# US 2023/0058076 A1 · Method and system for auto generating automotive data quality marker > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2023/0058076 A1 - Document ID: US20230058076A1 - Status: Published application - Kind: A1 - Application number: 17/687,181 - Filed: 2022-03-04 - Published: 2023-02-23 - Earliest claimed: 2021-08-18 - Assignees: Cerebrumx Labs Private Limited - Applicants: Cerebrumx Labs Private Limited - Inventors: Amit Gupta; Sandip Ranjhan; Sarika Gupta - Web page: https://patentgrep.com/patent/US20230058076A1 ## CPC - G06V 10/774: Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting - G07C 5/0841: {Registering performance data (recording measured values G01D; information storage G11B)} - G06F 18/2178: {based on feedback of a supervisor} - G06F 18/2415: based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate - G06F 18/243: relating to the number of classes - G06N 3/04: Architecture, e.g. interconnection topology - G06N 3/0464: Convolutional networks [CNN, ConvNet] - G06V 10/993: {Evaluation of the quality of the acquired pattern} - G06V 20/54: of traffic, e.g. cars on the road, trains or boats - G07C 5/008: {communicating information to a remotely located station (transmission systems for measured values G08C)} - G06N 3/082: modifying the architecture, e.g. adding, deleting or silencing nodes or connections - G06N 3/088: Non-supervised learning, e.g. competitive learning - G06N 3/09: Supervised learning - G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound - G06N 20/20: Ensemble learning - G06V 10/7715: {Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods} - G06V 10/82: using neural networks - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - G07C: TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract The present invention provides a robust and effective solution to an entity or an organization for creating and standardizing an Automotive Data Quality Marker (ADQM) to determine/evaluate/predict the quality of automotive data such as Telematics, Body Control, ADAS, Diagnostics, Dashcams, and In-Vehicle Infotainment but not limited to the like generated by the vehicle (i.e., data source) using a machine learning (ML) engine associated with a processing unit. The machine learning engine comprises an amalgamation of machine learning algorithms to determine ADQM for a particular dataset. Data pertaining to vehicles is huge and repetitive. Re-training of the model for improved accuracy is a requirement as automotive data can be augmented with additional signals and data received and stored as trip objects on a regular basis. ## Claims ### Claim 1 An auto generating automotive data quality marker (ADQM) system comprising: a plurality of Distributed Source Systems associated with a plurality of automotive data sources; a plurality of Distributed Storage Systems to store a set of data packets pertaining to any or a combination of numeric data generated in a vehicle during a trip and image dataset for example images captured through a dashboard camera during the trip; a processor; a data quality module coupled to the processor, the data quality module to: receive the set of data packets from the distributed storage systems, wherein the set of data packets is received at a specific frequency and in specific quantity/numbers; a Machine learning (ML) engine coupled to the processor, the ML engine to: extract a first set of attributes from the set of data packets received from the first computing devices associated with the plurality of automotive data sources, wherein the first set of attributes pertains to automotive numeric data set; extract a second set of attributes from the set of data packets received from the first computing devices associated with the plurality of automotive data sources, wherein the second set of attributes pertain to automotive image data set; and based on the first and the second set of attributes extracted, determine, an average score pertaining to quality of data in the set of data packets received, wherein the average score comprises the automotive data quality marker (ADQM); classify, based on the ADQM, the set of data packets related to the one or more automobiles among one or more pre-defined categories for quality marking. ### Claim 2 (depends on 1) The system as claimed in claim 1, wherein the ML engine combines supervised and unsupervised learning for predicting numeric data quality Numeric data Quality Marker (NDQM), wherein the ML engine further combines a convolution neural network (CNN) model for predicting image data quality by using Image Data Quality Marker (IDQM). ### Claim 3 (depends on 1) The system as claimed in claim 1, wherein the ML engine combines the Numeric Data Quality Marker (NDQM) and Image Data Quality Marker (IDQM) to calculate a Trip Data Quality Marker (TDQM) based on a predefined set of parameters. ### Claim 4 (depends on 1) The system as claimed in claim 1, wherein the ML engine classifies the ADQM calculated into the one or more predefined categories based on a predefined scale according to a predefined set of instructions. ### Claim 5 (depends on 1) The system as claimed in claim 1, wherein the ML engine is configured to use CNN modes for deriving the data quality marker for the image data, wherein the quality for numeric trip data is predicted using Linear regression, logistic regression, Naive Bayes, kNN, random forest, Principle Component Analysis (PCA) and a combination thereof. ### Claim 6 (depends on 1) The system as claimed in claim 1, wherein the data quality module updates a database every time and checks for an end-to-end data completeness for each data pipeline associated with the set of data packets. ### Claim 7 (depends on 1) The system as claimed in claim 1, wherein the key data features comprises automotive data pertaining to