Debate portals and similar web platforms constitute one of the main text sources in computational argumentation research and its applications. While the corpora built upon these sources are rich of argumentatively relevant content and structure, they also include text that is irrelevant, or even detrimental, to their purpose. In this paper, we present a precision-oriented approach to detecting such irrelevant text in a semi-supervised way. Given a few seed examples, the approach automatically learns basic lexical patterns of relevance and irrelevance and then incrementally bootstraps new patterns from sentences matching the patterns. In the existing args.me corpus with 400k argumentative texts, our approach detects almost 87k irrelevant sentences, at a precision of 0.97 according to manual evaluation. With low effort, the approach can be adapted to other web argument corpora, providing a generic way to improve corpus quality.
The Growing Gains and Pains of Iterative Web Corpora Crawling: Insights from South Slavic CLASSLA-web 2.0 Corpora
Jožef Stefan Institute（乔泽夫·斯塔芬研究所）；Faculty of Computer and Information Science, University of Ljubljana（卢布尔雅纳大学计算机与信息科学学院）；Institute of Contemporary History（当代历史研究所）；Natural Language Processing Centre, Masaryk University（马萨里克大学自然语言处理中心）
The algorithm of the creation texts parallel corpora was presented. The algorithm is based on the use of "key words" in text documents, and on the means of their automated translation. Key words were singled out by means of using Russian and Ukrainian morphological dictionaries, as well as dictionaries of the translation of nouns for the Russian and Ukrainianlanguages. Besides, to calculate the weights of the terms in the documents, empiric-statistic rules were used. The algorithm under consideration was realized in the form of a program complex, integrated into the content-monitoring InfoStre…
This paper evaluates large georeferenced corpora, taken from both web-crawled and social media sources, against ground-truth population and language-census datasets. The goal is to determine (i) which dataset best represents population demographics; (ii) in what parts of the world the datasets are most representative of actual populations; and (iii) how to weight the datasets to provide more accurate representations of underlying populations. The paper finds that the two datasets represent very different populations and that they correlate with actual populations with values of r=0.60 (social …
This paper presents Charon, a web tool for annotating multimodal corpora with FrameNet categories. Annotation can be made for corpora containing both static images and video sequences paired - or not - with text sequences. The pipeline features, besides the annotation interface, corpus import and pre-processing tools.
Recent machine translation algorithms mainly rely on parallel corpora. However, since the availability of parallel corpora remains limited, only some resource-rich language pairs can benefit from them. We constructed a parallel corpus for English-Japanese, for which the amount of publicly available parallel corpora is still limited. We constructed the parallel corpus by broadly crawling the web and automatically aligning parallel sentences. Our collected corpus, called JParaCrawl, amassed over 8.7 million sentence pairs. We show how it includes a broader range of domains and how a neural machi…
University of Oslo（奥斯陆大学）；University of Tyumen（秋明大学）
In this paper, we present a distributional word embedding model trained on one of the largest available Russian corpora: Araneum Russicum Maximum (over 10 billion words crawled from the web). We compare this model to the model trained on the Russian National Corpus (RNC). The two corpora are much different in their size and compilation procedures. We test these differences by evaluating the trained models against the Russian part of the Multilingual SimLex999 semantic similarity dataset. We detect and describe numerous issues in this dataset and publish a new corrected version. Aside from the …
University of Washington（华盛顿大学）；Hugging Face（抱抱脸）；Allen Institute for Artificial Intelligence（艾伦人工智能研究所）；Queer in AI（人工智能酷儿组织）
Large language models have led to remarkable progress on many NLP tasks, and researchers are turning to ever-larger text corpora to train them. Some of the largest corpora available are made by scraping significant portions of the internet, and are frequently introduced with only minimal documentation. In this work we provide some of the first documentation for the Colossal Clean Crawled Corpus (C4; Raffel et al., 2020), a dataset created by applying a set of filters to a single snapshot of Common Crawl. We begin by investigating where the data came from, and find a significant amount of text …
Unidata；Saint-Petersburg State University（圣彼得堡国立大学）
Corpora that contain tabular data such as WebTables are a vital resource for the academic community. Essentially, they are the backbone of any modern research in information management. They are used for various tasks of data extraction, knowledge base construction, question answering, column semantic type detection and many other. Such corpora are useful not only as a source of data, but also as a base for building test datasets. So far, there were no such corpora for the Russian language and this seriously hindered research in the aforementioned areas. In this paper, we present the first cor…
Medical University of Vienna（维也纳医科大学）；University of Alicante（阿利坎特大学）；Medical University of Graz（格拉茨医科大学）
