DCT
2:22-cv-00476
Queryly LLC v. Hitel LLC
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: Queryly, LLC (New Jersey)
- Defendant: Hitel Technologies, LLC (Texas) and IP Edge, LLC (Texas)
- Plaintiff’s Counsel: Lamkin IP Defense
- Case Identification: 2:22-cv-00476, E.D. Tex., 12/16/2022
- Venue Allegations: Venue is alleged to be proper in the Eastern District of Texas because Defendants are Texas corporations that purport to reside in the district.
- Core Dispute: Plaintiff, a search technology provider, seeks a declaratory judgment of non-infringement and invalidity of a patent related to dynamic search learning after its customer, the Boston Globe, received an infringement notice from Defendants.
- Technical Context: The technology at issue involves search engine systems that can dynamically learn associations between unknown user search terms and documents based on user navigational behavior.
- Key Procedural History: This is a declaratory judgment action filed by a technology supplier (Queryly) after its customer received a notice of infringement. The complaint makes extensive allegations that Defendant Hitel is a "sham" entity controlled by patent monetization firm IP Edge, LLC, to shield it from liability, citing other judicial proceedings involving IP Edge. The complaint also raises an invalidity defense based on the addition of new claim terms during patent prosecution and a cause of action for unenforceability due to inequitable conduct.
Case Timeline
| Date | Event |
|---|---|
| 2005-02-25 | '617 Patent Priority Date |
| 2010-03-30 | '617 Patent Issue Date |
| 2022-12-16 | Complaint Filing Date |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 7,689,617 - Dynamic Learning For Navigation Systems
- Patent Identification: U.S. Patent No. 7,689,617, issued March 30, 2010.
The Invention Explained
- Problem Addressed: The patent describes a limitation in traditional search systems that rely on pre-defined keywords. These systems may fail to return relevant documents if a user's query uses a term (e.g., "vintage car") that is not an existing keyword for a document that uses a similar term (e.g., "antique automobile") ('617 Patent, col. 2:1-7).
- The Patented Solution: The invention proposes a method where, upon receiving an "unknown word" from a user, the system prompts the user to navigate through a "graph" of documents. After the user selects a relevant document, the system "learns" by creating a new association between the unknown word and the keywords contained within the selected document. This new association is then stored and used to improve future search results for that unknown word ('617 Patent, Abstract; col. 2:56-65). The process is illustrated in flowcharts such as Figure 5, which depicts the steps of identifying an unknown word, allowing user navigation, and storing new associations if the user is satisfied ('617 Patent, Fig. 5).
- Technical Importance: The described technology aimed to create more flexible and intelligent search and navigation systems capable of dynamically expanding their knowledge base beyond a static set of keywords ('617 Patent, col. 1:12-17).
Key Claims at a Glance
- The complaint contests the validity and infringement of the patent generally, with a focus on language found in the independent claims, such as claim 1.
- Independent Claim 1 of the ’617 Patent recites the following essential elements:
- In a system with user navigable nodes, documents, and keywords, receiving an input from a user.
- Determining that the input contains an "unknown word."
- Navigating through the system to a "current node" based on at least one additional user input.
- Presenting a response to the user based on documents attached to that current node.
- Following the presentation, "learning one or more associations between the unknown word and one or more keywords from the documents attached to the current node," forming a direct relationship.
- The complaint does not specify which dependent claims may be at issue.
III. The Accused Instrumentality
Product Identification
- The "Queryly Accused Product," which is identified as Queryly's search engine technology provided to its customers (Compl. ¶¶ 18-19, 22).
Functionality and Market Context
- The complaint describes the accused product as a search engine used by third parties, such as the Boston Globe, on their websites (Compl. ¶ 1, 15). Functionally, the complaint alleges that if a user enters a misspelled word, the system "immediately associates that word with another word stored in the system and then performs the search query" (Compl. ¶ 42). It is alleged that the system does not "learn" from user navigation, because performing the same search a second time, after a user has navigated through the results, yields the exact same results as the initial search (Compl. ¶¶ 44, 48-49). A screenshot from Exhibit E purports to show that a second search for "colunms" yields the same 67 results as the initial search, offered as evidence that no learning occurred based on user navigation (Compl. ¶ 49, Exhibit E).
