Computational Linguistics Research Research Articles

Development of an intelligent information resource model based on modern natural language processing methods

<span lang="EN-US">Currently, there is an avalanche-like increase in the need for automatic text processing, respectively, new effective methods and tools for processing texts in natural language are emerging. Although these methods, tools and resources are mostly presented on the internet, many of them remain inaccessible to developers, since they are not systematized, distributed in various directories or on separate sites of both humanitarian and technical orientation. All this greatly complicates their search and practical use in conducting research in computational linguistics and developing applied systems for natural text processing. This paper is aimed at solving the need described above. The paper goal is to develop model of an intelligent information resource based on modern methods of natural language processing (IIR NLP). The main goal of IIR NLP is to render convenient valuable access for specialists in the field of computational linguistics. The originality of our proposed approach is that the developed ontology of the subject area “NLP” will be used to systematize all the above knowledge, data, information resources and organize meaningful access to them, and semantic web standards and technology tools will be used as a software basis.</span>

FALCoN: Detecting and classifying abusive language in social networks using context features and unlabeled data

Social networks have grown into a widespread form of communication that allows a large number of users to participate in conversations and consume information at any time. The casual nature of social media allows for nonstandard terminology, some of which may be considered rude and derogatory. As a result, a significant portion of social media users is found to express disrespectful language. This problem may intensify in certain developing countries where young children are granted unsupervised access to social media platforms. Furthermore, the sheer amount of social media data generated daily by millions of users makes it impractical for humans to monitor and regulate inappropriate content. If adolescents are exposed to these harmful language patterns without adequate supervision, they may feel obliged to adopt them. In addition, unrestricted aggression in online forums may result in cyberbullying and other dreadful occurrences. While computational linguistics research has addressed the difficulty of detecting abusive dialogues, issues remain unanswered for low-resource languages with little annotated data, leading the majority of supervised techniques to perform poorly. In addition, social media content is often presented in complex, context-rich formats that encourage creative user involvement. Therefore, we propose to improve the performance of abusive language detection and classification in a low-resource setting, using both the abundant unlabeled data and the context features via the co-training protocol that enables two machine learning models, each learning from an orthogonal set of features, to teach each other, resulting in an overall performance improvement. Empirical results reveal that our proposed framework achieves F1 values of 0.922 and 0.827, surpassing the state-of-the-art baselines by 3.32% and 45.85% for binary and fine-grained classification tasks, respectively. In addition to proving the efficacy of co-training in a low-resource situation for abusive language detection and classification tasks, the findings shed light on several opportunities to use unlabeled data and contextual characteristics of social networks in a variety of social computing applications.

A study on surprisal and semantic relatedness for eye-tracking data prediction.

Previous research in computational linguistics dedicated a lot of effort to using language modeling and/or distributional semantic models to predict metrics extracted from eye-tracking data. However, it is not clear whether the two components have a distinct contribution, with recent studies claiming that surprisal scores estimated with large-scale, deep learning-based language models subsume the semantic relatedness component. In our study, we propose a regression experiment for estimating different eye-tracking metrics on two English corpora, contrasting the quality of the predictions with and without the surprisal and the relatedness components. Different types of relatedness scores derived from both static and contextual models have also been tested. Our results suggest that both components play a role in the prediction, with semantic relatedness surprisingly contributing also to the prediction of function words. Moreover, they show that when the metric is computed with the contextual embeddings of the BERT model, it is able to explain a higher amount of variance.

Computational linguistics and discourse complexology: Paradigms and research methods

The dramatic expansion of modern linguistic research and enhanced accuracy of linguistic analysis have become a reality due to the ability of artificial neural networks not only to learn and adapt, but also carry out automate linguistic analysis, select, modify and compare texts of various types and genres. The purpose of this article and the journal issue as a whole is to present modern areas of research in computational linguistics and linguistic complexology, as well as to define a solid rationale for the new interdisciplinary field, i.e. discourse complexology. The review of trends in computational linguistics focuses on the following aspects of research: applied problems and methods, computational linguistic resources, contribution of theoretical linguistics to computational linguistics, and the use of deep learning neural networks. The special issue also addresses the problem of objective and relative text complexity and its assessment. We focus on the two main approaches to linguistic complexity assessment: “parametric approach” and machine learning. The findings of the studies published in this special issue indicate a major contribution of computational linguistics to discourse complexology, including new algorithms developed to solve discourse complexology problems. The issue outlines the research areas of linguistic complexology and provides a framework to guide its further development including a design of a complexity matrix for texts of various types and genres, refining the list of complexity predictors, validating new complexity criteria, and expanding databases for natural language.

