View source for Capital One Services, LLC (20240283822). LAYERED CYBERSECURITY USING SPURIOUS DATA SAMPLES simplified abstract

=LAYERED CYBERSECURITY USING SPURIOUS DATA SAMPLES=


==Organization Name==

[[:Category:Capital One Services, LLC|Capital One Services, LLC]]


[[Category:Capital One Services, LLC]]

==Inventor(s)==

[[:Category:Galen Rafferty of Mahomet IL (US)|Galen Rafferty of Mahomet IL (US)]][[Category:Galen Rafferty of Mahomet IL (US)]]

[[:Category:Samuel Sharpe of Cambridge MA (US)|Samuel Sharpe of Cambridge MA (US)]][[Category:Samuel Sharpe of Cambridge MA (US)]]

[[:Category:Brian Barr of Schenectady NY (US)|Brian Barr of Schenectady NY (US)]][[Category:Brian Barr of Schenectady NY (US)]]

[[:Category:Jeremy Goodsitt of Champaign IL (US)|Jeremy Goodsitt of Champaign IL (US)]][[Category:Jeremy Goodsitt of Champaign IL (US)]]

[[:Category:Michael Davis of Arington VA (US)|Michael Davis of Arington VA (US)]][[Category:Michael Davis of Arington VA (US)]]

[[:Category:Taylor Turner of Richmond VA (US)|Taylor Turner of Richmond VA (US)]][[Category:Taylor Turner of Richmond VA (US)]]

[[:Category:Justin Au-yeung of Somerville MA (US)|Justin Au-yeung of Somerville MA (US)]][[Category:Justin Au-yeung of Somerville MA (US)]]

[[:Category:Owen Reinert of Queens NY (US)|Owen Reinert of Queens NY (US)]][[Category:Owen Reinert of Queens NY (US)]]

==LAYERED CYBERSECURITY USING SPURIOUS DATA SAMPLES - A simplified explanation of the abstract==


This abstract first appeared for US patent application 20240283822 titled 'LAYERED CYBERSECURITY USING SPURIOUS DATA SAMPLES

'''Simplified Explanation''': 

The patent application describes a computing system that iterates between adding spurious data to a dataset and training a model on the dataset. If the model's performance does not drop significantly, more spurious data is added until the desired decrease in performance is achieved. The system can determine the impact of each feature on the model's output and generate spurious data samples by modifying impactful features. In case of a cybersecurity incident, the system can identify when the incident occurred based on the spurious data in the dataset.

'''Key Features and Innovation''':
* Iterative process of adding spurious data and training models
* Impact analysis of features on model output
* Generation of spurious data by modifying impactful features
* Detection of cybersecurity incidents based on spurious data

'''Potential Applications''':
This technology can be applied in cybersecurity for incident detection, in machine learning for model training, and in data analysis for feature impact assessment.

'''Problems Solved''':
This technology addresses the need for efficient model training, feature impact analysis, and cybersecurity incident detection.

'''Benefits''':
* Improved model performance through iterative training
* Enhanced understanding of feature impact on model output
* Efficient detection of cybersecurity incidents based on dataset analysis

'''Commercial Applications''':
Potential commercial applications include cybersecurity software, machine learning platforms, and data analysis tools for various industries.

'''Prior Art''''':
Prior art related to this technology may include research on iterative model training, feature impact analysis, and cybersecurity incident detection in datasets.

'''Frequently Updated Research''':
Stay updated on research related to iterative model training, feature impact analysis, and cybersecurity incident detection in datasets for the latest advancements in the field.

'''Questions about the Technology''':
1. How does the system determine the impact of each feature on the model's output?
2. What are the potential implications of using spurious data for cybersecurity incident detection?


==Original Abstract Submitted==

in some aspects, a computing system may iterate between adding spurious data to the dataset and training a model on the dataset. if the model's performance has not dropped by more than a threshold amount, then additional spurious data may be added to the dataset until the desired amount of performance decrease has been achieved. the computing system may determine the amount of impact each feature has on a model's output. the computing system may generate a spurious data sample by modifying values of features that are more impactful than other features. the computing system may repeatedly modify the spurious data that is stored in a dataset. if a cybersecurity incident occurs (e.g., the dataset is stolen or leaked), the system may identify when the cybersecurity incident took place based on the spurious data that is stored in the dataset.



[[Category:H04L9/40]]
[[Category:H04L41/16]]
[[Category:CPC_H04L63/1491]]