Title: Sparse Word Graphs: A Scalable Algorithm for Capturing Word Correlations in Topic Models
1Sparse Word GraphsA Scalable Algorithm for
Capturing Word Correlations in Topic Models
- Ramesh NallapatiJoint work with
- John Lafferty, Amr Ahmed,
- William Cohen and Eric Xing
- Machine Learning Department
- Carnegie Mellon University
2Introduction
- Statistical topic modeling an attractive
framework for topic discovery - Completely unsupervised
- Models text very well
- Lower perplexity compared to unigram models
- Reveals meaningful semantic patterns
- Can help summarize and visualize document
collections - e.g. PLSA, LDA, DPM, DTM, CTM, PA
3Introduction
- A common assumption in all the variants
- Exchangeability bag of words assumption
- Topics represented as a ranked list of words
- Consequences
- Word Correlation information is lost
- e.g. white-house vs. white and house
- Long distance correlations
4Introduction
- Objective
- To capture correlations between words within
topics - Motivation
- More interpretable representation of topics as a
network of words rather than a list - Helps better visualize and summarize document
collections - May reveal unexpected relationships and patterns
within topics
5Past Work Topic Models
- Bigram topic models Wallach, ICML 2006
- Requires KV(K-1) parameters
- Only captures local dependencies
- Does not model sparsity of correlations
- Does not capture within-topic correlations
6Past work Other approaches
- Hyperspace Analog to Language (HAL)
- Lund and Burges, Cog. Sci., 96
- Word pair correlation measured as a weighted
count of number of times they occur within a
fixed length window - Weight of an occurrence / 1/(mutual distance)
7Past work Other approaches
- Hyperspace Analog to Language (HAL)
- Lund and Burges, Cog. Sci., 96
- Plusses
- Sparse solutions, scalability
- Minuses
- Only unearths global correlations, not semantic
correlations - E.g. river bank, bank check
- Only local dependencies
8Past work Other approaches
- Query expansion in IR
- Similar in spirit finds words that highly
co-occur with the query words - However, not a corpus visualization tool
requires a context to operate on - Wordnet
- Semantic networks
- Human labeled not directly related to our goal
9Our approach
- L1 norm regularization
- Known to enforce sparse solutions
- Sparsity permits scalability
- Convex optimization problem
- Globally optimal solutions
- Recent advances in learning structure of
graphical models - L1 regularization framework asymptotically leads
to true structure
10BackgroundLASSO
- Example linear regression
- Regularization used to improve generalizability
- E.g.1 Ridge regression L2 norm regularization
- E.g.2 Lasso L1 norm regularization
11Background LASSO
- Lasso encourages sparse solutions
12Background Gaussian Random Fields
- Multivariate Gaussian distribution
- Random field structure G (V,E)
- V set of all variables X1,?,Xp
- (s,t) 2 E , ?-1st ? 0
- Xs ? Xu XN(s) where u ? N(s)
13Background Gaussian Random Fields
- Estimating the graph structure of GRF from data
Meinshausen and Buhlmann, Annals. Stats., 2006 - Regress each variable onto others imposing L1
penalty to encourage sparsity - Estimated neighborhood
14Background Gaussian Random Fields
Estimated graph
True Graph
Courtesy Meinshausen and Buhlmann, Annals.
Stats., 2006
15Background Gaussian Random Fields
- Application to topic models CTM
- Blei and Lafferty, NIPS, 2006
16Background Gaussian Random Fields
- Application to CTMBlei Lafferty, Annals.
Appl. Stats., 07
17Structure learning of an MRF
- Ising model
- L1 regularized conditional likelihood learns true
structure asymptotically - Wainwright, Ravikumar and Lafferty, NIPS06
18Structure learning of an MRF
Courtesy Wainwright, Ravikumar and Lafferty,
NIPS06
19Sparse Word Graphs
- Algorithm
- Run LDA on the document collection and obtain
topic assignments - Convert topic assignments for each document into
K binary vectors X - Assume an MRF for each topic with X as underlying
data - Apply structure learning for MRF using
regularized conditional likelihood
20Sparse Word Graphs
21Sparse Word Graphs Scalability
- We still run V logistic regression problems, each
of size V for each topic O(KV2) ! - However, each example is very sparse
- L1 penalty results in sparse solutions
- Can run each topic in parallel
- Efficient interior point based L1 regularized
logistic regression Koh, Kim Boyd, JMLR,07
22Experiments
- Small AP corpus
- 2.2K Docs, 10.5K unique words
- Ran 10 topic LDA model
- Used ? 0.1 in L1 logistic regression
- Took just 45 min. per topic
- Very sparse solutions
- Computes only under 0.1 of the total number of
possible edges
23Topic Business neighborhood of top LDA terms
24Topic Business neighborhood of top edges
25Topic War neighborhood of top LDA terms
26Topic War neighborhood of top edges
27Concluding remarks
- Pros
- A highly scalable algorithm for capturing within
topic word correlations - Captures both short distance and long distance
correlations - Makes topics more interpretable
- Cons
- Not a complete probabilistic model
- Significant modeling challenge since the
correlations are latent
28Concluding remarks
- Applications of Sparse Word Graphs
- Better document summarization and visualization
tool - Word sense disambiguation
- Semantic query expansion
- Future Work
- Evaluation on a real task
- Build a unified statistical model