Pub Date : 2022-05-01DOI: 10.1016/j.lana.2022.100223
Jeffrey W Brettler, Gloria P Giraldo Arcila, Teresa Aumala, Allana Best, Norm Rc Campbell, Shana Cyr, Angelo Gamarra, Marc G Jaffe, Mirna Jimenez De la Rosa, Javier Maldonado, Carolina Neira Ojeda, Modesta Haughton, Taraleen Malcolm, Vivian Perez, Gonzalo Rodriguez, Andres Rosende, Yamilé Valdés González, Peter W Wood, Eric Zúñiga, Pedro Ordunez
Background: Cardiovascular disease (CVD) is the leading cause of morbidity and mortality in the Americas, and hypertension is the most significant modifiable risk factor. However, hypertension control rates remain low, and CVD mortality is stagnant or rising after decades of continuing reduction. In 2016, the World Health Organization (WHO) launched the HEARTS technical package to improve hypertension control. The Pan American Health Organization (PAHO) designed the HEARTS in the Americas Initiative to improve CVD risk management, emphasizing hypertension control, to date implemented in 21 countries.
Methods: To advance implementation, an interdisciplinary group of practitioners was engaged to select the key evidence-based drivers of hypertension control and to design a comprehensive scorecard to monitor their implementation at primary care health facilities (PHC). The group studied high-performing health systems that achieve high hypertension control through quality improvement programs focusing on specific process measures, with regular feedback to providers at health facilities.
Findings: The final selected eight drivers were categorized into five main domains: (1) diagnosis (blood pressure measurement accuracy and CVD risk evaluation); (2) treatment (standardized treatment protocol and treatment intensification); (3) continuity of care and follow-up; (4) delivery system (team-based care, medication refill), and (5) system for performance evaluation. The drivers and recommendations were then translated into process measures, resulting in two interconnected scorecards integrated into the HEARTS in the Americas monitoring and evaluation system.
Interpretation: Focus on these key hypertension drivers and resulting scorecards, will guide the quality improvement process to achieve population control goals at the participating health centers in HEARTS implementing countries.
{"title":"Drivers and scorecards to improve hypertension control in primary care practice: Recommendations from the HEARTS in the Americas Innovation Group.","authors":"Jeffrey W Brettler, Gloria P Giraldo Arcila, Teresa Aumala, Allana Best, Norm Rc Campbell, Shana Cyr, Angelo Gamarra, Marc G Jaffe, Mirna Jimenez De la Rosa, Javier Maldonado, Carolina Neira Ojeda, Modesta Haughton, Taraleen Malcolm, Vivian Perez, Gonzalo Rodriguez, Andres Rosende, Yamilé Valdés González, Peter W Wood, Eric Zúñiga, Pedro Ordunez","doi":"10.1016/j.lana.2022.100223","DOIUrl":"10.1016/j.lana.2022.100223","url":null,"abstract":"<p><strong>Background: </strong>Cardiovascular disease (CVD) is the leading cause of morbidity and mortality in the Americas, and hypertension is the most significant modifiable risk factor. However, hypertension control rates remain low, and CVD mortality is stagnant or rising after decades of continuing reduction. In 2016, the World Health Organization (WHO) launched the HEARTS technical package to improve hypertension control. The Pan American Health Organization (PAHO) designed the HEARTS in the Americas Initiative to improve CVD risk management, emphasizing hypertension control, to date implemented in 21 countries.</p><p><strong>Methods: </strong>To advance implementation, an interdisciplinary group of practitioners was engaged to select the key evidence-based drivers of hypertension control and to design a comprehensive scorecard to monitor their implementation at primary care health facilities (PHC). The group studied high-performing health systems that achieve high hypertension control through quality improvement programs focusing on specific process measures, with regular feedback to providers at health facilities.</p><p><strong>Findings: </strong>The final selected eight drivers were categorized into five main domains: (1) diagnosis (blood pressure measurement accuracy and CVD risk evaluation); (2) treatment (standardized treatment protocol and treatment intensification); (3) continuity of care and follow-up; (4) delivery system (team-based care, medication refill), and (5) system for performance evaluation. The drivers and recommendations were then translated into process measures, resulting in two interconnected scorecards integrated into the HEARTS in the Americas monitoring and evaluation system.</p><p><strong>Interpretation: </strong>Focus on these key hypertension drivers and resulting scorecards, will guide the quality improvement process to achieve population control goals at the participating health centers in HEARTS implementing countries.</p><p><strong>Funding: </strong>No funding to declare.</p>","PeriodicalId":80307,"journal":{"name":"The Case manager","volume":"13 1 1","pages":"None"},"PeriodicalIF":7.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9121401/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81218169","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-01-01DOI: 10.18653/v1/2022.case-1.27
Arthur Müller, Andreas Dafnos
We participated in the Shared Task 1 at CASE 2021, Subtask 4 on protest event extraction from news articles and examined different techniques aimed at improving the performance of the winning system from the last competition round. We evaluated in-domain pre-training, task-specific pre-fine-tuning, alternative loss function, translation of the English training dataset into other target languages (i.e., Portuguese, Spanish, and Hindi) for the token classification task, and a simple data augmentation technique by random sentence reordering. This paper summarizes the results, showing that random sentence reordering leads to a consistent improvement of the model performance.
