Notice of Pre-AIA or AIA Status
1. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Objections
2. The claims are objected to because of the following informalities, and the following is suggested to overcome the informalities and to improve claim clarity:
Claim 15 recites the limitation, “acquiring contribution information indicating a contribution of each of the transaction log information and the time interval to determination of the fraud risk.” This limitation should be amended to correct the underlined grammatical mistake. For example, this limitation could be amended to state, “acquiring contribution information indicating a contribution of each of the transaction log information and the time interval to the determination of the fraud risk.”
Claim 16 recites the limitation, “inputting the input information set into a second model and acquiring contribution information indicating a contribution of each piece of input information in the input information set to determination of the fraud risk.” This limitation should be amended to correct the underlined grammatical mistake. For example, this limitation could be amended to state, “inputting the input information set into a second model and acquiring contribution information indicating a contribution of each piece of input information in the input information set to the determination of the fraud risk.”
Appropriate correction or clarification is requested.
Claim Rejections - 35 USC §112(b)
3. 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.
The following is a quotation of 35 U.S.C. §112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
4. Claim 15 is 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 15 recites the limitation, “acquiring contribution information indicating a contribution of each of the transaction log information and the time interval to determination of the fraud risk by inputting the first model, the transaction log information, and the time interval into a second model.” It is unclear, in light of the specification, exactly what the applicant is intending to convey by stating that the first model is “input” into a second model. For example, this could mean that the fraud risk output by the first model is input into a second model. However, this could also mean that other information corresponding to the functioning of the first model is input into a second model. Paragraphs 93 and 94 of the specification describe a process for acquiring the contribution information, “by inputting input information selected from the input information set 310 and the first model 350 into the second model 380.” This portion of the specification appears to state that the contribution information is acquired by inputting information selected from the “input data set” associated with the first model into the second model. However, it is unclear if this is the intended interpretation of this claim limitation. For the purpose of examination, this limitation has been interpreted as stating, “acquiring contribution information indicating a contribution of each of the transaction log information and the time interval to determination of the fraud risk by inputting an input information set associated with the first model, the transaction log information, and the time interval into a second model.” However, appropriate correction or clarification of this claim is required. No new matter may be added.
Claim Rejections - 35 USC § 101
5. 35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
6. Claims 1-20 are rejected under 35 U.S.C. §101 because the claimed invention recites and is directed to a judicial exception to patentability (i.e., a law of nature, a natural phenomenon, or an abstract idea) and does not include an inventive concept that is “significantly more” than the judicial exception under the January 2019 and October 2019 patentable subject matter eligibility guidance (2019 PEG) analysis which follows.
Step 1
7. Under the 2019 PEG step 1 analysis, it must first be determined whether the claims are directed to one of the four statutory categories of invention (i.e., process, machine, manufacture, or composition of matter). Applying step 1 of the analysis for patentable subject matter to the claims, it is determined that the claims are directed to the statutory category of a process (claims 1-18), a machine (claim 19) and a manufacture (claim 20); where the machine and the manufacture are substantially directed to the subject matter of the process (See e.g., MPEP §2106.03). Therefore, we proceed to step 2A, Prong 1.
Step 2A, Prong 1
8. Under the 2019 PEG step 2A, Prong 1 analysis, it must be determined whether the claims recite an abstract idea that falls within one or more designated categories of patent ineligible subject matter (i.e., organizing human activity, mathematical concepts, and mental processes) that amount to a judicial exception to patentability.
Claim 1 recites the abstract idea of:
A method performed by [[an apparatus comprising one or more processors and one or more memories configured to store at least one instruction executed by the one or more processors]], the method comprising:
receiving, from [[a first user terminal]] corresponding to a first user account, a withdrawal request for transmitting digital assets to an address of a second user account;
acquiring a fraud risk for the withdrawal request by inputting, [[into a first model]], transaction log information between the first user account and the second user account and a time interval between a time point of a last deposit in an address of the first user account and a time point of the withdrawal request;
performing a process for the withdrawal request based on the fraud risk.
Here, the recited abstract idea falls within one or more of the three enumerated 2019 PEG categories of patent ineligible subject matter, to wit: certain methods of organizing human activity, which includes fundamental economic practices or principles and/or commercial interactions (e.g., facilitating a withdrawal process).
Step 2A, Prong 2
9. Under the 2019 PEG step 2A, Prong 2 analysis, the identified abstract idea to which claim 1 is directed does not include limitations or additional elements that integrate the abstract idea into a practical application.
Besides reciting the abstract idea, the limitations of claim 1 also recite generic computer components (e.g., an apparatus comprising one or more processors and one or more memories configured to store at least one instruction executed by the one or more processors, a first user terminal, and a first model). In particular, the recited features of the abstract idea are merely being applied on a computer or computing device or via software programming that is simply being used as a tool (“apply it”) to implement the abstract idea. (See e.g., MPEP §2106.05(f)). Therefore, these additional elements are recited at a high level of generality such that they amount to no more than mere instructions to apply the exception using generic computer components. In other words, the additional elements are simply used as tools to perform the abstract idea.
