Analyzing and exploring images posted on social media

Patent metadata

Field	Value
Publication number	US-2017193333-A1
Application number	US-201514984542-A
Country	US
Kind code	A1
Filing date	Dec 30, 2015
Priority date	Dec 30, 2015
Publication date	Jul 6, 2017
Grant date	—

Abstract

Embodiments focus on the real-time processing and analysis of event-related images posts on social media. By computing the most posted images and grouping them in accordance to their similarity, these images can be organized in a visual interface to provide the users an image-driven social media topic detection system. The groups of images with the highest frequency represent the most prominent topics. By keeping track of the temporal occurrence of the images, it is possible to visualize the temporal evolution of the topics.

First claim

What is claimed is:

1 . A computer implemented method of analyzing image data, the method comprising:
receiving, using a processor system, image data of one or more images that have been posted to internet websites; analyzing, using the processor system, the image data to extract image feature data for each one of the one or more images; analyzing, using the processor system, the image data to extract metadata of the one or more images; creating multiple image files; wherein each one of the multiple image files comprises individual ones of the one or more images linked together based on overlaps between the image feature data of the individual ones of the one or more images; wherein each one of the multiple image files further comprises individual ones of the one or more images linked together based on overlaps between the metadata of the individual ones of the one or more images; indexing the multiple image files to form multiple indexed image files; wherein the indexing is based on the image feature data of the individual ones of the one or more images in an individual indexed image file; wherein the indexing is further based on the metadata of the individual ones of the one or more images in the individual indexed image file; and storing the multiple indexed image files in a memory of the processor system having a searchable indexed data storage structure.

2 . The computer implemented method of claim 1 further comprising;
applying the multiple indexed image files to a clustering algorithm to generate clustered results; and
displaying the clustered results on a display of the processor system.

3 . The computer implemented method of claim 1 , wherein:
the metadata comprises multiple discrete time frames; and the computer implemented method further comprises: searching the searchable indexed data storage structure of the memory to access the multiple indexed image files based on a predetermined subject matter and the multiple discrete time frames; and displaying results of the searching on a display of the processor system.

4 . The computer implemented method of claim 3 further comprising:
accessing sentiment data based on textual sentiment posted on internet web sites;
wherein the textual sentiment corresponds to the predetermined subject matter and the multiple discrete time frames; and
displaying the sentiment data with the results of the searching.

5 . The computer implemented method of claim 1 , wherein:
the processor system includes a machine learning processor; the analyzing of the image data to extract image features data for each one of the one or more images is performed using the machine learning processor; and the analyzing of the image data to extract metadata of the one or more images is performed using the machine learning processor.

6 . The computer implemented method of claim 5 wherein the image features data comprises image feature vectors.

7 . The computer implemented method of claim 5 wherein the metadata comprises metadata vectors.

8 . A computer system for analyzing image data, the system comprising:
a memory; and a processor system communicatively coupled to the memory; the processor system configured to perform a method comprising: receiving image data of one or more images that have been posted to internet websites; analyzing the image data to extract image feature data for each one of the one or more images; analyzing the image data to extract metadata of the one or more images; creating multiple image files; wherein each one of the multiple image files comprises individual ones of the one or more images linked together based on overlaps between the image feature data of the individual ones of the one or more images; wherein each one of the multiple image files further comprises individual ones of the one or more images linked together based on overlaps between the metadata of the individual ones of the one or more images; indexing the multiple image files to form multiple indexed image files; wherein the indexing is based on the image feature data of the individual ones of the one or more images in an individual indexed image file; wherein the indexing is further based on the metadata of the individual ones of the one or more images in the individual indexed image file; and storing the multiple indexed image files in the memory, wherein the memory includes a searchable indexed data storage structure.

9 . The computer system of claim 8 further comprising;
applying the multiple indexed image files to a clustering algorithm to generate clustered results; and
displaying the clustered results on a display of the processor system.

10 . The computer system of claim 8 , wherein:
the metadata comprises multiple discrete time frames; and the method performed by the processor system further comprises: searching the searchable indexed data storage structure of the memory to access the multiple indexed image files based on a predetermined subject matter and the multiple discrete time frames; and displaying results of the searching on a display of the processor system.

11 . The computer system of claim 10 further comprising:
accessing sentiment data based on textual sentiment posted on internet web sites;
wherein the textual sentiment corresponds to the predetermined subject matter and the multiple discrete time frames; and
displaying the sentiment data with the results of the searching.

12 . The computer system of claim 8 , wherein:
the processor system includes a machine learning processor; the analyzing of the image data to extract image features data for each one of the one or more images is performed using the machine learning processor; the analyzing of the image data to extract metadata of the one or more images is performed using the machine learning processor.

13 . The computer system of claim 12 wherein the image features data comprises image feature vectors.

14 . The computer system of claim 12 wherein the metadata comprises metadata vectors.

15 . A computer program product for analyzing image data, the computer program product comprising:
a computer readable storage medium having program instructions embodied therewith, wherein the computer readable storage medium is not a transitory signal per se, the program instructions readable by a processor system to cause the processor system to perform a method comprising: receiving, using the processor system, image data of one or more images that have been posted to internet websites; analyzing, using the processor system, the image data to extract image features data for each one of the one or more images; analyzing, using the processor system, the image data to extract metadata of the one or more images; creating multiple image files; wherein each one of the multiple image files comprises individual ones of the one or more images linked together based on overlaps between the image feature data of the individual ones of the one or more images; wherein each one of the multiple image files further comprises individual ones of the one or more images linked together based on overlaps between the metadata of the individual ones of the one or more images; indexing the multiple image files to form multiple indexed image files; wherein the indexing is based on the image feature data of the individual ones of the one or more images in an individual indexed image file; wherein the indexing is further based on the metadata of the individual ones of the one or more images in the individual indexed image file; and storing the multip

Assignees

• IBM

Inventors

• Cavalin Paulo Rodrigo
• Gatti Maira Athanazio De Cerqueira
• Mendes Carlos
• Moyano Luis Gregorio
• Pinhanez Claudio Santos

Classifications

• G06V20/70 Primary

  Labelling scene content, e.g. deriving syntactic or semantic representations · CPC title

• G06F18/23213

  with fixed number of clusters, e.g. K-means clustering · CPC title

• G06F18/22

  Matching criteria, e.g. proximity measures · CPC title

• G06F18/2413

  based on distances to training or reference patterns · CPC title

• G06K9/627

  Physics · mapped topic