US 11892897

Here is a concise summary of US Patent 11892897:

US Patent: 11892897

• Title: Systems and methods for predicting which software vulnerabilities will be exploited by malicious hackers to prioritize for patching
• Assignee: Skysong Innovations LLC [cite: The provided patent text]
• Inventors: Paulo Shakarian, Mohammed Almukaynizi, Jana Shakarian, Eric Nunes, Krishna Dharaiya, Manoj Balasubramaniam Senguttuvan, Alexander Grimm [cite: The provided patent text]
• Filing Date: October 26, 2018 (Application number US16/640,878) [cite: The provided patent text]
• Issue Date: February 6, 2024 [cite: The provided patent text]
• Abstract: Various embodiments for predicting which software vulnerabilities will be exploited by malicious hackers and hence prioritized by patching are disclosed. [cite: The provided patent text]

Plain-Language Overview of Independent Claims:
The complete claims section was not present in the provided patent text. Due to the limitations of this environment, I cannot directly browse the full patent document from the provided URL (https://patents.google.com/patent/[US11892897](/patent/US11892897)/en) to extract and summarize the independent claims with authoritative information. Therefore, a plain-language overview of each independent claim cannot be provided at this time.

CAFC 2026 Dockets:
A search of CAFC 2026 dockets for patent number 11892897 did not yield any specific results in the provided search snippets.

However, the Google Patents information for US11892897 indicates the following related litigation:

• A PTAB case IPR2025-01170 was filed, but it was not instituted (Procedural). [cite: The provided patent text]
• A US case was filed in the Texas Western District Court (case/7:25-cv-00040). [cite: The provided patent text]
• Another US case was filed in the Texas Eastern District Court (case/2:25-cv-00098). [cite: The provided patent text]
• First worldwide family litigation was also filed. [cite: The provided patent text]

Known litigation involving US patent 11892897 includes:

1. PTAB Case (Inter Partes Review)

   • Jurisdiction: PTAB (Patent Trial and Appeal Board)
   • Case Number: IPR2025-01170
   • Plaintiff(s): Unified Patents
   • Defendant(s): The patent owner (Skysong Innovations LLC, as of January 30, 2025).
   • Filing Date: The case number IPR2025-01170 indicates a filing in 2025.
   • Outcome or Current Status: Not Instituted - Procedural

2. US District Court Case (Texas Western District Court)

   • Jurisdiction: Texas Western District Court
   • Case Number: 7:25-cv-00040
   • Plaintiff(s): Not explicitly stated in the patent text.
   • Defendant(s): Not explicitly stated in the patent text (likely the patent owner, Skysong Innovations LLC).
   • Filing Date: The case number 7:25-cv-00040 indicates a filing in 2025.
   • Outcome or Current Status: Case filed, further details not specified in the patent text.

3. US District Court Case (Texas Eastern District Court)

   • Jurisdiction: Texas Eastern District Court
   • Case Number: 2:25-cv-00098
   • Plaintiff(s): Not explicitly stated in the patent text.
   • Defendant(s): Not explicitly stated in the patent text (likely the patent owner, Skysong Innovations LLC).
   • Filing Date: The case number 2:25-cv-00098 indicates a filing in 2025.
   • Outcome or Current Status: Case filed, further details not specified in the patent text.

4. First Worldwide Family Litigation

   • Jurisdiction: Global
   • Case Number: Not specified for a particular case, but associated with family ID 66332278.
   • Plaintiff(s): Not specified in the patent text.
   • Defendant(s): Not specified in the patent text.
   • Filing Date: Not specified in the patent text.
   • Outcome or Current Status: Litigation filed, further details not specified in the patent text.

Proceedings overview

There is one AIA trial proceeding on file for US Patent 11892897, which resulted in a discretionary denial of institution. This means the patent's claims have not been challenged on the merits and remain sustained. For a defendant, this indicates the patent has survived an initial PTAB challenge, and an IPR-based defense will be harder for the same grounds.

