Statistical analysis of word flow among five Indo-European languages
Facultad de Ciencias, Universidad Nacional Autónoma de México（墨西哥国立自治大学理学院）；Instituto de Física, Universidad Nacional Autónoma de México（墨西哥国立自治大学物理研究所）；Instituto de Investigaciones en Matemáticas Aplicadas y Sistemas, Universidad Nacional Autónoma de México（墨西哥国立自治大学应用数学与系统研究所）；Centro de Ciencias de la Complejidad, Universidad Nacional Autónoma de México（墨西哥国立自治大学复杂性科学中心）；Santa Fe Institute（圣塔菲研究所）
Josué Ely Molina, Jorge Flores, Carlos Gershenson, Carlos Pineda
A recent increase in data availability has allowed the possibility to perform different statistical linguistic studies. Here we use the Google Books Ngram dataset to analyze word flow among English, French, German, Italian, and Spanish. We study what we define as ``migrant words'', a type of loanwords that do not change their spelling. We quantify migrant words from one language to another for different decades, and notice that most migrant words can be aggregated in semantic fields and associated to historic events. We also study the statistical properties of accumulated migrant words and their rank dynamics. We propose a measure of use of migrant words that could be used as a proxy of cultural influence. Our methodology is not exempt of caveats, but our results are encouraging to promote further studies in this direction.
University of Oslo（奥斯陆大学）；Department of Informatics（信息学系）
The age of the root of the Indo-European language family has received much attention since the application of Bayesian phylogenetic methods by Gray and Atkinson(2003). The root age of the Indo-European family has tended to decrease from an age that supported the Anatolian origin hypothesis to an age that supports the Steppe origin hypothesis with the application of new models (Chang et al., 2015). However, none of the published work in the Indo-European phylogenetics studied the effect of tree priors on phylogenetic analyses of the Indo-European family. In this paper, I intend to fill this gap…
Using Phylogenetic Algebraic Geometry, we analyze computationally the phylogenetic tree of subfamilies of the Indo-European language family, using data of syntactic structures. The two main sources of syntactic data are the SSWL database and Longobardi's recent data of syntactic parameters. We compute phylogenetic invariants and likelihood functions for two sets of Germanic languages, a set of Romance languages, a set of Slavic languages and a set of early Indo-European languages, and we compare the results with what is known through historical linguistics.
This paper describes Charles University submission for Multilingual Low-Resource Translation for Indo-European Languages shared task at WMT21. We competed in translation from Catalan into Romanian, Italian and Occitan. Our systems are based on shared multilingual model. We show that using joint model for multiple similar language pairs improves upon translation quality in each pair. We also demonstrate that chararacter-level bilingual models are competitive for very similar language pairs (Catalan-Occitan) but less so for more distant pairs. We also describe our experiments with multi-task lea…
Natural Language Processing Lab（自然语言处理实验室）；The University of British Columbia（英属哥伦比亚大学）
In this work, we investigate methods for the challenging task of translating between low-resource language pairs that exhibit some level of similarity. In particular, we consider the utility of transfer learning for translating between several Indo-European low-resource languages from the Germanic and Romance language families. In particular, we build two main classes of transfer-based systems to study how relatedness can benefit the translation performance. The primary system fine-tunes a model pre-trained on a related language pair and the contrastive system fine-tunes one pre-trained on an …
It is reasonable to hypothesize that the divergence patterns formulated by historical linguists and typologists reflect constraints on human languages, and are thus consistent with Second Language Acquisition (SLA) in a certain way. In this paper, we validate this hypothesis on ten Indo-European languages. We formalize the delexicalized transfer as interpretable tree-to-string and tree-to-tree patterns which can be automatically induced from web data by applying neural syntactic parsing and grammar induction technologies. This allows us to quantitatively probe cross-linguistic transfer and ext…
