View source for Lin and Wu. 2009. Phrase Clustering for Discriminative Learning.

{{MyCiteconference | booktitle = Proceedings of the Annual Meeting of the Association for Computational Linguistics and the International Joint Conference on Natural Language Processing of the Asian Federation of Natural Language Processing| coauthors = X. Wu| date = 2009| first = D.| last = Lin| title = Phrase clustering for discriminative learning| url = http://www.aclweb.org/anthology/P/P09/P09-1116.pdf }}

This [[Category::Paper]] is available online [http://www.aclweb.org/anthology/P/P09/P09-1116.pdf].

== Summary ==

This paper makes use of phrase [[UsesMethod::clustering]] to improve on the state of the art for the [[AddressesProblem::Named Entity Recognition]] problem. They obtained 1.0 F-score improvement over NER systems on the CoNLL benchmark (in 2009). In their paper, they used phrases that occur more than 100 times in a 700 billion token web corpus ([[RelatedPaper::Lin et al., 2008]]).

This paper leveraged on large amount of unlabeled data to induce phrase [[clustering]], which provided an advantage over word [[clustering]] features used in current NER systems.

== Brief description of the method ==

Due to the large number of possible phrases, the authors used [[UsesMethod::Bloom filters]] to decide whether a sequence of tokens is considered a phrase.

=== Phrases as feature vectors ===

Each phrase is represented as a vector of its context words. The frequency count of words appearing within a fixed-sized window is aggregated and converted into [[UsesMethod::pointwise mutual information]](PMI) values.

=== Parallel K-Means using MapReduce ===

The phrase vectors are then clustered using [[UsesMethod::K-means]] [[clustering]] algorithm. The authors chose the [[K-means]] algorithm over other more advanced [[clustering]] algorithms because it is fast and can be very easily parallelized. Given the amount of data they have, this is a very important consideration. It took about 20 minutes on a 1000 machine cluster to perform K-means to convergence on their dataset. The "similarity" between two phrases is simply the Euclidean norm of the phrase vectors in the feature space. What this means is that phrases are considered to be more similar if they share more context words.

=== Soft clustering ===

In addition to just performing hard [[clustering]] (where each phrase can only belong to one cluster), they perform soft [[clustering]] as well by assigning phrases to cluster centroids that are within a threshold distance. The motivation behind doing soft [[clustering]] is that it may be better able to model the fact that phrases may contain several "senses".

== Experimental Result ==

The effectiveness of phrase [[clustering]] is evaluated on [[AddressesProblem::Named Entity Recognition]]. For [[NER]], they used a 1-word context window and hard [[clustering]], and a linear chain [[UsesMethod::CRF]] with standard features used in current [[NER]] systems. The baseline features contains a total of 48 feature templates.

The results on [[UsesDataset::CoNLL'03]] test set are as follows:

[[Image:conll_results.png]]

The authors evaluated their phrase clustering system on the [[UsesDataset::KDDCup 2005]] [[AddressesProblem::query classification]] competition. The task is to categorize 800k internet user search queries into 67 topical categories. They treated the problem as 67 separate binary classification task and trained [[UsesMethod::logistic regression]] classifiers with <math>L_2</math> regularization. Their system were on par with the winning KDDCup 2005 system.

[[Image:kddcup_results.png]]

== Related Papers ==

[[RelatedPaper::Lafferty_2001_Conditional_Random_Fields | Lafferty et al (2001)]] introduced [[CRF]] in their paper.

[[RelatedPaper::Brown et al, CL 1992]] introduced the [[UsesMethod::Brown clustering]] algorithm which is frequently used in NER for clustering words. It is based on the idea that one should minimize the bigram language model perplexity of a text corpus.

[[RelatedPaper::Miller et al, ACL 2004]] used a word clustering algorithm and discriminative training approach to the NER problem.