BACKGROUND: The amount of biomedical literature is rapidly growing and it is becoming increasingly difficult to keep manually curated knowledge bases and ontologies up-to-date. In this study we applied the word2vec deep learning toolkit to medical corpora to test its potential for identifying relationships from unstructured text. We evaluated the efficiency of word2vec in identifying properties of pharmaceuticals based on mid-sized, unstructured medical text corpora available on the web. Properties included relationships to diseases ('may treat') or physiological processes ('has physiological …
Recent studies have shown that macroscopic patterns of continuity and change over the course of centuries can be detected through the analysis of time series extracted from massive textual corpora. Similar data-driven approaches have already revolutionised the natural sciences, and are widely believed to hold similar potential for the humanities and social sciences, driven by the mass-digitisation projects that are currently under way, and coupled with the ever-increasing number of documents which are "born digital". As such, new interactive tools are required to discover and extract macroscop…
This paper presents the challenges in creating and managing large parallel corpora of 12 major Indian languages (which is soon to be extended to 23 languages) as part of a major consortium project funded by the Department of Information Technology (DIT), Govt. of India, and running parallel in 10 different universities of India. In order to efficiently manage the process of creation and dissemination of these huge corpora, the web-based (with a reduced stand-alone version also) annotation tool ILCIANN (Indian Languages Corpora Initiative Annotation Tool) has been developed. It was primarily de…
This paper measures similarity both within and between 84 language varieties across nine languages. These corpora are drawn from digital sources (the web and tweets), allowing us to evaluate whether such geo-referenced corpora are reliable for modelling linguistic variation. The basic idea is that, if each source adequately represents a single underlying language variety, then the similarity between these sources should be stable across all languages and countries. The paper shows that there is a consistent agreement between these sources using frequency-based corpus similarity measures. This …
This paper experiments with frequency-based corpus similarity measures across 39 languages using a register prediction task. The goal is to quantify (i) the distance between different corpora from the same language and (ii) the homogeneity of individual corpora. Both of these goals are essential for measuring how well corpus-based linguistic analysis generalizes from one dataset to another. The problem is that previous work has focused on Indo-European languages, raising the question of whether these measures are able to provide robust generalizations across diverse languages. This paper uses …
Virginia Tech（弗吉尼亚理工大学）；IBM Research Almaden（IBM阿尔马登研究院）；eBay Inc（eBay公司）
Recent advances in text representation have shown that training on large amounts of text is crucial for natural language understanding. However, models trained without predefined notions of topical interest typically require careful fine-tuning when transferred to specialized domains. When a sufficient amount of within-domain text may not be available, expanding a seed corpus of relevant documents from large-scale web data poses several challenges. First, corpus expansion requires scoring and ranking each document in the collection, an operation that can quickly become computationally expensiv…
Semantic Noise affects text analytics activities for the domain-specific industries significantly. It impedes the text understanding which holds prime importance in the critical decision making tasks. In this work, we formalize semantic noise as a sequence of terms that do not contribute to the narrative of the text. We look beyond the notion of standard statistically-based stop words and consider the semantics of terms to exclude the semantic noise. We present a novel Semantic Infusion technique to associate meta-data with the categorical corpus text and demonstrate its near-lossless nature. …
Pre-training text representations have led to significant improvements in many areas of natural language processing. The quality of these models benefits greatly from the size of the pretraining corpora as long as its quality is preserved. In this paper, we describe an automatic pipeline to extract massive high-quality monolingual datasets from Common Crawl for a variety of languages. Our pipeline follows the data processing introduced in fastText (Mikolov et al., 2017; Grave et al., 2018), that deduplicates documents and identifies their language. We augment this pipeline with a filtering ste…
Manual annotation of textual documents is a necessary task when constructing benchmark corpora for training and evaluating machine learning algorithms. We created a comprehensive directory of annotation tools that currently includes 93 tools. We analyzed the tools over a set of 31 features and implemented simple scripts and a Web application that filters the tools based on chosen criteria. We present two use cases using the directory and propose ideas for its maintenance. The directory, source codes for scripts, and link to the Web application are available at: https://github.com/mariananeves/…
