DCT
3:15-cv-03486
Word To Info Inc v. Google Inc
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: Word to Info, Inc. (Texas)
- Defendant: Google Inc. (Delaware)
- Plaintiff’s Counsel: Farney Daniels PC
- Case Identification: 3:15-cv-03486, N.D. Cal., 11/09/2015
- Venue Allegations: Venue is alleged as proper based on Google's systematic contacts with the Northern District of California and Google's prior acknowledgement of jurisdiction in a motion to transfer venue in the same case.
- Core Dispute: Plaintiff alleges that Defendant’s Google Knowledge Graph and Google Now products infringe a portfolio of seven patents related to systems and methods for natural language processing and storing and retrieving knowledge.
- Technical Context: The technology concerns computational systems for understanding human language by assigning semantic meaning to words and storing information in structured databases, a foundational concept for modern search engines and digital assistants.
- Key Procedural History: The complaint alleges Defendant was on notice of the asserted technology at least as early as October 2014, when the U.S. Patent and Trademark Office cited a publication claiming priority to the lead patent-in-suit during the prosecution of a Google patent application. The complaint further notes that an inventor on Google's application was involved in the development of the accused Knowledge Graph feature.
Case Timeline
| Date | Event |
|---|---|
| 1994-09-30 | Earliest Priority Date for all Patents-in-Suit |
| 1998-02-03 | U.S. Patent No. 5,715,468 Issued |
| 2000-10-24 | U.S. Patent No. 6,138,087 Issued |
| 2003-08-19 | U.S. Patent No. 6,609,091 Issued |
| 2008-03-25 | U.S. Patent No. 7,349,840 Issued |
| 2011-01-18 | U.S. Patent No. 7,873,509 Issued |
| 2012-05-16 | Google Knowledge Graph Launch Date (per cited source) |
| 2012-12-04 | U.S. Patent No. 8,326,603 Issued |
| 2013-12-30 | Google Patent Application 14/143,161 Filed |
| 2014-04-01 | U.S. Patent No. 8,688,436 Issued |
| 2014-10-10 | Notice of Allowability for Google's Application mailed, citing Plaintiff's technology |
| 2015-02-23 | Google files another application claiming benefit of the '161 application |
| 2015-11-09 | Complaint Filed |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 5,715,468 - "Memory System for Storing and Retrieving Experience and Knowledge with Natural Language"
The Invention Explained
- Problem Addressed: The patent's background describes prior art natural language processing systems as inadequate for understanding complex language, noting limitations in both "case frame" and "predicate calculus" approaches that struggled to represent and retrieve knowledge derived from natural language conversations (ʼ468 Patent, col. 2:50-3:24).
- The Patented Solution: The invention discloses a memory system that combines syntactic (grammatical) and semantic (meaning-based) analysis to understand language. The system uses a comprehensive dictionary where words are associated with "word sense numbers" to disambiguate meaning, along with "state representation data" to capture context. This information is stored in interconnected memory structures, including a "directed graph" of nodes, which represents knowledge and "experience" in a way that allows for retrieval based on relationships, not just keywords (’468 Patent, Abstract; col. 4:26-58).
- Technical Importance: The patent details a comprehensive architecture for enabling computers to move beyond simple keyword matching toward a contextual, meaning-based understanding of language, a critical step in the development of advanced search and AI systems (Compl. ¶8).
Key Claims at a Glance
The complaint alleges infringement of "one or more claims" without specifying which ones (Compl. ¶25). The allegations in paragraph 26 appear to track the elements of a system or method claim, such as independent method Claim 1.
- Claim 1 (Method):
- providing electronically encoded data which is representative of said natural language;
- providing a dictionary data base... [containing] syntax usage data, associated word sense numbers having associated state representation data and/or function codes;
- lexically processing said electronically encoded data to access said dictionary data base;
- providing a grammar specification;
- utilizing said syntax usage data... to produce output data representative of a grammatical parse of said natural language;
- said output data including said syntax usage.
U.S. Patent No. 6,138,087 - "Memory System for Storing and Retrieving Experience and Knowledge with Natural Language Utilizing State Representation Data, Word Sense Numbers, Function Codes and/or Directed Graphs"
The Invention Explained
- Problem Addressed: As a continuation of the application leading to the ’468 Patent, this patent addresses the same fundamental problem: the inadequacy of prior art systems in understanding and retrieving knowledge from complex natural language inputs (’087 Patent, col. 2:24-3:28).