{"title":"SPARTA at CASE 2021 Task 1: Evaluating Different Techniques to Improve Event Extraction","authors":"Arthur Müller, Andreas Dafnos","doi":"10.18653/v1/2022.case-1.27","DOIUrl":"https://doi.org/10.18653/v1/2022.case-1.27","url":null,"abstract":"We participated in the Shared Task 1 at CASE 2021, Subtask 4 on protest event extraction from news articles and examined different techniques aimed at improving the performance of the winning system from the last competition round. We evaluated in-domain pre-training, task-specific pre-fine-tuning, alternative loss function, translation of the English training dataset into other target languages (i.e., Portuguese, Spanish, and Hindi) for the token classification task, and a simple data augmentation technique by random sentence reordering. This paper summarizes the results, showing that random sentence reordering leads to a consistent improvement of the model performance.","PeriodicalId":80307,"journal":{"name":"The Case manager","volume":"2 1","pages":"189-194"},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81812790","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}
Pub Date : 2022-01-01DOI: 10.18653/v1/2022.case-1.15
Abigail Sticha, P. Brenner
Social and political researchers require robust event datasets to conduct data-driven analysis, an example being the need for trigger event datasets to analyze under what conditions and in what patterns certain trigger-type events increase the probability of mass killings. Fortunately, NLP and ML can be leveraged to create these robust datasets. In this paper we (i) outline a robust ML framework that prioritizes understandability through visualizations and generalizability through the ability to implement different ML algorithms, (ii) perform a comparative analysis of these ML tools within the framework for the coup trigger, (iii) leverage our ML framework along with a unique combination of NLP tools, such as NER and knowledge graphs, to produce a dataset for the the assassination trigger, and (iv) make this comprehensive, consolidated, and cohesive assassination dataset publicly available to provide temporal data for understanding political violence as well as training data for further socio-political research.
{"title":"Hybrid Knowledge Engineering Leveraging a Robust ML Framework to Produce an Assassination Dataset","authors":"Abigail Sticha, P. Brenner","doi":"10.18653/v1/2022.case-1.15","DOIUrl":"https://doi.org/10.18653/v1/2022.case-1.15","url":null,"abstract":"Social and political researchers require robust event datasets to conduct data-driven analysis, an example being the need for trigger event datasets to analyze under what conditions and in what patterns certain trigger-type events increase the probability of mass killings. Fortunately, NLP and ML can be leveraged to create these robust datasets. In this paper we (i) outline a robust ML framework that prioritizes understandability through visualizations and generalizability through the ability to implement different ML algorithms, (ii) perform a comparative analysis of these ML tools within the framework for the coup trigger, (iii) leverage our ML framework along with a unique combination of NLP tools, such as NER and knowledge graphs, to produce a dataset for the the assassination trigger, and (iv) make this comprehensive, consolidated, and cohesive assassination dataset publicly available to provide temporal data for understanding political violence as well as training data for further socio-political research.","PeriodicalId":80307,"journal":{"name":"The Case manager","volume":"38 1","pages":"106-116"},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87973572","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}
Socio-political protests often lead to grave consequences when they occur. The early detection of such protests is very important for taking early precautionary measures. However, the main shortcoming of protest event detection is the scarcity of sufficient training data for specific language categories, which makes it difficult to train data-hungry deep learning models effectively. Therefore, cross-lingual and zero-shot learning models are needed to detect events in various low-resource languages. This paper proposes a multi-lingual cross-document level event detection approach using pre-trained transformer models developed for Shared Task 1 at CASE 2022. The shared task constituted four subtasks for event detection at different granularity levels, i.e., document level to token level, spread over multiple languages (English, Spanish, Portuguese, Turkish, Urdu, and Mandarin). Our system achieves an average F1 score of 0.73 for document-level event detection tasks. Our approach secured 2nd position for the Hindi language in subtask 1 with an F1 score of 0.80. While for Spanish, we secure 4th position with an F1 score of 0.69. Our code is available at https://github.com/nehapspathak/campros/.
