Pub Date : 2023-03-30DOI: 10.48550/arXiv.2303.17683
Aarohi Srivastava, David Chiang
In this work, we induce character-level noise in various forms when fine-tuning BERT to enable zero-shot cross-lingual transfer to unseen dialects and languages. We fine-tune BERT on three sentence-level classification tasks and evaluate our approach on an assortment of unseen dialects and languages. We find that character-level noise can be an extremely effective agent of cross-lingual transfer under certain conditions, while it is not as helpful in others. Specifically, we explore these differences in terms of the nature of the task and the relationships between source and target languages, finding that introduction of character-level noise during fine-tuning is particularly helpful when a task draws on surface level cues and the source-target cross-lingual pair has a relatively high lexical overlap with shorter (i.e., less meaningful) unseen tokens on average.
{"title":"Fine-Tuning BERT with Character-Level Noise for Zero-Shot Transfer to Dialects and Closely-Related Languages","authors":"Aarohi Srivastava, David Chiang","doi":"10.48550/arXiv.2303.17683","DOIUrl":"https://doi.org/10.48550/arXiv.2303.17683","url":null,"abstract":"In this work, we induce character-level noise in various forms when fine-tuning BERT to enable zero-shot cross-lingual transfer to unseen dialects and languages. We fine-tune BERT on three sentence-level classification tasks and evaluate our approach on an assortment of unseen dialects and languages. We find that character-level noise can be an extremely effective agent of cross-lingual transfer under certain conditions, while it is not as helpful in others. Specifically, we explore these differences in terms of the nature of the task and the relationships between source and target languages, finding that introduction of character-level noise during fine-tuning is particularly helpful when a task draws on surface level cues and the source-target cross-lingual pair has a relatively high lexical overlap with shorter (i.e., less meaningful) unseen tokens on average.","PeriodicalId":167439,"journal":{"name":"Workshop on NLP for Similar Languages, Varieties and Dialects","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132412383","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fernando Benites, M. Hürlimann, Pius von Däniken, Mark Cieliebak
We describe our approaches for the Social Media Geolocation (SMG) task at the VarDial Evaluation Campaign 2020. The goal was to predict geographical location (latitudes and longitudes) given an input text. There were three subtasks corresponding to German-speaking Switzerland (CH), Germany and Austria (DE-AT), and Croatia, Bosnia and Herzegovina, Montenegro and Serbia (BCMS). We submitted solutions to all subtasks but focused our development efforts on the CH subtask, where we achieved third place out of 16 submissions with a median distance of 15.93 km and had the best result of 14 unconstrained systems. In the DE-AT subtask, we ranked sixth out of ten submissions (fourth of 8 unconstrained systems) and for BCMS we achieved fourth place out of 13 submissions (second of 11 unconstrained systems).
{"title":"ZHAW-InIT - Social Media Geolocation at VarDial 2020","authors":"Fernando Benites, M. Hürlimann, Pius von Däniken, Mark Cieliebak","doi":"10.21256/ZHAW-21551","DOIUrl":"https://doi.org/10.21256/ZHAW-21551","url":null,"abstract":"We describe our approaches for the Social Media Geolocation (SMG) task at the VarDial Evaluation Campaign 2020. The goal was to predict geographical location (latitudes and longitudes) given an input text. There were three subtasks corresponding to German-speaking Switzerland (CH), Germany and Austria (DE-AT), and Croatia, Bosnia and Herzegovina, Montenegro and Serbia (BCMS). We submitted solutions to all subtasks but focused our development efforts on the CH subtask, where we achieved third place out of 16 submissions with a median distance of 15.93 km and had the best result of 14 unconstrained systems. In the DE-AT subtask, we ranked sixth out of ten submissions (fourth of 8 unconstrained systems) and for BCMS we achieved fourth place out of 13 submissions (second of 11 unconstrained systems).","PeriodicalId":167439,"journal":{"name":"Workshop on NLP for Similar Languages, Varieties and Dialects","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122659501","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Our submissions for the GDI 2017 Shared Task are the results from three different types of classifiers: Naïve Bayes, Conditional Random Fields (CRF), and Support Vector Machine (SVM). Our CRF-based run achieves a weighted F1 score of 65% (third rank) being beaten by the best system by 0.9%. Measured by classification accuracy, our ensemble run (Naïve Bayes, CRF, SVM) reaches 67% (second rank) being 1% lower than the best system. We also describe our experiments with Recurrent Neural Network (RNN) architectures. Since they performed worse than our non-neural approaches we did not include them in the submission.
