1 Building Hierarchical Classifiers Using Class Proximity Ke Wang Senqiang Zhou Shiang Chen Liew National University of Singapore.

1

Building Hierarchical Classifiers Using Class Proximity

Ke Wang

Senqiang Zhou

Shiang Chen Liew

National University of Singapore

VLDB99, Sept 6-10, Edinburgh

2

Hierarchical classification• Given

– a class hierarchy– a collection of pre-classified documents – a document is a set of terms

• Build– a classifier that assigns a relevant class to a new

document

• Key– extract features of classes


3

Yahoo classes

Yahoo

recreation science

automotivesports

skatingcycling


4

ACM classes

Hardware

General Memory_structure

General Design_style

Cache_memories

Level 1

Level 2

Level 3

Level 4


5

Existing local approaches

• build one classifier at each split of the

class hierarchy

• determine features locally at each node

• classify a document by going through a

path of classifiers starting from the root


6

Diminishing of high level structure

• rely on classification at high levels

• but high level structures usually weak, i.e., divergence of topics

• e.g., “car” is a feature at Recreation: Automotive, but not at Recreation


7

Bias of misclassification

• sibling classes Vs. nephew classes

• misclassification at high levels Vs. at low levels

• specialisation Vs. generalisation


8

Features should be

• determined wrt the target class

• determined at all concept levels

• correlated

The solution: generalised association rules (SA95, HF95)

{sql, IO} DB

{language, performance} CS


9

Our approach

• class proximity

• global classifier

• term hierarchy

• use the “best” generalised association rule

T C to determine the class


10

Rank association rules

• Biased confidence

• Biased J-measure


11

An example

author story

writer editor fiction poem

Music Literature

A_Music A_Literature

Arts

Term hierarchy

Class hierarchy

... ...


12

Term hierarchy(T)=YesClass proximity(B)=Yes

• R0: author,storyLiterature (ConfB=1,Clist=d6,d7)

• R1: authorLiterature (ConfB=1)

• R2: storyLiterature (ConfB=0.67, Wlist=d5(1))

• R4: hallMusic (ConfB=0.4, Clist=d1,d2, Wlist=d3(1))

• R3: StatesA_Literature (ConfB=0.33, Clist=d4,d5)


13


14

Experiment I

• http://www.acm.org/dl/toc.html/

• 26,515 papers, 78 classes, 14,754 terms

• class hierarchy=Level-1 and level-2 categories

• term hierarchy=Level-3 and level-4 categories

• document=Title and level-4 categories


15

Best rules found by (B,T)• CSO:

– vector,stream,processor,parallelProcessor_Architectures– multiple_instruction_streamProcessor_Architectures– data_flow,architecturProcessor_Architectures– internet, architecturComputer_Communication_Networks – mode,atmComputer_Communication_Networks – network,circuit_switching Computer_Communication_Networks – tecniqu, model, attributPerformance_of_Systems

• Software:– program,function, applicationProgramming_Techniques– object_oriented_programmingProgramming_Techniques– reusable_softwareSoftware_Engineering– software,methodologieSoftware_Engineering– organization, distributed_systemOperating_Systems


16

() --- | (T) --- (B) --- (B,T) --- (CDAR97,T) --- (CDAR97) ---


17


18

Experiment II

• http://dir.yahoo.com/recreation/sports

• 7,550 documents

• 367 classes, 7 levels

• 10,747 terms

• 90% of the terms occur in no more than 10 documents and many documents contain only such terms


19

Best rules found by (B,T)• Sports:Cycling:

– page,mountain Mountain_Biking– product,bikeMountain_Biking– mtb,mountain Mountain_Biking– held,bicyclRaces– classic,bicyclRaces– trip,tourTravelogues– trip,canada Travelogues– bicycl,alaskaTravelogues

• Sports:Auto_Racing:– team,result,driverFormula_one– model,featurTracks_and_Speedways– ovalTracks_and_Speedways– racewayTracks_and_Speedways


20


21

1 Building Hierarchical Classifiers Using Class Proximity Ke Wang Senqiang Zhou Shiang Chen Liew National University of Singapore.

Documents

class slide

recreation slide

cs slide

generalisation slide

root slide

d5 slide

terms class hierarchy

memories level