US 10,592,541 B2Granted

Technologies for dynamic automated content discovery

AssigneeIntel Corporation

InventorsElliot Smith, Max Waterman, Plamena Manolova, Karolina Kret, Mikael Metthey, Alok Barsode

Life of this patent

Estimating the term…

Loading patent…

Abstract

Technologies for dynamic automated content discovery include a computing device that determines a contextual part of a document selected by a user and extracts one or more key terms from the contextual part of the document using an automated key phrase extraction algorithm. The computing device may perform a syntactic algorithm, named entity recognition, or the TextRank algorithm. The computing device may calculate a vagueness score for terms of the document by querying a semantic database and select the key terms based on the corresponding vagueness scores. The computing device performs a content search based on the key terms to generate one or more search results and presents the search results to the user. The computing device may associate each of the search results with the corresponding key term of the contextual part of the document, for example by visually highlighting the key term. Other embodiments are described and claimed.

Claims9

Dependent claims

A computing device for automatic content discovery, the computing device comprising:

a document context module to determine a contextual part of a document, wherein the contextual part of the document is currently accessible by a user of the computing device, and wherein to determine the contextual part of the document comprises to identify a part of the document visible in an application viewport of the computing device;
a key phrase extraction module to extract a key term from the contextual part of the document based on an importance value associated with the key term as determined by an automated key phrase extraction algorithm in response to a determination of the contextual part of the document, wherein the document comprises a document selected by the user of the computing device;
a vagueness ranking module configured to:
(i) calculate a vagueness score for each term of a plurality of terms of the document by a query of a semantic database, wherein to calculate the vagueness score comprises to calculate an average of a plurality of vagueness measures, wherein each vagueness measure is produced by an associated query of the semantic database,
(ii) determine a ranking score for each term as a function of the corresponding importance value and the corresponding vagueness score,
(iii) rank each term based on the corresponding ranking score,
(iv) re-rank the plurality of terms based on the corresponding vagueness score of each term to generate an adjusted ranked list of terms, and
(v) select the key term from the adjusted ranked list of terms based on a rank of each term;
a content search module to perform an automated content search based on the key term to generate one or more search results;
and
a user interface module to present the one or more search results to the user.

8 more elements

One or more computer-readable storage media comprising a plurality of instructions that in response to being executed cause a computing device to:

determine a contextual part of a document, wherein the contextual part of the document is currently accessible by a user of the computing device, and wherein to determine the contextual part of the document comprises to identify a part of the document visible in an application viewport of the computing device;
extract a key term from the contextual part of the document based on an importance value associated with the key term as determined by an automated key phrase extraction algorithm in response to determining the contextual part of the document, wherein the document comprises a document selected by the user of the computing device;
establish the computing device configured to:
(i) calculate a vagueness score for each term of a plurality of terms of the document by a query of a semantic database, wherein to calculate the vagueness score comprises to calculate an average of a plurality of vagueness measures, wherein each vagueness measure is produced by an associated query of the semantic database;
(ii) determine a ranking score for each term as a function of the corresponding importance value and the corresponding vagueness score;
(iii) rank each term based on the corresponding ranking score;
(iv) re-rank the plurality of terms based on the corresponding vagueness score of each term to generate an adjusted ranked list of terms;
and
(v) select the key term from the adjusted ranked list of terms based on ranking each term;
perform an automated content search based on the key term to generate one or more search results;
and
present the one or more search results to the user.

9 more elements

Dependencies are inferred from the claim text.