Indonesian is an agglutinative language since it has a compounding process of word-formation. Therefore, the translation model of this language requires a mechanism that is even lower than the word level, referred to as the sub-word level. This compounding process leads to a rare word problem since the number of vocabulary explodes. We propose a strategy to address the unique word problem of the neural machine translation (NMT) system, which uses Indonesian as a pair language. Our approach uses a rule-based method to transform a word into its roots and accompanied affixes to retain its meaning…
Quantitative Measurement of Language Distance among Closely Related Indo-European Languages Using Pretrained Language Models: A Case Study on the North Germanic Branch
We study the persistent homology of the data set of syntactic parameters of the world languages. We show that, while homology generators behave erratically over the whole data set, non-trivial persistent homology appears when one restricts to specific language families. Different families exhibit different persistent homology. We focus on the cases of the Indo-European and the Niger-Congo families, for which we compare persistent homology over different cluster filtering values. We investigate the possible significance, in historical linguistic terms, of the presence of persistent generators o…
MICE is a corpus of emotion words in four languages which is currently working progress. There are two sections to this study, Part I: Emotion word corpus and Part II: Emotion word survey. In Part 1, the method of how the emotion data is culled for each of the four languages will be described and very preliminary data will be presented. In total, we identified 3,750 emotion expressions in Malay, 6,657 in Indonesian, 3,347 in Mandarin Chinese and 8,683 in English. We are currently evaluating and double checking the corpus and doing further analysis on the distribution of these emotion expressio…
A simple machine learning model of pluralisation as a linear regression problem minimising a p-adic metric substantially outperforms even the most robust of Euclidean-space regressors on languages in the Indo-European, Austronesian, Trans New-Guinea, Sino-Tibetan, Nilo-Saharan, Oto-Meanguean and Atlantic-Congo language families. There is insufficient evidence to support modelling distinct noun declensions as a p-adic neighbourhood even in Indo-European languages.
University of Hildesheim（希尔德斯海姆大学）；LDRP-ITR；University of Duisburg-Essen（杜伊斯堡-埃森大学）；University of Bedfordshire（贝德福德大学）；University of Leeds（利兹大学）；University of Bamberg（班贝格大学）；Dalhousie University（达尔豪斯大学）；DA-IICT
With the growth of social media, the spread of hate speech is also increasing rapidly. Social media are widely used in many countries. Also Hate Speech is spreading in these countries. This brings a need for multilingual Hate Speech detection algorithms. Much research in this area is dedicated to English at the moment. The HASOC track intends to provide a platform to develop and optimize Hate Speech detection algorithms for Hindi, German and English. The dataset is collected from a Twitter archive and pre-classified by a machine learning system. HASOC has two sub-task for all three languages: …
Prosa Solusi Cerdas（普罗萨智能解决方案公司）；School of Electrical Engineering and Informatics（电气与信息工程学院）；Institut Teknologi Bandung（万隆理工学院）
Researches on Indonesian named entity (NE) tagger have been conducted since years ago. However, most did not use deep learning and instead employed traditional machine learning algorithms such as association rule, support vector machine, random forest, naïve bayes, etc. In those researches, word lists as gazetteers or clue words were provided to enhance the accuracy. Here, we attempt to employ deep learning in our Indonesian NE tagger. We use long short-term memory (LSTM) as the topology since it is the state-of-the-art of NE tagger. By using LSTM, we do not need a word list in order to enhanc…
Institute of Neuroinformatics, University of Zürich and ETH Zürich（苏黎世大学与苏黎世联邦理工学院神经信息学研究所）