MEANING AS A MARKER OF METAPHORICITY TOWARDS A COMPUTATIONAL IDENTIFICATION OF METAPHOR IN THE EVER-GLORIOUS QUR’ĀN

This work represents a novel direction for computational linguistics research on metaphor in the Ever-Glorious Qur’ān. The present study proposed a basic/non-basic meaning criterion as a marker for the computational identification of metaphor in the Ever-Glorious Qur’ān. The corpus was Sūrat Hūd, where manual identification for candidate metaphors was conducted by referring to four authentic exegeses, namely Tafsīr Aṭ-Ṭabarī (2010), Az-Zamakhshari (2007), Al-Qurṭubi (1964) and Ash-Sha’rāwi (1997) as well as Yūsuf ‘Ali interpretation of the meanings of the Ever-Glorious Qur’ān (1992). 22 metaphors were identified and classified based on the proposed semantic criterion. The data were analyzed both qualitatively and quantitatively, and candidates were classified into three categories depending on their meaning and their degree of metaphoricity. The study drew on the conceptual theory of metaphor by Lackoff Johnson (1980), studies on metaphors in the Ever-Glorious Qur’ān, as well as computational studies of metaphor, in general, to finally arrive at a criterion that describes the meaning of metaphor towards a computational identification of metaphors in the Ever-Glorious Qur’ān. The study proposed a software input to detect metaphors by identifying their semantic features.

Gender bias recognition in political news articles

Gender bias exists not only in the stereotypes used to portray male and female figures, but more importantly, in the word choice media outlets use to describe political figures. Research in computational linguistics has not yet focused on the problem of identifying gender bias in political news from a variety of leanings, despite decades of research from the political science perspective. In this work, we introduce a new dataset – News-Bias – and demonstrate through extensive experimentation that there exists significant gender bias in political news, even with all gendered terms, and personally identifiable information removed. We show that bias persists through the document embeddings, sentiment and word choice across news outlets and political leanings.

Compilation, Analysis and Application of a Comprehensive Bangla Corpus KUMono

Research in Natural Language Processing (NLP) and computational linguistics highly depends on a good quality representative corpus of any specific language. Bangla is one of the most spoken languages in the world but Bangla NLP research is in its early stage of development due to the lack of quality public corpus. This article describes the detailed compilation methodology of a comprehensive monolingual Bangla corpus, KUMono. The newly developed corpus consists of more than 350 million word tokens and more than one million unique tokens from 18 major text categories of online Bangla websites. We have conducted several word-level and character-level linguistic phenomenon analyses based on empirical studies of the developed corpus. The corpus follows Zipf’s curve and hapax legomena rule. The quality of the corpus is also assessed by analyzing and comparing the inherent sparseness of the corpus with existing Bangla corpora, by analyzing the distribution of function words of the corpus and vocabulary growth rate. We have developed a Bangla article categorization application based on the KUMono corpus and received compelling results by comparing to the state-of-the-art models.

Emotion norms for 6000 Polish word meanings with a direct mapping to the Polish wordnet

Emotion lexicons are useful in research across various disciplines, but the availability of such resources remains limited for most languages. While existing emotion lexicons typically comprise words, it is a particular meaning of a word (rather than the word itself) that conveys emotion. To mitigate this issue, we present the Emotion Meanings dataset, a novel dataset of 6000 Polish word meanings. The word meanings are derived from the Polish wordnet (plWordNet), a large semantic network interlinking words by means of lexical and conceptual relations. The word meanings were manually rated for valence and arousal, along with a variety of basic emotion categories (anger, disgust, fear, sadness, anticipation, happiness, surprise, and trust). The annotations were found to be highly reliable, as demonstrated by the similarity between data collected in two independent samples: unsupervised (n = 21,317) and supervised (n = 561). Although we found the annotations to be relatively stable for female, male, younger, and older participants, we share both summary data and individual data to enable emotion research on different demographically specific subgroups. The word meanings are further accompanied by the relevant metadata, derived from open-source linguistic resources. Direct mapping to Princeton WordNet makes the dataset suitable for research on multiple languages. Altogether, this dataset provides a versatile resource that can be employed for emotion research in psychology, cognitive science, psycholinguistics, computational linguistics, and natural language processing.

Normalized Datasets of Harnack’s Reconstruction of Marcion’s <i>Gospel</i>

These two datasets are the first born-digital, normalized, peer-reviewed datasets of Harnack’s classic reconstruction of Marcion’s <em>Gospel</em>. The first consists of human-readable postclassical Greek, the second of lemmatized and morphologically tagged text following the openly licensed BibleWorks Greek Morphology schema. The recent deluge of critical editions of Marcion’s <em>Gospel</em> makes Harnack’s public domain work even more relevant as scholars turn from theology- to text-based approaches to restore Marcion’s <em>Gospel</em> and account for its place in the editorial history of early canonical and non-canonical Gospels. These datasets resource Marcion’s <em>Gospel</em> becoming a major topic of interest in Computational Linguistics research.

