US-12625888-B2 - Methods and system for determining, navigating, and presenting complex relationships in content

US12625888B2US 12625888 B2US12625888 B2US 12625888B2US-12625888-B2

Abstract

The present teaching relates to identifying related content and significance thereof with respect to some specified goal. Initial content is obtained based on a specified goal. Associated content is identified based on the semantics of the initial content and semantic-based temporal information. Different temporal relations are determined based on the specified goal and are extracted from content comprising the initial and associated content. With respect to each extracted temporal relation, determine a semantic analysis to be performed based on the specified goal. A summary is automatically generated for each of the semantic analysis results and is visualized to reveal semantic and temporal relations among different related content pieces. The visualization can be adjusted according to user input.

Inventors

Wen Ruan
Joseph Anthony Manico
William Yurich Fowlkes
Young No

Assignees

IP.COM I, LLC

Dates

Publication Date: 20260512
Application Date: 20241011

Claims (20)

1 . A method comprising: receiving, from a user, an input specifying one of multiple different goals; searching initial content with respect to the specified goal; performing semantics analysis on the initial content to determine semantics with respect to the specified goal; obtaining associated content relating to the initial content, wherein the associated content includes additional semantically related content identified based on the semantics of the content, and semantic-based temporal information; analyzing the initial content and the associated content to identify one or more temporal relations between different pieces of content with the initial content and the associated content; with respect to each of the one or more temporal relations, determining one of multiple different types of semantic analysis with respect to the specified goal based on the temporal relation, wherein the different types of semantic analysis are provided for achieving multiple different goals and are applied to analyze respective portions of the initial content determined with respect to each of the multiple different goals, carrying out the type of semantic analysis determined based on the temporal relation; generating a summary for each of the semantic analyses based on one or more metrics computed therefor; visualizing the summary for each of the semantic analyses to visually present to the user the content in at least some of the groups in terms of the specified goal and linked based on the one or more temporal relations; and dynamically adjusting the visualization based on an input received from the user.
2 . The method of claim 1 , wherein the specified goal includes at least one of: a first goal relates to obtaining knowledge regarding intellectual properties (IP) associated with some specified target and relationship among different pieces of the IP; a second goal relates to gathering publicly available information associated with an academic field; and a third goal relates to revealing survey information associated with questionaries directed to a study.
3 . The method of claim 2 , wherein the specified target includes one or more of a business entity; an individual; a product; a technological field; and a textual description.
4 . The method of claim 1 , wherein the associated content includes: temporally related content linked to the initial content based on some temporal relationship defined according to a first criterion; and semantically related content linked to the initial content based on some semantic relationship defined according to a second criterion.
5 . The method of claim 1 , wherein the content includes one or more groups of content, which includes at least one of: one or more semantic groups, each of which comprises multiple pieces of the content that are semantically similar according to some pre-defined semantic-relevant criterion; one or more groups of temporally relevant content, each of which comprises multiple pieces of the content that are temporally associated according to some pre-defined temporal relationship; and one or more temporally associated semantically relevant groups, each of which comprises multiple pieces of the content that are semantically similar and temporally associated.
6 . The method of claim 4 , wherein the temporal relationship includes at least one of: an event-based temporal relationship; and a reference-based temporal relationship.
7 . The method of claim 6 , wherein the event-based temporal relationship is defined based on a temporal sequence of events of at least one pre-defined type occurred over time.
8 . The method of claim 7 , wherein the at least one pre-defined type of event includes: submission of a document with a modified portion as compared with an original version of the document; and submission of a document already submitted to a first source to a different source.
9 . The method of claim 6 , wherein the reference-based temporal relationship is defined based on an action carried out over time.
10 . The method of claim 9 , wherein the action includes at least one of a citation and a reference to another document.
11 . The method of claim 1 , wherein the summary for each of the semantic analyses comprises: one or more statistics characterizing group content included in the group; and a representation of connections among the multiple pieces of content included in the group.
12 . A machine readable and non-transitory medium having information recorded thereon, wherein the information, when read by the machine, causes the machine to perform the following steps: receiving, from a user, an input specifying one of multiple different goals; searching initial content with respect to the specified goal; performing semantics analysis on the initial content to determine semantics with respect to the specified goal; obtaining associated content relating to the initial content, wherein the associated content includes additional semantically related content identified based on the semantics of the content, and semantic-based temporal information; analyzing the initial content and the associated content to identify one or more temporal relations between different pieces of content with the initial content and the associated content; with respect to each of the one or more temporal relations, determining one of multiple different types of semantic analysis with respect to the specified goal based on the temporal relation, wherein the different types of semantic analysis are provided for achieving multiple different goals and are applied to analyze respective portions of the initial content determined with respect to each of the multiple different goals, carrying out the type of semantic analysis determined based on the temporal relation; generating a summary for each of the semantic analyses based on one or more metrics computed therefor; visualizing the summary for each of the semantic analyses to visually present to the user the content in at least some of the groups in terms of the specified goal and linked based on the one or more temporal relations; and dynamically adjusting the visualization based on an input received from the user.
13 . The medium of claim 12 , wherein the specified goal includes at least one of: a first goal relates to obtaining knowledge regarding intellectual properties (IP) associated with some specified target and relationship among different pieces of the IP; a second goal relates to gathering publicly available information associated with an academic field; and a third goal relates to revealing survey information associated with questionaries directed to a study.
14 . The medium of claim 13 , wherein the specified target includes one or more of a business entity; an individual; a product; a technological field; and a textual description.
15 . The medium of claim 12 , wherein the associated content includes: temporally related content linked to the initial content based on some temporal relationship defined according to a first criterion; and semantically related content linked to the initial content based on some semantic relationship defined according to a second criterion.
16 . The medium of claim 12 , wherein the content includes one or more groups of content, which includes at least one of: one or more semantic groups, each of which comprises multiple pieces of the content that are semantically similar according to some pre-defined semantic-relevant criterion; one or more groups of temporally relevant content, each of which comprises multiple pieces of the content that are temporally associated according to some pre-defined temporal relationship; and one or more temporally associated semantically relevant groups, each of which comprises multiple pieces of the content that are semantically similar and temporally associated.
17 . The method of claim 15 , wherein the temporal relationship includes at least one of: an event-based temporal relationship; and a reference-based temporal relationship.
18 . The medium of claim 17 , wherein the event-based temporal relationship is defined based on a temporal sequence of events of at least one pre-defined type occurred over time.
19 . The medium of claim 18 , wherein the at least one pre-defined type of event includes: submission of a document with a modified portion as compared with an original version of the document; and submission of a document already submitted to a first source to a different source.
20 . The medium of claim 17 , wherein the reference-based temporal relationship is defined based on an action carried out over time.