- The Patented Solution: The invention describes a memory system where knowledge and experience are stored in an "experience and knowledge data base which is comprised of directed graphs." These graphs consist of nodes connected by paths, where each node is associated with a "clause implying word sense numbers." The system retrieves information by finding a path through the graph where the "access conditions" of the nodes are met, which allows for context-sensitive information retrieval (’087 Patent, Abstract; col. 13:1-25).
- Technical Importance: This patent further elaborates on the use of directed graphs for knowledge representation, a data structure paradigm that has become central to modern semantic search technologies.
Key Claims at a Glance
The complaint alleges infringement of "one or more claims" without specifying them (Compl. ¶34). The infringement allegations in paragraph 35 appear to map to independent method Claim 23.
- Claim 23 (Method):
- providing an experience and knowledge data base which is comprised of directed graphs comprised of nodes with associated clause implying word sense numbers organized into paths of said nodes,
- processing said associated clause implying word sense numbers of said experience and knowledge data base.
U.S. Patent No. 6,609,091 - "Memory System for Storing and Retrieving Experience and Knowledge with Natural Language Utilizing State Representation Data, Word Sense Numbers, Function Codes and/or Directed Graphs"
- Technology Synopsis: This patent describes a natural language processing system that uses a dictionary containing syntax data and word sense numbers. A key feature is a "data base of requirements" which must be met by the state representation data of a word sense number for it to be selected, adding a layer of rule-based validation to the language interpretation process (’091 Patent, Abstract).
- Asserted Claims: The complaint alleges infringement of "one or more claims" (Compl. ¶38).
- Accused Features: The accused features include the system for electronically encoding data and using a dictionary and a requirements database to select and utilize word sense numbers within Google's Knowledge Graph and Google Now products (Compl. ¶39).
U.S. Patent No. 7,349,840 - "Memory System for Storing and Retrieving Experience and Knowledge with Natural Language Utilizing State Representation Data, Word Sense Numbers, Function Codes, Directed Graphs and/or Context Memory"
- Technology Synopsis: This patent adds the concept of a "context data base" to the natural language processing system. The system utilizes this context database, in conjunction with a dictionary and a grammar specification, to select appropriate word sense numbers, thereby grounding the interpretation of language in the broader conversational or situational context (’840 Patent, Abstract).
- Asserted Claims: The complaint alleges infringement of "one or more claims" (Compl. ¶42).
- Accused Features: The accused features involve the use of a dictionary database and a context database to select word sense numbers and function codes in Google's Knowledge Graph and Google Now products (Compl. ¶43).
U.S. Patent No. 7,873,509 - "Memory System for Storing and Retrieving Experience and Knowledge with Natural Language Utilizing State Representation Data, Word Sense Numbers, Function Codes, Directed Graphs, Context Memory, and/or Purpose Relations"
- Technology Synopsis: This patent focuses on identifying the "purpose" behind natural language statements. It describes a system with a knowledge base of directed graphs that uses "purpose relation path identification processes" to find paths between nodes, allowing the system to infer user intent or the objective of a statement. It also provides criteria for selecting among multiple potential paths (’509 Patent, Abstract).
- Asserted Claims: The complaint alleges infringement of "one or more claims" (Compl. ¶49).
- Accused Features: The accused functionality is the use of an experience and knowledge database with directed graphs and "purpose relation path identification processes" to find and select paths within Google's Knowledge Graph and Google Now products (Compl. ¶50).
U.S. Patent No. 8,326,603 - "Memory System for Storing and Retrieving Experience and Knowledge with Natural Language Queries"
- Technology Synopsis: This patent claims a system that uses a natural language processor to handle queries. The system includes an experience and knowledge database structured as directed graphs, where nodes are organized into paths with specific access conditions. A "purpose relation path identification processor" is used to identify paths based on the query, enabling the system to answer natural language questions (’603 Patent, Abstract).
- Asserted Claims: The complaint alleges infringement of "one or more claims" (Compl. ¶53).