{"title":"CLUZH at VarDial GDI 2017: Testing a Variety of Machine Learning Tools for the Classification of Swiss German Dialects","authors":"S. Clematide, Peter Makarov","doi":"10.18653/v1/W17-1221","DOIUrl":"https://doi.org/10.18653/v1/W17-1221","url":null,"abstract":"Our submissions for the GDI 2017 Shared Task are the results from three different types of classifiers: Naïve Bayes, Conditional Random Fields (CRF), and Support Vector Machine (SVM). Our CRF-based run achieves a weighted F1 score of 65% (third rank) being beaten by the best system by 0.9%. Measured by classification accuracy, our ensemble run (Naïve Bayes, CRF, SVM) reaches 67% (second rank) being 1% lower than the best system. We also describe our experiments with Recurrent Neural Network (RNN) architectures. Since they performed worse than our non-neural approaches we did not include them in the submission.","PeriodicalId":167439,"journal":{"name":"Workshop on NLP for Similar Languages, Varieties and Dialects","volume":"90 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122315794","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The present study has examined the similarity and the mutual intelligibility between Amharic and Tigrigna using three tools namely Levenshtein distance, intelligibility test and questionnaires. The study has shown that both Tigrigna varieties have almost equal phonetic and lexical distances from Amharic. The study also indicated that Amharic speakers understand less than 50% of the two varieties. Furthermore, the study showed that Amharic speakers are more positive about the Ethiopian Tigrigna variety than the Eritrean Variety. However, their attitude towards the two varieties does not have an impact on their intelligibility. The Amharic speakers’ familiarity to the Tigrigna varieties is largely dependent on the genealogical relation between Amharic and the two Tigrigna varieties.
{"title":"The similarity and Mutual Intelligibility between Amharic and Tigrigna Varieties","authors":"Tekabe Legesse Feleke","doi":"10.18653/v1/W17-1206","DOIUrl":"https://doi.org/10.18653/v1/W17-1206","url":null,"abstract":"The present study has examined the similarity and the mutual intelligibility between Amharic and Tigrigna using three tools namely Levenshtein distance, intelligibility test and questionnaires. The study has shown that both Tigrigna varieties have almost equal phonetic and lexical distances from Amharic. The study also indicated that Amharic speakers understand less than 50% of the two varieties. Furthermore, the study showed that Amharic speakers are more positive about the Ethiopian Tigrigna variety than the Eritrean Variety. However, their attitude towards the two varieties does not have an impact on their intelligibility. The Amharic speakers’ familiarity to the Tigrigna varieties is largely dependent on the genealogical relation between Amharic and the two Tigrigna varieties.","PeriodicalId":167439,"journal":{"name":"Workshop on NLP for Similar Languages, Varieties and Dialects","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126672106","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper describes the system developed by the Centre for English Corpus Linguistics (CECL) to discriminating similar languages, language varieties and dialects. Based on a SVM with character and POStag n-grams as features and the BM25 weighting scheme, it achieved 92.7% accuracy in the Discriminating between Similar Languages (DSL) task, ranking first among eleven systems but with a lead over the next three teams of only 0.2%. A simpler version of the system ranked second in the German Dialect Identification (GDI) task thanks to several ad hoc postprocessing steps. Complementary analyses carried out by a cross-validation procedure suggest that the BM25 weighting scheme could be competitive in this type of tasks, at least in comparison with the sublinear TF-IDF. POStag n-grams also improved the system performance.
{"title":"Improving the Character Ngram Model for the DSL Task with BM25 Weighting and Less Frequently Used Feature Sets","authors":"Yves Bestgen","doi":"10.18653/v1/W17-1214","DOIUrl":"https://doi.org/10.18653/v1/W17-1214","url":null,"abstract":"This paper describes the system developed by the Centre for English Corpus Linguistics (CECL) to discriminating similar languages, language varieties and dialects. Based on a SVM with character and POStag n-grams as features and the BM25 weighting scheme, it achieved 92.7% accuracy in the Discriminating between Similar Languages (DSL) task, ranking first among eleven systems but with a lead over the next three teams of only 0.2%. A simpler version of the system ranked second in the German Dialect Identification (GDI) task thanks to several ad hoc postprocessing steps. Complementary analyses carried out by a cross-validation procedure suggest that the BM25 weighting scheme could be competitive in this type of tasks, at least in comparison with the sublinear TF-IDF. POStag n-grams also improved the system performance.","PeriodicalId":167439,"journal":{"name":"Workshop on NLP for Similar Languages, Varieties and Dialects","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131589157","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Marcos Zampieri, S. Malmasi, Nikola Ljubesic, Preslav Nakov, Ahmed Ali, J. Tiedemann, Yves Scherrer, Noëmi Aepli
We present the results of the VarDial Evaluation Campaign on Natural Language Processing (NLP) for Similar Languages, Varieties and Dialects, which we organized as part of the fourth edition of the VarDial workshop at EACL’2017. This year, we included four shared tasks: Discriminating between Similar Languages (DSL), Arabic Dialect Identification (ADI), German Dialect Identification (GDI), and Cross-lingual Dependency Parsing (CLP). A total of 19 teams submitted runs across the four tasks, and 15 of them wrote system description papers.