Claim 1 also recites the limitations, “generating training information corresponding to the withdrawal request based on the fraud risk; and training the first model based on the training information.” These limitations simply state that the first model is trained based on training information corresponding to the withdrawal request. However, the claim does not provide significant technical detail regrading how the model is trained and/or how the training information is generated. Therefore, such limitations amount to no more than merely applying generic machine learning technology and techniques to implement the abstract idea on a computer.
Thus, claim 1 does not include any limitations or additional elements that integrate the abstract idea into a practical application. As a result, claim 1 is directed to an abstract idea.
Step 2B
10. Under the 2019 PEG step 2B analysis, the additional elements of claim 1 are evaluated to determine whether they amount to something “significantly more” than the recited abstract idea. (i.e., an innovative concept). Here, the recited additional elements (e.g., an apparatus comprising one or more processors and one or more memories configured to store at least one instruction executed by the one or more processors, a first user terminal, and a first model), do not amount to an innovative concept since, as stated above in the Step 2A, Prong 2 analysis, the claims are simply using the additional elements as a tool to carry out the abstract idea (i.e., “apply it”) on a computer or computing device and/or via software programming (See e.g., MPEP §2106.05(f)). The additional elements are specified at a high level of generality such that they are being used in the claims to simply implement the abstract idea and are not themselves being technologically improved (See e.g., MPEP 2106.05(I)(A)); (See also applicant’s Specification at least Paragraphs 75-82).
Thus, claim 1 does not recite any additional elements that amount to “significantly more” than the abstract idea.
Additional Independent Claims
11. Independent claims 19 and 20 are similarly rejected under 35 U.S.C. 101 for the reasons described below:
Claim 19 recites limitations that are substantially similar to those recited in claim 1. However, the primary difference between claims 19 and 1 is that claim 19 is drafted as a system rather than a method. Similarly, as described above regarding claim 1, claim 19 recites generic computer components (e.g., an apparatus comprising one or more processors and one or more memories, a first user terminal; and a first model) that are simply being used as a tool (“apply it”) to implement the abstract idea. Therefore, since the same analysis should be used for claims 1 and 19, claim 19 is not patent eligible (See Alice Corp. Pty. Ltd. V. CLS Bank Int’l, 134 S. Ct. 2347, 2354 (2014)).
Claim 20 recites limitations that are substantially similar to those recited in claim 1. However, the primary difference between claims 20 and 1 is that claim 20 is drafted as a computer-readable medium rather than as a method. Similarly, as described above regarding claim 1, claim 20 recites generic computer components (e.g., a non-transitory computer-readable recording medium recording at least one instruction executed by one or more processors, a first user terminal, and a first model) that are simply being used as a tool (“apply it”) to implement the abstract idea. Therefore, since the same analysis should be used for claims 1 and 20, claim 20 is not patent eligible (See Alice Corp. Pty. Ltd. V. CLS Bank Int’l, 134 S. Ct. 2347, 2354 (2014)).
Dependent Claims
12. Dependent claims 2-18 are also rejected under 35 U.S.C. 101 for the reasons described below:
Claim 2 simply provides further description to the “first model” introduced in claim 1. Specifically, claim 2 states that the first model is trained based on transaction log information between two user accounts, and that the training information comprises label information. However, such limitations do not provide significant technical detail regrading how the model is trained. In other words, simply describing the type of data that is used to train the model does not amount to a technical improvement corresponding to how the model is trained and implemented. Therefore, such limitations amount to no more than merely applying generic machine learning technology and techniques to implement the abstract idea on a computer.
Claim 3 simply refines the abstract idea because it recites process steps (e.g., determining whether the fraud risk is higher than or equal to a threshold, and rejecting/approving the request based on the result) that fall under the category of organizing human activity, as described above regarding claim 1.
Claim 4 simply provides further description to the “training information” introduced in claim 1. Specifically, claim 4 states that the training information corresponds to the withdrawal request and is based on the determination that the fraud risk is higher than or equal to the first threshold risk. However, simply defining the type of data that is used to train the model does not amount to a technical improvement corresponding to how the model is trained and implemented. Therefore, such limitations amount to no more than merely applying generic machine learning technology and techniques to implement the abstract idea on a computer.
Claim 5 simply refines the abstract idea because it recites process steps (e.g., determining whether the fraud risk is higher than or equal to a threshold, performing an authentication procedure based on the determination, and rejecting/approving the request based on the result) that fall under the category of organizing human activity, as described above regarding claim 1.
Claim 6 simply provides further description to the “training information” introduced in claim 1. Specifically, claim 6 states that the training information is based on the determination that the withdrawal request is not authenticated based on the authentication procedure. However, simply defining the type of data that is used to train the model does not amount to a technical improvement corresponding to how the model is trained and implemented. Therefore, such limitations amount to no more than merely applying generic machine learning technology and techniques to implement the abstract idea on a computer.