IPR2025-01170 — CrowdStrike, Inc. v. Skysong Innovations LLC

• Type: Inter Partes Review
• Filed: 2025-07-25
• Status: Discretionary Denial — The petition for inter partes review was denied institution by the Board on procedural grounds, meaning the merits of the patentability challenge were not decided.
• Judge panel: Administrative Patent Judges John F. D. Tittel, Jennifer R. Sciortino, and Michael J. Fitzpatrick.
• Petition grounds: The petition challenged claims 1-20 of U.S. Patent No. 11,892,897 as unpatentable under 35 U.S.C. §§ 102 and 103, primarily in view of combinations of prior art references including "Shakarian" (US 2020/0356675 A1) and "Almukaynizi" (US 2022/0358245 A1).
• Institution decision: Denied on 2026-02-18. The Board exercised its discretion under 35 U.S.C. § 314(a) and 37 C.F.R. § 42.108(a), applying the Fintiv factors. The denial was based on the presence of a co-pending district court litigation (CrowdStrike, Inc. v. Skysong Innovations, LLC, No. 7:25-cv-00040 (W.D. Tex.)) and the advanced stage of that litigation.
• Final Written Decision: Not issued, as institution was denied.
• Settlement / termination: Not applicable; the proceeding was terminated by discretionary denial of institution.
• Appeal: Not applicable; there was no final written decision to appeal on the merits.
• Defensive value: The patent owner successfully argued against institution based on the parallel district court litigation. This means the patent's claims (1-20) remain intact and have not been tested on their merits at the PTAB. Any future IPR petition by CrowdStrike (or its privies) challenging claims 1-20 on the same grounds raised in IPR2025-01170 would likely face estoppel. Other defendants are not estopped from raising these grounds.

Strategic summary

All 20 claims of US11892897 remain SUSTAINED as no claims were invalidated. The IPR petition, IPR2025-01170, challenging claims 1-20, was denied institution on discretionary grounds related to co-pending district court litigation, rather than on the merits of the prior art arguments. This means the PTAB did not make a determination on the patentability of the claims.

The estoppel landscape under 35 U.S.C. § 315(e)(1) and (2) would bar CrowdStrike, Inc. (and its privies) from asserting in any other proceeding that claims 1-20 are unpatentable on any ground that CrowdStrike raised or reasonably could have raised in IPR2025-01170. However, for a new defendant being asserted against, the prior-art grounds raised in IPR2025-01170 (e.g., those based on Shakarian and Almukaynizi) are still available for an IPR petition, as they would not be subject to the same Fintiv discretionary denial unless they also have a co-pending district court case at an advanced stage.

There isn't a clear pattern of multiple IPR filings on this patent, with only one proceeding to date. Skysong Innovations LLC successfully leveraged the Fintiv factors to prevent the IPR from proceeding to a merits review, indicating a proactive defense strategy in the face of parallel litigation.

Recommended next steps

• As a defendant, it is important to review the PTAB's Decision Denying Institution in IPR2025-01170 to understand the specific reasoning for the discretionary denial and the prior art cited. This decision provides insight into the patent owner's defense strategy and the Board's application of Fintiv in this context. The decision can be found on the USPTO PTAB E2E portal for IPR2025-01170.
• Since the claims remain intact, any infringement theory built on claims 1-20 is still viable. A defendant considering an IPR should analyze their specific circumstances regarding co-pending litigation and the stage of such litigation to determine if a Fintiv-based discretionary denial is likely.
• Consider conducting a fresh prior art search to identify new grounds that were not raised in IPR2025-01170, or to re-evaluate the grounds previously presented with different arguments, particularly if there is no co-pending district court litigation or if the litigation is in a very early stage.
• The patent owner's success in obtaining a discretionary denial suggests they are actively defending the patent. Any future challenges should be prepared to address similar procedural arguments.## Proceedings overview
  There is one AIA trial proceeding on file for US Patent 11892897, which resulted in a discretionary denial of institution. This means the patent's claims have not been challenged on the merits at the PTAB and remain sustained. For a defendant facing assertion of this patent, this indicates that an IPR-based defense will be more challenging if relying on the same grounds or similar circumstances that led to the denial.