Description

CROSS REFERENCE TO RELATED APPLICATION The present application claims the benefit and priority of U.S. Provisional Patent Application No. 63/589,809, filed on Oct. 12, 2023, entitled “METHODS AND SYSTEM FOR DETERMINING, NAVIGATING, AND PRESENTING COMPLEX RELATIONSHIPS IN CONTENT”, the contents of which are hereby incorporated by reference in its entirety. BACKGROUND 1. Technical Field The present teaching generally relates to data processing. More specifically, the present teaching relates to artificial intelligence (AI) based data processing and applications thereof. 2. Technical Background Identifying document(s) of interest from a large collection of documents can be a laborious, subjective, and potentially inaccurate process. Even when the data set is reduced or narrowed by various search strategies to a smaller data set, it is still tedious to read through the documents and understand the content of each document and relationships between each other with subjectivity and inaccuracy. When it comes to analyzing a technology field or a large company's portfolio, it is inevitable to deal with a fairly large number of documents that will be difficult to review them manually. Identifying the most critical documents in a data set and determining what they reveal is a problem that needs to be solved for many applications, which include but are not limited to competitive intelligence analytics, due diligence for business transactions such as a merger and acquisition (M&A) or in capital investment, new product development, and patent related operations such as prior art search, patentability study, patent prosecution/maintenance, freedom to operate study, invalidity/infringement analysis, etc. This situation may be exacerbated in patent related applications due to large amount of related collections, cross-referenced patent families that may include hundreds of individual family members, related patent applications such as co-filed or cross-referenced applications, continuations, divisionals, continuations-in-part, and patents/patent applications linked via terminal disclaimers or regulatory extensions. Individual patents/patent applications and those within large patent families may also have hundreds of forward, backward, or horizontal/cross citations with varying degrees of relevance to each other. Some documents may be related in some fashion, such as via some temporal relations, procedural relations, semantic relations. Citations may be another type of relation that may be used to link different documents. Citations exist in many types of documents, e.g., scientific literature, technical and product documents, webpages, patent documents, etc., and are used to establish some kind of relevance beyond temporal relationships between documents, assuming that only related works are referenced. Conventionally, empirically derived citation network diagrams may capture temporal relationships between a document of interest and its citations in both directions, but such diagrams do not recognize the level of relevance that may exhibit in reality. For example, an inventor may cite numerous patents of dubious relevance that a patent practitioner must review manually to identify the most relevant cited patents. On the other hand, not all relevant or related documents are cited by a document due to various limits. A highly relevant patent may not be cited by an inventor or by an examiner which could pose a potential challenge as to the validity of the patent. Besides relevance, a conventional citation analysis, e.g., the main path or key route analysis which solely focuses on various link traversal frequencies, may also overlook the quality of the document itself. Individual patent documents within a patent family may have an original parent patent with hundreds of pages of specification and numerous divisionals, continuations, and continuation-in-part (CIP) with very similar if not identical specifications. The children of the parent patent application may have claim sets covering different protection scopes directed to various alternative embodiments and specialized features derived from the text and illustrations of the parent specification. With conventional methods, individual claim sets and possibly their supporting specifications need to be reviewed to determine the relevance to a patent of interest, a patent portfolio, a product, a service, or some inventive concepts, which again can be a laborious task. BRIEF DESCRIPTION OF THE DRAWINGS The methods, systems and/or programming described herein are further described in terms of exemplary embodiments. These exemplary embodiments are described in detail with reference to the drawings. These embodiments are non-limiting exemplary embodiments, in which like reference numerals represent similar structures throughout the several views of the drawings, and wherein: FIG. 1A illustrates contrasting situations between having to deal with a pile of unorganized documents as op