Claim 7 simply states that the process of generating the second training information includes determining label information indicating that the transaction log information, the time interval, and the withdrawal request are fraudulent. However, the claim does not provide significant technical detail regarding how the label information is determined and/or applied to the model training process. Therefore, such limitations simply define the type of data used to train the model. Such limitations amount to no more than merely applying generic machine learning technology and techniques to implement the abstract idea on a computer.
Claim 8 recites the limitation, “determining whether a transaction pattern corresponding to the transaction log information, the time interval, and the withdrawal request is at least one of predetermined fraud patterns.” This limitation simply refines the abstract idea because it recites a process step (e.g., determining whether a transaction pattern matches a predetermined fraud pattern) that falls under the category of organizing human activity, as described above regarding claim 1.
Additionally, claim 8 recites the limitation, “determining, as the second training information, label information indicating that the transaction log information, the time interval, and the withdrawal request are fraudulent based on a determination that the transaction pattern does not correspond to the fraud patterns.” This limitation simply states that the process of generating the second training information includes determining label information indicating that the transaction log information, the time interval, and the withdrawal request are fraudulent. However, the claim does not provide significant technical detail regarding how the label information is determined and/or applied to the model training process. Therefore, such limitations simply define the type of data used to train the model. Such limitations amount to no more than merely applying generic machine learning technology and techniques to implement the abstract idea on a computer.
Claim 9 simply states that the process for acquiring the fraud risk comprises inputting information into the first model. However, the claim does not provide significant technical detail regarding how the model functions to process the input data. Therefore, such limitations amount to no more than merely applying generic machine-learning technology and techniques to implement the abstract idea on a computer.
Claims 10 and 11 simply refines the abstract idea because they recite process steps (e.g., determining the specificity of the withdrawal request, wherein the specificity of the withdrawal request indicates a degree of dissimilarity between the withdrawal request and the deposit-withdrawal pattern) that falls under the category of organizing human activity, as described above regarding claim 1.
Claim 12 simply refines the abstract idea because it recites a process step (e.g., determining a time section corresponding to the time point of the withdrawal request) that falls under the category of organizing human activity, as described above regarding claim 1. Additionally, claim 12 states that the fraud risk is acquired by inputting the determined time section into the first model. However, the claim does not provide significant technical detail regarding how the model functions to process the input data. Therefore, such limitations amount to no more than merely applying generic machine-learning technology and techniques to implement the abstract idea on a computer.
Claim 13 simply refines the abstract idea because it recites a process step (e.g., determining a digital asset classification corresponding to the digital assets requested to be withdrawn) that falls under the category of organizing human activity, as described above regarding claim 1. Additionally, claim 13 states that the fraud risk is acquired by inputting the determined digital asset classification into the first model. However, the claim does not provide significant technical detail regarding how the model functions to process the input data. Therefore, such limitations amount to no more than merely applying generic machine-learning technology and techniques to implement the abstract idea on a computer.
Claim 14 simply refines the abstract idea because it recites a process step (e.g., determining a wallet classification corresponding to an address of the second user account) that falls under the category of organizing human activity, as described above regarding claim 1. Additionally, claim 14 states that the fraud risk is acquired by inputting the determined wallet classification into the first model. However, the claim does not provide significant technical detail regarding how the model functions to process the input data. Therefore, such limitations amount to no more than merely applying generic machine-learning technology and techniques to implement the abstract idea on a computer.
Claim 15 recites the limitation, “acquiring contribution information indicating a contribution of each of the transaction log information and the time interval to determination of the fraud risk by inputting the first model, the transaction log information, and the time interval into a second model; and transmitting the contribution information to the first user terminal.” This limitation simply refines the abstract idea because it recites process steps (e.g., acquiring contribution information corresponding to the contribution of the transaction log information and the time interval to the determination of the fraud risk, and transmitting the contribution information to the first user) that fall under the category of organizing human activity, as described above regarding claim 1. Additionally, the claim does not provide significant technical detail regarding how the second model functions to make this determination. Therefore, simply applying the second model, as recited in the claim, amounts to no more than merely applying generic machine-learning technology and techniques to implement the abstract idea on a computer.
Additionally, claim 15 recites the limitation, “wherein the second model is a model trained to input a plurality of combinations of the transaction log information and the time interval into the first model and acquire contribution information of each of the transaction log information and the time interval, based on a fraud risk acquired for each of the combinations.” This limitation simply states that the second model is trained based on providing input to the first model. However, the claim does not provide significant technical detail regarding how the training process is performed and/or how the second model function to produce the desired output. Therefore, such limitations also amount to no more than merely applying generic machine-learning technology and techniques to implement the abstract idea on a computer.