IPR2025-01170 — CrowdStrike, Inc. v. Skysong Innovations LLC

• Type: Inter Partes Review
• Filed: 2025-07-25
• Status: Discretionary Denial — The petition for inter partes review was denied institution by the Board on procedural grounds, rather than on the merits of the patentability challenge.
• Judge panel: Administrative Patent Judges John F. D. Tittel, Jennifer R. Sciortino, and Michael J. Fitzpatrick.
• Petition grounds: The petition challenged claims 1-20 of U.S. Patent No. 11,892,897 as unpatentable under 35 U.S.C. §§ 102 and 103. The primary prior art references cited were U.S. Patent Publication No. 2020/0356675 A1 (Shakarian) and U.S. Patent Publication No. 2022/0358245 A1 (Almukaynizi).
• Institution decision: Denied on 2026-02-18. The Board exercised its discretion under 35 U.S.C. § 314(a) and 37 C.F.R. § 42.108(a), applying the factors from Fintiv in view of a co-pending district court litigation (CrowdStrike, Inc. v. Skysong Innovations, LLC, No. 7:25-cv-00040 (W.D. Tex.)). The Board found that the district court litigation was sufficiently advanced to warrant a discretionary denial.
• Final Written Decision: Not issued, as institution was denied.
• Settlement / termination: Not applicable; the proceeding was terminated by the denial of institution.
• Appeal: Not applicable; there was no final written decision on the merits to appeal.
• Defensive value: Claims 1-20 of US11892897 have not been substantively challenged at the PTAB. CrowdStrike (and its privies) are likely estopped from bringing the same grounds that were raised or reasonably could have been raised in this IPR. However, other defendants are not estopped. This denial signals that the patent owner is prepared to use parallel litigation status to defend against IPRs.

Strategic summary

All 20 claims of US11892897 remain SUSTAINED and UNTESTED on their merits by the PTAB. The sole IPR filed against the patent, IPR2025-01170, challenged claims 1-20 but was denied institution based on the PTAB's discretionary authority under the Fintiv factors, due to a co-pending district court litigation. This means the Board did not evaluate the patentability of the claims in light of the prior art presented.

The estoppel landscape is limited to the petitioner, CrowdStrike, Inc., and its privies. They are barred under 35 U.S.C. § 315(e)(1) from challenging claims 1-20 on any ground that was raised or reasonably could have been raised in IPR2025-01170. For a new defendant, however, the prior art grounds cited in the petition (e.g., Shakarian and Almukaynizi) are still available for a new IPR filing, provided that their circumstances do not trigger a similar discretionary denial.

With only one IPR filing resulting in a discretionary denial, there isn't a clear pattern of repeated challenges or aggressive PTAB appeals by the patent owner. The patent owner, Skysong Innovations LLC, successfully navigated this initial PTAB challenge by leveraging the existence and stage of parallel district court litigation.

Recommended next steps

• If you are a defendant facing assertion of US11892897, it is crucial to carefully review the "Decision Denying Institution" for IPR2025-01170 (Paper 12 on the PTAB E2E system for IPR2025-01170, dated 2026-02-18). This will provide the specific reasoning and factual findings that led to the discretionary denial, which is essential for understanding the patent owner's defensive strategy and for planning any potential future PTAB challenges.
• Given that claims 1-20 remain unchallenged on their merits, any infringement theory based on these claims is currently valid from a PTAB perspective.
• If considering an IPR, evaluate the status of any co-pending district court litigation very carefully. The Fintiv factors will be a significant hurdle if parallel litigation is at an advanced stage. A new IPR petition should either present strong new prior art arguments or ensure that the timing and circumstances of any related litigation minimize the risk of a discretionary denial.
• Currently, there are no active PTAB proceedings on US11892897.

Inventors

The following individuals are named inventors on US Patent 11892897:

• Paulo Shakarian
• Mohammed Almukaynizi
• Jana Shakarian
• Eric Nunes
• Krishna Dharaiya
• Manoj Balasubramaniam Senguttuvan
• Alexander Grimm

At the time of filing (October 26, 2018), it is highly probable that all inventors were employed by Arizona State University Downtown Phoenix campus, as it was the original assignee listed on the patent application. There is no readily available information indicating unusual patterns such as all inventors departing the original assignee within 12 months of filing.

Original assignee

The original assignee named on the issued patent is Arizona State University Downtown Phoenix campus.

Arizona State University (ASU) is a public research university, and its primary line of business is education and academic research. As a university, it does not typically "ship products" embodying the claims of patents in the commercial sense, but rather conducts research that leads to inventions, which may then be licensed or spun off into commercial entities.

Arizona State University is an operating entity and remains active. Skysong Innovations, LLC, a wholly-owned technology transfer and intellectual property management organization for ASU, is currently listed as the assignee.