Grief-Stricken in a Crowd: The Language of Bereavement and Distress in Social Media

People turn to social media to express their emotions surrounding major life events. Death of a loved one is one scenario in which people share their feelings in the semi-public space of social networking sites. In this paper, we present the results of a two-part investigation of grief and distress in the context of messages posted to the profiles of deceased MySpace users. We present coding system for identifying emotion distressed content, followed by a detailed analysis of language use that lays a foundation for natural language processing (NLP) tasks, such as automatic detection of bereavement-related distress. Our findings suggest that in addition to words bearing positive or negative sentiment, linguistic style can be an indicator of messages that demonstrate distress in the space of post-mortem social media content. These results contribute to research in computational linguistics by identifying linguistic features that can be used for automatic classification as well as to research on death and bereavement by enumerating attributes of distressed self-expression in a post-mortem context.

Distributional Semantics and Linguistic Theory

Distributional semantics provides multidimensional, graded, empirically induced word representations that successfully capture many aspects of meaning in natural languages, as shown by a large body of research in computational linguistics; yet, its impact in theoretical linguistics has so far been limited. This review provides a critical discussion of the literature on distributional semantics, with an emphasis on methods and results that are relevant for theoretical linguistics, in three areas: semantic change, polysemy and composition, and the grammar–semantics interface (specifically, the interface of semantics with syntax and with derivational morphology). The goal of this review is to foster greater cross-fertilization of theoretical and computational approaches to language as a means to advance our collective knowledge of how it works.

Model evaluation in computational historical linguistics

Abstract This is a reply to the comments by Hammarström et al. (This volume) and List (This volume) on the target article Computational Historical Linguistics (This volume). There I proposed several methodological principles for research in Computational Historical Linguistics pertaining to suitable techniques for model fitting and model evaluation. Hammarström et al. debate the usefulness of these principles, and List proposes a novel evaluation measure specifically aimed at the task of proto-form reconstruction. This reply will focus on the role of model evaluation in our field.

Crowdsourcing evaluation of the quality of automatically generated questions for supporting computer-assisted language teaching

AbstractHow can state-of-the-art computational linguistic technology reduce the workload and increase the efficiency of language teachers? To address this question, we combine insights from research in second language acquisition and computational linguistics to automatically generate text-based questions to a given text. The questions are designed to draw the learner’s attention to target linguistic forms – phrasal verbs, in this particular case – by requiring them to use the forms or their paraphrases in the answer. Such questions help learners create form-meaning connections and are well suited for both practice and testing. We discuss the generation of a novel type of question combining a wh- question with a gapped sentence, and report the results of two crowdsourcing evaluation studies investigating how well automatically generated questions compare to those written by a language teacher. The first study compares our system output to gold standard human-written questions via crowdsourcing rating. An equivalence test shows that automatically generated questions are comparable to human-written ones. The second crowdsourcing study investigates two types of questions (wh- questions with and without a gapped sentence), their perceived quality, and the responses they elicit. Finally, we discuss the challenges and limitations of creating and evaluating question-generation systems for language learners.

Measuring common knowledge: latent semantic analysis, linguistic synchrony, and common knowledge in international relations

Common knowledge, also called intersubjectivity, is a core theme in the study of international cooperation and diplomacy. Yet International Relations (IR) lacks a method to systematically measure the degree of common knowledge. Drawing from research in computational linguistics, psychology, and communication, we introduce latent semantic analysis (LSA) to measure common knowledge in specific communicative exchanges between actors. We argue that the extent to which speaking partners use words in the same way and get in synch linguistically can be used to measure the degree of common knowledge, and this can be measured by the LSA method. We outline several ways LSA can be valuable to IR scholars and provide an empirical illustration of using this method in the case of Bretton Woods negotiations. The LSA method promises to help IR scholars seize the research opportunities offered by the digital age and build a bridge between qualitative and quantitative methods.

Sentiment disagreement and bitcoin price fluctuations: a psycholinguistic approach

ABSTRACTWe investigate the extent to which Bitcoin price fluctuations are associated with investors’ sentiment disagreement. We employ three textual sentiment analysis techniques: 1) a Python library offered by the Computational Linguistics and Psycholinguistics Research Center; 2) Loughran and McDonald’s (2011) dictionary; and 3) semantic orientation by the point-wise mutual information method. The results show that investors’ attention and sentiment disagreement induce extremely high volatility and jumps in Bitcoin prices. These findings complement existing studies on how investors’ sentiment manifests in asset prices.