Claim 16 recites the limitation, “inputting the input information set into a second model and acquiring contribution information indicating a contribution of each piece of input information in the input information set to determination of the fraud risk; and transmitting the contribution information to the first user terminal.” This limitation simply refines the abstract idea because it recites process steps (e.g., acquiring contribution information indicating the contribution of each piece of input information to the determination of the fraud risk, and transmitting the contribution information to the first user) that fall under the category of organizing human activity, as described above regarding claim 1. Additionally, the claim does not provide significant technical detail regarding how the second model functions to make this determination. Therefore, simply applying the second model, as recited in the claim, amounts to no more than merely applying generic machine-learning technology and techniques to implement the abstract idea on a computer.
Additionally, claim 16 recites the limitation, “wherein the second model is a model trained to acquire contribution information for each piece of the input information, based on a fraud risk acquired for each of a plurality of combinations of the input information by inputting the plurality of combinations of the input information into the first model.” This limitation simply states that the second model is trained to acquire contribution information based on inputting information into the first model. However, the claim does not provide significant technical detail regarding how the training process is performed. Rather, such limitations simply describe the type of information used to train the second model. Therefore, such limitations also amount to no more than merely applying generic machine-learning technology and techniques to implement the abstract idea on a computer.
Claim 17 simply recites process steps for generating third training information, and training the first model based on the third training information. However, the claim does not provide significant technical detail regarding how the training process is performed. Therefore, such limitations also amount to no more than merely applying generic machine-learning technology and techniques to implement the abstract idea on a computer.
Claim 18 simply further describe the “third training information” recited in claim 17. Simply stating that the third training data is assigned a weight based on the contribution information does not amount to a technical improvement to machine learning technology or any other technical field. The claims do not provide significant technical detail regarding how the weights are determined and/or how the weights are applied to facilitate the training process. Therefore, such limitations amount to no more than merely defining the type of data that is used to train the first model.
Thus, the dependent claims do not add any additional element or subject matter that provides a technological improvement (i.e., an integration into a practical application) that results in the claims being directed to patent eligible subject matter or include an element or feature that is significantly more than the recited abstract idea (i.e., a technological inventive concept under Step 2B).
Claim Rejections - 35 USC § 103
13. In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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, 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.
14. Claims 1, 2, 9, 19, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Kadhim (U.S. Patent No. 12423703) in view of Silva (U.S. Pre-Grant Publication No. 20040177035) and Langford (U.S. Patent No. 11934384).
Claim 1
Regarding Claim 1, Kadhim teaches:
A method performed by an apparatus comprising one or more processors and one or more memories configured to store at least one instruction executed by the one or more processors, the method comprising (See at least Col. 55, Lines 23-65: Describes a system/method comprising a processor and memory):
receiving, from a first user terminal corresponding to a first user account, a withdrawal request for transmitting digital assets to an address of a second user account (See at least Col. 22, Lines 4-16: A first user device corresponding to a first user may receive a request to send funds to a receiving user [i.e., a withdrawal request]. The request to send funds may correspond to a request to send an amount of cryptocurrency from an address of the first user to a destination address corresponding to the second user);
acquiring a fraud risk for the withdrawal request by inputting, into a first model, [[transaction log information between the first user account and the second user account and a time interval between a time point of a last deposit in an address of the first user account and a time point of the withdrawal request]] (See at least Col. 22, Lines 30-51: The system may utilize a machine learning model to determine whether the requested transaction is associated with a suspected fraudulent account [i.e., a fraud risk]. Examiner's Note: Kadhim does not explicitly teach the use of "transaction log" and "time interval" information for determining the fraud risk However, the use of these specific information sets is described by Silva and Langford as described below);
performing a process for the withdrawal request based on the fraud risk (See at least Col. 22, Lines 52-67: The system may either process or hold the requested transaction [i.e., perform a process] based on whether the transaction is associated with a suspicious destination address);
generating training information corresponding to the withdrawal request based on the fraud risk (See at least Paragraphs Col. 27, Line 16 – Col. 28, Line 11: The system may utilize real-time transaction data, such as data associated with the transaction described in Col. 22, Lines 4-67, to train a machine learning model to identify fraudulent/scam transactions. The system may may receive input regarding the fraudulent activity and input corresponding to the transaction [i.e., training information] to a training process for the machine learning model); and
training the first model based on the training information (See at least Col. 27, Line 66 – Col. 28, Line 11: The system may train the machine learning model based on the input data).
Regarding Claim 1, Kadhim does not explicitly teach, but Silva, however, does teach:
acquiring a fraud risk for the withdrawal request by inputting, into a first model, transaction log information between the first user account and the second user account and a time interval between a time point of a last deposit in an address of the first user account and a time point of the withdrawal request (See at least Paragraph 7: Describes a process for determining fraud associated with deposits and withdrawals from a user account. The system may determine the time between a large deposit and a withdrawal request to identify a fraudulent withdrawal).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the application, to combine the teachings of Kadhim and Silva in order to provide an improved system/method for allowing financial institutions to monitor accounts in an attempt to detect fraud, money laundering, and other suspicious activities (Silva: Paragraphs 2-6).
