# US 2021/0173916 A1 · Systems and methods for dynamic passphrases

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0173916 A1
- Document ID: US20210173916A1
- Status: Published application
- Kind: A1
- Application number: 17/129,631
- Filed: 2020-12-21
- Published: 2021-06-10
- Earliest claimed: 2018-07-24
- Assignees: Royal Bank of Canada
- Applicants: Royal Bank of Canada
- Inventors: Edison U. ORTIZ; Mohammad Abuzar SHAIKH; Margaret Inez SALTER; Sarah Rachel Waigh Yean WILKINSON; Arya POURTABATABAIE; Iustina-Miruna VINTILA; Steven FERNANDES; Sumit Kumar JHA
- Web page: https://patentgrep.com/patent/US20210173916A1

## CPC

- G06F 21/32: using biometric data, e.g. fingerprints, iris scans or voiceprints
- G06F 18/2113: by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation
- G06F 18/22: Matching criteria, e.g. proximity measures
- G06F 18/24137: {Distances to cluster centroïds}
- G06F 18/251: {of input or preprocessed data}
- G06F 21/31: User authentication
- G06F 21/33: using certificates
- G06F 21/46: by designing passwords or checking the strength of passwords
- G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
- G06N 3/045: Combinations of networks
- G06N 3/0455: Auto-encoder networks; Encoder-decoder networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/0475: Generative networks
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G06N 3/094: Adversarial learning
- G06N 3/096: Transfer learning
- G06N 20/00: Machine learning
- G06Q 20/108: {Remote banking, e.g. home banking}
- G06Q 20/1085: {involving automatic teller machines [ATMs]}
- G06Q 20/202: {Interconnection or interaction of plural electronic cash registers [ECR] or to host computer, e.g. network details, transfer of information from host to ECR or from ECR to ECR}
- G06Q 20/206: {comprising security or operator identification provisions, e.g. password entry}
- G06Q 20/322: {Aspects of commerce using mobile devices [M-devices]}
- G06Q 20/3267: {In-app payments}
- G06Q 20/3276: {using a pictured code, e.g. barcode or QR-code, being read by the M-device}
- G06Q 20/36: using electronic wallets or electronic money safes
- G06Q 20/38215: {Use of certificates or encrypted proofs of transaction rights}
- G06Q 20/3825: {Use of electronic signatures}
- G06Q 20/385: {using an alias or single-use codes}
- G06Q 20/4014: {Identity check for transactions}
- G06Q 20/40145: {Biometric identity checks}
- G06Q 20/4016: {involving fraud or risk level assessment in transaction processing}
- G06V 10/17: {using hand-held instruments}
- G06V 10/454: {Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN]}
- G06V 10/764: using classification, e.g. of video objects
- G06V 10/771: Feature selection, e.g. selecting representative features from a multi-dimensional feature space
- G06V 10/776: Validation; Performance evaluation
- G06V 10/803: {of input or preprocessed data}
- G06V 10/82: using neural networks
- G06V 40/165: {using facial parts and geometric relationships}
- G06V 40/171: {Local features and components; Facial parts (eye characteristics G06V40/18); Occluding parts, e.g. glasses; Geometrical relationships}
- G06V 40/172: {Classification, e.g. identification}
- G06V 40/20: Movements or behaviour, e.g. gesture recognition (recognition of facial expressions G06V40/16)
- G06V 40/40: Spoof detection, e.g. liveness detection
- G07G 1/009: {the reader being an RFID reader}
- G10L 15/25: using position of the lips, movement of the lips or face analysis
- G10L 17/24: the user being prompted to utter a password or a predefined phrase
- H04L 9/0866: {involving user or device identifiers, e.g. serial number, physical or biometrical information, DNA, hand-signature or measurable physical characteristics}
- H04L 9/3213: {using tickets or tokens, e.g. Kerberos (network architectures or network communication protocols for entities authentication using tickets in a packet data network H04L63/0807)}
- H04L 9/3226: {using a predetermined code, e.g. password, passphrase or PIN (network architectures or network communication protocols for supporting authentication of entities using passwords in a packet data network H04L63/083)}
- H04L 9/3231: {Biological data, e.g. fingerprint, voice or retina (network architectures or network communication protocols for supporting authentication of entities using biometrical features in a packet data network H04L63/0861)}
- H04L 9/3271: {using challenge-response}
- H04L 63/0838: {using one-time-passwords}
- H04L 63/0876: {based on the identity of the terminal or configuration, e.g. MAC address, hardware or software configuration or device fingerprint}
- G06N 3/047: Probabilistic or stochastic networks
- G10L 17/14: Use of phonemic categorisation or speech recognition prior to speaker recognition or verification
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G07G: REGISTERING THE RECEIPT OF CASH, VALUABLES, OR TOKENS
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

Systems, devices, methods, and computer readable media are provided in various embodiments relating to generating a dynamic challenge passphrase data object. The method includes establishing, a plurality of data record clusters, representing a mutually exclusive set of structured data records of an individual, ranking the plurality of feature data fields based on a determined contribution value of each feature data field relative to the establishing of the data record cluster, and identifying, using the ranked plurality of feature data fields, a first and a second feature data field of the plurality of feature data fields. The method includes generating the dynamic challenge passphrase data object, wherein the first or the second feature data field is used to establish a statement string portion, and a remaining one of the first or the second feature data field is used to establish a question string portion and a correct response string.