Linguistic matrix theory

Recent research in computational linguistics has developed algorithms which associate matrices with adjectives and verbs, based on the distribution of words in a corpus of text. These matrices are linear operators on a vector space of context words. They are used to construct the meaning of composite expressions from that of the elementary constituents, forming part of a compositional distributional approach to semantics. We propose a Matrix Theory approach to this data, based on permutation symmetry along with Gaussian weights and their perturbations. A simple Gaussian model is tested against word matrices created from a large corpus of text. We characterize the cubic and quartic departures from the model, which we propose, alongside the Gaussian parameters, as signatures for comparison of linguistic corpora. We propose that perturbed Gaussian models with permutation symmetry provide a promising framework for characterizing the nature of universality in the statistical properties of word matrices. The matrix theory framework developed here exploits the view of statistics as zero dimensional perturbative quantum field theory. It perceives language as a physical system realizing a universality class of matrix statistics characterized by permutation symmetry.

Register in computational language research

AbstractShlomo Argamon is Professor of Computer Science and Director of the Master of Data Science Program at the Illinois Institute of Technology (USA). In this article, he reflects on the current and potential relationship between register and the field of computational linguistics. He applies his expertise in computational linguistics and machine learning to a variety of problems in natural language processing. These include stylistic variation, forensic linguistics, authorship attribution, and biomedical informatics. He is particularly interested in the linguistic structures used by speakers and writers, including linguistic choices that are influenced by social variables such as age, gender, and register, as well as linguistic choices that are unique or distinctive to the style of individual authors. Argamon has been a pioneer in computational linguistics and NLP research in his efforts to account for and explore register variation. His computational linguistic research on register draws inspiration from Systemic Functional Linguistics, Biber’s multi-dimensional approach to register variation, as well as his own extensive experience accounting for variation within and across text types and authors. Argamon has applied computational methods to text classification and description across registers – including blogs, academic disciplines, and news writing – as well as the interaction between register and other social variables, such as age and gender. His cutting-edge research in these areas is certain to have a lasting impact on the future of computational linguistics and NLP.

Reproducibility in Computational Linguistics: Are We Willing to Share?

This study focuses on an essential precondition for reproducibility in computational linguistics: the willingness of authors to share relevant source code and data. Ten years after Ted Pedersen’s influential “Last Words” contribution in Computational Linguistics, we investigate to what extent researchers in computational linguistics are willing and able to share their data and code. We surveyed all 395 full papers presented at the 2011 and 2016 ACL Annual Meetings, and identified whether links to data and code were provided. If working links were not provided, authors were requested to provide this information. Although data were often available, code was shared less often. When working links to code or data were not provided in the paper, authors provided the code in about one third of cases. For a selection of ten papers, we attempted to reproduce the results using the provided data and code. We were able to reproduce the results approximately for six papers. For only a single paper did we obtain the exact same results. Our findings show that even though the situation appears to have improved comparing 2016 to 2011, empiricism in computational linguistics still largely remains a matter of faith. Nevertheless, we are somewhat optimistic about the future. Ensuring reproducibility is not only important for the field as a whole, but also seems worthwhile for individual researchers: The median citation count for studies with working links to the source code is higher.

CLiC-it 2017: A Retrospective

The Fourth Italian Conference on Computational Linguistics (CLiC-it 2017) took place in Rome, in December 2017. As in previous editions, it served as the prime forum in Italy for discussing research in computational linguistics and Natural Language Processing. As General Chairs, we offer a retrospective over the conference, highlighting its international flavour and its attention to students and young researchers, with a particular focus on the innovations that were introduced at the 2017 edition.

Arabic Part-of-Speech Tagger, an Approach Based on Neural Network Modelling

POS-tagging gained the interest of researchers in computational linguistics sciences in the recent years. Part-of-speech tagging systems assign the proper grammatical tag or morpho-syntactical category labels automatically to every word in the corpus per its appearance on the text. POS-tagging serves as a fundamental and preliminary step in linguistic analysis which can help in developing many natural language processing applications such as: word processing systems, spell checking systems, building dictionaries and in parsing systems. Arabic language gained the interest of researchers which led to increasing demand for Arabic natural language processing systems. Artificial neural networks has been applied in many applications such as speech recognition and part of speech prediction, but it is considered as a new approach in Part-of-speech tagging. In this research, we developed an Arabic POS-tagger using artificial neural network. A corpus of 20,620 words, which were manually assigned to the appropriate tags was developed and used to train the artificial neural network and to test the part of speech tagger systems’ overall performance. The accuracy of the developed tagger reaches 89.04% using the testing dataset. While, it reaches 98.94% using the training dataset. By combining the two datasets, the accuracy rate for the whole system is 96.96%.