Afsaneh Fazly

A Probabilistic Computational Model of Cross-Situational Word Learning

Words are the essence of communication: They are the building blocks of any language. Learning the meaning of words is thus one of the most important aspects of language acquisition: Children must first learn words before they can combine them into complex utterances. Many theories have been developed to explain the impressive efficiency of young children in acquiring the vocabulary of their language, as well as the developmental patterns observed in the course of lexical acquisition. A major source of disagreement among the different theories is whether children are equipped with special mechanisms and biases for word learning, or their general cognitive abilities are adequate for the task. We present a novel computational model of early word learning to shed light on the mechanisms that might be at work in this process. The model learns word meanings as probabilistic associations between words and semantic elements, using an incremental and probabilistic learning mechanism, and drawing only on general cognitive abilities. The results presented here demonstrate that much about word meanings can be learned from naturally occurring child-directed utterances (paired with meaning representations), without using any special biases or constraints, and without any explicit developmental changes in the underlying learning mechanism. Furthermore, our model provides explanations for the occasionally contradictory child experimental data, and offers predictions for the behavior of young word learners in novel situations.

Unsupervised type and token identification of idiomatic expressions

Idiomatic expressions are plentiful in everyday language, yet they remain mysterious, as it is not clear exactly how people learn and understand them. They are of special interest to linguists, psycholinguists, and lexicographers, mainly because of their syntactic and semantic idiosyncrasies as well as their unclear lexical status. Despite a great deal of research on the properties of idioms in the linguistics literature, there is not much agreement on which properties are characteristic of these expressions. Because of their peculiarities, idiomatic expressions have mostly been overlooked by researchers in computational linguistics. In this article, we look into the usefulness of some of the identified linguistic properties of idioms for their automatic recognition. Specifically, we develop statistical measures that each model a specific property of idiomatic expressions by looking at their actual usage patterns in text. We use these statistical measures in a type-based classification task where we automatically separate idiomatic expressions (expressions with a possible idiomatic interpretation) from similar-on-the-surface literal phrases (for which no idiomatic interpretation is possible). In addition, we use some of the measures in a token identification task where we distinguish idiomatic and literal usages of potentially idiomatic expressions in context.

Pulling their Weight: Exploiting Syntactic Forms for the Automatic Identification of Idiomatic Expressions in Context

Much work on idioms has focused on type identification, i.e., determining whether a sequence of words can form an idiomatic expression. Since an idiom type often has a literal interpretation as well, token classification of potential idioms in context is critical for NLP. We explore the use of informative prior knowledge about the overall syntactic behaviour of a potentially-idiomatic expression (type-based knowledge) to determine whether an instance of the expression is used idiomatically or literally (token-based knowledge). We develop unsupervised methods for the task, and show that their performance is comparable to that of state-of-the-art supervised techniques.

Distinguishing Subtypes of Multiword Expressions Using Linguistically-Motivated Statistical Measures

We identify several classes of multiword expressions that each require a different encoding in a (computational) lexicon, as well as a different treatment within a computational system. We examine linguistic properties pertaining to the degree of semantic idiosyncrasy of these classes of expressions. Accordingly, we propose statistical measures to quantify each property, and use the measures to automatically distinguish the classes.

Statistical measures of the semi-productivity of light verb constructions

We propose a statistical measure for the degree of acceptability of light verb constructions, such as take a walk, based on their linguistic properties. Our measure shows good correlations with human ratings on unseen test data. Moreover, we find that our measure correlates more strongly when the potential complements of the construction (such as walk, stroll, or run) are separated into semantically similar classes. Our analysis demonstrates the systematic nature of the semi-productivity of these constructions.

Testing the efficacy of part-of-speech information in word completion

We investigate the effect of incorporating syntactic information into a word-completion algorithm. We introduce two new algorithms that combine part-of-speech tag trigrams with word bigrams, and evaluate them with a test-bench constructed for the purpose. The results show a small but statistically significant improvement in keystroke savings for one of our algorithms over baselines that use only word n-grams.

An Incremental Bayesian Model for Learning Syntactic Categories

We present an incremental Bayesian model for the unsupervised learning of syntactic categories from raw text. The model draws information from the distributional cues of words within an utterance, while explicitly bootstrapping its development on its own partially-learned knowledge of syntactic categories. Testing our model on actual child-directed data, we demonstrate that it is robust to noise, learns reasonable categories, manages lexical ambiguity, and in general shows learning behaviours similar to those observed in children.

Learning Structured Appearance Models from Captioned Images of Cluttered Scenes

Given an unstructured collection of captioned images of cluttered scenes featuring a variety of objects, our goal is to learn both the names and appearances of the objects. Only a small number of local features within any given image are associated with a particular caption word. We describe a connected graph appearance model where vertices represent local features and edges encode spatial relationships. We use the repetition of feature neighborhoods across training images and a measure of correspondence with caption words to guide the search for meaningful feature configurations. We demonstrate improved results on a dataset to which an unstructured object model was previously applied. We also apply the new method to a more challenging collection of captioned images from the Web, detecting and annotating objects within highly cluttered realistic scenes.

Using Language to Learn Structured Appearance Models for Image Annotation

Given an unstructured collection of captioned images of cluttered scenes featuring a variety of objects, our goal is to simultaneously learn the names and appearances of the objects. Only a small fraction of local features within any given image are associated with a particular caption word, and captions may contain irrelevant words not associated with any image object. We propose a novel algorithm that uses the repetition of feature neighborhoods across training images and a measure of correspondence with caption words to learn meaningful feature configurations (representing named objects). We also introduce a graph-based appearance model that captures some of the structure of an object by encoding the spatial relationships among the local visual features. In an iterative procedure, we use language (the words) to drive a perceptual grouping process that assembles an appearance model for a named object. Results of applying our method to three data sets in a variety of conditions demonstrate that, from complex, cluttered, real-world scenes with noisy captions, we can learn both the names and appearances of objects, resulting in a set of models invariant to translation, scale, orientation, occlusion, and minor changes in viewpoint or articulation. These named models, in turn, are used to automatically annotate new, uncaptioned images, thereby facilitating keyword-based image retrieval.

Automatically learning semantic knowledge about multiword predicates

Highly frequent and highly polysemous verbs, such as give, take, and make, pose a challenge to automatic lexical acquisition methods. These verbs widely participate in multiword predicates (such as light verb constructions, or LVCs), in which they contribute a broad range of figurative meanings that must be recognized. Here we focus on two properties that are key to the computational treatment of LVCs. First, we consider the degree of figurativeness of the semantic contribution of such a verb to the various LVCs it participates in. Second, we explore the patterns of acceptability of LVCs, and their productivity over semantically related combinations. To assess these properties, we develop statistical measures of figurativeness and acceptability that draw on linguistic properties of LVCs. We demonstrate that these corpus-based measures correlate well with human judgments of the relevant property. We also use the acceptability measure to estimate the degree to which a semantic class of nouns can productively form LVCs with a given verb. The linguistically-motivated measures outperform a standard measure for capturing the strength of collocation of these multiword expressions.