Identifying Intention Posts in Discussion Forums Meichun Hsu - - PowerPoint PPT Presentation

identifying intention posts in discussion forums
SMART_READER_LITE
LIVE PREVIEW

Identifying Intention Posts in Discussion Forums Meichun Hsu - - PowerPoint PPT Presentation

Identifying Intention Posts in Discussion Forums Meichun Hsu Zhiyuan (Brett) Chen Malu Castellanos Bing Liu Riddhiman Ghosh What is Intention? Example Hello, I am going to buy a new high-end gaming laptop my budget is below than 1500$ ram


slide-1
SLIDE 1

Identifying Intention Posts in Discussion Forums

Zhiyuan (Brett) Chen Bing Liu Meichun Hsu Malu Castellanos Riddhiman Ghosh

slide-2
SLIDE 2

What is Intention?

slide-3
SLIDE 3

Example

Hello, I am going to buy a new high-end gaming laptop my budget is below than 1500$ ram should be more than 6gb,graphics card must be more than 2gb and the processor should be intel core i7 3rd generations or better.

slide-4
SLIDE 4

Explicit Intention

I plan to buy this book. I am looking for a new car. I am going to travel to Atlanta.

slide-5
SLIDE 5

Implicit Intention

Anyone knows the battery life of iPhone? What are the components in this laptop?

slide-6
SLIDE 6

Identifying Intention

A totally NEW problem Explicit intention Applications like advertisement

slide-7
SLIDE 7

Problem Definition

slide-8
SLIDE 8

Two-Class Post Classification Transfer Learning: Use labeled

data from other domains (source domains) to classify target domain

slide-9
SLIDE 9
slide-10
SLIDE 10

The ways to express an intention are similar in different domains.

slide-11
SLIDE 11

Motivation Examples

I want to buy a car. I want to buy a camera. I want to buy the tickets.

slide-12
SLIDE 12

Special Difficulties

slide-13
SLIDE 13

Noise

I read many reviews of two Canon models which I'm considering for purchase, the Canon PowerShot S2 IS and the Canon PowerShot S3 IS. This is my second digital camera, my first being a Kodak EasyShare from about 4 years ago. …

slide-14
SLIDE 14

Imbalanced Shared Features

slide-15
SLIDE 15

EM Algorithm with NB

(Nigam et al., 2000)

E-step M-step Naïve Bayes

slide-16
SLIDE 16

Proposed Models

FS-EM Co-Class

slide-17
SLIDE 17

FS-EM

Incorporates feature selection into EM iterations Selects features from both labeled source (domain) data and unlabeled target (domain) data

slide-18
SLIDE 18
slide-19
SLIDE 19

Co-Class

Builds two classifiers based on FS-EM Solves the imbalanced shared feature problem

slide-20
SLIDE 20
slide-21
SLIDE 21

Features & Feature Selection

N-grams

slide-22
SLIDE 22

Evaluation

slide-23
SLIDE 23

Datasets

4 Domains from different forums (http://www.cs.uic.edu/~zchen/) Human annotation Cross-validation

slide-24
SLIDE 24

Evaluation Measures

slide-25
SLIDE 25

Supervised Learning (One Domain)

slide-26
SLIDE 26

Model Comparisons

3TR-1TE (Aue & Gamon, 2005) EM (Nigam, et al., 2000) ANB (Tan et al., 2009) FS-EM1 FS-EM2 Co-Class

slide-27
SLIDE 27

Model Comparisons

slide-28
SLIDE 28

Effects of Source Domains

slide-29
SLIDE 29

Conclusions

Novel problem of identifying intention Suitable for transfer learning Two special difficulties Effectiveness of Co-Class

slide-30
SLIDE 30

Future Directions

Sentence-level classification Extract intention components

slide-31
SLIDE 31

Q & A

slide-32
SLIDE 32

Dataset Download Link: http://www.cs.uic.edu/~zchen/