DCT
6:23-cv-00108
Ask Sydney LLC v. Amazon.com Services LLC
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: Ask Sydney, LLC (New York)
- Defendant: Amazon.com Services, LLC (Delaware)
- Plaintiff’s Counsel: Ramey LLP
- Case Identification: 6:23-cv-00108, W.D. Tex., 02/13/2023
- Venue Allegations: Plaintiff alleges venue is proper because Defendant has regular and established places of business in the district, including a specific corporate office located at 11501 Alterra Parkway, Austin, Texas.
- Core Dispute: Plaintiff alleges that Defendant’s Amazon.com online retail and web service platform infringes two patents related to iterative, feedback-driven methods for visually guiding a user to a product or item of current interest.
- Technical Context: The technology at issue involves interactive search algorithms where a user's sequential feedback on presented images is used to process associated metadata "tags" and refine subsequent results, thereby narrowing in on a user's specific, contemporaneous preference.
- Key Procedural History: The asserted U.S. Patent No. 10,474,705 is a continuation of the family that produced the asserted U.S. Patent No. 9,323,786, indicating a shared specification and subject matter. The '705 Patent is subject to a terminal disclaimer, which may limit its enforceable term to that of the earlier '786 Patent.
Case Timeline
| Date | Event |
|---|---|
| 2014-08-15 | Earliest Priority Date for '786 and '705 Patents |
| 2016-04-26 | U.S. Patent No. 9,323,786 Issued |
| 2019-11-12 | U.S. Patent No. 10,474,705 Issued |
| 2023-02-13 | Complaint Filed |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 9,323,786 - "System and computer method for visually guiding a user to a current interest," issued April 26, 2016
The Invention Explained
- Problem Addressed: The patent describes a deficiency in recommendation systems that rely on a user's long-term historical preferences. Such systems are often too general to identify a user's specific, granular, and potentially fleeting "current interest" or "craving" at a particular moment in time (’786 Patent, col. 1:26-44).
- The Patented Solution: The invention discloses a guided, iterative search method. A system presents a user with an image representing a physical object (e.g., a food dish). The image is associated with a set of descriptive "tags." The user provides feedback indicating a preference (e.g., like/dislike), which the system uses to process the tags and determine a new set of tags. Based on this new tag set, a different image is presented, and the process repeats, generating a sequence of images that narrows in on the user's current interest (’786 Patent, Abstract; col. 2:1-20).
- Technical Importance: This method sought to improve upon static, filter-based e-commerce searches by creating a dynamic, interactive dialogue between the user and the system within a single session to overcome "information overload" (’786 Patent, col. 5:28-40).
Key Claims at a Glance
- The complaint asserts independent claim 1 (Compl. ¶18).
- Essential Elements of Claim 1 (Method):
- Receiving an indication of an application instance on a user's electronic device.
- Determining a plurality of tags specific to the user based on a user profile.
- Transmitting an electronic image representing a physical object, where the image is associated with a set of tags from the user-specific plurality.
- Receiving a user input indicating a preference for the physical object.
- Processing the plurality of tags based on the preference to determine a next set of tags.
- Determining a next electronic image associated with the next set of tags, which represents a different physical object.
- Generating a sequence of images by repeating the transmitting, receiving, and processing steps.
- The complaint reserves the right to assert additional claims (Compl. ¶18).
U.S. Patent No. 10,474,705 - "Iterative image search algorithm informed by continuous human-machine input feedback," issued November 12, 2019
The Invention Explained
- Problem Addressed: As with its parent patent, the '705 Patent addresses the problem that recommendation systems based on static user history cannot accurately predict a user's specific, "current, specific preference" (’705 Patent, col. 1:32-46).
- The Patented Solution: The patent describes a similar iterative search method but focuses on the mechanics of filtering results. The method involves presenting images, receiving either a "favorable indication" or an "unfavorable indication" from the user, and then "adjusting weights" of the tags associated with the image. This weight adjustment, based on user feedback, informs the selection of the subsequent image in the sequence, allowing the search to "iteratively resolve toward" a desired item (’705 Patent, Abstract; col. 26:21-49).
- Technical Importance: This approach refines the guided search concept by articulating a specific mechanism—tag weight adjustment—for learning from both positive and negative user feedback within a single search session (’705 Patent, col. 15:30-42).
Key Claims at a Glance
- The complaint asserts independent claim 1 (Compl. ¶25).
- Essential Elements of Claim 1 (Method):
- Determining a plurality of digital images to present in a sequence, each associated with a plurality of tags.
- Receiving an "unfavorable indication of a disinclination" for an image's features.
- Analyzing tags to determine a next set of tags.
- "Adjusting weights" of tags based on an association relative to tags in the same category.
- Transitioning from the current digital image to a subsequent digital image.
- Receiving a further input for the subsequent image.
- The complaint reserves the right to assert additional claims (Compl. ¶25).
III. The Accused Instrumentality
Product Identification
- The complaint identifies the accused instrumentality broadly as "Amazon.com, is an online retailer and web service provider" (Compl. ¶16).
Functionality and Market Context
- The complaint does not describe the specific functionality of Amazon.com that is alleged to infringe. It alleges that Defendant "derives substantial revenue from goods and services" and commercializes infringing products in the district and throughout the United States (Compl. ¶7, ¶21, ¶28). The infringement theory appears to be centered on the user experience of browsing for products on the Amazon.com platform, where user interactions with product images allegedly trigger a back-end process of refining subsequent product recommendations in a manner that reads on the patent claims. The complaint references claim chart exhibits that were not available for this analysis (Compl. ¶23, ¶30).
