A k nearest neighbor based multi instance multi label learning algorithm
This presentation is the property of its rightful owner.
Sponsored Links
1 / 17

A k-Nearest Neighbor Based Multi-Instance Multi-Label Learning Algorithm PowerPoint PPT Presentation

  • Uploaded on
  • Presentation posted in: General

A k-Nearest Neighbor Based Multi-Instance Multi-Label Learning Algorithm. kobe chai [email protected] כריית מידע,המכללה האקדמית להנדסה ירושלים. Mountains. Multi-label learning. Trees. Lake. ? What is Multi-Label Objects. example: natural scene image.

Download Presentation

A k-Nearest Neighbor Based Multi-Instance Multi-Label Learning Algorithm

An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.

- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -

Presentation Transcript

A k-Nearest Neighbor Based Multi-InstanceMulti-Label Learning Algorithm

kobe chai

[email protected]



Multi-label learning



?What is Multi-Label Objects

example: natural scene image

(a) Traditional supervised learning

(b) Multi-instance learning

(c) Multi-label learning

(d) Multi-instance multi-label learning

MIML explanations

In multi-instance multi-label learning (i.e. MIML),

each example is not only represented by multiple instances but

also associated with multiple labels

Using example - MIML

Using example - MIML

person name

movie character name

movie name

game name

book name





Friend network


Search behavior


To Address the Ambiguity

Multiple labels

Sport? Fans? Football? Training?

The Problems In MIML

Identification process may lose useful information encoded in

training examples and therefore be harmful to the learning algorithms performance.

information loss during degeneration process!

An instance of a car the MIML algorithm cant recognize this specific car

Because of the change below.

kNN-MIML will consider the citer (blue door) and will learn the new car.


MIML-kNN is proposed for MIML by utilizing the popular k nearest neighbor techniques.

Given a test example, MIML-kNN not only considers its neighbors, but also considers its citers which regard it as their own neighbors.


Equation #1:

Equation #2:


Equation #5:


Process Diagram

Test Point


Data (D)






How it works?

Learns one binary classifier for each label

Outputs the union of their predictions

Can do ranking if classifier outputs scores

LimitationDoes not consider label relationships

+1 or -1?

Experiment Summery

The performance of MIML-kNN is compared with MIMLBOOST

and MIMLSVM on two real-world MIML tasks.

The scene classification data contains 2,000 natural scene images.

All the possible class labels are desert, mountains, sea, sunset and trees.

average number of labels per image is 1.240.44.


  • Label Set #Images | Label Set #Images | Label Set #Images----------------------------------------------------------------------------------------------------- desert 340 | desert+sunset 21 | sunset+trees 28 mountains 268 | desert+trees 20 | desert+mountains+sunset 1 sea 341 | mountains+sea 38 | desert+sunset+trees 3 sunset 216 | mountains+sunset 19 | mountains+sea+trees 6 trees 378 | mountains+trees 106 | mountains+sunset+trees 1desert+mountains 19 | sea+sunset 172 | sea+sunset+trees 4desert+sea 5 | sea+trees 14 | Total 2,000-----------------------------------------------------------------------------------------------------------------

  • Login