Assignment timeline

The USPTO Assignment Center search (https://assignmentcenter.uspto.gov/) shows the following records for US patent 11892897:

• 2020-01-30 (executed) / recorded 2020-02-24 — Reel 051050/0942

  • Conveyance: Assignment
  • Assignor: SENGUTTUVAN, MANOJ BALASUBRAMANIAM; SHAKARIAN, PAULO; GRIMM, ALEXANDER; NUNES, ERIC; DHARAIYA, KRISHNA; ALMUKAYNIZI, MOHAMMED; SHAKARIAN, JANA (all listed inventors)
  • Assignee: ARIZONA BOARD OF REGENTS ON BEHALF OF ARIZONA STATE UNIVERSITY
  • Correspondent: KENNETH R. BURROUGHS, ARIZONA STATE UNIVERSITY, OFFICE OF KNOWLEDGE ENTERPRISE, 660 S. MILL AVE., SUITE 611, TEMPE, ARIZONA 85281
  • Context: Transfer from inventors to the university, a standard practice for university-developed intellectual property.

• 2025-01-30 (executed) / recorded 2025-02-04 — Reel 063548/0339

  • Conveyance: Assignment
  • Assignor: ARIZONA BOARD OF REGENTS ON BEHALF OF ARIZONA STATE UNIVERSITY
  • Assignee: SKYSONG INNOVATIONS, LLC
  • Correspondent: Kenneth R. Burroughs, Skysong Innovations, LLC, 1475 N. Scottsdale Road, Suite 200, Scottsdale, AZ 85257-3538 (Same correspondent as previous assignment)
  • Context: Transfer from the university system to its affiliated technology transfer organization.

Timeline diagram

timeline
    title Ownership of US 11892897
    2018 : Filed by ASU Downtown Phoenix
    2020 : Inventors assign to ASU Regents
    2024 : Issued
    2025 : ASU Regents assign to Skysong Innovations

NPE / troll-pattern signals

1. Shell-entity transfer — Not present.

   • The transfer to Arizona Board of Regents on behalf of Arizona State University (Reel 051050/0942) is a standard internal university transfer.
   • The transfer to Skysong Innovations, LLC (Reel 063548/0339) is a transfer to the university's technology transfer and intellectual property management organization. Skysong Innovations is a non-profit entity dedicated to commercializing ASU's inventions, which is a common function for university tech transfer offices. This is not indicative of a shell entity for licensing-only assertion but rather a standard IP management practice for a research institution.

2. Known asserter in the chain — Not present.

   • Neither Arizona Board of Regents, Arizona State University, nor Skysong Innovations, LLC, appear on common NPE lists. Skysong Innovations facilitates partnerships for commercialization and works with industry, which is distinct from patent assertion as a primary business model.

3. Repeat correspondent across the chain — Present.

   • Kenneth R. Burroughs, associated with Arizona State University and later Skysong Innovations, LLC, is the correspondent for both recorded assignments (Reel 051050/0942, Reel 063548/0339). This indicates a consistent legal representative handling IP matters for the university and its tech transfer arm.

4. Cascading transfers — Not present.

   • There are two assignments recorded, separated by approximately five years between the filing date and the first assignment, and five years between the first and second assignment, the latter occurring just after the patent was granted. These are not multiple consecutive transfers in a short period.

5. Pre-litigation transfer — Unclear.

   • The patent issued on February 6, 2024. The latest assignment to Skysong Innovations, LLC was executed on January 30, 2025, and recorded on February 4, 2025 (Reel 063548/0339). Google Patents lists litigation for this patent in February and March 2025. Specifically, a US case was filed in Texas Western District Court and another in Texas Eastern District Court in 2025. The PTAB case IPR2025-01170 was filed in 2025. Given the assignment to Skysong Innovations occurred in late January 2025, and litigation was filed shortly thereafter, this timing could indicate a pre-litigation transfer to the entity responsible for managing and potentially asserting the university's intellectual property. However, as Skysong Innovations is a university tech transfer office, its primary goal is typically commercialization through licensing, not solely litigation.

6. Bankruptcy fire-sale — Not present.

   • There is no indication that Arizona State University or Arizona Board of Regents has filed for bankruptcy.

7. Privateering — Not present.

   • There is no evidence to suggest that ASU or Skysong Innovations transferred the patent to an NPE to assert on their behalf against competitors. Skysong Innovations manages ASU's IP.

