Graph-based framework for multi-task multi-view learning

Patent metadata

Field	Value
Publication number	US-8990128-B2
Application number	US-201213488885-A
Country	US
Kind code	B2
Filing date	Jun 5, 2012
Priority date	Jun 5, 2012
Publication date	Mar 24, 2015
Grant date	Mar 24, 2015

Abstract

A system and method a Multi-Task Multi-View (M 2 TV) learning problem. The method uses the label information from related tasks to make up for the lack of labeled data in a single task. The method further uses the consistency among different views to improve the performance. It is tailored for the above complicated dual heterogeneous problems where multiple related tasks have both shared and task-specific views (features), since it makes full use of the available information.

First claim

What is claimed is:

1. A method for classifying entities from multiple channels in multi-task multi-view learning problems, wherein entities of different tasks are related with each other through shared or common features in multiple views, and a single learning task relating to a task specific feature in multiple views said method comprising:
generating a bi-partite graph-based model relating one or more entities and features in each said view;
forming an objective function to…

Assignees

• He Jingrui
• Gondek David C
• Lawrence Richard D
• Vijil Enara C
• IBM

Inventors

• He Jingrui
• Gondek David C
• Lawrence Richard D
• Vijil Enara C

Classifications

• G06F18/2431 Primary

  Physics · mapped topic

• G06K9/628 Primary

  Physics · mapped topic