Spain

As in any field of inquiry that depends on experiments, the verifiability of experimental studies is important in computational linguistics. Despite increased attention to verification of empirical results, the practices in the field are unclear. Furthermore, we argue, certain traditions and practices that are seemingly useful for verification may in fact be counterproductive. We demonstrate this through a set of multi-lingual experiments on parsing Universal Dependencies treebanks. In particular, we show that emphasis on exact replication leads to practices (some of which are now well established) that hide the variation in experimental results, effectively hindering verifiability with a false sense of certainty. The purpose of the present paper is to highlight the magnitude of the issues resulting from these common practices with the hope of instigating further discussion. Once we, as a community, are convinced about the importance of the problems, the solutions are rather obvious, although not necessarily easy to implement. \end{abstract}


COLING 2020

Verification, Reproduction and Replication of NLP Experiments: a Case Study on Parsing Universal Dependencies

stages 

COLING, the International Conference on Computational Linguistics, is one of the premier conferences for natural language processing and computational linguistics. Often grouped within the field of artificial intelligence, but actually pre-dating the development of artificial intelligence, advances in computational linguistics and natural language processing are now some of the major drivers behind the use of artificial intelligence for commercial and social applications – for example, on-line search, machine translation and with voice-assisted conversational devices.

First established in 1965, the biennial COLING conference is held in diverse parts of the globe and attracts participants from both top-ranked research centers and emerging countries. Today, the most important developments in our field are taking place not only in universities and academic research institutes, but also in industrial research departments and in technological startups. COLING conferences provide opportunities for all these communities to showcase their exciting developments.

COLING, the International Conference on Computational Linguistics, is one of the premier conferences for natural language processing and computational linguistics. Often grouped within the field of artificial intelligence, but actually pre-dating the development of artificial intelligence, advances in computational linguistics and natural language processing are now some of the major drivers behind the use of artificial intelligence for commercial and social applications – for example, on-line search, machine translation and with voice-assisted conversational devices.

technical paper

The topic of this presentation is a rule-based pipeline for converting constituency treebanks based on the Penn Treebank format to Universal Dependencies (UD). We describe the conversion process, the methods used to deliver a fully automated UD corpus and complications involved. An Icelandic constituency treebank is converted to a UD corpus, and the converter extended to convert a Faroese constituency treebank. The result is an open-source conversion tool, published under an Apache 2.0 license, applicable to a Penn-style treebank for conversion to a UD corpus, along with two new UD corpora, an Icelandic one and a Faroese one. Both are included in version 2.7 of UD.

A Universal Dependencies Conversion Pipeline for a Penn-format Constituency Treebank

To investigate issues that arise in the process of developing a Universal Dependency (UD) treebank for Korean and Japanese, we begin by addressing the typological characteristics of Korean and Japanese. Both Korean and Japanese are agglutinative and head-final languages. And the principle of word segmentation for both languages is different from English, which makes it difficult to apply UD guidelines. Following the typological characteristics of the two languages and the issue of UD application, we review the application of UPOS and DEPREL schemes to the two languages. The annotation principles for AUX, ADJ, DET, ADP and PART are discussed for the UPOS scheme, and the annotation principles for case, aux, iobj, and obl are discussed for the DEPREL scheme.

Annotation Issues in Universal Dependencies for Korean and Japanese

We present a method for identifying and analyzing adnominal possessive constructions in 63 Universal Dependencies treebanks. We classify adpossessive constructions in terms of their morphological type (locus of marking) and present a workflow for detecting and analyzing them typologically. Based on a preliminary evaluation, the algorithm works fairly reliably in adpossessive constructions that are morphologically marked. However, it performs rather poorly in adpossessive constructions that are not marked morphologically, so-called zero-marked constructions, because of difficulties in identifying these constructions with the current annotation. The research focuses on one well-circumscribed and universal construction in the hope of generating more interest in using UD for cross-linguistic comparison.

Variation in Universal Dependencies annotation: A token based typological case study on adpossessive constructions

Predicate-argument structure analysis is a central component in meaning representations of text. The fact that some arguments are not explicitly mentioned in a sentence gives rise to ambiguity in language understanding, and renders it difficult for machines to interpret text correctly. However, only few resources represent implicit roles for NLU, and existing studies in NLP only make coarse distinctions between categories of arguments omitted from linguistic form. This paper proposes a typology for fine-grained implicit argument annotation on top of Universal Conceptual Cognitive Annotation's foundational layer. The proposed implicit argument categorisation is driven by theories of implicit role interpretation and consists of six types: Deictic, Generic, Genre-based, Type-identifiable, Non-specific, and Iterated-set. We exemplify our design by revisiting part of the UCCA EWT corpus, providing a new dataset annotated with the refinement layer, and making a comparative analysis with other schemes.

