US 9,552,548 Using classified text and deep learning algorithms to identify risk and provide early warning
3 defendants · 3 suits · updated October 6, 2026
Also cited as U.S. Patent No. 9,552,548 · US9552548B1 · US 9,552,548 B1 · 9552548 · the '548 patent
Arc Link LLC has filed 3 patent-infringement suits on US 9,552,548, "Using classified text and deep learning algorithms to identify risk and provide early warning" since June 8, 2026, all in the E.D. Tex., against 3 distinct defendants including Fortinet Inc., Check Point Software Technologies Ltd. and Sophos Limited. The patent is held by Ip3 2023 Series 923 Of Allied Security Trust I and was granted January 24, 2017 on a priority date of July 1, 2016. Its abstract opens: "Deep learning is used to identify specific, potential risks to an enterprise (of which litigation is the prime example) while such risks are still internal electronic communications." The examiner cited 5 references during prosecution.
- Assignee
- Ip3 2023 Series 923 Of Allied Security Trust I
- Priority date
- July 1, 2016
- Granted
- January 24, 2017
- Expires
- July 1, 2036
- Status
- Active
- Claims
- 20 (3 independent)
- Inventors
- Nelson E. Brestoff
- Full text
- Google Patents
The suits
3 cases · E.D. Tex. (3)
| Filed | Defendant | Plaintiff | Court | Docket |
|---|---|---|---|---|
| 2026-06-08 | Fortinet Inc. | Arc Link LLC | E.D. Tex. | 2:26-cv-00451 |
| 2026-06-17 | Check Point Software Technologies Ltd. | Arc Link LLC | E.D. Tex. | 2:26-cv-00476 |
| 2026-06-18 | Sophos Limited | Arc Link LLC | E.D. Tex. | 2:26-cv-00490 |
Docket links open the public record on CourtListener. Filing dates are the complaint's.
Independent claims
What is actually asserted, verbatim.
1. A method of using classified text and deep learning algorithms to identify risk and provide early warning comprising: creating one or more training datasets for textual data corresponding to one or more risk classifications, wherein said risk classification comprises one or more threats or risks of interest; training one or more deep learning algorithms using said one or more training datasets; extracting an internal electronic communication of an enterprise; applying said one or more deep learning algorithms to said internal electronic communication to identify and report any one of said one or more threats or risks of interest; determining if said identified one of said one or more threats or risks of interest is a false positive or a true positive; re-training said one or more deep learning algorithms if said identified one of said one or more threats or risks of interest is a false positive; and saving said internal electronic communication in a true positive database if said identified one of said one or more threats or risks of interest is a true positive.
12. A method of using classified text and deep learning algorithms to identify risk and provide early warning comprising: creating one or more training datasets by mining litigation database for textual data corresponding to one or more threats or risks of interest; training one or more deep learning algorithms using said one or more training datasets; extracting and indexing one or more internal electronic communications; applying said one or more deep learning algorithms to said one or more internal electronic communications to identify and report any one of said one or more threats or risks of interest; determining if said identified one of said one or more threats or risks of interest is a false positive or a true positive; re-training said one or more deep learning algorithms if said identified one of said one or more threats or risks of interest is a false positive; and saving said internal electronic communication in a true positive database for evaluation if said identified one of said one or more threats or risks of interest is a true positive, wherein said true positive database provides early warning.
17. A method of using classified text and deep learning algorithms to identify risk and provide early warning comprising: creating one or more training datasets for textual data corresponding to one or more risk classifications, wherein said risk classification comprises one or more threats or risks of interest; training one or more deep learning algorithms using said one or more training datasets; extracting and indexing one or more internal electronic communications of an enterprise; applying said one or more deep learning algorithms to said one or more internal electronic communications to identify any one of said one or more threats or risks of interest; and reporting said identified one of said one or more threats or risks of interest to an existing case management system for action by the enterprise.
Art before the examiner
11 references of record, 5 cited by the examiner.
Everything listed here was before the Office and is what a challenge must go beyond. Art the examiner never saw is where an invalidity search starts.
Examiner-cited
- US7574409B2 — Method, apparatus, and system for clustering and classification (Vericept Corporation)
- US8473442B1 — System and method for intelligent state management (Mcafee, Inc.)
- US8706709B2 — System and method for intelligent term grouping (Mcafee, Inc.)
- US9292493B2 — Systems and methods for automatically detecting deception in human communications expressed in digital form (The Trustees Of The Stevens Institute Of Technology)
- US9342794B2 — Non-linear classification of text samples (Bazaarvoice, Inc.)
Applicant-cited and other (6)
- US20040024775A1 — Electronic management and distribution of legal information
- US20040205661A1 — System and method of reviewing and revising business documents
- US20100325171A1 — System and method for gathering and analyzing litigation marketplace intelligence
- US20130297519A1 — System and method for identifying potential legal liability and providing early warning in an enterprise
- US20140244524A1 — System and method for identifying potential legal liability and providing early warning in an enterprise
- US9324022B2 — Classifying data with deep learning neural records incrementally refined through expert input
Classification
- G06Q10/0635primary — Risk analysis of enterprise or organisation activities
- G06F16/35 — Clustering; Classification
- G06F18/214 — Generating training patterns; Bootstrap methods, e.g. bagging or boosting
- G06F40/10 — Text processing
- G06N20/00 — Machine learning
- G06N3/04 — Architecture, e.g. interconnection topology
- G06N3/044 — Recurrent networks, e.g. Hopfield networks
- G06N3/0442 — Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
- G06N3/08 — Learning methods
- G06N3/09 — Supervised learning
- G06Q10/107 — Computer-aided management of electronic mailing [e-mailing]
- G06Q50/18 — Legal services
- H04L9/50 — Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols using hash chains, e.g. blockchains or hash trees
Family
- US9552548B1 — B1 (2017)
- US10095992B1 — B1 (2018)
- US9760850B1 — B1 (2017)
Abstract
Deep learning is used to identify specific, potential risks to an enterprise (of which litigation is the prime example) while such risks are still internal electronic communications. The system involves mining and using existing classifications of data (e.g., from a litigation database) to train one or more deep learning algorithms, and then examining the internal electronic communications with the trained algorithm, to generate a scored output that will enable enterprise personnel to be alerted to risks and take action in time to prevent the risks from resulting in harm to the enterprise or others.
Also asserted by Arc Link LLC
- US 11,563,655 — Network monitoring apparatus and method thereof in programmable network virtualization (3 defendants)
- US 8,200,592 — System and method for modeling multilabel classification and ranking (3 defendants)
- US 9,549,328 — Method to optimize the communication parameters between an access point and at least one client device (3 defendants)
Named in one of these suits? A prior-art search is commissioned once and serves every defendant who shares it. The examiner's own search record is free on the tools page, and your first preliminary search is free — one claim, back within 24 hours — so you know whether there is art the Office never saw before anyone pays for more. Flat fees: $2,500$899 for the independent claims, $5,000$1,500 for every asserted claim.
This page is assembled from public records — federal court dockets and the patent's own file — and is updated as the case list changes. It states what the record shows and is not an opinion on the validity or infringement of any claim. Patent data via Google Patents; docket data via CourtListener.