{"title":"CamPros at CASE 2022 Task 1: Transformer-based Multilingual Protest News Detection","authors":"Kumari Neha, Mrinal Anand, Tushar Mohan, P. Kumaraguru, Arun Balaji Buduru","doi":"10.18653/v1/2022.case-1.24","DOIUrl":"https://doi.org/10.18653/v1/2022.case-1.24","url":null,"abstract":"Socio-political protests often lead to grave consequences when they occur. The early detection of such protests is very important for taking early precautionary measures. However, the main shortcoming of protest event detection is the scarcity of sufficient training data for specific language categories, which makes it difficult to train data-hungry deep learning models effectively. Therefore, cross-lingual and zero-shot learning models are needed to detect events in various low-resource languages. This paper proposes a multi-lingual cross-document level event detection approach using pre-trained transformer models developed for Shared Task 1 at CASE 2022. The shared task constituted four subtasks for event detection at different granularity levels, i.e., document level to token level, spread over multiple languages (English, Spanish, Portuguese, Turkish, Urdu, and Mandarin). Our system achieves an average F1 score of 0.73 for document-level event detection tasks. Our approach secured 2nd position for the Hindi language in subtask 1 with an F1 score of 0.80. While for Spanish, we secure 4th position with an F1 score of 0.69. Our code is available at https://github.com/nehapspathak/campros/.","PeriodicalId":80307,"journal":{"name":"The Case manager","volume":"7 1","pages":"169-174"},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82708990","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}
Pub Date : 2022-01-01DOI: 10.18653/v1/2022.case-1.1
Surendrabikram Thapa, Aditya Shah, F. Jafri, Usman Naseem, Imran Razzak
This paper presents a new multi-modal dataset for identifying hateful content on social media, consisting of 5,680 text-image pairs collected from Twitter, labeled across two labels. Experimental analysis of the presented dataset has shown that understanding both modalities is essential for detecting these techniques. It is confirmed in our experiments with several state-of-the-art multi-modal models. In future work, we plan to extend the dataset in size. We further plan to develop new multi-modal models tailored explicitly to hate-speech detection, aiming for a deeper understanding of the text and image relation. It would also be interesting to perform experiments in a direction that explores what social entities the given hate speech tweet targets.
{"title":"A Multi-Modal Dataset for Hate Speech Detection on Social Media: Case-study of Russia-Ukraine Conflict","authors":"Surendrabikram Thapa, Aditya Shah, F. Jafri, Usman Naseem, Imran Razzak","doi":"10.18653/v1/2022.case-1.1","DOIUrl":"https://doi.org/10.18653/v1/2022.case-1.1","url":null,"abstract":"This paper presents a new multi-modal dataset for identifying hateful content on social media, consisting of 5,680 text-image pairs collected from Twitter, labeled across two labels. Experimental analysis of the presented dataset has shown that understanding both modalities is essential for detecting these techniques. It is confirmed in our experiments with several state-of-the-art multi-modal models. In future work, we plan to extend the dataset in size. We further plan to develop new multi-modal models tailored explicitly to hate-speech detection, aiming for a deeper understanding of the text and image relation. It would also be interesting to perform experiments in a direction that explores what social entities the given hate speech tweet targets.","PeriodicalId":80307,"journal":{"name":"The Case manager","volume":"5 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89591250","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}
Pub Date : 2022-01-01DOI: 10.18653/v1/2022.case-1.16
Yaoyao Dai, Benjamin J. Radford, Andrew Halterman
We report on the current status of an effort to produce political event data from unstructured text via a Transformer language model. Compelled by the current lack of publicly available and up-to-date event coding software, we seek to train a model that can produce structured political event records at the sentence level. Our approach differs from previous efforts in that we conceptualize this task as one of text-to-text sequence generation. We motivate this choice by outlining desirable properties of text generation models for the needs of event coding. To overcome the lack of sufficient training data, we also describe a method for generating synthetic text and event record pairs that we use to fit our model.