Regarding Claim 1, the combination of Kadhim and Silva does not explicitly teach, but Langford, however, does teach:
acquiring a fraud risk for the withdrawal request by inputting, into a first model, transaction log information between the first user account and the second user account and a time interval between a time point of a last deposit in an address of the first user account and a time point of the withdrawal request (See at least Col. 10, Lines 13-29: Describes a system for determining whether a transaction is fraudulent. The system may train a machine learning model to predict whether a given transaction is fraudulent based on the number of previous transactions between two accounts [i.e., transaction log information]).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the application, to combine the teachings of Kadhim, Silva, and Langford in order to provide a trained machine learning model capably of predicting whether a given transaction is fraudulent (Langford: Col. 10, Lines 13-29).
Claim 2
Regarding Claim 2, Kadhim teaches:
wherein the first model is a model trained based on training information having, as input information, [[transaction log information between two user accounts among a plurality of user accounts and a time interval between a time point of a last deposit in an address of one of the two user accounts and a time point of a withdrawal request]] (See at least Col. 27, Line 66 – Col. 28, line 11: The system may train the machine learning model based on the input data. Examiner's Note: Kadhim does not explicitly state that the model is trained based on "transaction log information" and a "time interval." However, as described above regarding claim 1, Silva and Langford do teach using such information for detecting fraud), and
receiving, from a first user terminal corresponding to a first user account, a withdrawal request for transmitting digital assets to an address of a second user account (See at least Col. 27, Lines 43-55: The input data may comprise a report indicating the destination address associated with the transaction is associated with fraudulent activity [i.e., label information indicating whether the transaction is fraudulent]).
Claim 9
Regarding Claim 9, Kadhim teaches:
wherein the acquiring the fraud risk for the withdrawal request includes: acquiring the fraud risk by inputting an input information set related to the withdrawal request into the first model (See at least Col. 27, Lines 31-55: Determining that the transaction is associated with a suspected fraudulent account may comprise utilizing a machine learning model that is trained using various inputs associated with the transaction. The transaction is analyzed by the machine learning model based on these inputs [See Col. 22, Lines 24-51]), and
wherein the input information set includes at least one of input information selected from: a time point of the withdrawal request, a type of digital asset requested to be withdrawn, a specificity of the withdrawal request, an amount obtained by converting digital assets requested to be withdrawn into a value of legal tender, a type of a wallet of the first user account, a number of logins to the first user account, deposit information in the address of the first user account for a predetermined period, or withdrawal information from the address of the first user account for a predetermined period (See at least Col. 27, Lines 31-55: The inputs provided to the machine learning model may include the time that associated transactions are initiated or received [i.e., a time point of the withdrawal request]).
Claim 19
Regarding Claim 19, Kadhim teaches:
An apparatus comprises: one or more processors; and one or more memories configured to store at least one instruction executed by the one or more processors, wherein the one or more processors are configured to execute the at least one instruction to (See at least Col. 55, Lines 23-65: Describes a system/method comprising a processor and memory):
receive, from a first user terminal corresponding to a first user account, a withdrawal request for transmitting digital assets to an address of a second user account (See at least Col. 22, Lines 4-16: A first user device corresponding to a first user may receive a request to send funds to a receiving user [i.e., a withdrawal request]. The request to send funds may correspond to a request to send an amount of cryptocurrency from an address of the first user to a destination address corresponding to the second user);
acquire a fraud risk for the withdrawal request by inputting, into a first model, [[transaction log information between the first user account and the second user account and a time point of a last deposit in an address of the first user account]] (See at least Col. 22, Lines 30-51: The system may utilize a machine learning model to determine whether the requested transaction is associated with a suspected fraudulent account [i.e., a fraud risk]. Examiner's Note: Kadhim does not explicitly teach the use of "transaction log" and "time interval" information for determining the fraud risk However, the use of these specific information sets is described by Silva and Langford as described below);
perform a process for the withdrawal request based on the fraud risk (See at least Col. 22, Lines 52-67: The system may either process or hold the requested transaction [i.e., perform a process] based on whether the transaction is associated with a suspicious destination address);
generate training information corresponding to the withdrawal request based on the fraud risk (See at least Paragraphs Col. 27, Line 16 – Col. 28, Line 11: The system may utilize real-time transaction data, such as data associated with the transaction described in Col. 22, Lines 4-67, to train a machine learning model to identify fraudulent/scam transactions. The system may may receive input regarding the fraudulent activity and input corresponding to the transaction [i.e., training information] to a training process for the machine learning model); and
train the first model based on the training information (See at least Col. 27, Line 66 – Col. 28, Line 11: The system may train the machine learning model based on the input data).
