Towards Automated Semantic Explainability of Multimedia Feature Graphs

Multimedia feature graphs are employed to represent features of images, video, audio, or text. Various techniques exist to extract such features from multimedia objects. In this paper, we describe the extension of such a feature graph to represent the meaning of such multimedia features and introduce a formal context-free PS-grammar (Phrase Structure grammar) to automatically generate human-understandable natural language expressions based on such features. To achieve this, we define a semantic extension to syntactic multimedia feature graphs and introduce a set of production rules for phrases of natural language English expressions. This explainability, which is founded on a semantic model provides the opportunity to represent any multimedia feature in a human-readable and human-understandable form, which largely closes the gap between the technical representation of such features and their semantics. We show how this explainability can be formally defined and demonstrate the corresponding implementation based on our generic multimedia analysis framework. Furthermore, we show how this semantic extension can be employed to increase the effectiveness in precision and recall experiments.

Twentieth Century Linguistics at Closing Time

This chapter appraises the state of linguistics at the end of the twentieth century in the wake of the Generative/Interpretive Semantics episode. The period saw a huge upswing in Noam Chomsky’s influence with the dominance of his Government and Binding/Principles and Parameters model, but also the development of multiple other competing and intersecting formal models, all of which did away with Chomsky’s totemic concept, the transformation: Relational Grammar (RG), Lexical-Functional Grammar (LFG), Generalized Phrase Structure Grammar (GPSG), and so many more that Frederick Newmeyer tagged the lot of them Alphabet Grammars (AGs). Alongside these frameworks came George Lakoff’s most far-reaching and influential development, with philosopher, Mark Johnson, “Conceptual Metaphor Theory” (a label the author rejects).

Pronominal Feature Re-assembly: L1 and L2 Pronoun Resolution of Spanish Epicene and Common Gender Antecedents

In this paper, we propose the implementation of a full-fledged feature-based lexicalist syntactic theory as a way to represent the possible configurations of features in the learner’s interlanguage and formalize a theory of acquisition based in feature reassembly. We describe gender agreement pronominal coindexation in Spanish using Head-driven Phrase Structure Grammar (HPSG) and use it to analyze the results of a self-paced reading test with L1 and L2 speakers. We find that the specification of the gender feature value at the syntactic level in epicene antecedents facilitates pronominal resolution in L1 Spanish speakers. Conversely, there is a cognitive cost when the gender feature is underspecified at the syntactic level in common gender antecedents; this cost is not found among L2 speakers. The detailed descriptions in terms of feature specification in the HPSG framework allow us to observe differences between the L1 and L2 grammars in fine-grained detail and represent optionality at the lexical level.

What participles are a mixture of

It is commonly assumed that participles show a mixture of verbal and adjectival properties, but the issue of how this mixed nature can best be captured is anything but settled. Analyses range from the purely adjectival to the purely verbal with various shades in between. This lack of consensus is at least partly due to the fact that participles are used in a variety of ways and that an analysis which fits one of them is not necessarily equally plausible for the other. In an effort to overcome the resulting fragmentation this paper proposes an analysis that covers all uses of the participles, from the adnominal over the predicative to the free adjunct uses, including also the nominalized ones. To keep it feasible we focus on one language, namely Dutch. With the help of a treebank we first identify the uses of the Dutch participles and describe their properties in informal terms. In a second step we provide an analysis in terms of the notation of Head-driven Phrase Structure Grammar. A key property of the analysis is the differentiation between core uses and grammaticalized uses. The treatment of the latter is influenced by insights from Grammaticalization Theory.

NPs, not DPs: The NP vs. DP debate in the context of dependency grammar

This paper considers the NP vs. DP debate from the perspective of dependency grammar (DG). The message is delivered that given DG assumptions about sentence structure, the traditional NP-analysis of nominal groups is preferable over the DP-analysis. The debate is also considered from the perspective of phrase structure grammar (PSG). While many of the issues discussed here do not directly support NP over DP given PSG assumptions, some do. More importantly, one has to accept the widespread presence of null determiner heads for the DP analysis to be plausible on PSG assumptions. The argument developed at length here is that the traditional NP-analysis of nominal groups is both more accurate and simpler than the DP-analysis, in part because it does not rely on the frequent occurrence of null determiners.

Adjectives as roots of nominal groups: the big mess construction in dependency grammar

The so-called ‘Big Mess Construction’ (BMC) frustrates standard assumptions about the structure of nominal groups. The normal position of an attributive adjective is after the determiner and before the noun, but in the BMC, the adjective precedes the determiner, e.g. that strange a sound, so big a scandal, too lame an excuse. Previous accounts of the BMC are couched in ‘Phrase Structure Grammar’ (PSG) and view the noun or the determiner (or the preposition of) as the root/head of the BMC phrase. In contrast, the current approach, which is couched in a ‘Dependency Grammar’ (DG) model, argues that the adjective is in fact the root/head of the phrase. A number of insights point to the adjective as the root/head, the most important of which is the optional appearance of the preposition of, e.g. that strange of a sound, so big of a scandal, too lame of an excuse.

