Utilizing online content to suggest item attribute importance

Patent metadata

Field	Value
Publication number	US-10528633-B2
Application number	US-201715412585-A
Country	US
Kind code	B2
Filing date	Jan 23, 2017
Priority date	Jan 23, 2017
Publication date	Jan 7, 2020
Grant date	Jan 7, 2020

Abstract

Utilizing social media and articles to suggest item attribute importance to improve site level searching for a user.

First claim

What is claimed is:

1. A method of utilizing online content to suggest attribute importance for an item to improve site level searching comprising the steps of:
the computer searching for and monitoring the online content for attributes of an item to determine attributes referenced for the item;
the computer monitoring the online content to determine how often attributes of the item are referenced;
the computer syntactically and semantically analyzing, the online content;
the computer receiving previously analyzed electronic documents with respect to a Web crawler associated with previous user searches;
the computer determining, based on said previously analyzed electronic documents, a contextual meaning of terms of said online content;
the computer analyzing a plurality of natural languages associated with said online content;
the computer identifying lexemes within strings of characters of the online content;
the computer converting, based on said lexemes, a sequence of said characters of the online content into a sequence of associated tokens;
the computer identifying syntactic relationships within a passage composed of a group of tokens of said of said sequence of associated tokens;
the computer categorizing sentiments associated with said group of tokens;
the computer identifying and annotating, within text passages of said online content, keywords preselected as indicators of sentiment polarity associated with said group of tokens;
the computer tagging the tokens based on speech categories of said online content;
in response to the syntactically and semantically analyzing, the analyzing, the converting, the identifying and annotating, and the tagging, the computer adjusting a presence of an attribute from a value pair attribute associated with the item based on the determination of how often attributes of the item are referenced; and
the computer storing an output of a natural language processor associated with said sequence of associated tokens, as an information corpus within data warehouses, information corpora, data models, and document repositories such that said information corpus stores and retrieves data as a copy of ingested and parsed references with respect to said attribute and a candidate set of structured attributes for said item, wherein said data stored within said information corpus is structured such that analytical requirements are addressed.

2. The method of claim 1 , wherein the online content searched for and monitored is selected from the group consisting of: news articles, expert testimony regarding the item, blogs, forums, and support documents.

3. The method of claim 1 , wherein natural language processing is used to search for and monitor the online content.

4. The method of claim 1 , further comprising the step of: the computer sends a notification to an administrator regarding removal of the attribute from the value pair attribute associated with the item.

5. The method of claim 1 , wherein the presence of the attribute is adjusted by removing the attributed from the value pair attribute when the attributes are referenced at number of times below a threshold.

6. The method of claim 1 , wherein the presence of the attribute is adjusted by adding the attributed from the value pair attribute when the attributes are referenced at number of times greater than a threshold.

7. A computer program product for utilizing online content to suggest attribute importance for an item to improve site level searching, a computer comprising at least one processor, one or more memories, one or more computer readable storage media, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by the computer to perform a method comprising:
searching for and monitoring, by the computer, the online content for attributes of an item to determine attributes referenced for the item;
monitoring, by the computer, the online content to determine how often attributes of the item are referenced;
syntactically and semantically analyzing, by the computer, the online content;
receiving, by the computer, previously analyzed electronic documents with respect to a Web crawler associated with previous user searches;
determining, by the computer based on said previously analyzed electronic documents, a contextual meaning of terms of said online content;
analyzing, by the computer, a plurality of natural languages associated with said online content;
identifying, by the computer, lexemes within strings of characters of the online content;
converting, by the computer based on said lexemes, a sequence of said characters of the online content into a sequence of associated tokens;
identifying, by the computer, syntactic relationships within a passage composed of a group of tokens of said of said sequence of associated tokens;
categorizing, by the computer, sentiments associated with said group of tokens;
identifying and annotating, by the computer within text passages of said online content, keywords preselected as indicators of sentiment polarity associated with said group of tokens;
tagging, by the computer, the tokens based on speech categories of said online content;
in response to the syntactically and semantically analyzing, the analyzing, the converting, the identifying and annotating, and the tagging, adjusting, by the computer, a presence of an attribute from a value pair attribute associated with the item based on the determination of how often attributes of the item are referenced; and
storing an output of a natural language processor associated with said sequence of associated tokens, as an information corpus within data warehouses, information corpora, data models, and document repositories such that said information corpus stores and retrieves data as a copy of ingested and parsed references with respect to said attribute and a candidate set of structured attributes for said item, wherein said data stored within said information corpus is structured such that analytical requirements are addressed.

8. The computer program product of claim 7 , wherein the online content searched for and monitored is selected from the group consisting of: news articles, expert testimony regarding the item, blogs, forums, and support documents.

9. The computer program product of claim 7 , wherein natural language processing is used to search for and monitor the online content.

10. The computer program product of claim 7 , further comprising program instructions sending, by the computer, a notification to an administrator regarding removal of the attribute from the value pair attribute associated with the item.

11. The computer program product of claim 7 , wherein the presence of the attribute is adjusted, by the computer, by removing the attributed from the value pair attribute when the attributes are referenced at number of times below a threshold.

12. The computer program product of claim 7 , wherein the presence of the attribute is adjusted, by the computer, by adding the attributed from the value pair attribute when the attributes are referenced at number of times greater than a threshold.

13. A computer system for utilizing online content to suggest attribute importance for an item to improve site level searching, the computer system comprising a computer comprising at least one processor, one or more memories, one or more computer readable storage media having program instructions executable by the computer to perform the program instructions comprising:
searching for and monitoring, by the computer, the online content for attributes of an item t

Assignees

• IBM

Inventors

• Altaf Faheem
• Deluca Lisa Seacat
• Srinivas Raghuram

Classifications

• G06F16/313

  Selection or weighting of terms for indexing · CPC title

• G06F16/9535 Primary

  Search customisation based on user profiles and personalisation · CPC title

• G06F16/9536 Primary

  Search customisation based on social or collaborative filtering · CPC title