## Claims

### Claim 1

A computer system for generating a dynamic challenge passphrase data object, the computer system comprising a non-transitory computer readable storage device, computer memory, and a processor configured to:
retrieve, from one or more data sources, one or more data objects storing a plurality of structured data records associated with an individual, each structured data record of the plurality of structured data records including at least a plurality of data values corresponding to a feature data field of a plurality of feature data fields;
establish, using an unsupervised machine learning model data architecture, a plurality of data record clusters, each data record cluster representing a mutually exclusive set of structured data records of the plurality of structured data records;
for a data record cluster having a largest number of structured data records, rank the plurality of feature data fields based on a determined contribution value of each feature data field relative to the establishing of the data record cluster having the largest number of structured data records;
identify, using the ranked plurality of feature data fields, a first and a second feature data field of the plurality of feature data fields having the two highest determined contribution values; and
generate the dynamic challenge passphrase data object, wherein the first or the second feature data field is used to establish a statement string portion, and a remaining one of the first or the second feature data field is used to establish a question string portion and a correct response string;
wherein the dynamic challenge passphrase data object is used to validate a challenge response data structure using the correct response string; and
wherein the dynamic challenge passphrase data object is automatically generated from one or more data sources such that an individual is not required to establish the dynamic challenge passphrase data object during an initial registration.

### Claim 2 (depends on 1)

The system of claim 1, wherein for each of the plurality of structured data records, the processor is configured to transform textual data elements into numerical vectors having numerical values corresponding to a level of semantic similarity between underlying words represented by the numerical vectors.

### Claim 3 (depends on 1)

The system of claim 1, wherein the processor is configured to first generate a baseline statement string using the first or the second feature data field, and then the dynamic challenge passphrase data object is generated based on one of a plurality of question strings generated based on the baseline statement string.

### Claim 4 (depends on 3)

The system of claim 3, wherein the plurality of question strings generated based on the baseline statement string are each based on removal of a word from the baseline statement string, and the correct response string corresponds to the removed word from the baseline statement string.

### Claim 5 (depends on 1)

The system of claim 1, wherein the correct response string is used to validate a response data object to provision access to an electronic resource.

### Claim 6 (depends on 5)

The system of claim 5, wherein the processor is configured to:
receive a script-reading video data set capturing a portion of or an entirety of the individual's face while the individual is speaking words corresponding to a script data structure, the script data structure having a sequence of pre-identified phonemes or phoneme transitions, the pre-identified phonemes or phoneme transitions including at least one overlapping phoneme or phoneme transition required to be spoken when speaking words of the correct response string;
extract, from the script-reading video data set, a data subset representing the one or more facial or lip features of the requesting individual corresponding to each phoneme or phoneme transition corresponding to the sequence of pre-identified phonemes or phoneme transitions;
train, one or more baseline machine learning data model architectures, each baseline machine learning data model architecture of the one or more baseline machine learning data model architectures corresponding to a corresponding pre-identified phoneme or phoneme transition of the script data structure such that parameters of the baseline machine learning data model architectures are tuned based on the corresponding one or more facial or lip features;
receive an answer-reading video data set capturing a portion of or an entirety of the individual's face while the individual is speaking the words corresponding to the correct response string; and
process, the answer-reading video data set, using the one or more baseline machine learning data model architectures corresponding to the at least one overlapping phoneme or phoneme transition to determine an overall classification similarity score;
wherein the provisioning of access to the electronic resource only occurs if the correct response string has been validated against the correct response string and the overall classification similarity score is greater than a pre-defined threshold similarity score.

### Claim 7 (depends on 6)

The system of claim 6, wherein the script data structure includes words corresponding to a phonetic pangram or a holo-alphabetic sentence.

### Claim 8 (depends on 1)

The system of claim 1, wherein the dynamic challenge passphrase data object is processed to control rendering of a graphical user interface where the statement string portion and the question string portion are displayed as textual display elements on a computer display.

### Claim 9 (depends on 5)

The system of claim 5, wherein the secure electronic resource is a secure webpage.

### Claim 10 (depends on 9)

The system of claim 9, wherein the secure webpage is an online banking website.