Refining Implicit Argument Annotation for UCCA

Starting with the goal and motivation in choosing this topic, a brief overview of eye-tracking is given. The corpus used in all the experiments is described along with the pre-processing done on the data. The two parsers used in all experiments namely, the sequence labelling parser and graph-based parser along with the different experimental setups are mentioned. Finally the results of the experiments are shared.

Eyes on the Parse: Using Gaze Features in Syntactic Parsing

While many languages use adpositions to encode semantic relationships between content words in a sentence (e.g., agentivity or temporality), the details of how adpositions work vary widely across languages with respect to both form and meaning. In this paper, we empirically adapt the SNACS framework (Schneider et al., 2018) to Korean, a language that is typologically distant from English—the language SNACS was originally designed upon. We apply the SNACS framework to annotate the highly popular novella The Little Prince with semantic supersense labels over all Korean postpositions. Thus, we introduce the first broad-coverage corpus annotated with Korean postposition semantics and provide a detailed analysis of the corpus with an apples-to-apples comparison between Korean and English annotations.

K-SNACS: Annotating Korean Adposition Semantics

This paper introduces a representation and annotation scheme for argument structure constructions that are used metaphorically with verbs in different semantic domains. We aim to contribute to the study of constructional metaphors which has received little attention in theoretical and computational linguistics. The proposed representation consists of a systematic mapping between the constructional and verbal event structures in two domains. It reveals the semantic motivations that lead to constructions being metaphorically extended. We demonstrate this representation on argument structure constructions with Transfer of Possession verbs and test the viability of this scheme with an annotation exercise.

Representing constructional metaphors

Abstract Meaning Representation (AMR) is a simple, expressive semantic framework whose emphasis on predicate-argument structure is effective for many tasks. Nevertheless, AMR lacks a systematic treatment of projection phenomena, making its translation into logical form problematic. We present a translation function from AMR to first order logic using continuation semantics, which allows us to capture the semantic context of an expression in the form of an argument. This is a natural extension of AMR’s original design principles, allowing us to easily model basic projection phenomena such as quantification and negation as well as complex phenomena such as bound variables and donkey anaphora.

A Continuation Semantics for Abstract Meaning
Representation

This paper examines how Abstract Meaning Representation (AMR) can be utilized for finding answers to research questions in medical scientific documents, in particular, to advance the study of UV (ultraviolet) inactivation of the novel coronavirus that causes the disease COVID-19. We describe the development of a proof-of-concept prototype tool, InfoForager, which uses AMR to conduct a semantic search, targeting the meaning of the user question, and matching this to
sentences in medical documents that may contain information to answer that question. This work was conducted as a sprint over a period of six weeks, and reveals both promising results and challenges in reducing the user search time relating to COVID-19 research, and in general, domain adaption of AMR for this task.

InfoForager: Leveraging Semantic Search with AMR for COVID-19 Research

This paper presents a “road map” for the annotation of semantic categories in typologically diverse languages, with potentially few linguistic resources, and often no existing computational resources. Past semantic annotation efforts have focused largely on high-resource languages, or relatively low-resource languages with a large number of native speakers. However, there are certain typological traits, namely the synthesis of multiple concepts into a single word, that are more common in languages with a smaller speech community. For example, what is expressed as a sentence in a more analytic language like English, may be expressed as a single word in a more synthetic language like Arapaho. This paper proposes solutions for annotating analytic and synthetic languages in a comparable way based on existing typological research, and introduces a road map for the annotation of languages with a dearth of resources.

Verification, Reproduction and Replication of NLP Experiments: a Case Study on Parsing Universal Dependencies

Downloads

Next from COLING 2020

A Universal Dependencies Conversion Pipeline for a Penn-format Constituency Treebank

Similar lecture

Ethics and Bias

Verification, Reproduction and Replication of NLP Experiments: a Case Study on Parsing Universal Dependencies

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from COLING 2020

A Universal Dependencies Conversion Pipeline for a Penn-format Constituency Treebank

Similar lecture

Ethics and Bias

Downloads