- Accused Features: The accused system utilizes a natural language processor and a knowledge database of directed graphs, along with a purpose relation path identification processor, to process queries in Google's Knowledge Graph and Google Now (Compl. ¶54).
U.S. Patent No. 8,688,436 - "Memory System for Storing and Retrieving Experience and Knowledge by Utilizing Natural Language Responses"
- Technology Synopsis: This patent introduces a "natural language plausibility and expectedness processor." This component evaluates the likelihood or sensicality of a given interpretation of natural language. The system uses this processor to initiate access to dictionary entries associated with the language, improving the accuracy of the interpretation (’436 Patent, Abstract).
- Asserted Claims: The complaint alleges infringement of "one or more claims" (Compl. ¶57).
- Accused Features: The infringement allegation targets the use of a system that processes natural language using a dictionary database and a "natural language plausibility and expectedness processor" in Google's Knowledge Graph and Google Now products (Compl. ¶58).
III. The Accused Instrumentality
- Product Identification: The accused instrumentalities are Google's "Knowledge Graph feature of its internet search engine product" and/or its "Google Now product" (Compl. ¶25).
- Functionality and Market Context: The complaint alleges these products operate by processing natural language to understand user intent and retrieve relevant information (Compl. ¶26). The Knowledge Graph is identified as a significant feature of Google's search engine, with the complaint citing a Google blog post from 2012 announcing its launch and describing its function as understanding real-world entities and their relationships (Compl. ¶21). The complaint asserts that these products are operated throughout the United States and form a core part of Google's widely used internet search services (Compl. ¶25).
No probative visual evidence provided in complaint.
IV. Analysis of Infringement Allegations
U.S. Patent No. 5,715,468 Infringement Allegations
The complaint does not map its allegations to a specific claim. The allegations in paragraph 26 are summarized below against the elements of independent Claim 1.
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| A method of processing natural language, which comprises steps: providing electronically encoded data which is representative of said natural language | Google provides a system that electronically encodes data representative of natural language. | ¶26 | col. 367:1-3 |
| providing a dictionary data base wherein said dictionary data base contains a plurality of entries which are comprised of one or more of syntax usage data, associated word sense numbers having associated state representation data and/or function codes | The system includes a database containing syntax usage data entries, associated word sense numbers with associated state representation data, and/or function codes. | ¶26 | col. 367:4-9 |
| lexically processing said electronically encoded data to access said dictionary data base | The system lexically processes the electronically encoded data and accesses the dictionary data base. | ¶26 | col. 367:10-11 |
| providing a grammar specification | The system provides a grammar specification. | ¶26 | col. 367:12-12 |
| utilizing said syntax usage data which are from entries of said dictionary data base and which are associated with said word sense numbers with reference to said grammar specification to produce output data representative of a grammatical parse of said natural language | The system uses syntax usage data from the database with reference to the grammar specification to produce an output representative of a grammatical parse. | ¶26 | col. 367:13-18 |
| said output data including said syntax usage. | The output data from the system includes syntax usage. | ¶26 | col. 367:19-20 |
- Identified Points of Contention:
- Technical Questions: A primary question will be whether Google's modern, likely machine-learning-based, NLP systems utilize the specific data structures recited in the claim, such as a "dictionary data base" with "word sense numbers" and "state representation data" as those terms are defined in the patent's extensive specification. The complaint does not provide technical evidence showing that Google's internal architecture maps to this specific claimed structure.
- Scope Questions: The dispute may center on whether the term "word sense number," which is described in the patent with a specific, hierarchical structure, can be construed to read on any numerical identifier Google might use for semantic disambiguation, or if it is limited to the patent's detailed embodiments.
U.S. Patent No. 6,138,087 Infringement Allegations
The complaint does not map its allegations to a specific claim. The allegations in paragraph 35 are summarized below against the elements of independent Claim 23.
| Claim Element (from Independent Claim 23) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| A method of processing experience and knowledge, which comprises steps: providing an experience and knowledge data base which is comprised of directed graphs comprised of nodes with associated clause implying word sense numbers organized into paths of said nodes | Google provides a system that uses an experience and knowledge data base which is comprised of directed graphs comprised of nodes with associated clause implying word sense numbers organized into paths of said nodes. | ¶35 | col. 369:3-6 |
| processing said associated clause implying word sense numbers of said experience and knowledge data base. | The system processes said associated clause implying word sense numbers of said experience and knowledge database. | ¶35 | col. 370:1-3 |
- Identified Points of Contention:
- Technical Questions: The central technical question is whether Google's Knowledge Graph, as a product, constitutes the "experience and knowledge data base" as claimed. Evidence will be required to show that it is not merely a database of facts but one that stores "experience" and is processed by traversing paths between nodes based on "access conditions" as described in the patent.