Regarding Claim 19, Kadhim does not explicitly teach, but Silva, however, does teach:
acquire a fraud risk for the withdrawal request by inputting, into a first model, transaction log information between the first user account and the second user account and a time point of a last deposit in an address of the first user account (See at least Paragraph 7: Describes a process for determining fraud associated with deposits and withdrawals from a user account. The system may determine the time between a large deposit and a withdrawal request to identify a fraudulent withdrawal).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the application, to combine the teachings of Kadhim and Silva in order to provide an improved system/method for allowing financial institutions to monitor accounts in an attempt to detect fraud, money laundering, and other suspicious activities (Silva: Paragraphs 2-6).
Regarding Claim 19, the combination of Kadhim and Silva does not explicitly teach, but Langford, however, does teach:
acquire a fraud risk for the withdrawal request by inputting, into a first model, transaction log information between the first user account and the second user account and a time point of a last deposit in an address of the first user account (See at least Col. 10, Lines 13-29: Describes a system for determining whether a transaction is fraudulent. The system may train a machine learning model to predict whether a given transaction is fraudulent based on the number of previous transactions between two accounts [i.e., transaction log information]).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the application, to combine the teachings of Kadhim, Silva, and Langford in order to provide a trained machine learning model capably of predicting whether a given transaction is fraudulent (Langford: Col. 10, Lines 13-29).
Claim 20
Regarding Claim 20, Kadhim teaches:
A non-transitory computer-readable recording medium recording at least one instruction executed by one or more processors, wherein the at least one instruction causes the one or more processors to (See at least Col. 55, Lines 23-65: Describes a system/method comprising a processor and memory):
receive, from a first user terminal corresponding to a first user account, a withdrawal request for transmitting digital assets to an address of a second user account (See at least Col. 22, Lines 4-16: A first user device corresponding to a first user may receive a request to send funds to a receiving user [i.e., a withdrawal request]. The request to send funds may correspond to a request to send an amount of cryptocurrency from an address of the first user to a destination address corresponding to the second user);
acquire a fraud risk for the withdrawal request by inputting, into a first model, [[transaction log information between the first user account and the second user account and a time point of a last deposit in an address of the first user account]] (See at least Col. 22, Lines 30-51: The system may utilize a machine learning model to determine whether the requested transaction is associated with a suspected fraudulent account [i.e., a fraud risk]. Examiner's Note: Kadhim does not explicitly teach the use of "transaction log" and "time interval" information for determining the fraud risk However, the use of these specific information sets is described by Silva and Langford as described below);
perform a process for the withdrawal request based on the fraud risk (See at least Col. 22, Lines 52-67: The system may either process or hold the requested transaction [i.e., perform a process] based on whether the transaction is associated with a suspicious destination address);
generate training information corresponding to the withdrawal request based on the fraud risk (See at least Paragraphs Col. 27, Line 16 – Col. 28, Line 11: The system may utilize real-time transaction data, such as data associated with the transaction described in Col. 22, Lines 4-67, to train a machine learning model to identify fraudulent/scam transactions. The system may may receive input regarding the fraudulent activity and input corresponding to the transaction [i.e., training information] to a training process for the machine learning model); and
train the first model based on the training information (See at least Col. 27, Line 66 – Col. 28, Line 11: The system may train the machine learning model based on the input data).
Regarding Claim 20, Kadhim does not explicitly teach, but Silva, however, does teach:
acquire a fraud risk for the withdrawal request by inputting, into a first model, transaction log information between the first user account and the second user account and a time point of a last deposit in an address of the first user account (See at least Paragraph 7: Describes a process for determining fraud associated with deposits and withdrawals from a user account. The system may determine the time between a large deposit and a withdrawal request to identify a fraudulent withdrawal).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the application, to combine the teachings of Kadhim and Silva in order to provide an improved system/method for allowing financial institutions to monitor accounts in an attempt to detect fraud, money laundering, and other suspicious activities (Silva: Paragraphs 2-6).
Regarding Claim 20, the combination of Kadhim and Silva does not explicitly teach, but Langford, however, does teach:
acquire a fraud risk for the withdrawal request by inputting, into a first model, transaction log information between the first user account and the second user account and a time point of a last deposit in an address of the first user account (See at least Col. 10, Lines 13-29: Describes a system for determining whether a transaction is fraudulent. The system may train a machine learning model to predict whether a given transaction is fraudulent based on the number of previous transactions between two accounts [i.e., transaction log information]).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the application, to combine the teachings of Kadhim, Silva, and Langford in order to provide a trained machine learning model capably of predicting whether a given transaction is fraudulent (Langford: Col. 10, Lines 13-29).
15. Claims 3-5 are rejected under 35 U.S.C. 103 as being unpatentable over Kadhim (U.S. Patent No. 12423703) in view of Silva (U.S. Pre-Grant Publication No. 20040177035) and Langford (U.S. Patent No. 11934384), and in further view of Hanis (U.S. Pre-Grant Publication No. 20180253737).