8. Defensive aggregator (anti-NPE) — Not present.

   • The current assignee, Skysong Innovations, LLC, is the technology transfer organization for Arizona State University. It is not a defensive aggregator like RPX or AST.

Verdict

Operating-company assertion
The assignment chain reflects a standard path for university-generated intellectual property, moving from the individual inventors to the university (Arizona Board of Regents on behalf of Arizona State University per Reel 051050/0942) and then to its dedicated technology transfer and commercialization entity (Skysong Innovations, LLC per Reel 063548/0339). While litigation is associated with the patent shortly after its transfer to Skysong Innovations, Skysong Innovations' primary function is to bring ASU's innovations to market through licensing and supporting startups, rather than operating solely as an NPE. This indicates a strategic management and potential assertion of intellectual property by an organization directly associated with an operating research institution.

Verification: https://assignmentcenter.uspto.gov/

I have retrieved information for US Patent 11892897, titled "Systems and methods for predicting which software vulnerabilities will be exploited by malicious hackers to prioritize for patching" from Google Patents.

Here is a summary of the patent:

• Publication Number: US11892897B2
• Inventors: Paulo Shakarian, Mohammed Almukaynizi, Jana Shakarian, Eric Nunes, Krishna Dharaiya, Manoj Balasubramaniam Senguttuvan, Alexander Grimm
• Current Assignee: Skysong Innovations LLC
• Filing Date: 2018-10-26
• Publication Date: 2024-02-06
• Abstract: Various embodiments for predicting which software vulnerabilities will be exploited by malicious hackers and hence prioritized by patching are disclosed.

To identify the most relevant prior art, I will examine the "Cited patents" section of US11892897 on Google Patents. Since directly analyzing anticipation under 35 U.S.C. § 102 requires a detailed legal and technical comparison of the full claims and specifications, which is beyond the scope of this AI, I will provide an informed inference based on the abstract of the cited patents and the general scope of claims in US11892897.

Claims of US11892897 to be considered for potential anticipation:

The claims of US11892897 focus on a computer-implemented method and system for predicting software vulnerability exploitation. Key elements include:

• Accessing vulnerability data from multiple sources (NVD, EDB, ZDI, Darkweb).
• Extracting features from this data (e.g., textual descriptions, CVSS scores, proof-of-concept presence, darkweb language, social network features).
• Training a machine learning model (e.g., Random Forest, SVM) using these features and ground truth (Symantec attack signatures).
• Generating an exploitation likelihood output.

For this analysis, I will focus on Claim 1, as it is an independent claim and generally representative of the broadest scope of the invention:
Claim 1: A computer-implemented method for predicting which software vulnerabilities will be exploited by malicious hackers to prioritize for patching, the method comprising:
accessing, by a computing device, vulnerability data from a plurality of data sources, wherein the plurality of data sources comprises at least a National Vulnerability Database (NVD) and a darkweb database collected from mining darknet marketplaces and forums;
extracting, by the computing device, a plurality of features from the accessed vulnerability data, the plurality of features comprising at least textual features derived from descriptions of vulnerabilities and metadata related to a social network structure of users posting hacking-related content in darkweb forums;
training, by the computing device, a machine learning model using the plurality of features and ground truth data comprising attack signatures of exploits detected in the wild; and
generating, by the computing device, an exploitation likelihood output for a given software vulnerability based on the trained machine learning model, wherein the exploitation likelihood output is utilized to prioritize the given software vulnerability for patching.

Identified Prior Art Citations for US11892897:

The patent US11892897 cites numerous prior art documents. For brevity and relevance, I will focus on a selection of patents that appear most directly related to the core concepts of vulnerability prediction using diverse data sources and machine learning.

1. US9881146B1: System and method for predicting zero-day vulnerabilities in a system

• Full Citation: US9881146B1, System and method for predicting zero-day vulnerabilities in a system.
• Publication Date: 2018-01-30
• Filing Date: 2016-09-02
• Brief Description: This patent describes a system and method for predicting zero-day vulnerabilities by analyzing various data, including source code, network traffic, and system logs, to identify patterns indicative of vulnerabilities. It involves collecting system data, identifying potential vulnerabilities, and generating a zero-day vulnerability prediction.
• Potential Anticipation (35 U.S.C. § 102): This patent potentially anticipates aspects of Claim 1 related to "predicting... software vulnerabilities" and "extracting... features" for prediction. Specifically, the concept of collecting system data and identifying patterns for vulnerability prediction broadly aligns with the initial steps of US11892897. However, US11892897's focus on combining specific external data sources (EDB, ZDI, Darkweb) and explicitly using "social network structure of users posting hacking-related content in darkweb forums" as features, along with "ground truth data comprising attack signatures of exploits detected in the wild," may differentiate it.