Google Research（谷歌研究院）；Masakhane NLP（Masakhane自然语言处理组织）；Turkic Interlingua（突厥语国际语组织）；Haverford College（哈弗福德学院）；RobotsMali（马里机器人组织）；Allen Institute for Artificial Intelligence（艾伦人工智能研究所）；University of Zambia（赞比亚大学）；Google（谷歌公司）；AIMS-AMMI（非洲数学科学研究所-非洲机器学习硕士项目）；Inria（法国国家信息与自动化研究所）；University of Zurich（苏黎世大学）；Stanford University（斯坦福大学）；Kwame Nkrumah University of Science and Technology（夸梅·恩克鲁玛科技大学）；Sorbonne Université（索邦大学）；Niger-Volta LTI（尼日尔-沃尔特语言技术研究所）；University of Waterloo（滑铁卢大学）；University of Electronic Science and Technology of China（电子科技大学）；University of Notre Dame（圣母大学）；Bayero University Kano（卡诺巴耶罗大学）
With the success of large-scale pre-training and multilingual modeling in Natural Language Processing (NLP), recent years have seen a proliferation of large, web-mined text datasets covering hundreds of languages. We manually audit the quality of 205 language-specific corpora released with five major public datasets (CCAligned, ParaCrawl, WikiMatrix, OSCAR, mC4). Lower-resource corpora have systematic issues: At least 15 corpora have no usable text, and a significant fraction contains less than 50% sentences of acceptable quality. In addition, many are mislabeled or use nonstandard/ambiguous l…
Large text corpora are increasingly important for a wide variety of Natural Language Processing (NLP) tasks, and automatic language identification (LangID) is a core technology needed to collect such datasets in a multilingual context. LangID is largely treated as solved in the literature, with models reported that achieve over 90% average F1 on as many as 1,366 languages. We train LangID models on up to 1,629 languages with comparable quality on held-out test sets, but find that human-judged LangID accuracy for web-crawl text corpora created using these models is only around 5% for many lower…
Lurio University（卢里奥大学）；University of Cambridge（剑桥大学）
Major advancement in the performance of machine translation models has been made possible in part thanks to the availability of large-scale parallel corpora. But for most languages in the world, the existence of such corpora is rare. Emakhuwa, a language spoken in Mozambique, is like most African languages low-resource in NLP terms. It lacks both computational and linguistic resources and, to the best of our knowledge, few parallel corpora including Emakhuwa already exist. In this paper we describe the creation of the Emakhuwa-Portuguese parallel corpus, which is a collection of texts from the…
Bitdefender（比特梵德）；University of Bucharest（布加勒斯特大学）
The DarkWeb represents a hotbed for illicit activity, where users communicate on different market forums in order to exchange goods and services. Law enforcement agencies benefit from forensic tools that perform authorship analysis, in order to identify and profile users based on their textual content. However, authorship analysis has been traditionally studied using corpora featuring literary texts such as fragments from novels or fan fiction, which may not be suitable in a cybercrime context. Moreover, the few works that employ authorship analysis tools for cybercrime prevention usually empl…
Urgench State University（乌尔根奇国立大学）；Urgench Branch of Tashkent university of Information Technologies Named After Muhammad al-Khwarizmi（穆罕默德·花剌子米塔什干信息技术大学乌尔根奇分校）
Nowadays, creation of the tagged corpora is becoming one of the most important tasks of Natural Language Processing (NLP). There are not enough tagged corpora to build machine learning models for the low-resource Uzbek language. In this paper, we tried to fill that gap by developing a novel Part Of Speech (POS) and syntactic tagset for creating the syntactic and morphologically tagged corpus of the Uzbek language. This work also includes detailed description and presentation of a web-based application to work on a tagging as well. Based on the developed annotation tool and the software, we sha…
University of Pittsburgh（匹兹堡大学）；Carnegie Mellon University（卡内基梅隆大学）；The Ohio State University（俄亥俄州立大学）
Embeddings of words and concepts capture syntactic and semantic regularities of language; however, they have seen limited use as tools to study characteristics of different corpora and how they relate to one another. We introduce TextEssence, an interactive system designed to enable comparative analysis of corpora using embeddings. TextEssence includes visual, neighbor-based, and similarity-based modes of embedding analysis in a lightweight, web-based interface. We further propose a new measure of embedding confidence based on nearest neighborhood overlap, to assist in identifying high-quality…
Università degli Studi di Cagliari（卡利亚里大学）；Università degli Studi di Torino（都灵大学）；Dublin City University（都柏林城市大学）；University of Stuttgart（斯图加特大学）；Universitat Politècnica de València（瓦伦西亚理工大学）；University of Mannheim（曼海姆大学）；Leibniz-Institut für Deutsche Sprache Mannheim（曼海姆莱布尼茨德语研究所）；INRIA Paris（法国国家信息与自动化研究所巴黎中心）；Georgetown University（乔治城大学）
This article presents a discussion on the main linguistic phenomena which cause difficulties in the analysis of user-generated texts found on the web and in social media, and proposes a set of annotation guidelines for their treatment within the Universal Dependencies (UD) framework of syntactic analysis. Given on the one hand the increasing number of treebanks featuring user-generated content, and its somewhat inconsistent treatment in these resources on the other, the aim of this article is twofold: (1) to provide a condensed, though comprehensive, overview of such treebanks -- based on avai…
Centre for Translation Studies, University of Leeds（利兹大学翻译研究中心）
This paper proposes a novel framework for digital curation of Web corpora in order to provide robust estimation of their parameters, such as their composition and the lexicon. In recent years language models pre-trained on large corpora emerged as clear winners in numerous NLP tasks, but no proper analysis of the corpora which led to their success has been conducted. The paper presents a procedure for robust frequency estimation, which helps in establishing the core lexicon for a given corpus, as well as a procedure for estimating the corpus composition via unsupervised topic models and via su…