{"title":"Findings of the VarDial Evaluation Campaign 2017","authors":"Marcos Zampieri, S. Malmasi, Nikola Ljubesic, Preslav Nakov, Ahmed Ali, J. Tiedemann, Yves Scherrer, Noëmi Aepli","doi":"10.18653/v1/W17-1201","DOIUrl":"https://doi.org/10.18653/v1/W17-1201","url":null,"abstract":"We present the results of the VarDial Evaluation Campaign on Natural Language Processing (NLP) for Similar Languages, Varieties and Dialects, which we organized as part of the fourth edition of the VarDial workshop at EACL’2017. This year, we included four shared tasks: Discriminating between Similar Languages (DSL), Arabic Dialect Identification (ADI), German Dialect Identification (GDI), and Cross-lingual Dependency Parsing (CLP). A total of 19 teams submitted runs across the four tasks, and 15 of them wrote system description papers.","PeriodicalId":167439,"journal":{"name":"Workshop on NLP for Similar Languages, Varieties and Dialects","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127195579","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper investigates diatopic variation in a historical corpus of German. Based on equivalent word forms from different language areas, replacement rules and mappings are derived which describe the relations between these word forms. These rules and mappings are then interpreted as reflections of morphological, phonological or graphemic variation. Based on sample rules and mappings, we show that our approach can replicate results from historical linguistics. While previous studies were restricted to predefined word lists, or confined to single authors or texts, our approach uses a much wider range of data available in historical corpora.
{"title":"Investigating Diatopic Variation in a Historical Corpus","authors":"Stefanie Dipper, Sandra Waldenberger","doi":"10.18653/v1/W17-1204","DOIUrl":"https://doi.org/10.18653/v1/W17-1204","url":null,"abstract":"This paper investigates diatopic variation in a historical corpus of German. Based on equivalent word forms from different language areas, replacement rules and mappings are derived which describe the relations between these word forms. These rules and mappings are then interpreted as reflections of morphological, phonological or graphemic variation. Based on sample rules and mappings, we show that our approach can replicate results from historical linguistics. While previous studies were restricted to predefined word lists, or confined to single authors or texts, our approach uses a much wider range of data available in historical corpora.","PeriodicalId":167439,"journal":{"name":"Workshop on NLP for Similar Languages, Varieties and Dialects","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124974357","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We present a machine learning approach for the Arabic Dialect Identification (ADI) and the German Dialect Identification (GDI) Closed Shared Tasks of the DSL 2017 Challenge. The proposed approach combines several kernels using multiple kernel learning. While most of our kernels are based on character p-grams (also known as n-grams) extracted from speech transcripts, we also use a kernel based on i-vectors, a low-dimensional representation of audio recordings, provided only for the Arabic data. In the learning stage, we independently employ Kernel Discriminant Analysis (KDA) and Kernel Ridge Regression (KRR). Our approach is shallow and simple, but the empirical results obtained in the shared tasks prove that it achieves very good results. Indeed, we ranked on the first place in the ADI Shared Task with a weighted F1 score of 76.32% (4.62% above the second place) and on the fifth place in the GDI Shared Task with a weighted F1 score of 63.67% (2.57% below the first place).