### Claim 11

A method for generating a dynamic challenge passphrase data object, the method comprising:
retrieving, from one or more data sources, one or more data objects storing a plurality of structured data records associated with an individual, each structured data record of the plurality of structured data records including at least a plurality of data values corresponding to a feature data field of a plurality of feature data fields;
establishing, using an unsupervised machine learning model data architecture, a plurality of data record clusters, each data record cluster representing a mutually exclusive set of structured data records of the plurality of structured data records;
for a data record cluster having a largest number of structured data records, ranking the plurality of feature data fields based on a determined contribution value of each feature data field relative to the establishing of the data record cluster having the largest number of structured data records;
identifying, using the ranked plurality of feature data fields, a first and a second feature data field of the plurality of feature data fields having the two highest determined contribution values; and
generating the dynamic challenge passphrase data object, wherein the first or the second feature data field is used to establish a statement string portion, and a remaining one of the first or the second feature data field is used to establish a question string portion and a correct response string;
wherein the dynamic challenge passphrase data object is used to validate a challenge response data structure using the correct response string; and
wherein the dynamic challenge passphrase data object is automatically generated from one or more data sources such that an individual is not required to establish the dynamic challenge passphrase data object during an initial registration.

### Claim 12 (depends on 11)

The method of claim 11, further comprising, for each of the plurality of structured data records, transforming textual data elements into numerical vectors having numerical values corresponding to a level of semantic similarity between underlying words represented by the numerical vectors.

### Claim 13 (depends on 11)

The method of claim 11, further comprising generating a baseline statement string using the first or the second feature data field, and then the dynamic challenge passphrase data object is generated based on one of a plurality of question strings generated based on the baseline statement string.

### Claim 14 (depends on 13)

The method of claim 13, wherein the plurality of question strings generated based on the baseline statement string are each based on removal of a word from the baseline statement string, and the correct response string corresponds to the removed word from the baseline statement string.

### Claim 15 (depends on 11)

The method of claim 11, wherein the correct response string is used to validate a response data object to provision access to an electronic resource.

### Claim 16 (depends on 15)

The method of claim 15, the method further comprising:
receiving a script-reading video data set capturing a portion of or an entirety of the individual's face while the individual is speaking words corresponding to a script data structure, the script data structure having a sequence of pre-identified phonemes or phoneme transitions, the pre-identified phonemes or phoneme transitions including at least one overlapping phoneme or phoneme transition required to be spoken when speaking words of the correct response string;
extracting, from the script-reading video data set, a data subset representing the one or more facial or lip features of the requesting individual corresponding to each phoneme or phoneme transition corresponding to the sequence of pre-identified phonemes or phoneme transitions;
training, one or more baseline machine learning data model architectures, each baseline machine learning data model architecture of the one or more baseline machine learning data model architectures corresponding to a corresponding pre-identified phoneme or phoneme transition of the script data structure such that parameters of the baseline machine learning data model architectures are tuned based on the corresponding one or more facial or lip features;
receiving an answer-reading video data set capturing a portion of or an entirety of the individual's face while the individual is speaking the words corresponding to the correct response string; and
processing, the answer-reading video data set, using the one or more baseline machine learning data model architectures corresponding to the at least one overlapping phoneme or phoneme transition to determine an overall classification similarity score;
wherein the provisioning of access to the electronic resource only occurs if the correct response string has been validated against the correct response string and the overall classification similarity score is greater than a pre-defined threshold similarity score.

### Claim 17 (depends on 16)

The method of claim 16, wherein the script data structure includes words corresponding to a phonetic pangram or a holo-alphabetic sentence.

### Claim 18 (depends on 11)

The method of claim 11, wherein the dynamic challenge passphrase data object is processed to control rendering of a graphical user interface where the statement string portion and the question string portion are displayed as textual display elements on a computer display.

### Claim 19 (depends on 15)

The method of claim 15, wherein the secure electronic resource is a secure webpage.

### Claim 20

A non-transitory computer readable medium storing machine interpretable instructions which when executed by a processor, cause the processor to:
retrieve, from one or more data sources, one or more data objects storing a plurality of structured data records associated with an individual, each structured data record of the plurality of structured data records including at least a plurality of data values corresponding to a feature data field of a plurality of feature data fields;
establish, using an unsupervised machine learning model data architecture, a plurality of data record clusters, each data record cluster representing a mutually exclusive set of structured data records of the plurality of structured data records;
for a data record cluster having a largest number of structured data records, rank the plurality of feature data fields based on a determined contribution value of each feature data field relative to the establishing of the data record cluster having the largest number of structured data records;
identify, using the ranked plurality of feature data fields, a first and a second feature data field of the plurality of feature data fields having the two highest determined contribution values; and
generate the dynamic challenge passphrase data object, wherein the first or the second feature data field is used to establish a statement string portion, and a remaining one of the first or the second feature data field is used to establish a question string portion and a correct response string;
wherein the dynamic challenge passphrase data object is used to validate a challenge response data structure using the correct response string; and
wherein the dynamic challenge passphrase data object is automatically generated from one or more data sources such that an individual is not required to establish the dynamic challenge passphrase data object during an initial registration.
