Publications

Abstract

Students who are learning to write in a foreign language, often want feedback on the grammatical quality of the sentences they produce. The usual NLP approach to this problem is based on parsing student-generated text. Here, we propose a generation-based ap- proach aiming at preventing errors ("scaffolding"). In our ICALL system, the student constructs sentences by composing syntactic trees out of lexically anchored "treelets" via a graphical drag & drop user interface. A natural-language generator computes all possible grammatically well-formed sentences entailed by the student-composed tree. It provides positive feedback if the student-composed tree belongs to the well-formed set, and negative feedback otherwise. If so requested by the student, it can substantiate the positive or negative feedback based on a comparison between the student-composed tree and its own trees (informative feedback on demand). In case of negative feedback, the system refuses to build the structure attempted by the student. Frequently occurring errors are handled in terms of "malrules." The system we describe is a prototype (implemented in JAVA and C++) which can be parameterized with respect to L1 and L2, the size of the lexicon, and the level of detail of the visually presented grammatical structures.

Abstract

This paper investigates the time and space complexity of word order computation in the psycholinguistically motivated grammar formalism of Performance Grammar (PG). In PG, the first stage of syntax assembly yields an unordered tree ('mobile') consisting of a hierarchy of lexical frames (lexically anchored elementary trees). Associated with each lexica l frame is a linearizer—a Finite-State Automaton that locally computes the left-to-right order of the branches of the frame. Linearization takes place after the promotion component may have raised certain constituents (e.g. Wh- or focused phrases) into the domain of lexical frames higher up in the syntactic mobile. We show that the worst-case time and space complexity of analyzing input strings of length n is O(n5) and O(n4), respectively. This result compares favorably with the time complexity of word-order computations in Tree Adjoining Grammar (TAG). A comparison with Head-Driven Phrase Structure Grammar (HPSG) reveals that PG yields a more declarative linearization method, provided that the FSA is rewritten as an equivalent regular expression.

Abstract

We present the design, implementation and simulation results of a psycholinguistic model of human syntactic processing that meets major empirical criteria. The parser operates in conjunction with a lexicalist grammar and is driven by syntactic information associated with heads of phrases. The dynamics of the model are based on competition by lateral inhibition ('competitive inhibition'). Input words activate lexical frames (i.e. elementary trees anchored to input words) in the mental lexicon, and a network of candidate 'unification links' is set up between frame nodes. These links represent tentative attachments that are graded rather than all-or-none. Candidate links that, due to grammatical or 'treehood' constraints, are incompatible, compete for inclusion in the final syntactic tree by sending each other inhibitory signals that reduce the competitor's attachment strength. The outcome of these local and simultaneous competitions is controlled by dynamic parameters, in particular by the Entry Activation and the Activation Decay rate of syntactic nodes, and by the Strength and Strength Build-up rate of Unification links. In case of a successful parse, a single syntactic tree is returned that covers the whole input string and consists of lexical frames connected by winning Unification links. Simulations are reported of a significant range of psycholinguistic parsing phenomena in both normal and aphasic speakers of English: (i) various effects of linguistic complexity (single versus double, center versus right-hand self-embeddings of relative clauses; the difference between relative clauses with subject and object extraction; the contrast between a complement clause embedded within a relative clause versus a relative clause embedded within a complement clause); (ii) effects of local and global ambiguity, and of word-class and syntactic ambiguity (including recency and length effects); (iii) certain difficulty-of-reanalysis effects (contrasts between local ambiguities that are easy to resolve versus ones that lead to serious garden-path effects); (iv) effects of agrammatism on parsing performance, in particular the performance of various groups of aphasic patients on several sentence types.

Abstract

Four probe latency experiments show that the ‘constituent boundary effect’ (transitions between constituents are more difficult than within constituents) is a retrieval and not a storage phenomenon. The experimental logic used is called paraphrastic reproduction: after verbatim memorization of some sentences, subjects were instructed to reproduce them both in their original wording and in the form of sentences that, whilst preserving the original meaning, embodied different syntactic constructions. Syntactic constructions are defined as pairs which consist of a pattern of conceptual information and a syntactic scheme, i.e. a sequence of syntactic word categories and function words. For example, the sequence noun + finite intransitive main verb (‘John runs’) expresses a conceptual actor-action relationship. It is proposed that for each overlearned and simple syntactic construction there exists a retrieval plan which does the following. It searches through the long-term memory information that has been designated as the conceptual content of the utterance(s) to be produced, looking for a token of its conceptual pattern. The retrieved information is then cast into the format of its syntactic scheme. The organization of such plans is held responsible for the constituent boundary effect.