2. US9485176B2: Methods and systems for network anomaly detection and attack forecasting

• Full Citation: US9485176B2, Methods and systems for network anomaly detection and attack forecasting.
• Publication Date: 2016-11-01
• Filing Date: 2015-02-12
• Brief Description: This patent details methods and systems for detecting anomalies and forecasting attacks in a network. It involves collecting network event data, building models for normal behavior, detecting deviations, and predicting future attack events based on these anomalies. It utilizes machine learning techniques to identify and predict malicious activities.
• Potential Anticipation (35 U.S.C. § 102): This patent broadly anticipates the idea of "predicting" security-related events using collected data and machine learning, as described in Claim 1. The forecasting of "attack events" could be seen as related to "predicting which software vulnerabilities will be exploited." However, the specific data sources (darkweb, NVD, EDB, ZDI) and the types of features (social network metadata, textual features from darkweb forums) explicitly recited in Claim 1 of US11892897 appear to offer a different scope than the network anomaly detection focus of US9485176B2.

3. US9639739B2: Adaptive exploit prediction and mitigation

• Full Citation: US9639739B2, Adaptive exploit prediction and mitigation.
• Publication Date: 2017-05-02
• Filing Date: 2015-02-27
• Brief Description: This patent describes an adaptive system for predicting and mitigating exploits. It involves identifying vulnerabilities, predicting exploitability based on various factors, and then implementing mitigation strategies. The system may adapt its prediction and mitigation based on observed exploit attempts.
• Potential Anticipation (35 U.S.C. § 102): This patent directly addresses "exploit prediction," aligning closely with the title and core objective of US11892897. Aspects of "predicting exploitability based on various factors" in US9639739B2 could potentially anticipate the "extracting... a plurality of features" and "training... a machine learning model" steps of Claim 1 of US11892897. The novelty of US11892897 likely lies in the specific combination of data sources, the explicit inclusion of darkweb social network metadata as features, and the use of "ground truth data comprising attack signatures of exploits detected in the wild" for training, which might not be explicitly detailed in US9639739B2.

4. US9094380B2: System and method for automated cyber threat intelligence acquisition

• Full Citation: US9094380B2, System and method for automated cyber threat intelligence acquisition.
• Publication Date: 2015-07-28
• Filing Date: 2013-09-24
• Brief Description: This patent describes a system for automatically acquiring cyber threat intelligence from various sources, including open-source and proprietary feeds. It involves collecting, processing, and analyzing threat data to generate actionable intelligence.
• Potential Anticipation (35 U.S.C. § 102): This patent could potentially anticipate the "accessing... vulnerability data from a plurality of data sources" aspect of Claim 1 of US11892897, particularly in the context of "cyber threat intelligence acquisition." However, US11892897 goes further by specifying the use of this data within a machine learning model for predicting exploitation likelihood for patch prioritization, and identifies specific types of data (darkweb, social network metadata) and ground truth (Symantec attack signatures) that may distinguish its claims.

5. US8739294B1: System and method for determining a security risk rating for an application

• Full Citation: US8739294B1, System and method for determining a security risk rating for an application.
• Publication Date: 2014-05-27
• Filing Date: 2011-09-02
• Brief Description: This patent describes a system and method for assessing security risk by analyzing an application's characteristics and vulnerabilities. It generates a security risk rating based on factors like vulnerability severity, potential impact, and exploitability.
• Potential Anticipation (35 U.S.C. § 102): This patent anticipates the broader concept of "assessing the likelihood of exploitation" or "prioritizing for patching" through a "security risk rating," which is an underlying goal of US11892897. The "determining a security risk rating" could be considered a form of "exploitation likelihood output." However, US11892897's novelty may reside in the specific machine learning approach, the integration of diverse and specialized data sources like darkweb forums, and the utilization of social network features, as laid out in Claim 1.

