250 likes | 371 Vues
This research paper explores the innovative use of parallel dependency trees in machine translation (MT). The author, Yuan Ding from the University of Pennsylvania, discusses the limitations of traditional statistical MT approaches and presents a new alignment algorithm that accommodates syntax and semantics. The framework employs heuristics for mapping sentence structures across languages and provides a walkthrough example. The evaluation section addresses the performance of the proposed method, highlighting its effectiveness in managing structural divergence.
E N D
Better MT Using Parallel Dependency Trees Yuan Ding University of Pennsylvania 2003 (c) University of Pennsylvania
Outline • Motivation • The alignment algorithm • Algorithm at a glance • The framework • Heuristics • Walking through an example • Evaluation • Conclusion 2003 (c) University of Pennsylvania
Motivation (1)Statistical MT Approaches • Statistical MT approaches • Pioneered by (Brown et al., 1990, 1993) • Leverage large training corpus • Outperform traditional transfer based approaches • Major Criticism • No internal representation, syntax/semantics 2003 (c) University of Pennsylvania
Motivation (2) Hybrid Approaches • Hybrid approaches • (Wu, 1997) (Alshawi et al., 2000) (Yamada and Knight, 2001, 2002) (Gildea 2003) • Applying statistical learning to structured data • Problems with Hybrid MT Approaches • Structural Divergence (Dorr, 1994) • Vagaries of loose translations in real corpora 2003 (c) University of Pennsylvania
Motivation (3) • Holy grail: • Syntax based MT which captures structural divergence • Accomplished work • A new approach to the alignment of parallel dependency trees (paper published at MT summit IX) • Allowing non-isomorphism of dependency trees 2003 (c) University of Pennsylvania
We are here… 2003 (c) University of Pennsylvania
Outline • Motivation • The alignment algorithm • Algorithm at a glance • The framework • Heuristics • Walking through an example • Evaluation • Conclusion 2003 (c) University of Pennsylvania
Define the Alignment Problem • Define the alignment problem • In natural language: find word mappings between English and Foreign sentences • In math: DefinitionFor each , find a labeling ,where 2003 (c) University of Pennsylvania
The IBM Models • The IBM way • Model 1: Orders of words don’t matter, i.e. “bag of words” model • Model 2: Condition the probabilities on the length and position • Model 3, 4, 5: • A. generate fertility of each english word • B. generate the identity • C. generate the position • Gradually adding positioning information 2003 (c) University of Pennsylvania
Using Dependency Trees • Positioning information can be acquired from parse trees • Parsers: (Collins, 1999) (Bikel, 2002) • Problems with using parse trees directly • Two types of nodes • Unlexicalized non-terminals control the domain • Using dependency trees • (Fox, 2002): best* phrasal cohesion properties • (Xia, 2001): constructing dependency trees from parse trees using the Tree Adjoining Grammar 2003 (c) University of Pennsylvania
The Framework (1) • Step 1: train IBM model 1 for lexical mapping probabilities • Step 2: find and fix high confidence mappings according to a heuristic functionh(f, e) The girl kissed her kitty cat The girl gave a kiss to her cat A pseudo-translation example 2003 (c) University of Pennsylvania
The Framework (2) • Step 3: • Partition the dependency trees on both sides w.r.t. fixed mappings • One fixed mapping creates one new “treelet” • Create a new set of parallel dependency structures 2003 (c) University of Pennsylvania
The Framework (3) • Step 4: Go back to Step 1 unless enough nodes fixed • Algorithm properties • An iterative algorithm • Time complexity O(n * T(h)), where T(h) is the time for the heuristic function in Step 2. • P(f |e) in IBM Model 1 has a unique global maximun • Guaranteed convergence • Results only depend on the heuristic function h(f, e) 2003 (c) University of Pennsylvania
Heuristics • Heuristic functions for Step 2 • Objective: find out the confidence of a mapping between a pair of words • First Heuristic: Entropy • Intuition: model probability distribution shape • Second heuristic: Inside-outside probability • Idea borrowed from PCFG parsing • Fertility threshold: rule out unlikely fertility ratio (>2.0) 2003 (c) University of Pennsylvania
Outline • Motivation • The alignment algorithm • Algorithm at a glance • The framework • Heuristics • Walking through an example • Evaluation • Conclusion 2003 (c) University of Pennsylvania
Walking through an Example (1) • [English] I have been here since 1947. • [Chinese] 1947 nian yilai wo yizhi zhu zai zheli. • Iteration 1: • One dependency tree pair. Align “I” and “wo” 2003 (c) University of Pennsylvania
Walking through an Example (2) • Iteration 2: • Partition and form two treelet pairs. • Align “since” and “yilai” 2003 (c) University of Pennsylvania
Walking through an Example (3) • Iteration 3: • Partition and form three treelet pairs. • Align “1947” and “1947”, “here” and “zheli” 2003 (c) University of Pennsylvania
Outline • Motivation • The alignment algorithm • Algorithm at a glance • The framework • Heuristics • Walking through an example • Evaluation • Conclusion 2003 (c) University of Pennsylvania
Evaluation • Training: • LDC Xinhua newswire Chinese – English parallel corpus • Filtered roughly 50%, 60K+ sentence pairs used • The parser generated 53130 parsed sentence pairs. • Evaluation: • 500 sentence pairs provided by Microsoft Research Asia. • Word level aligned by hand. • F-score: • A: set of word pairs aligned by automatic alignment • G: set of word pairs aligned in the gold file. 2003 (c) University of Pennsylvania
Results (1) • Results for IBM Model 1 to Model 4 (GIZA) • Bootstrapped from Model 1 to Model 4 • Signs of overfitting • Suspect caused by difference b/w genres in training/testing 2003 (c) University of Pennsylvania
Results (2) • Results for our algorithm: • Heuristic h1: (entropy) • Heuristic h2: (inside-outside probability) • The table shows results after one iteration, M1 = IBM model 1 • Overfitting problem • mainly caused by violation of the partition assumption in fine-grained dependency structures. 2003 (c) University of Pennsylvania
Outline • Motivation • Algorithm at a glance • The framework • Heuristics • Walking through an example • Evaluation • Conclusion 2003 (c) University of Pennsylvania
Conclusion • Model based on partitioning sentences according to their dependency structure • Without the unrealistic isomorphism assumption • Outperforms the unstructured IBM models on a large data set. • “Orthogonal” to the IBM models • uses syntactic structure but no linear ordering information. 2003 (c) University of Pennsylvania
Thank You! 2003 (c) University of Pennsylvania