Character-level Neural Machine Translation (NMT) models have recently achieved impressive results on many language pairs. They mainly do well for Indo-European language pairs, where the languages share the same writing system. However, for translating between Chinese and English, the gap between the two different writing systems poses a major challenge because of a lack of systematic correspondence between the individual linguistic units. In this paper, we enable character-level NMT for Chinese, by breaking down Chinese characters into linguistic units similar to that of Indo-European language…
European Commission, Joint Research Centre (JRC)（欧盟委员会联合研究中心）；Universitat Pompeu Fabra（庞培法布拉大学）；KAJO
This paper describes a prototype system that integrates social media analysis into the European Flood Awareness System (EFAS). This integration allows the collection of social media data to be automatically triggered by flood risk warnings determined by a hydro-meteorological model. Then, we adopt a multi-lingual approach to find flood-related messages by employing two state-of-the-art methodologies: language-agnostic word embeddings and language-aligned word embeddings. Both approaches can be used to bootstrap a classifier of social media messages for a new language with little or no labeled …
Facebook AI Research（Facebook人工智能研究院）；Tel Aviv University（特拉维夫大学）
Unsupervised word translation from non-parallel inter-lingual corpora has attracted much research interest. Very recently, neural network methods trained with adversarial loss functions achieved high accuracy on this task. Despite the impressive success of the recent techniques, they suffer from the typical drawbacks of generative adversarial models: sensitivity to hyper-parameters, long training time and lack of interpretability. In this paper, we make the observation that two sufficiently similar distributions can be aligned correctly with iterative matching methods. We present a novel metho…
University of Bucharest（布加勒斯特大学）；University of Wolverhampton（伍尔弗汉普顿大学）；Harvard Medical School（哈佛医学院）；Saarland University（萨尔大学）
In this paper we present a system based on SVM ensembles trained on characters and words to discriminate between five similar languages of the Indo-Aryan family: Hindi, Braj Bhasha, Awadhi, Bhojpuri, and Magahi. We investigate the performance of individual features and combine the output of single classifiers to maximize performance. The system competed in the Indo-Aryan Language Identification (ILI) shared task organized within the VarDial Evaluation Campaign 2018. Our best entry in the competition, named ILIdentification, scored 88:95% F1 score and it was ranked 3rd out of 8 teams.
University of Trento（特伦托大学）；Fondazione Bruno Kessler（布鲁诺·凯斯勒基金会）；MMT Srl（MMT有限公司）
Both research and commercial machine translation have so far neglected the importance of properly handling the spelling, lexical and grammar divergences occurring among language varieties. Notable cases are standard national varieties such as Brazilian and European Portuguese, and Canadian and European French, which popular online machine translation services are not keeping distinct. We show that an evident side effect of modeling such varieties as unique classes is the generation of inconsistent translations. In this work, we investigate the problem of training neural machine translation fro…
Nanyang Technological University（南洋理工大学）；Salesforce Research Asia（赛富时亚洲研究院）
Adversarial training has shown impressive success in learning bilingual dictionary without any parallel data by mapping monolingual embeddings to a shared space. However, recent work has shown superior performance for non-adversarial methods in more challenging language pairs. In this work, we revisit adversarial autoencoder for unsupervised word translation and propose two novel extensions to it that yield more stable training and improved results. Our method includes regularization terms to enforce cycle consistency and input reconstruction, and puts the target encoders as an adversary again…
This article is devoted to the verification of the empirical Heaps law in European languages using Google Books Ngram corpus data. The connection between word distribution frequency and expected dependence of individual word number on text size is analysed in terms of a simple probability model of text generation. It is shown that the Heaps exponent varies significantly within characteristic time intervals of 60-100 years.