Adjectives as roots of nominal groups: the big mess construction in dependency grammar

The so-called ‘Big Mess Construction’ (BMC) frustrates standard assumptions about the structure of nominal groups. The normal position of an attributive adjective is after the determiner and before the noun, but in the BMC, the adjective precedes the determiner, e.g. that strange a sound, so big a scandal, too lame an excuse. Previous accounts of the BMC are couched in ‘Phrase Structure Grammar’ (PSG) and view the noun or the determiner (or the preposition of) as the root/head of the BMC phrase. In contrast, the current approach, which is couched in a ‘Dependency Grammar’ (DG) model, argues that the adjective is in fact the root/head of the phrase. A number of insights point to the adjective as the root/head, the most important of which is the optional appearance of the preposition of, e.g. that strange of a sound, so big of a scandal, too lame of an excuse.

A Comparative Study of Deep and Shallow Parsing Approaches to Automated Grammaticality Evaluation

The concept of automated grammar evaluation of natural language texts is one that has attracted significant interests in the natural language processing community. It is the examination of natural language text for grammatical accuracy using computer software. The current work is a comparative study of different deep and shallow parsing techniques that have been applied to lexical analysis and grammaticality evaluation of natural language texts. The comparative analysis was based on data gathered from numerous related works. Shallow parsing using induced grammars was first examined along with its two main sub-categories, the probabilistic statistical parsers and the connectionist approach using neural networks. Deep parsing using handcrafted grammar was subsequently examined along with several of it‟s subcategories including Transformational Grammars, Feature Based Grammars, Lexical Functional Grammar (LFG), Definite Clause Grammar (DCG), Property Grammar (PG), Categorial Grammar (CG), Generalized Phrase Structure Grammar (GPSG), and Head-driven Phrase Structure Grammar (HPSG). Based on facts gathered from literature on the different aforementioned formalisms, a comparative analysis of the deep and shallow parsing techniques was performed. The comparative analysis showed among other things that while the shallow parsing approach was usually domain dependent, influenced by sentence length and lexical frequency and employed machine learning to induce grammar rules, the deep parsing approach were not domain dependent, not influenced by sentence length nor lexical frequency, and they made use of well spelt out set of precise linguistic rules. The deep parsing techniques proved to be a more labour intensive approach while the induced grammar rules were usually faster and reliability increased with size, accuracy and coverage of training data. The shallow parsing approach has gained immense popularity owing to availability of large corpora for different languages, and has therefore become the most accepted and adopted approach in recent times. Keywords: Grammaticality, Natural language processing, Deep parsing, Shallow parsing, Handcrafted grammar, Precision grammar, Induced grammar, Automated scoring, Computational linguistics, Comparative study.

Learning of Hierarchical Temporal Structures for Guided Improvisation

This article focuses on learning the hierarchical structure of what we call a “temporal scenario” (for instance, a chord progression) to perform automatic improvisation consistently over several different time scales. We first present a way to represent hierarchical structures with a phrase structure grammar. Such a grammar enables us to analyze a scenario at several levels of organization, creating a “multilevel scenario.” We then develop a method to automatically induce this grammar from a corpus, based on sequence selection with mutual information. We applied this method to a corpus of transcribed improvisations based on the chord sequence, also with chord substitutions, from George Gershwin's “I Got Rhythm.” From these we obtained multilevel scenarios similar to the analyses performed by professional musicians. We then present a novel heuristic approach, exploiting the multilevel structure of a scenario to guide the improvisation with anticipatory behavior in an improvisation paradigm driven by a factor oracle. This method ensures consistency of the improvisation with regard to the global form, and it opens up possibilities when playing on chords that do not exist in memory. This system was evaluated by professional improvisers during listening sessions and received excellent feedback.

Phrase Structure

The phrase structure of English has been a central concern for most approaches to syntax, including various forms of Transformational Grammar, Head-driven Phrase Structure Grammar, Lexical Functional Grammar, and the earlier Generalized Phrase Structure Grammar framework. They have developed detailed analyses of verb phrases, nominal phases, clauses of various kinds, including unbounded dependency clauses and elliptical clauses, and adjective phrases and prepositional phrases, and coordinate structures. There are similarities and differences between the various approaches in all these areas. They differ in whether or not they are confined to binary branching, whether or not they assume that all phrases are headed, and in the extent to which they assume heads which are phonologically empty. More generally they vary in how complex they take phrase structures to be and in how much variety they see in the local trees that they consist of.