{"title":"Political Event Coding as Text-to-Text Sequence Generation","authors":"Yaoyao Dai, Benjamin J. Radford, Andrew Halterman","doi":"10.18653/v1/2022.case-1.16","DOIUrl":"https://doi.org/10.18653/v1/2022.case-1.16","url":null,"abstract":"We report on the current status of an effort to produce political event data from unstructured text via a Transformer language model. Compelled by the current lack of publicly available and up-to-date event coding software, we seek to train a model that can produce structured political event records at the sentence level. Our approach differs from previous efforts in that we conceptualize this task as one of text-to-text sequence generation. We motivate this choice by outlining desirable properties of text generation models for the needs of event coding. To overcome the lack of sufficient training data, we also describe a method for generating synthetic text and event record pairs that we use to fit our model.","PeriodicalId":80307,"journal":{"name":"The Case manager","volume":"33 1 1","pages":"117-123"},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83764770","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}
Pub Date : 2022-01-01DOI: 10.18653/v1/2022.case-1.8
R. Raj, Kajsa Andreasson, Tobias Norlund, Richard Johansson, Aron Lagerberg
Most of today’s systems for socio-political event detection are text-based, while an increasing amount of information published on the web is multi-modal. We seek to bridge this gap by proposing a method that utilizes existing annotated unimodal data to perform event detection in another data modality, zero-shot. Specifically, we focus on protest detection in text and images, and show that a pretrained vision-and-language alignment model (CLIP) can be leveraged towards this end. In particular, our results suggest that annotated protest text data can act supplementarily for detecting protests in images, but significant transfer is demonstrated in the opposite direction as well.
{"title":"Cross-modal Transfer Between Vision and Language for Protest Detection","authors":"R. Raj, Kajsa Andreasson, Tobias Norlund, Richard Johansson, Aron Lagerberg","doi":"10.18653/v1/2022.case-1.8","DOIUrl":"https://doi.org/10.18653/v1/2022.case-1.8","url":null,"abstract":"Most of today’s systems for socio-political event detection are text-based, while an increasing amount of information published on the web is multi-modal. We seek to bridge this gap by proposing a method that utilizes existing annotated unimodal data to perform event detection in another data modality, zero-shot. Specifically, we focus on protest detection in text and images, and show that a pretrained vision-and-language alignment model (CLIP) can be leveraged towards this end. In particular, our results suggest that annotated protest text data can act supplementarily for detecting protests in images, but significant transfer is demonstrated in the opposite direction as well.","PeriodicalId":80307,"journal":{"name":"The Case manager","volume":"1 1","pages":"56-60"},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87813541","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}
Pub Date : 2022-01-01DOI: 10.18653/v1/2022.case-1.4
Thierry Desot, Orphée De Clercq, Veronique Hoste
Event and argument role detection are frequently conceived as separate tasks. In this work we conceive both processes as one taskin a hybrid event detection approach. Its main component is based on automatic keyword extraction (AKE) using the self-attention mechanism of a BERT transformer model. As a bottleneck for AKE is defining the threshold of the attention values, we propose a novel method for automatic self-attention thresholdselection. It is fueled by core event information, or simply the verb and its arguments as the backbone of an event. These are outputted by a knowledge-based syntactic parser. In a secondstep the event core is enriched with other semantically salient words provided by the transformer model. Furthermore, we propose an automatic self-attention layer and head selectionmechanism, by analyzing which self-attention cells in the BERT transformer contribute most to the hybrid event detection and which linguistic tasks they represent. This approach was integrated in a pipeline event extraction approachand outperforms three state of the art multi-task event extraction methods.