University of São Paulo（圣保罗大学）；Institute of Mathematics and Computer Sciences（数学与计算机科学学院）；Federal University of São Carlos（圣卡洛斯联邦大学）
Word embeddings have been found to provide meaningful representations for words in an efficient way; therefore, they have become common in Natural Language Processing sys- tems. In this paper, we evaluated different word embedding models trained on a large Portuguese corpus, including both Brazilian and European variants. We trained 31 word embedding models using FastText, GloVe, Wang2Vec and Word2Vec. We evaluated them intrinsically on syntactic and semantic analogies and extrinsically on POS tagging and sentence semantic similarity tasks. The obtained results suggest that word analogies are …
Smoothing is one technique to overcome data sparsity in statistical language model. Although in its mathematical definition there is no explicit dependency upon specific natural language, different natures of natural languages result in different effects of smoothing techniques. This is true for Russian language as shown by Whittaker (1998). In this paper, We compared Modified Kneser-Ney and Witten-Bell smoothing techniques in statistical language model of Bahasa Indonesia. We used train sets of totally 22M words that we extracted from Indonesian version of Wikipedia. As far as we know, this i…
The most studied and most successful language models were developed and evaluated mainly for English and other close European languages, such as French, German, etc. It is important to study applicability of these models to other languages. The use of vector space models for Russian was recently studied for multiple corpora, such as Wikipedia, RuWac, lib.ru. These models were evaluated against word semantic similarity task. For our knowledge Twitter was not considered as a corpus for this task, with this work we fill the gap. Results for vectors trained on Twitter corpus are comparable in accu…
Insight Centre for Data Analytics（洞察数据分析中心）；Data Science Institute（数据科学研究所）；National University of Ireland Galway（爱尔兰国立高威大学）
Princeton WordNet is one of the most important resources for natural language processing, but is only available for English. While it has been translated using the expand approach to many other languages, this is an expensive manual process. Therefore it would be beneficial to have a high-quality automatic translation approach that would support NLP techniques, which rely on WordNet in new languages. The translation of wordnets is fundamentally complex because of the need to translate all senses of a word including low frequency senses, which is very challenging for current machine translation…
We describe a cross-lingual transfer method for dependency parsing that takes into account the problem of word order differences between source and target languages. Our model only relies on the Bible, a considerably smaller parallel data than the commonly used parallel data in transfer methods. We use the concatenation of projected trees from the Bible corpus, and the gold-standard treebanks in multiple source languages along with cross-lingual word representations. We demonstrate that reordering the source treebanks before training on them for a target language improves the accuracy of langu…
How does word frequency in pre-training data affect the behavior of similarity metrics in contextualized BERT embeddings? Are there systematic ways in which some word relationships are exaggerated or understated? In this work, we explore the geometric characteristics of contextualized word embeddings with two novel tools: (1) an identity probe that predicts the identity of a word using its embedding; (2) the minimal bounding sphere for a word's contextualized representations. Our results reveal that words of high and low frequency differ significantly with respect to their representational geo…
How do neural networks "perceive" speech sounds from unknown languages? Does the typological similarity between the model's training language (L1) and an unknown language (L2) have an impact on the model representations of L2 speech signals? To answer these questions, we present a novel experimental design based on representational similarity analysis (RSA) to analyze acoustic word embeddings (AWEs) -- vector representations of variable-duration spoken-word segments. First, we train monolingual AWE models on seven Indo-European languages with various degrees of typological similarity. We then …
Crime & Security Research Institute（犯罪与安全研究院）；School of Computer Science and Informatics（计算机科学与信息学院）；Cardiff University（卡迪夫大学）
This paper demonstrates a two-stage method for deriving insights from social media data relating to disinformation by applying a combination of geospatial classification and embedding-based language modelling across multiple languages. In particular, the analysis in centered on Twitter and disinformation for three European languages: English, French and Spanish. Firstly, Twitter data is classified into European and non-European sets using BERT. Secondly, Word2vec is applied to the classified texts resulting in Eurocentric, non-Eurocentric and global representations of the data for the three ta…
National Technical University of Athens（雅典国立技术大学）；NCSR Demokritos（德谟克利特国家科学研究中心）；National Observatory of Athens（雅典国家天文台）；University of Athens（雅典大学）；Athena R.C.（雅典娜研究中心）
We study the frequency distributions and correlations of the word lengths of ten European languages. Our findings indicate that a) the word-length distribution of short words quantified by the mean value and the entropy distinguishes the Uralic (Finnish) corpus from the others, b) the tails at long words, manifested in the high-order moments of the distributions, differentiate the Germanic languages (except for English) from the Romanic languages and Greek and c) the correlations between nearby word lengths measured by the comparison of the real entropies with those of the shuffled texts are f…
ENEA, Italian National Agency for New Technologies, Energy and Sustainable Economic Development（意大利国家新技术、能源与可持续经济发展署（ENEA））；Università di Roma “Sapienza”（罗马第一大学（Sapienza 罗马大学））；Centre for Statistical Mechanics and Complexity (SMC), CNR-INFM（意大利国家研究委员会-国家物质物理研究所 统计力学与复杂性中心（SMC））；INFN（意大利国家核物理研究所（INFN））；Università di Bari（巴里大学）
We classify twenty-one Indo-European languages starting from written text. We use neural networks in order to define a distance among different languages, construct a dendrogram and analyze the ultrametric structure that emerges. Four or five subgroups of languages are identified, according to the "cut" of the dendrogram, drawn with an entropic criterion. The results and the method are discussed.
