Title: Collective%20Classification%20%20A%20brief%20overview%20and%20possible%20connections%20to%20email-acts%20classification
1Collective Classification A brief overview and
possible connections to email-acts classification
- Vitor R. Carvalho
- Text Learning Group Meetings,
- Carnegie Mellon University
- November 10th 2004
2Data Representation
spam
Not spam
- Flat Data
- Object email msgs
- Attributes words, sender, etc
- Class spam/not spam
- Usually assumed IID
- Sequential Data
- Object words in text
- Attr capitalized, number, dict
- Class POS (or name/not)
- Relational Data
- classattributes
- links(relations)
- Example webpages
spam
spam
Not spam
pron
name
det
name
verb
3J. Neville et al., 2003
4Relational Data and Collective Classification
- Different objects interact
- Different types of relations (links)
- Attributes may be correlated
- Examples
- actors, directors, movies, companies
- papers, authors, conferences, citations
- company, employee, customer,
Classify objects collectively
Use prediction on some objects to improve
prediction on related objects
5Collective Classification Methods
- Relational Probability Trees (RPT)
- Iterative methods (Relaxation-based Methods)
- Relational Dependency Networks (RDN)
- Relational Bayesian Networks (RBN/PRM)
- Relational Markov Networks (RMN)
- Other models (ILP based, Vector Space based,
etc) - Overall
- Lack of direct comparison among methods
- Results are usually compared to flat model
- Splitting data into train/test sets can be an
issue
6Relational Probability Trees
- Decision Trees applied to Relational data
- Predicts the target class label based on
- same object attributes
- attributes links in relational neighborhood
(one link away) - counts of attributes and links in the
neighborhood - Enhanced feature selection (Chi-square, pruning,
randomization tests) - Results were not exciting
- Neville et al. KDD2003, related work from
Blockeel et al. (Artificial Intelligence, 1998),
Kramer AAAI-96
7Iterative Methods
- Predicts the target class label based on
- Same object attributes
- Attributes and links of relational neighborhood
- CLASS LABEL of neighborhood
- Features derived from CLASS LABELS
- Different update strategies
- By threshold in prediction confidence
- By top-N most confident predictions
- Heuristic-based
- Slattery Mitchell, ICML-2000Neville Jensen,
AAAI-2000 Chakrabarti et al. ACM-SIGMOD-98 - Some results with Email-acts
8Relational Bayesian Networks (RBN/PRM)
- Bayes Net extended to Relational domain
- Given an instantiation, it induces a bayes-net
that specifies a joint probability distribution
over all attributes of all entities - Directed graphical model, with acyclicity
constraint. - Exact model - Closed form for parameter
estimation Products of conditional
probabilities - Was applied to simple domains, since the
acyclicity constraints is very restrictive to
most relational applications - Friedman et al, IJCAI-99 Getoor et al.,
ICML-2001 Taskar et al. IJCAI-2001
9Relational Markov Networks (RMN)
- Extension of CRF idea to Relational Domain
- Given an instantiation, it induces a Markov
Network that specifies a probability distribution
of labels, given links and attributes - Undirected, Discriminative model
- Parameter estimation is expensive, requires
approximate probabilistic inference (belief
propagation) - Taskar et al., UAI2002
10Relational Dependency Networks (RDN)
- Dependency Networks extended to Relational
domain - P(X) p Prob (Xi Neighbor(Xi))
- Given an instantiation, it induces a DN that
specifies an approximate joint probability
distribution over all attributes of all objects - Undirected graphical model, no acyclicity
constraint. - Approximate model - Simple parameter estimation
approximate inference (Gibbs sampling) - Neville Jensen, KDD-MRDM-2003
11Other Models
From Neville et al., 2003
12Comparing Some Results
PRM
- Comparing PRM, RMN, SVM and M3N
- Diff PRM and RMN
- Diff mSVM and RMN
- RN (Relational Neighbor) is a very simple
Relational Classifier - RN (Macskassy et al., 2003)
- M3N(Taskar et al., 2003)
RMN
13End of overviewnow, the email-act problem
- Strong correlation with previous and next message
Commit
Proposal
Request
Request
Request
Proposal
Delivery
Commit
Request
Acknowled
Delivery
- A verb has little or no correlation with other
verbs of same message
- Flat data?
- Sequential data?
Commit
Delivery
Time