- Scope Questions: The definition of "directed graph" may be a point of contention. While Google's product is called a "Knowledge Graph," the parties may dispute whether its technical implementation meets the specific structural and functional requirements of the "directed graph" described and claimed in the patent.
V. Key Claim Terms for Construction
The Term: "word sense number" (from '468 Patent, Claim 1)
Context and Importance: This term is fundamental to the patents' method of disambiguating word meaning. Whether Google's system uses an equivalent structure will be critical to the infringement analysis. Practitioners may focus on this term because the patent's specification provides a highly detailed and structured definition, which could limit the claim's scope relative to modern, potentially more abstract, methods of representing word meaning.
Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The claim language itself refers to "associated word sense numbers," which could arguably encompass any identifier linked to a specific meaning of a word (’468 Patent, col. 367:7).
- Evidence for a Narrower Interpretation: The specification provides a detailed format for word sense numbers, including components for a "Word Sense Identification Number," "Type Number," "Specificity Number," and "Experience Number" (’468 Patent, Fig. 17A, col. 147:1-40). This detailed embodiment may be used to argue that the term requires this specific, multi-component structure.
The Term: "experience and knowledge data base" (from '087 Patent, Claim 23)
Context and Importance: This term defines the nature of the information being processed. The infringement case against Google's Knowledge Graph depends on whether a database of interconnected facts about real-world entities qualifies as a database of "experience and knowledge" as contemplated by the inventor.
Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The patent states that a purpose includes "all related experience or knowledge such as information on advantages, an activity... a plan... a causal path... a result," which could be argued to cover a wide range of interconnected information types (’087 Patent, col. 4:5-10).
- Evidence for a Narrower Interpretation: The specification repeatedly discusses the storage of information derived from a "conversation" and using that context (’087 Patent, col. 13:11-20). This could support an interpretation that the "experience" must be conversational or learned through interaction, potentially distinguishing it from a pre-compiled database of general facts.
VI. Other Allegations
- Indirect Infringement: The complaint exclusively alleges "direct[] infringing" under 35 U.S.C. § 271(a) for each asserted patent and does not plead facts to support claims of induced or contributory infringement (Compl. ¶¶ 25, 34, 38, 42, 49, 53, 57).
- Willful Infringement: Willfulness is alleged for the '468 and '840 patents. The complaint bases this allegation on Google's alleged pre-suit knowledge, asserting that Google was made aware of the '468 patent at least as early as October 10, 2014, through a Notice of Allowability for its own patent application which cited the technology (Compl. ¶¶ 30, 32, 45, 47). This same event is alleged to have provided notice of the '840 patent, which issued from a publication also cited in that Notice of Allowability (Compl. ¶¶ 19-20, 45).
VII. Analyst’s Conclusion: Key Questions for the Case
This case appears to present a foundational challenge to a core technology of a major internet company, based on patents with very early priority dates in the field of natural language processing. The resolution will likely depend on the following key questions:
A core issue will be one of technical implementation: Do Google's modern Knowledge Graph and Google Now, which may be built on machine learning and other contemporary AI architectures, practice the specific, rule-based, and highly structured methods of processing language recited in the claims from the 1990s? The Plaintiff's success may depend on its ability to show that, despite advancements, the accused systems operate on the same fundamental principles.
A second central issue will be one of claim construction: Can foundational terms like "word sense number" and "experience and knowledge data base" be interpreted broadly enough to cover the data structures and knowledge repositories used in a modern semantic search engine, or will they be confined to the very detailed and specific embodiments described across the voluminous patent specifications?
Finally, if infringement is found, a key question will be willfulness: Does a patent examiner's citation of an applicant's foundational technology during the prosecution of the applicant's own later patent application constitute the kind of pre-suit notice required to support a finding of willful infringement?