Claim 3
Regarding Claim 3, Kadhim teaches:
wherein the performing the process for the withdrawal request includes: determining whether the fraud risk is higher than or equal to [[a first threshold risk]] (See at least Col. 22, lines 30-51: The system may utilize a machine learning model to determine whether the requested transaction is associated with a suspected fraudulent account [i.e., a fraud risk]. Examiner's Note: Kadhim does not explicitly teach the use of a "threshold" for determining whether the transaction is fraudulent. Rather, Kadhim simply states that the system determines whether the transaction is associated with a fraudulent destination address. While it would be obvious to one of ordinary skill in the art that there would be some form of "threshold" to indicate whether the transaction is fraudulent, Hanis has been referenced to more explicitly teach the use of a threshold value for making this determination); and
transmitting, to the first user terminal, information indicating that the withdrawal request is rejected, based on a determination that the fraud risk is higher than or equal to the first threshold risk (See at least Col. 22, Line 53 – Col. 23, Line 12: If the request does include a suspicious destination address, the payment service system may hold the requested transaction. The payment service system may send a notification of fraudulent activity associated with the destination address to the first user device), and
transmitting digital assets corresponding to the withdrawal request to the address of the second user account, based on a determination that the fraud risk is lower than the first threshold risk (See at least Col. 22, Lines 52-57: If the request does not include a suspicious destination address, the payment service system may process the requested transaction. The payment service system may transfer the funds from the account of the sending user to the receiving user).
Regarding Claim 3, the combination of Kadhim, Silva, and Langford does not explicitly teach, but Hanis, however, does teach:
wherein the performing the process for the withdrawal request includes: determining whether the fraud risk is higher than or equal to a first threshold risk (See at least Paragraph 61: Describes a system for determining the risk of fraud associated with a transaction. The system may determine a risk score for the transaction and compare the risk score to a threshold. The system may perform different processes based on whether the risk score exceeds a threshold).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the application, to combine the teachings of Kadhim, Silva, Langford, and Hanis in order to provide a system that allows an enterprise to effectively detect fraud, and avoid the often significant losses associated with fraud (Hanis: Paragraphs 1-3)
Claim 4
Regarding Claim 4, Kadhim teaches:
wherein the generating the training information corresponding to the withdrawal request includes: generating first training information corresponding to the withdrawal request based on the determination that the fraud risk is higher than or equal to [[the first threshold risk]] (See at least Col. 27, Line 16 – Col. 28, Line 11: The system may utilize real-time transaction data, such as data associated with the transaction described in Col. 22, Lines 4-67, to train a machine learning model to identify fraudulent/scam transactions. The system may may receive input regarding the fraudulent activity and input corresponding to the transaction [i.e., training information] to a training process for the machine learning model. Examiner's Note: As described above, Kadhim does not explicitly teach the use of a "threshold" for determining that the transaction is associated with fraud. However, this limitation is disclosed by Hanis as described above), and
wherein the training of the first model, based on the training information, comprises training the first model based on the first training information (See at least Col. 27, Line 66 – Col. 28, line 11: The system may train the machine learning model based on the input data).
Claim 5
Regarding Claim 5, Kadhim teaches:
performing an authentication procedure for the withdrawal request [[based on a determination that the fraud risk is higher than or equal to the second threshold risk]] (See at least Col. 23, Lines 35-54: The payment service system may request to verify the identity of the receiving user to receive the funds [i.e., perform an authentication procedure] in response to determining that the request is associated with a suspicious destination address. Examiner's Note: Kadhim does not explicitly teach the use of a "second threshold" for determining whether an authentication procedure should be initiated. However, this limitation is disclosed by Hanis as described above); and
determining whether the withdrawal request is authenticated based on the authentication procedure (See at least Col. 23, Line 65 – Col. 24, Line 10: The payment service system may determine whether the identity verification data is valid and otherwise acceptable along with degree of confidence in the assessment), and
transmitting, to the first user terminal, information indicating the withdrawal request is rejected, based on a determination that the withdrawal request is not authenticated based on the authentication procedure (See at least Col. 24, lines 11-26: If the payment service system determines that the account has not been verified, then the payment service system may at step return funds to the first user [i.e., returning the funds corresponds to an indication that the transaction has not been authenticated]).
transmitting digital assets corresponding to the withdrawal request to the address of the second user account based on a determination that the withdrawal request is authenticated based on the authentication procedure (See at least Col. 24, lines 11-26: If the payment service system determines that the account has been verified, then the payment service system may process the transaction and send the funds to the destination address associated with the receiving user).