an overall trip data, vehicle profile, driver profile, driving pattern, safety data, media data, critical events, vehicle health, driver behavior and a combination thereof. ### Claim 8 (depends on 1) The system as claimed in claim 1, wherein the data mode comprises of data collected and shared in real-time, batch, historical or a combination thereof. ### Claim 9 (depends on 1) The system as claimed in claim 1, wherein the data validity comprises data boundary checks to remove out of bound values, signal correlation and dependency. ### Claim 10 (depends on 1) The system as claimed in claim 1, wherein the data skewness comprises balanced data to draw inferences from trip duration, vehicle types, driver Profile, location and a combination thereof. ### Claim 11 A method facilitating auto generating automotive data quality marker, said method comprising: receiving by a data quality module coupled to a processor, a set of data packets from a plurality of distributed storage systems associated with a plurality of automotive data sources, wherein the set of data packets is received at a specific frequency and in specific quantity/numbers; extracting, by an ML engine coupled to the processor, a first set of attributes from the set of data packets received from a plurality of first computing devices associated with the plurality of automotive data sources, wherein the first set of attributes pertain to automotive numeric data set; extracting, by the ML engine coupled to the processor, a second set of attributes from the set of data packets received from the first computing devices associated with the plurality of automotive data sources, wherein the second set of attributes pertain to automotive image data set; and based on the first and the second set of attributes extracted, determining, by the ML engine, an average score pertaining to quality of data in the set of data packets received, wherein the average score comprises the automotive data quality marker (ADQM); and classifying, based on the ADQM, the set of data packets related to the one or more automobiles among one or more pre-defined categories for quality marking. ### Claim 12 (depends on 11) The method as claimed in claim 11 wherein the method comprises combining by the ML engine, supervised and unsupervised learning for predicting numeric data quality Numeric data Quality Marker (NDQM), wherein method further comprises combining a convolution neural network (CNN) model for predicting image data quality by using Image Data Quality Marker (IDQM). ### Claim 13 (depends on 11) The method as claimed in claim 11, wherein the method comprises combining by the ML engine, the Numeric Data Quality Marker (NDQM) and Image Data Quality Marker (IDQM) to calculate a Trip Data Quality Marker (TDQM) based on a predefined set of parameters. ### Claim 14 (depends on 11) The method as claimed in claim 11, wherein the method further comprises: classifying by the ML engine, the ADQM calculated into the one or more predefined categories based on a predefined scale according to a predefined set of instructions. ### Claim 15 (depends on 11) The method as claimed in claim 11, wherein the key data features comprises automotive data pertaining to an overall trip data, vehicle profile, driver profile, driving pattern, safety data, media data, critical events, vehicle health, driver behavior and a combination thereof. ### Claim 16 (depends on 11) The method as claimed in claim 11, wherein the data mode comprises of data collected and shared in real-time, batch, historical or a combination thereof. ### Claim 17 (depends on 11) The method as claimed in claim 11, wherein the data validity comprises data boundary checks to remove out of bound values, signal correlation and dependency. ### Claim 18 A non-transitory computer readable medium comprising machine executable instructions that are executable by a processor to: receive by a data quality module coupled to the processor, a set of data packets from a plurality of distributed storage systems associated with a plurality of automotive data sources, wherein the set of data packets is received at a specific frequency and in specific quantity/numbers; extract by an ML engine coupled to the processor, a first set of attributes from the set of data packets received from a plurality of first computing devices associated with the plurality of automotive data sources, wherein the first set of attributes pertain to automotive numeric data set; extract, by the ML engine coupled to the processor, a second set of attributes from the set of data packets received from the first computing devices associated with the plurality of automotive data sources, wherein the second set of attributes pertain to automotive image data set; and based on the first and the second set of attributes extracted, determine, by the ML engine, an average score pertaining to quality of data in the set of data packets received, wherein the average score comprises the automotive data quality marker (ADQM); classify, based on the ADQM, the set of data packets related to the one or more automobiles among one or more pre-defined categories for quality marking. ### Claim 19 (depends on 18) The non-transitory computer readable medium as claimed in claim 18, wherein the ML engine combines supervised and unsupervised learning for predicting numeric data quality Numeric data Quality Marker (NDQM), wherein the ML engine combines a convolution neural network (CNN) model for predicting image data quality by using Image Data Quality Marker (IDQM). ### Claim 20 (depends on 18) The non-transitory computer readable medium as claimed in claim 18, wherein the ML engine combines the Numeric Data Quality Marker (NDQM) and Image Data Quality Marker (IDQM) to calculate a Trip Data Quality Marker (TDQM) based on a predefined set of parameters.