IV. Analysis of Infringement Allegations
'617 Patent Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality (as characterized by Plaintiff for non-infringement) | Complaint Citation | Patent Citation |
|---|---|---|---|
| receiving an input from a user; | A user enters a misspelled query, "colunms," into the search bar on the Boston Globe website. A screenshot in Exhibit A shows this user input. | ¶45 | col. 13:52 |
| determining that the input contains an unknown word; | The complaint alleges that an infringing product must learn to associate "unknown words" with keywords in documents. | ¶41 | col. 13:53-54 |
| subsequent to the determining, navigating through the system to a current node...based upon at least one additional input from the user... | The infringement theory being challenged alleges that a user selecting an article from the search results constitutes navigation. | ¶47 | col. 13:55-60 |
| presenting at least one response to the user based upon the documents attached to the current node... | The system presents a list of 67 search results in response to the initial query. | ¶45 | col. 13:61-64 |
| learning one or more associations between the unknown word and one or more keywords from the documents attached to the current node... | The complaint alleges this element is not met because the accused system "never 'learns' to associate a misspelled word and another word based upon user navigation." It contends the system does not change its search outcomes after user navigation, as shown by a second, identical search result. | ¶¶43, 49 | col. 13:65-col. 14:2 |
Identified Points of Contention
- Technical Questions: The central technical dispute is whether the accused system's handling of a misspelled query constitutes the claimed "learning" process. The complaint puts forth evidence suggesting the system performs a one-time correction rather than a dynamic learning process based on user navigation (Compl. ¶¶ 42, 49). A screenshot from Exhibit A shows the initial search results for the misspelled query "colunms" on the Boston Globe website (Compl. ¶ 45, Exhibit A). The key question is what evidence will demonstrate that the accused system's internal operations match the specific learning mechanism recited in the claims.
- Scope Questions: The dispute raises the question of whether the claimed "learning" process, which involves forming new associations with keywords from a selected document, can be read to cover a system that allegedly uses a pre-set dictionary or algorithm to correct misspellings without altering its associations based on user behavior (Compl. ¶¶ 42-44).
V. Key Claim Terms for Construction
The Term: "learning one or more associations"
- Context and Importance: This term is the core of the infringement dispute. Queryly's non-infringement defense hinges on its argument that its system does not "learn" in the manner required by the patent. Practitioners may focus on this term because its definition will determine whether a system that performs static spell-correction infringes a patent directed to dynamic, behavior-based learning.
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The patent speaks generally of creating "associations" and "learning meanings for words," which a party could argue covers any method of linking a user's input to a more useful search term ('617 Patent, col. 2:18-20).
- Evidence for a Narrower Interpretation: The specification repeatedly describes "learning" as a specific multi-step process: an unknown word is entered, the user navigates to and selects a document, and only then does the system create an association between that unknown word and the keywords from the selected document ('617 Patent, col. 7:1-4). Claim 1 itself requires learning "following presentation of the at least one response."
The Term: "user navigable nodes"
- Context and Importance: This term is central to Queryly's invalidity challenge. The complaint alleges the term was added during prosecution to overcome prior art and lacks adequate support in the original specification, raising a written description issue under 35 U.S.C. § 112 (Compl. ¶¶ 59-63).
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: A party might argue that a person of ordinary skill in the art would understand "user navigable nodes" in the context of the patent's disclosure of graphs, websites, and menu trees, even if that exact phrase is not used verbatim.
- Evidence for a Narrower Interpretation: The complaint alleges that neither the term "user navigable nodes" nor a description of such a structure appears in the original patent specification as filed, which could support an argument that the term introduces new matter (Compl. ¶¶ 60-62).
The Term: "unknown word"
- Context and Importance: The infringement scenario described in the complaint involves a misspelled word, "colunms" (Compl. ¶ 45). Whether this constitutes an "unknown word" under the patent's definition is a potential point of dispute.
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The term could be construed broadly to mean any word not recognized as a keyword by the system.
- Evidence for a Narrower Interpretation: The patent provides a specific definition: "a word that is neither a keyword or an existing synonym, either generated in advance or through the learning method itself" ('617 Patent, col. 2:65-col. 3:2). A party could argue that an automatically corrected misspelling is treated as an "existing synonym" and therefore does not trigger the claimed learning process for "unknown words."
VI. Other Allegations
Inequitable Conduct and Sham Entity Allegations
- The complaint includes a cause of action for a declaration of unenforceability of the '617 Patent due to inequitable conduct (Compl. ¶¶ 66-86). The allegations are based on the assertion that employees of IP Edge filed a fraudulent assignment with the U.S. Patent and Trademark Office, purporting to transfer "all right, title, and interest" to Hitel while IP Edge allegedly retained material rights and control (Compl. ¶¶ 70, 75). The complaint alleges this was done with an intent to defraud the USPTO and the courts by hiding the true owner of the patent (Compl. ¶ 86). These allegations are linked to broader claims that Hitel is a "sham plaintiff" created by IP Edge to insulate itself from sanctions and attorney's fees awards (Compl. ¶¶ 3, 8, 74).
VII. Analyst’s Conclusion: Key Questions for the Case
- A key infringement question will be one of operational function: Does the accused search engine's method for handling misspellings operate as the claimed "learning" process, which requires creating new associations based on post-query user navigation, or does it perform a distinct, pre-determined correction that falls outside the claim scope?
- A central validity challenge will turn on written description: Does the original patent specification provide adequate support for the claim term "user navigable nodes," which was allegedly added during prosecution, or does the term lack the requisite description under 35 U.S.C. § 112?
- The case presents a significant question of corporate form and litigation conduct: Can the Plaintiff prove its allegations that Hitel is a sham entity for IP Edge and that the patent assignment was fraudulent, which could impact the Defendants' standing to assert the patent and potentially render the case "exceptional" for the purpose of awarding attorneys' fees?