{"title":"Learning to Identify Arabic and German Dialects using Multiple Kernels","authors":"Radu Tudor Ionescu, Andrei M. Butnaru","doi":"10.18653/v1/W17-1225","DOIUrl":"https://doi.org/10.18653/v1/W17-1225","url":null,"abstract":"We present a machine learning approach for the Arabic Dialect Identification (ADI) and the German Dialect Identification (GDI) Closed Shared Tasks of the DSL 2017 Challenge. The proposed approach combines several kernels using multiple kernel learning. While most of our kernels are based on character p-grams (also known as n-grams) extracted from speech transcripts, we also use a kernel based on i-vectors, a low-dimensional representation of audio recordings, provided only for the Arabic data. In the learning stage, we independently employ Kernel Discriminant Analysis (KDA) and Kernel Ridge Regression (KRR). Our approach is shallow and simple, but the empirical results obtained in the shared tasks prove that it achieves very good results. Indeed, we ranked on the first place in the ADI Shared Task with a weighted F1 score of 76.32% (4.62% above the second place) and on the fifth place in the GDI Shared Task with a weighted F1 score of 63.67% (2.57% below the first place).","PeriodicalId":167439,"journal":{"name":"Workshop on NLP for Similar Languages, Varieties and Dialects","volume":"153 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120949442","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This research suggests a method for machine translation among two Kurdish dialects. We chose the two widely spoken dialects, Kurmanji and Sorani, which are considered to be mutually unintelligible. Also, despite being spoken by about 30 million people in different countries, Kurdish is among less-resourced languages. The research used bi-dialectal dictionaries and showed that the lack of parallel corpora is not a major obstacle in machine translation between the two dialects. The experiments showed that the machine translated texts are comprehensible to those who do not speak the dialect. The research is the first attempt for inter-dialect machine translation in Kurdish and particularly could help in making online texts in one dialect comprehensible to those who only speak the target dialect. The results showed that the translated texts are in 71% and 79% cases rated as understandable for Kurmanji and Sorani respectively. They are rated as slightly-understandable in 29% cases for Kurmanji and 21% for Sorani.
{"title":"Kurdish Interdialect Machine Translation","authors":"Hossein Hassani","doi":"10.18653/v1/W17-1208","DOIUrl":"https://doi.org/10.18653/v1/W17-1208","url":null,"abstract":"This research suggests a method for machine translation among two Kurdish dialects. We chose the two widely spoken dialects, Kurmanji and Sorani, which are considered to be mutually unintelligible. Also, despite being spoken by about 30 million people in different countries, Kurdish is among less-resourced languages. The research used bi-dialectal dictionaries and showed that the lack of parallel corpora is not a major obstacle in machine translation between the two dialects. The experiments showed that the machine translated texts are comprehensible to those who do not speak the dialect. The research is the first attempt for inter-dialect machine translation in Kurdish and particularly could help in making online texts in one dialect comprehensible to those who only speak the target dialect. The results showed that the translated texts are in 71% and 79% cases rated as understandable for Kurmanji and Sorani respectively. They are rated as slightly-understandable in 29% cases for Kurmanji and 21% for Sorani.","PeriodicalId":167439,"journal":{"name":"Workshop on NLP for Similar Languages, Varieties and Dialects","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133575145","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Helena Gómez-Adorno, I. Markov, J. Baptista, G. Sidorov, David Pinto
This paper presents the cic_ualg’s system that took part in the Discriminating between Similar Languages (DSL) shared task, held at the VarDial 2017 Workshop. This year’s task aims at identifying 14 languages across 6 language groups using a corpus of excerpts of journalistic texts. Two classification approaches were compared: a single-step (all languages) approach and a two-step (language group and then languages within the group) approach. Features exploited include lexical features (unigrams of words) and character n-grams. Besides traditional (untyped) character n-grams, we introduce typed character n-grams in the DSL task. Experiments were carried out with different feature representation methods (binary and raw term frequency), frequency threshold values, and machine-learning algorithms – Support Vector Machines (SVM) and Multinomial Naive Bayes (MNB). Our best run in the DSL task achieved 91.46% accuracy.
{"title":"Discriminating between Similar Languages Using a Combination of Typed and Untyped Character N-grams and Words","authors":"Helena Gómez-Adorno, I. Markov, J. Baptista, G. Sidorov, David Pinto","doi":"10.18653/v1/W17-1217","DOIUrl":"https://doi.org/10.18653/v1/W17-1217","url":null,"abstract":"This paper presents the cic_ualg’s system that took part in the Discriminating between Similar Languages (DSL) shared task, held at the VarDial 2017 Workshop. This year’s task aims at identifying 14 languages across 6 language groups using a corpus of excerpts of journalistic texts. Two classification approaches were compared: a single-step (all languages) approach and a two-step (language group and then languages within the group) approach. Features exploited include lexical features (unigrams of words) and character n-grams. Besides traditional (untyped) character n-grams, we introduce typed character n-grams in the DSL task. Experiments were carried out with different feature representation methods (binary and raw term frequency), frequency threshold values, and machine-learning algorithms – Support Vector Machines (SVM) and Multinomial Naive Bayes (MNB). Our best run in the DSL task achieved 91.46% accuracy.","PeriodicalId":167439,"journal":{"name":"Workshop on NLP for Similar Languages, Varieties and Dialects","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134509536","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：