Regarding Claim 5, the combination of Kadhim, Silva, and Langford does not explicitly teach, but Hanis, however, does teach:
wherein the performing the process for the withdrawal request includes: determining whether the fraud risk is higher than or equal to a second threshold risk (See at least Paragraph 61: The system may determine if the risk score is below a second threshold. If the risk score is below the second threshold, the system may request additional authentication information from the requestor).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the application, to combine the teachings of Kadhim, Silva, Langford, and Hanis in order to provide a system that allows an enterprise to effectively detect fraud, and avoid the often significant losses associated with fraud (Hanis: Paragraphs 1-3)
16. Claims 10 and 11 are rejected under 35 U.S.C. 103 as being unpatentable over Kadhim (U.S. Patent No. 12423703) in view of Silva (U.S. Pre-Grant Publication No. 20040177035) and Langford (U.S. Patent No. 11934384), and in further view of Bishop (U.S. Pre-Grant Publication No. 20110184845).
Claim 10
Regarding Claim 10, the combination of Kadhim, Silva, and Langford does not explicitly teach, but Bishop, however, does teach:
wherein the acquiring the fraud risk for the withdrawal request further includes: determining the specificity of the withdrawal request based on a deposit-withdrawal pattern shown in a deposit-withdrawal record of the first user account for a preset period (See at least Paragraph 50: Describes a system for assigning "weightings" [i.e., a specificity] to activities initiated via a user account, such as deposits and withdrawals. The system may identify patterns associated with the account, and weightings may be assigned to particular activities based on how closely they correlate to the predetermined patterns).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the application, to combine the teachings of Kadhim, Silva, Langford, and Bishop in order to provide an improved system/method that reveals fraudulent activity and/or account deterioration (Bishop: Paragraph 50).
Claim 11
Regarding Claim 11, the combination of Kadhim, Silva, and Langford does not explicitly teach, but Bishop, however, does teach:
wherein the specificity of the withdrawal request indicates a degree of dissimilarity between the withdrawal request and the deposit-withdrawal pattern, and wherein the fraud risk acquired from the first model increases in proportion to the specificity of the withdrawal request (See at least Paragraph 50: The weightings are representative of how closely a particular activity correlates to a predetermined pattern. For example, a negative weighting may be assigned to a comparison that reveals activity uncharacteristic of a pattern, and a positive weighting may be assigned to a comparison that reveals activity characteristic of a pattern).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the application, to combine the teachings of Kadhim, Silva, Langford, and Bishop in order to provide an improved system/method that reveals fraudulent activity and/or account deterioration (Bishop: Paragraph 50).
17. Claim 12 is rejected under 35 U.S.C. 103 as being unpatentable over Kadhim (U.S. Patent No. 12423703) in view of Silva (U.S. Pre-Grant Publication No. 20040177035) and Langford (U.S. Patent No. 11934384), and in further view of Mossoba (U.S. Pre-Grant Publication No. 20200027092).
Claim 12
Regarding Claim 12, the combination of Kadhim, Silva, and Langford does not explicitly teach, but Mossoba, however, does teach:
wherein the acquiring the fraud risk for the withdrawal request further includes: determining a time section corresponding to the time point of the withdrawal request among a plurality of time sections (See at least Paragraph 47: Describes a system for detecting fraudulent transactions. The system may identify “safe zone rules” based an analysis of the user’s transactions. For example, the system may determine that any transaction under $15 between the hours of 8 a.m. and 10 a.m. [i.e., a time section] are not fraudulent. A plurality of these rules may be identified and implemented); and
acquiring the fraud risk by further inputting information indicating the determined time section into the first model (See at least Paragraph 47: A machine learning model may be implemented to analyze the user’s transactions and implement the determined rules).
Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the application, to combine the teachings of Kadhim, Silva, Langford, and Mossoba in order to provide an improved system for analyzing a user’s transaction history to create rules regarding the detection of fraud in the user’s transactions (Mossoba: Paragraphs 32 and 44).
Citation of Pertinent Prior Art
18. The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Ronca (U.S. Pre-Grant Publication No. 20150363770): Describes systems that relate generally to cryptocurrency transactions, and more specifically, to a cryptocurrency transaction payment system. The system may identify suspicious or seemingly fraudulent past transactions associated with a user profile.
Patel (U.S. Pre-Grant Publication No. 20220101192): Describes systems that relate generally to transactions and, more specifically, to a machine learning framework for detecting fraudulent transactions.
Gelda (U.S. Pre-Grant Publication No. 20220414665): Describes systems that relate generally to determining fraud and, in some non-limiting embodiments or aspects, to systems, methods, and computer program products for determining fraud.
Osborn (U.S. Pre-Grant Publication No. 20230298016): Describes systems and methods for validating asset destinations in blockchain networks and more particularly to improved destination blockchain address verification for digital asset transfers using machine learning.
Tekle (U.S. Patent No. 10108968): Describes systems and methods that facilitate the detection of suspected fraudulent advertising accounts in computing systems and networks.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to WILLIAM D NEWLON whose telephone number is (571)272-4407. The examiner can normally be reached Mon - Fri 8:30 - 4:30.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Matthew Gart can be reached at (571) 272-3955. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/WILLIAM D NEWLON/Examiner, Art Unit 3696
/MATTHEW S GART/Supervisory Patent Examiner, Art Unit 3696
Prosecution Insights