Hitel Tech LLC v. Tapestry Inc

I. Executive Summary and Procedural Information

• Parties & Counsel:
  • Plaintiff: Hitel Technologies LLC (Texas)
  • Defendant: Tapestry, Inc. (Maryland)
  • Plaintiff’s Counsel: The Mort Law Firm, PLLC
• Case Identification: 6:22-cv-00113, W.D. Tex., 01/31/2022
• Venue Allegations: Venue is alleged to be proper based on Defendant having committed acts of infringement and maintaining a regular and established place of business within the Western District of Texas.
• Core Dispute: Plaintiff alleges that Defendant’s Coach brand e-commerce website infringes a patent related to dynamic learning systems for website navigation and search.
• Technical Context: The technology concerns methods for information retrieval systems to learn from user behavior, particularly when a user query contains a term not recognized by the system, in order to improve future search results.
• Key Procedural History: The complaint does not mention any prior litigation, Inter Partes Review (IPR) proceedings, or specific licensing history related to the patent-in-suit.

Case Timeline

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, "DYNAMIC LEARNING FOR NAVIGATION SYSTEMS", issued March 30, 2010.

The Invention Explained

• Problem Addressed: The patent describes a limitation in conventional search and navigation systems that rely strictly on pre-defined keywords. Such systems fail to return relevant results if a user’s query uses a synonym or related term not explicitly present in the system's data (e.g., searching for "vintage car" when a document only contains "antique automobile") ('617 Patent, col. 1:64-col. 2:2).
• The Patented Solution: The invention proposes a method where, if a user enters an "unknown word," the system allows the user to navigate to a desired document or node. The system then "learns" an association between the previously unknown word and the keywords contained within the user-selected document. This new association is stored and used to improve subsequent searches by other users who might use the same "unknown word" ('617 Patent, col. 3:1-14; Fig. 5).
• Technical Importance: This approach sought to create more flexible and intelligent navigation systems that could adapt to user vocabulary and intent, moving beyond the rigidity of traditional keyword-based search engines ('617 Patent, col. 2:8-20).

Key Claims at a Glance

• The complaint asserts infringement of at least Claim 1 ('617 Patent, col. 13:45-col. 14:18; Compl. ¶18).
• Independent Claim 1 recites the essential elements of a method comprising:
  • Receiving an input from a user at a user navigable node.
  • 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 at least one response to the user based on documents attached to the current node.
  • Learning one or more associations between the unknown word and keywords from the documents attached to the current node, forming a direct relationship.
• The complaint does not explicitly reserve the right to assert dependent claims.

III. The Accused Instrumentality

Product Identification

• The accused instrumentality is the Coach Website, located at https://www.coach.com (the "Accused Website") (Compl. ¶7).

Functionality and Market Context

• The complaint focuses on the search and navigation functionality of the Accused Website, which allows users to search for items using keywords in a search box (Compl. ¶8).
• The core of the allegation centers on the website's behavior when a user enters a search term that does not directly match an existing category, such as a typo (Compl. ¶9). The complaint alleges that the website identifies the term as "unknown," suggests a corrected term, and upon user selection of the suggestion, presents corresponding product results (Compl. ¶10-11). The complaint includes a screenshot showing that a search for "Bgs" results in a prompt stating, "No Results Found for 'Bgs'. Did you mean Business? Showing 17 results for Business" (Compl. p. 4). This functionality is alleged to form the basis of the infringement.

IV. Analysis of Infringement Allegations

U.S. Patent No. 7,689,617 Infringement Allegations

• Identified Points of Contention:
  • Technical Questions: A primary question is whether the Accused Website's typo-correction feature performs the specific "learning" process required by the claim. The final claim element requires forming and storing a "direct relationship" for future use. The complaint alleges this occurs but provides no direct evidence of a persistent association being created, raising the question of whether the website is merely performing a one-time string-similarity check or a true learning function as described in the patent.
  • Scope Questions: The dispute may turn on whether a typo like "Bgs" qualifies as an "unknown word" in the context of the patent. The patent appears to contemplate semantically distinct words (like "vintage car" vs. "antique automobile") rather than simple misspellings ('617 Patent, col. 1:64-col. 2:2). The interpretation of "learning" will also be critical—does a temporary, session-based correction meet the claim limitation, or must the system permanently update its associations as the patent's specification suggests? ('617 Patent, col. 7:6-9).

V. Key Claim Terms for Construction

• The Term: "unknown word"

• Context and Importance: This term is foundational to the claim, as the entire method is triggered by its detection. The parties will likely dispute whether a simple typo, which can be algorithmically corrected via string distance analysis, constitutes an "unknown word" in the manner envisioned by the patent, which seems to focus more on conceptual or semantic novelty.

• Intrinsic Evidence for Interpretation:

  • Evidence for a Broader Interpretation: The specification states that an unknown word is "a word that is neither a keyword or an existing synonym" ('617 Patent, col. 2:63-65), a definition that could arguably include a misspelling not found in the system's dictionary.
  • Evidence for a Narrower Interpretation: The patent's own examples, such as "vintage car" for "antique automobile," suggest the term refers to semantically meaningful words that are simply absent from the keyword list, not typographical errors ('617 Patent, col. 1:64-col. 2:2).

• The Term: "learning one or more associations"

• Context and Importance: This term is the functional core of the invention. Its construction will determine what level of technical proof is required to show infringement. The central issue is whether "learning" requires a persistent change to the system's knowledge base, as opposed to a transient, in-session response to a query.

• Intrinsic Evidence for Interpretation:

  • Evidence for a Broader Interpretation: The term itself is not explicitly limited to permanent storage, and one could argue that any process that associates the input with a result for the duration of a user action constitutes "learning" in a broad sense.
  • Evidence for a Narrower Interpretation: The specification repeatedly describes a process where new associations are stored for future use, implying a persistent, systemic change. For example, it states, "This new association is stored by the system for future use" and "the process will work as before from then on" ('617 Patent, col. 7:8-9; col. 12:29-31). The flowchart in Figure 5 also shows a final step to "[s]tore the new associations" (Fig. 5, step 550).

VI. Other Allegations

• Indirect Infringement: The complaint alleges that Defendant infringes by "encouraging customers (and potential customers) to use the Accused Website to practice the claims" (Compl. ¶16). This appears to be a claim for induced infringement, based on the theory that Defendant provides the website and implicitly instructs users on how to operate its search function, which allegedly performs the patented method.

VII. Analyst’s Conclusion: Key Questions for the Case

• A core issue will be one of definitional scope: does a common e-commerce typo-correction feature, which suggests an alternative based on string similarity (e.g., "Bgs" -> "Business"), fall within the patent's definition of a system that handles an "unknown word" by "learning" a new conceptual association for future use?

• A key evidentiary question will be one of technical implementation: what evidence can be produced to show that the Accused Website performs the final, critical step of Claim 1—that it actually "learns" and stores a persistent "direct relationship" between a typo and a category's keywords, rather than simply executing a transient, algorithmic correction for a single user session? The complaint's reliance on "information and belief" for this element suggests this will be a central point of discovery.