Indian Institute of Technology Bombay（印度理工学院孟买分校）；IITB-Monash Research Academy（印度理工学院孟买分校-莫纳什大学联合研究学院）；Monash University（莫纳什大学）
Automatic Cognate Detection (ACD) is a challenging task which has been utilized to help NLP applications like Machine Translation, Information Retrieval and Computational Phylogenetics. Unidentified cognate pairs can pose a challenge to these applications and result in a degradation of performance. In this paper, we detect cognate word pairs among ten Indian languages with Hindi and use deep learning methodologies to predict whether a word pair is cognate or not. We identify IndoWordnet as a potential resource to detect cognate word pairs based on orthographic similarity-based methods and trai…
Universidad Nacional Autónoma de México（墨西哥国立自治大学）
Statistical studies of languages have focused on the rank-frequency distribution of words. Instead, we introduce here a measure of how word ranks change in time and call this distribution \emph{rank diversity}. We calculate this diversity for books published in six European languages since 1800, and find that it follows a universal lognormal distribution. Based on the mean and standard deviation associated with the lognormal distribution, we define three different word regimes of languages: "heads" consist of words which almost do not change their rank in time, "bodies" are words of general us…
PRHLT Research Center, Universitat Politècnica de València（PRHLT研究中心，瓦伦西亚理工大学）；AUSY R&D（奥西研发中心）；IRIT, CNRS, Université de Toulouse（图卢兹大学计算机研究所，法国国家科学研究中心）
This paper proposes the first multilingual (French, English and Arabic) and multicultural (Indo-European languages vs. less culturally close languages) irony detection system. We employ both feature-based models and neural architectures using monolingual word representation. We compare the performance of these systems with state-of-the-art systems to identify their capabilities. We show that these monolingual models trained separately on different languages using multilingual word representation or text-based features can open the door to irony detection in languages that lack of annotated dat…
Bilingual lexicons and phrase tables are critical resources for modern Machine Translation systems. Although recent results show that without any seed lexicon or parallel data, highly accurate bilingual lexicons can be learned using unsupervised methods, such methods rely on the existence of large, clean monolingual corpora. In this work, we utilize a single Skip-gram model trained on a multilingual corpus yielding polyglot word embeddings, and present a novel finding that a surprisingly simple constrained nearest-neighbor sampling technique in this embedding space can retrieve bilingual lexic…
George Washington University（乔治华盛顿大学）；University of Washington（华盛顿大学）
Does the grammatical gender of a language interfere when measuring the semantic gender information captured by its word embeddings? A number of anomalous gender bias measurements in the embeddings of gendered languages suggest this possibility. We demonstrate that word embeddings learn the association between a noun and its grammatical gender in grammatically gendered languages, which can skew social gender bias measurements. Consequently, word embedding post-processing methods are introduced to quantify, disentangle, and evaluate grammatical gender signals. The evaluation is performed on five…
Institute of Polish Language, Polish Academy of Sciences（波兰科学院波兰语言研究所）；University of Tartu（塔尔图大学）；Institute of Czech Literature, Czech Academy of Sciences（捷克科学院捷克文学研究所）；Leiden University（莱顿大学）
Recent advances in cultural analytics and large-scale computational studies of art, literature and film often show that long-term change in the features of artistic works happens gradually. These findings suggest that conservative forces that shape creative domains might be underestimated. To this end, we provide the first large-scale formal evidence of the persistent association between poetic meter and semantics in 18-19th European literatures, using Czech, German and Russian collections with additional data from English poetry and early modern Dutch songs. Our study traces this association …
