluster model comprises generating a new cluster model. (See at least 0035)
With regard to Claim 9, this claim is essentially identical to Claim 1 and is obvious for the same reasons as set forth above with respect to that claim.
With regard to Claim 10, this claim is essentially identical to Claim 8 and is obvious for the same reasons as set forth above with respect to that claim.
With regard to Claim 11, this claim is essentially identical to Claim 2 and is obvious for the same reasons as set forth above with respect to that claim.
With regard to Claim 12, this claim is essentially identical to Claim 3 and is obvious for the same reasons as set forth above with respect to that claim.
With regard to Claim 13, this claim is essentially identical to Claim 4 and is obvious for the same reasons as set forth above with respect to that claim.
With regard to Claim 14, this claim is essentially identical to Claim 5 and is obvious for the same reasons as set forth above with respect to that claim.
With regard to Claim 15, this claim is essentially identical to Claim 6 and is obvious for the same reasons as set forth above with respect to that claim.
With regard to Claim 16, this claim is essentially identical to Claim 7 and is obvious for the same reasons as set forth above with respect to that claim.
With regard to Claim 17, this claim is essentially identical to Claim 1 and is obvious for the same reasons as set forth above with respect to that claim.
With regard to Claim 18, this claim is essentially identical to Claim 8 and is obvious for the same reasons as set forth above with respect to that claim.
With regard to Claim 19, this claim is essentially identical to Claim 2 and is obvious for the same reasons as set forth above with respect to that claim.
With regard to Claim 20, this claim is essentially identical to Claim 3 and is obvious for the same reasons as set forth above with respect to that claim.
Conclusion
5. Applicant should carefully consider the following in connection with this Office Action:
A. Search and Prior Art
The search conducted in connection with this Office Action, as well as any previous Actions, encompassed the inventive concepts as defined in the Applicant’s specification. That is, the search(es) included concepts and features which are defined by the pending claims but also pertinent to significant although unclaimed subject matter. Accordingly, such search(es) were directed to the defined invention as well as the general state of the art, including references which are in the same field of endeavor as the present application as well as related fields (e.g. the use of machine learning models to detect fraud in financial transactions) Indeed, there is a plethora of prior art in these fields – sponsored search.
Therefore, in addition to prior art references cited and applied in connection with this and any previous Office Actions, the following prior art is also made of record but not relied upon in the current rejection:
U.S. Patent No. 11,016,997 to Huang et al. This reference relates to the concept of clustering unstructured data.
U.S. Patent Publication No. 2021/0326888 to Adjaoute. This reference relates to the concept of the use of feature vectors.
U.S. Patent Publication No. 2019/0065550 to Stankiewicz et al. This reference relates to the concept of the use of cluster ID’s.
U.S. Patent Publication No. 2022/0058341 to Lambert et al. This reference relates to the concept of transforming unstructured data.
Non-Patent Literature to Collados et al., “From Word to Sense Embeddings: A Survey on Vector Representations of Meaning,” Journal of Artificial Intelligence Research 63 (2008) pp. 743-788
B. Responding to this Office Action
In view of the foregoing explanation of the scope of searches conducted in connection with the examination of this application, in preparing any response to this Action, Applicant is encouraged to carefully review the entire disclosures of the above-cited, unapplied references, as well as any previously cited references. It is likely that one or more such references disclose or suggest features which Applicant may seek to claim. Moreover, for the same reasons, Applicant is encouraged to review the entire disclosures of the references applied in the foregoing rejections and not just the sections mentioned.
C. Interviews and Compact Prosecution
The Office strongly encourages interviews as an important aspect of compact prosecution. Statistics and studies have shown that prosecution can be greatly advanced by way of interviews. Indeed, in many instances, during the course of one or more interviews, the Examiner and Applicant may reach an agreement on eligible and allowable subject matter that is supported by the specification.
Interviews are especially welcomed by this examiner at any stage of the prosecution process. Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool (e.g. TEAMS).
To facilitate the scheduling of an interview, the Examiner requests the use of the AIR form as follows:
USPTO Automated Interview Request http://www.uspto.gov/interviewpractice.
Other forms of interview requests filed in this application may result in a delay in scheduling the interview because of the time required to appear on the Examiner's docket. Thus, the use of the AIR form is strongly encouraged.
D. Communicating with the Office
Any inquiry concerning this communication or earlier communications from the examiner should be directed to WILLIAM BUNKER whose telephone number is (571)272-0017. The examiner can normally be reached on M - F 8:30AM - 5:30PM, Pacific.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Abhishek Vyas, can be reached at 571-270-1836. Information regarding the status of an application, whether published or unpublished, may be obtained from the “Patent Center” system. For more information about the Patent Center system, https://patentcenter.uspto.gov/
/William (Bill) Bunker/
U.S. Patent Examiner
AU 3691
william.bunker@uspto.gov
(571) 272-0017
April 25, 2026
/ABHISHEK VYAS/Supervisory Patent Examiner, Art Unit 3691
Prosecution Insights