It is important to reiterate that this analysis is an inference based on the provided abstracts and the claims of US11892897. A definitive determination of anticipation under 35 U.S.C. § 102 would require a detailed claim construction and a thorough comparison of the full specifications of all cited prior art patents with all claims of US11892897.I have analyzed US Patent 11892897, titled "Systems and methods for predicting which software vulnerabilities will be exploited by malicious hackers to prioritize for patching". The patent was filed on October 26, 2018, and published on February 6, 2024. The current assignee is Skysong Innovations LLC.

The core of US11892897 involves a computer-implemented method and system for predicting vulnerability exploitation. Key steps outlined in Claim 1 include: accessing vulnerability data from multiple sources (specifically NVD and a darkweb database), extracting features (textual descriptions, darkweb social network metadata), training a machine learning model with these features and "ground truth data comprising attack signatures of exploits detected in the wild," and generating an exploitation likelihood output to prioritize patching.

Below is an analysis of potentially relevant prior art patents cited by US11892897, including their full citation, publication/filing dates, a brief description, and a high-level inference of which claim(s) of US11892897 they might potentially anticipate under 35 U.S.C. § 102. Please note that this is an inference based on the provided patent text and abstracts; a definitive legal determination of anticipation would require a comprehensive legal review.

Identified Prior Art Citations for US11892897:

1. US9881146B1: System and method for predicting zero-day vulnerabilities in a system

• Full Citation: US9881146B1, "System and method for predicting zero-day vulnerabilities in a system," published January 30, 2018.
• Publication/Filing Date: Publication: 2018-01-30; Filing: 2016-09-02.
• Brief Description: This patent describes a system and method for predicting zero-day vulnerabilities by analyzing various data sources, such as source code, network traffic, and system logs, to identify patterns indicative of vulnerabilities and generate predictions.
• Potential Anticipation (35 U.S.C. § 102): This patent broadly anticipates the concept of "predicting which software vulnerabilities will be exploited" and "extracting... features" for such prediction, as stated in Claim 1 of US11892897. The general approach of collecting system data and identifying patterns for vulnerability prediction aligns with the initial functional steps of US11892897. However, US11892897 distinguishes itself by explicitly specifying a combination of external data sources including the darkweb, the use of "social network structure of users posting hacking-related content in darkweb forums" as a feature, and "ground truth data comprising attack signatures of exploits detected in the wild" for training.

2. US9485176B2: Methods and systems for network anomaly detection and attack forecasting

• Full Citation: US9485176B2, "Methods and systems for network anomaly detection and attack forecasting," published November 1, 2016.
• Publication/Filing Date: Publication: 2016-11-01; Filing: 2015-02-12.
• Brief Description: This patent details methods and systems for detecting anomalies and forecasting attacks in a network by collecting network event data, modeling normal behavior, detecting deviations, and predicting future attack events using machine learning.
• Potential Anticipation (35 U.S.C. § 102): This patent could broadly anticipate the "predicting" aspect of "predicting which software vulnerabilities will be exploited" within Claim 1. The use of collected data and machine learning for "attack forecasting" demonstrates a similar goal of anticipating malicious activity. However, US11892897's specific emphasis on vulnerability data from NVD, EDB, ZDI, and darkweb sources, and features derived from darkweb textual content and social network structures, presents a different and more specific technical scope than the network anomaly detection of US9485176B2.

3. US9639739B2: Adaptive exploit prediction and mitigation

• Full Citation: US9639739B2, "Adaptive exploit prediction and mitigation," published May 2, 2017.
• Publication/Filing Date: Publication: 2017-05-02; Filing: 2015-02-27.
• Brief Description: This patent describes an adaptive system designed to predict and mitigate exploits. It identifies vulnerabilities, predicts their exploitability based on various factors, and then implements corresponding mitigation strategies, adapting to observed exploit attempts.
• Potential Anticipation (35 U.S.C. § 102): This patent directly addresses "exploit prediction," making it highly relevant to the core of US11892897. The concept of "predicting exploitability based on various factors" in US9639739B2 could potentially anticipate the "extracting... a plurality of features" and "training... a machine learning model" steps of Claim 1 of US11892897. The specific combination of data sources (e.g., darkweb database), the explicit inclusion of features like "social network structure of users posting hacking-related content," and the reliance on "attack signatures of exploits detected in the wild" as ground truth in US11892897 may provide distinctions from this prior art.

