# US 2017/0286554 A1 · Detecting relevant facets by leveraging diagram identification, social media and statistical analysis software

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2017/0286554 A1
- Document ID: US20170286554A1
- Status: Published application
- Kind: A1
- Application number: 15/626,211
- Filed: 2017-06-19
- Published: 2017-10-05
- Earliest claimed: 2015-11-04
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: Xiao Xian Bei; Zhili Guo; Peng Hui Jiang; Fang Ming Li; Gang Zhan
- Web page: https://patentgrep.com/patent/US20170286554A1

## CPC

- G06F 16/9536: Search customisation based on social or collaborative filtering
- G06F 16/24575: {using context}
- G06F 16/248: Presentation of query results
- G06F 16/29: Geographical information databases
- G06F 16/3331: Query processing
- G06F 16/335: Filtering based on additional data, e.g. user or group profiles (filtering in web context G06F16/9535, G06F16/9536)
- G06F 16/338: Presentation of query results
- G06F 16/9535: Search customisation based on user profiles and personalisation
- G06F 16/583: using metadata automatically derived from the content
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

An approach to finding facets of big data across multiple subject fields which have mutual relevance and provide useful information and insights. The approach is based on an analysis of diagrams associated with facets found in big data, such as that available on the internet, and the results of the analysis are further refined through a regional analysis and a search of social media data for events pertaining to the facets. Embodiments can further make use of statistical analysis software to provide statistical evidence related to the facets that can validate the results and findings of facet relevancy.

## Claims

### Claim 1

A method for determining relevant facets of big data, the method comprising:
searching a first one or more data sources for one or more facets, wherein the one or more facets comprise at least one of facet data and facet diagrams;
creating one or more facet diagrams for the at least one or more facet data;
comparing the one or more facet diagrams to a base facet diagram to generate a first group of relevant facet diagrams based on a diagrammatic analysis;
filtering the first group of relevant facet diagrams based on at least one of a regional analysis and a search of social media to generate a second group of relevant facet diagrams; and
outputting the second group of relevant facet diagrams.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
presenting a list of one or more second data sources, which provide data associated with the second group of relevant facet diagrams, for statistical analysis via statistical analysis software.

### Claim 3 (depends on 1)

The method of claim 1, wherein the one or more facets comprise at least one of structured and unstructured data.

### Claim 4 (depends on 1)

The method of claim 1, wherein the searching the first one or more data sources is based on a preconfigured search criteria.

### Claim 5 (depends on 1)

The method of claim 1, wherein the diagrammatic analysis comprises comparing trends of the one or more facet diagrams across a plurality of corresponding time segments for a predetermined threshold.

### Claim 6 (depends on 1)

The method of claim 1, wherein the regional analysis comprises retrieving similar facets, associated with the first group of relevant facet diagrams, from one or more geographical regions other than those associated with the first group of relevant facet diagrams, comparing the similar facets based on the diagrammatic analysis and filtering the first group of relevant facet diagrams based on the comparison of the similar facets, according to a predetermined criteria.

### Claim 7 (depends on 1)

The method of claim 1, wherein the filtering of the first group of relevant facet diagrams based on the search of social media comprises a retrieval of one or more social media events, associated with the first group of relevant facet diagrams, within a predetermined threshold.