Daniel Klein

Accurate Unlexicalized Parsing

We demonstrate that an unlexicalized PCFG can parse much more accurately than previously shown, by making use of simple, linguistically motivated state splits, which break down false independence assumptions latent in a vanilla treebank grammar. Indeed, its performance of 86.36% (LP/LR F1) is better than that of early lexicalized PCFG models, and surprisingly close to the current state-of-the-art. This result has potential uses beyond establishing a strong lower bound on the maximum possible accuracy of unlexicalized models: an unlexicalized PCFG is much more compact, easier to replicate, and easier to interpret than more complex lexical models, and the parsing algorithms are simpler, more widely understood, of lower asymptotic complexity, and easier to optimize.

Feature-rich part-of-speech tagging with a cyclic dependency network

We present a new part-of-speech tagger that demonstrates the following ideas: (i) explicit use of both preceding and following tag contexts via a dependency network representation, (ii) broad use of lexical features, including jointly conditioning on multiple consecutive words, (iii) effective use of priors in conditional loglinear models, and (iv) fine-grained modeling of unknown word features. Using these ideas together, the resulting tagger gives a 97.24% accuracy on the Penn Treebank WSJ, an error reduction of 4.4% on the best previous single automatically learned tagging result.

Learning Accurate, Compact, and Interpretable Tree Annotation

We present an automatic approach to tree annotation in which basic nonterminal symbols are alternately split and merged to maximize the likelihood of a training treebank. Starting with a simple X-bar grammar, we learn a new grammar whose nonterminals are subsymbols of the original nonterminals. In contrast with previous work, we are able to split various terminals to different degrees, as appropriate to the actual complexity in the data. Our grammars automatically learn the kinds of linguistic distinctions exhibited in previous work on manual tree annotation. On the other hand, our grammars are much more compact and substantially more accurate than previous work on automatic annotation. Despite its simplicity, our best grammar achieves an F1 of 90.2% on the Penn Treebank, higher than fully lexicalized systems.

Alignment by Agreement

We present an unsupervised approach to symmetric word alignment in which two simple asymmetric models are trained jointly to maximize a combination of data likelihood and agreement between the models. Compared to the standard practice of intersecting predictions of independently-trained models, joint training provides a 32% reduction in AER. Moreover, a simple and efficient pair of HMM aligners provides a 29% reduction in AER over symmetrized IBM model 4 predictions.

Corpus-Based Induction of Syntactic Structure: Models of Dependency and Constituency

We present a generative model for the unsupervised learning of dependency structures. We also describe the multiplicative combination of this dependency model with a model of linear constituency. The product model outperforms both components on their respective evaluation metrics, giving the best published figures for unsupervised dependency parsing and unsupervised constituency parsing. We also demonstrate that the combined model works and is robust cross-linguistically, being able to exploit either attachment or distributional regularities that are salient in the data.

Learning Dependency-Based Compositional Semantics

Suppose we want to build a system that answers a natural language question by representing its semantics as a logical forxm and computing the answer given a structured database of facts. The core part of such a system is the semantic parser that maps questions to logical forms. Semantic parsers are typically trained from examples of questions annotated with their target logical forms, but this type of annotation is expensive.Our goal is to instead learn a semantic parser from question–answer pairs, where the logical form is modeled as a latent variable. We develop a new semantic formalism, dependency-based compositional semantics (DCS) and define a log-linear distribution over DCS logical forms. The model parameters are estimated using a simple procedure that alternates between beam search and numerical optimization. On two standard semantic parsing benchmarks, we show that our system obtains comparable accuracies to even state-of-the-art systems that do require annotated logical forms.

An End-to-End Discriminative Approach to Machine Translation

We present a perceptron-style discriminative approach to machine translation in which large feature sets can be exploited. Unlike discriminative reranking approaches, our system can take advantage of learned features in all stages of decoding. We first discuss several challenges to error-driven discriminative approaches. In particular, we explore different ways of updating parameters given a training example. We find that making frequent but smaller updates is preferable to making fewer but larger updates. Then, we discuss an array of features and show both how they quantitatively increase BLEU score and how they qualitatively interact on specific examples. One particular feature we investigate is a novel way to introduce learning into the initial phrase extraction process, which has previously been entirely heuristic.

Evaluating strategies for similarity search on the web

Finding pages on the Web that are similar to a query page (Related Pages) is an important component of modern search engines. A variety of strategies have been proposed for answering Related Pages queries, but comparative evaluation by user studies is expensive, especially when large strategy spaces must be searched (e.g., when tuning parameters). We present a technique for automatically evaluating strategies using Web hierarchies, such as Open Directory, in place of user feedback. We apply this evaluation methodology to a mix of document representation strategies, including the use of text, anchor-text, and links. We discuss the relative advantages and disadvantages of the various approaches examined. Finally, we describe how to efficiently construct a similarity index out of our chosen strategies, and provide sample results from our index.

A parsing: fast exact Viterbi parse selection

We present an extension of the classic A* search procedure to tabular PCFG parsing. The use of A* search can dramatically reduce the time required to find a best parse by conservatively estimating the probabilities of parse completions. We discuss various estimates and give efficient algorithms for computing them. On average-length Penn treebank sentences, our most detailed estimate reduces the total number of edges processed to less than 3% of that required by exhaustive parsing, and a simpler estimate, which requires less than a minute of pre-computation, reduces the work to less than 5%. Un-like best-first and finite-beam methods for achieving this kind of speed-up, an A* method is guaranteed to find the most likely parse, not just an approximation. Our parser, which is simpler to implement than an upward-propagating best-first parser, is correct for a wide range of parser control strategies and maintains worst-case cubic time.

Prototype-Driven Learning for Sequence Models

We investigate prototype-driven learning for primarily unsupervised sequence modeling. Prior knowledge is specified declaratively, by providing a few canonical examples of each target annotation label. This sparse prototype information is then propagated across a corpus using distributional similarity features in a log-linear generative model. On part-of-speech induction in English and Chinese, as well as an information extraction task, prototype features provide substantial error rate reductions over competitive baselines and outperform previous work. For example, we can achieve an English part-of-speech tagging accuracy of 80.5% using only three examples of each tag and no dictionary constraints. We also compare to semi-supervised learning and discuss the systems error trends.