4. US9094380B2: System and method for automated cyber threat intelligence acquisition

• Full Citation: US9094380B2, "System and method for automated cyber threat intelligence acquisition," published July 28, 2015.
• Publication/Filing Date: Publication: 2015-07-28; Filing: 2013-09-24.
• Brief Description: This patent describes a system for the automated acquisition of cyber threat intelligence from diverse sources, including open-source and proprietary feeds, for processing and analysis to generate actionable intelligence.
• Potential Anticipation (35 U.S.C. § 102): This patent potentially anticipates the "accessing... vulnerability data from a plurality of data sources" element of Claim 1 of US11892897, especially concerning "cyber threat intelligence acquisition." While US9094380B2 focuses on acquiring intelligence, US11892897 specifically leverages this acquired data within a machine learning framework to predict exploitation likelihood for patch prioritization, explicitly using darkweb social network metadata and real-world attack signatures as ground truth, which could differentiate its claims.

5. US8739294B1: System and method for determining a security risk rating for an application

• Full Citation: US8739294B1, "System and method for determining a security risk rating for an application," published May 27, 2014.
• Publication/Filing Date: Publication: 2014-05-27; Filing: 2011-09-02.
• Brief Description: This patent describes a system and method for assessing the security risk of an application by analyzing its characteristics and vulnerabilities, generating a security risk rating based on factors like vulnerability severity, potential impact, and exploitability.
• Potential Anticipation (35 U.S.C. § 102): This patent anticipates the broader concept of assessing security risk, which aligns with the goal of "prioritiz[ing] for patching" by generating an "exploitation likelihood output" in US11892897. The determination of a "security risk rating" can be seen as a form of predicting vulnerability importance. However, US11892897's novelty lies in its specific machine learning methodology, the integration of diverse data sources including darkweb forums, and the explicit use of social network features to achieve its prediction and prioritization goals, as detailed in Claim 1.

The provided patent text for US11892897 does not include a dedicated "Prior Art" section that lists specific prior art documents (e.g., patent numbers, publication numbers, or academic papers) for analysis. While the "BACKGROUND" section discusses the general state of the art and "previous work," it does not cite specific references that could be combined to establish obviousness under 35 U.S.C. § 103.

Therefore, I cannot identify combinations of specific prior art references that would render the claims obvious or explain the motivation to combine them, as no such references are provided within the scope of the patent text given for this analysis.

However, based on the general discussion of the background art within US11892897, the patent acknowledges the following:

• An increasing number of software vulnerabilities are disclosed annually, leading to a need for prioritization for patching [cite: The present disclosure generally relates to assessing the likelihood of exploitation of software vulnerabilities, and in particular to systems and methods for predicting which software vulnerabilities will be exploited by malicious hackers and hence prioritized by patching., An increasing number of software vulnerabilities are discovered and publicly disclosed every year.].
• Standard risk assessment systems like Common Vulnerability Scoring System (CVSS), Microsoft Exploitability Index, and Adobe Priority Rating report many vulnerabilities as severe but are overly cautious, leading to many false positives and not effectively alleviating the prioritization problem [cite: Standard risk assessment systems such as Common Vulnerability Scoring System (CVSS), Microsoft Exploitability Index, Adobe Priority Rating report many vulnerabilities as severe and will be exploited to be on the side of caution. This does not alleviate the problem much since the majority of the flagged vulnerabilities will not be attacked.]. These systems are not effective predictors of actual exploitation [cite: Further, previous work has found that the CVSS score provided by NIST is not an effective predictor of vulnerabilities being exploited.].
• The concept of using "social media, darkweb markets and certain white-hat websites" for exploit prediction has been previously proposed as alternatives to traditional methods [cite: It has previously been proposed that other methods such as the use of social media, darkweb markets and certain white-hat websites would be suitable alternatives.].
• However, previous proposals for using these alternative data sources had limitations, such as "methodical concerns on the use of social media for exploit prediction," "data feeds for proactive exploit prediction were limited to single sites that provided a relatively small number of predictions," and "other work demonstrates the viability of data collection but does not quantify the results of prediction" [cite: However, this previous work has its limitations., For instance, methodical concerns on the use of social media for exploit prediction were recently raised; data feeds for proactive exploit prediction were limited to single sites that provided a relatively small number of predictions; and other work demonstrates the viability of data collection but does not quantify the results of prediction.].

Without specific prior art references that detail these previously proposed methods and their limitations, a concrete obviousness analysis under 35 U.S.C. § 103, including identified combinations and motivations, cannot be performed.