The use of attributed quotes is the most direct and least filtered pathway of information propagation in news. Consequently, quotes play a central role in the conception, reception, and analysis of news stories. Since quotes provide a more direct window into a speaker's mind than regular reporting, they are a valuable resource for journalists and researchers alike. While substantial research efforts have been devoted to methods for the automated extraction of quotes from news and their attribution to speakers, few comprehensive corpora of attributed quotes from contemporary sources are availab…
Mila/McGill University（Mila/麦吉尔大学）；Microsoft Research Montreal（微软研究院蒙特利尔分院）
The Winograd Schema Challenge (WSC) and variants inspired by it have become important benchmarks for common-sense reasoning (CSR). Model performance on the WSC has quickly progressed from chance-level to near-human using neural language models trained on massive corpora. In this paper, we analyze the effects of varying degrees of overlap between these training corpora and the test instances in WSC-style tasks. We find that a large number of test instances overlap considerably with the corpora on which state-of-the-art models are (pre)trained, and that a significant drop in classification accur…
Improving the quality of Web-mined Parallel Corpora of Low-Resource Languages using Debiasing Heuristics
Dept. of Computer Science and Engineering, University of Moratuwa（moratuwa大学计算机科学与工程系）；School of Mathematical and Computational Sciences, Massey University（massey大学数学与计算科学学院）
The University of Western Australia（西澳大利亚大学）；The University of Melbourne（墨尔本大学）
Entity extraction is an important task in text mining and natural language processing. A popular method for entity extraction is by comparing substrings from free text against a dictionary of entities. In this paper, we present several techniques as a post-processing step for improving the effectiveness of the existing entity extraction technique. These techniques utilise models trained with the web-scale corpora which makes our techniques robust and versatile. Experiments show that our techniques bring a notable improvement on efficiency and effectiveness.
iCoSys Institute, University of Applied Sciences and Arts Western Switzerland（瑞士西部应用科学与艺术大学iCoSys研究所）；University of Fribourg（弗里堡大学）；Swisscom AG（瑞士电信公司）
This paper presents SwissCrawl, the largest Swiss German text corpus to date. Composed of more than half a million sentences, it was generated using a customized web scraping tool that could be applied to other low-resource languages as well. The approach demonstrates how freely available web pages can be used to construct comprehensive text corpora, which are of fundamental importance for natural language processing. In an experimental evaluation, we show that using the new corpus leads to significant improvements for the task of language modeling. To capture new content, our approach will ru…
Princeton University（普林斯顿大学）；University of Bath（巴斯大学）
Artificial intelligence and machine learning are in a period of astounding growth. However, there are concerns that these technologies may be used, either with or without intention, to perpetuate the prejudice and unfairness that unfortunately characterizes many human institutions. Here we show for the first time that human-like semantic biases result from the application of standard machine learning to ordinary language---the same sort of language humans are exposed to every day. We replicate a spectrum of standard human biases as exposed by the Implicit Association Test and other well-known …
Large parallel corpora that are automatically obtained from the web, documents or elsewhere often exhibit many corrupted parts that are bound to negatively affect the quality of the systems and models that learn from these corpora. This paper describes frequent problems found in data and such data affects neural machine translation systems, as well as how to identify and deal with them. The solutions are summarised in a set of scripts that remove problematic sentences from input corpora.
The most effective paradigm for word sense disambiguation, supervised learning, seems to be stuck because of the knowledge acquisition bottleneck. In this paper we take an in-depth study of the performance of decision lists on two publicly available corpora and an additional corpus automatically acquired from the Web, using the fine-grained highly polysemous senses in WordNet. Decision lists are shown a versatile state-of-the-art technique. The experiments reveal, among other facts, that SemCor can be an acceptable (0.7 precision for polysemous words) starting point for an all-words system. Th…
Indian Institute of Technology Bombay（孟买印度理工学院）；Google Research（谷歌研究院）
Recent research in multilingual language models (LM) has demonstrated their ability to effectively handle multiple languages in a single model. This holds promise for low web-resource languages (LRL) as multilingual models can enable transfer of supervision from high resource languages to LRLs. However, incorporating a new language in an LM still remains a challenge, particularly for languages with limited corpora and in unseen scripts. In this paper we argue that relatedness among languages in a language family may be exploited to overcome some of the corpora limitations of LRLs, and propose …