Insight Centre for Data Analytics, National University of Ireland Galway（爱尔兰国立大学戈尔韦分校Insight数据分析中心）；University of Kurdistan Hewlêr（库尔德斯坦埃尔比勒大学）
Morphological analysis is the study of the formation and structure of words. It plays a crucial role in various tasks in Natural Language Processing (NLP) and Computational Linguistics (CL) such as machine translation and text and speech generation. Kurdish is a less-resourced multi-dialect Indo-European language with highly inflectional morphology. In this paper, as the first attempt of its kind, the morphology of the Kurdish language (Sorani dialect) is described from a computational point of view. We extract morphological rules which are transformed into finite-state transducers for generat…
More than 2 billion mobile users worldwide type in multiple languages in the soft keyboard. On a monolingual keyboard, 38% of falsely auto-corrected words are valid in another language. This can be easily avoided by detecting the language of typed words and then validating it in its respective language. Language detection is a well-known problem in natural language processing. In this paper, we present a fast, light-weight and accurate Language Detection Engine (LDE) for multilingual typing that dynamically adapts to user intended language in real-time. We propose a novel approach where the fu…
Social bias in language - towards genders, ethnicities, ages, and other social groups - poses a problem with ethical impact for many NLP applications. Recent research has shown that machine learning models trained on respective data may not only adopt, but even amplify the bias. So far, however, little attention has been paid to bias in computational argumentation. In this paper, we study the existence of social biases in large English debate portals. In particular, we train word embedding models on portal-specific corpora and systematically evaluate their bias using WEAT, an existing metric t…
Indian Institute of Technology (BHU)（印度理工学院（瓦拉纳西印度大学分校））
The ambiguities introduced by the recombination of morphemes constructing several possible inflections for a word makes the prediction of syntactic traits in Morphologically Rich Languages (MRLs) a notoriously complicated task. We propose the Multi Task Deep Morphological analyzer (MT-DMA), a character-level neural morphological analyzer based on multitask learning of word-level tag markers for Hindi and Urdu. MT-DMA predicts a set of six morphological tags for words of Indo-Aryan languages: Parts-of-speech (POS), Gender (G), Number (N), Person (P), Case (C), Tense-Aspect-Modality (TAM) marker…
On the Cross-lingual Transferability of Pre-trained wav2vec2-based Models
Department of Informatics（信息学院）；Pontifical Catholic University of Rio de Janeiro（里约热内卢天主教大学）；National School of Statistical Sciences（国家统计科学学校）；Brazilian Institute of Geography and Statistics（巴西地理与统计研究所）；Institute of Systems and Robotics（系统与机器人研究所）；University of Coimbra（科英布拉大学）
Institute of Nuclear Physics, Polish Academy of Sciences（波兰科学院核物理研究所）；Cracow University of Technology（克拉科夫理工大学）
From a grammar point of view, the role of punctuation marks in a sentence is formally defined and well understood. In semantic analysis punctuation plays also a crucial role as a method of avoiding ambiguity of the meaning. A different situation can be observed in the statistical analyses of language samples, where the decision on whether the punctuation marks should be considered or should be neglected is seen rather as arbitrary and at present it belongs to a researcher's preference. An objective of this work is to shed some light onto this problem by providing us with an answer to the quest…