IV. Analysis of Infringement Allegations
The complaint references but does not include Exhibits C and D, which contain the detailed infringement claim charts (Compl. ¶23, ¶30). The analysis below is based on the complaint's general allegations and a plausible mapping of the asserted claims to the functionality of a large-scale e-commerce platform.
No probative visual evidence provided in complaint.
’786 Patent Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| determining, by the one or more computer devices, a plurality of tags specific to the user... based on each tag of the plurality of tags specific to the user being associated with a profile of the user | The Amazon.com system allegedly accesses a set of product-related tags (e.g., categories, brands, attributes) determined to be relevant to a user based on their account profile or browsing history. | ¶18 | col. 12:42-54 |
| transmitting... one electronic image... representing a physical object... and being associated with one set of tags | Amazon's servers transmit a product image to a user's browser or app, with the product being associated with metadata tags (e.g., "running shoe," "Nike," "blue"). | ¶18 | col. 8:51-58 |
| receiving... an input from the user indicating a preference for the physical object | The system receives an indication of user preference, such as a user clicking on a specific product image to view its detail page. | ¶18 | col. 7:1-6 |
| processing... the plurality of tags specific to the user based on the preference and the one set of tags to determine a next set of tags | The system allegedly processes the tags of the selected product to determine which tags are relevant for selecting the next products to display (e.g., in a "related items" or "customers also viewed" section). | ¶18 | col. 14:10-21 |
| determining, by the one or more computer devices, a next electronic image from the plurality of electronic images associated with the next set of tags | Based on the processed tags, the system selects the next product image to display to the user as a recommendation. | ¶18 | col. 17:12-20 |
| generating a sequence of electronic images... by repeating the transmitting, the causing, the receiving of the input, the processing, and the determining | The alleged infringement arises from the continuous process where a user navigates from one product to another, with the system updating its recommendations at each step, creating an iterative sequence. | ¶18 | col. 17:49-55 |
Identified Points of Contention
- Technical Question: Does the complaint provide evidence that Amazon’s recommendation algorithm functions via the specific, session-based tag processing method claimed, or does it use a more complex machine learning model that considers factors beyond the sequence of interactions in a single session?
- Scope Question: Does a user’s general browsing path on an e-commerce website constitute the "generating a sequence of electronic images" as contemplated by the patent, which describes a more structured, guided process for resolving a "current interest"?
V. Key Claim Terms for Construction
The Term: "plurality of tags specific to the user" (’786 Patent, Claim 1)
- Context and Importance: The definition of this term is critical for determining when infringement begins. The dispute may center on whether this requires a predefined set of tags formally associated with a user's static profile, or if it can be a dynamic set of tags deemed relevant to the user's context at the start of a browsing session.
- Intrinsic Evidence for a Broader Interpretation: The specification discusses tracking user interactions to "mine trends across the users," which could support an interpretation where tags become "specific to the user" dynamically based on behavior, not just a static profile (’786 Patent, col. 12:1-6).
- Intrinsic Evidence for a Narrower Interpretation: The claim explicitly links these tags to being "associated with a profile of the user," and the specification describes "user profiles" that store information like preferred tags and liked images, suggesting a more formally defined and persistent data structure (’786 Patent, col. 12:42-54).
The Term: "adjusting weights" (’705 Patent, Claim 1)
- Context and Importance: This term is central to the alleged inventive aspect of the ’705 Patent. Practitioners may focus on this term because the infringement case depends on whether Amazon's algorithm performs a function that can be characterized as "adjusting weights" in response to user feedback.
- Intrinsic Evidence for a Broader Interpretation: The claim language itself does not specify a particular mathematical formula, only "adjusting weights... based on an association relative to tags." This could be argued to cover any algorithmic promotion or demotion of tag-related products in response to user input (’705 Patent, col. 26:37-41).
- Intrinsic Evidence for a Narrower Interpretation: The specification describes a process of determining a "next set of tags" based on a "threshold association" and selecting tags with the "highest weighting," which may suggest a more structured, numerical weighting system rather than a generalized algorithmic change (’705 Patent, col. 16:1-14).
VI. Other Allegations
- Willful Infringement: The complaint does not explicitly use the word "willful." However, it alleges that Defendant "has made no attempt to design around the claims" of the patents-in-suit and "did not have a reasonable basis for believing that the claims... were invalid" (Compl. ¶19-20, ¶26-27). Furthermore, the prayer for relief requests a judgment that this is an "exceptional case" under 35 U.S.C. § 285, which is the statutory basis for awarding enhanced damages and attorney's fees, often in cases of willful infringement (Compl., Prayer for Relief ¶C). These allegations lay a foundation for a potential future claim of willfulness.
VII. Analyst’s Conclusion: Key Questions for the Case
- A key evidentiary question will be one of algorithmic correspondence: can the Plaintiff demonstrate through discovery that Amazon's highly complex and likely multi-faceted recommendation engine performs the specific, sequential tag-processing and weight-adjustment steps required by the patent claims, or is there a fundamental mismatch in technical operation?
- A central legal issue will be one of definitional scope: can the term "sequence of electronic images", which in the patent context describes a focused, iterative process to resolve a single "current interest," be construed broadly enough to read on a user's more general and potentially disjointed browsing activity on a vast e-commerce platform?
- A third question will concern damages and claim scope: given the breadth of the accused instrumentality ("Amazon.com"), the case may involve significant dispute over the appropriate royalty base and whether the patented features drive demand for the entire platform or only for a small, isolatable component of the recommendation functionality.