{"title":"A Hybrid Knowledge and Transformer-Based Model for Event Detection with Automatic Self-Attention Threshold, Layer and Head Selection","authors":"Thierry Desot, Orphée De Clercq, Veronique Hoste","doi":"10.18653/v1/2022.case-1.4","DOIUrl":"https://doi.org/10.18653/v1/2022.case-1.4","url":null,"abstract":"Event and argument role detection are frequently conceived as separate tasks. In this work we conceive both processes as one taskin a hybrid event detection approach. Its main component is based on automatic keyword extraction (AKE) using the self-attention mechanism of a BERT transformer model. As a bottleneck for AKE is defining the threshold of the attention values, we propose a novel method for automatic self-attention thresholdselection. It is fueled by core event information, or simply the verb and its arguments as the backbone of an event. These are outputted by a knowledge-based syntactic parser. In a secondstep the event core is enriched with other semantically salient words provided by the transformer model. Furthermore, we propose an automatic self-attention layer and head selectionmechanism, by analyzing which self-attention cells in the BERT transformer contribute most to the hybrid event detection and which linguistic tasks they represent. This approach was integrated in a pipeline event extraction approachand outperforms three state of the art multi-task event extraction methods.","PeriodicalId":80307,"journal":{"name":"The Case manager","volume":"17 1","pages":"21-31"},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87860592","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}
Automated socio-political protest event detection is a challenging task when multiple languages are considered. In CASE 2022 Task 1, we propose ensemble learning methods for multilingual protest event detection in four subtasks with different granularity levels from document-level to entity-level. We develop an ensemble of fine-tuned Transformer-based language models, along with a post-processing step to regularize the predictions of our ensembles. Our approach places the first place in 6 out of 16 leaderboards organized in seven languages including English, Mandarin, and Turkish.
{"title":"ARC-NLP at CASE 2022 Task 1: Ensemble Learning for Multilingual Protest Event Detection","authors":"Umitcan Sahin, Oguzhan Ozcelik, Izzet Emre Kucukkaya, Cagri Toraman","doi":"10.18653/v1/2022.case-1.25","DOIUrl":"https://doi.org/10.18653/v1/2022.case-1.25","url":null,"abstract":"Automated socio-political protest event detection is a challenging task when multiple languages are considered. In CASE 2022 Task 1, we propose ensemble learning methods for multilingual protest event detection in four subtasks with different granularity levels from document-level to entity-level. We develop an ensemble of fine-tuned Transformer-based language models, along with a post-processing step to regularize the predictions of our ensembles. Our approach places the first place in 6 out of 16 leaderboards organized in seven languages including English, Mandarin, and Turkish.","PeriodicalId":80307,"journal":{"name":"The Case manager","volume":"87 1","pages":"175-183"},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88942589","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}
Pub Date : 2022-01-01DOI: 10.18653/v1/2022.case-1.23
M. Suri, Krishna Chopra, Adwita Arora
Event detection, specifically in the socio-political domain, has posed a long-standing challenge to researchers in the NLP domain. Therefore, the creation of automated techniques that perform classification of the large amounts of accessible data on the Internet becomes imperative. This paper is a summary of the efforts we made in participating in Task 1 of CASE 2022. We use state-of-art multilingual BERT (mBERT) with further fine-tuning to perform document classification in English, Portuguese, Spanish, Urdu, Hindi, Turkish and Mandarin. In the document classification subtask, we were able to achieve F1 scores of 0.8062, 0.6445, 0.7302, 0.5671, 0.6555, 0.7545 and 0.6702 in English, Spanish, Portuguese, Hindi, Urdu, Mandarin and Turkish respectively achieving a rank of 5 in English and 7 on the remaining language tasks.
{"title":"NSUT-NLP at CASE 2022 Task 1: Multilingual Protest Event Detection using Transformer-based Models","authors":"M. Suri, Krishna Chopra, Adwita Arora","doi":"10.18653/v1/2022.case-1.23","DOIUrl":"https://doi.org/10.18653/v1/2022.case-1.23","url":null,"abstract":"Event detection, specifically in the socio-political domain, has posed a long-standing challenge to researchers in the NLP domain. Therefore, the creation of automated techniques that perform classification of the large amounts of accessible data on the Internet becomes imperative. This paper is a summary of the efforts we made in participating in Task 1 of CASE 2022. We use state-of-art multilingual BERT (mBERT) with further fine-tuning to perform document classification in English, Portuguese, Spanish, Urdu, Hindi, Turkish and Mandarin. In the document classification subtask, we were able to achieve F1 scores of 0.8062, 0.6445, 0.7302, 0.5671, 0.6555, 0.7545 and 0.6702 in English, Spanish, Portuguese, Hindi, Urdu, Mandarin and Turkish respectively achieving a rank of 5 in English and 7 on the remaining language tasks.","PeriodicalId":80307,"journal":{"name":"The Case manager","volume":"36 1","pages":"161-168"},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86674377","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号