Publications

Ph.D. Thesis

Sudipta Kar

Dept. of Computer Science, University of Houston

title={Automatic Story Characterization},
        author={Kar, Sudipta},
        year={2020},
        group = "thesis",
        school={University of Houston},
        url={https://uh-ir.tdl.org/bitstream/handle/10657/6709/KAR-DOCTORALTHESISEDD-2020.pdf

Computerized systems capable of generating high-level story descriptions have many potential real-life applications. However, enabling computers to do so requires teaching computers to obtain an abstract understanding of natural language stories algorithmically, which is one of the non-trivial problems in Artificial Intelligence and Natural Language Processing.
In this dissertation, we tackle the challenge of automatically characterizing stories at a high-level by generating a set of tags from narrative texts written in English. We start by presenting a background study on the problem, discuss the required resources for research, and propose a new corpus to facilitate research on high-level story understanding by selecting tag prediction for movies as an application of this problem. Then, we focus on designing methods for high-level story understanding from written narratives and predicting tags for movies from the written plot synopses. First, we employ a wide range of linguistic features to design a machine learning approach for generating descriptive tags for stories from narrative texts. At the next step, we design a neural methodology for modeling the flow of emotions throughout stories and enhance a system that uses a high-level representation of narrative texts to predict tags. We furthermore exploit the hierarchical structure of text documents to encode the synopses and strengthen the tag prediction mechanism. In the final part of this dissertation, we demonstrate a technique utilizing user reviews to generate tags for characterizing stories at a high-level. We made the new dataset, the source code of the systems, and a live tag prediction system publicly available to the community to encourage further exploration in the direction of automatic story characterization.

Patent

Book analysis and recommendation

Holly Lynn Payne, Mark Fielding Bregman, Bogart Vargas, Thamar Solorio, Suraj Maharjan, Sudipta Kar

title={Book analysis and recommendation},
      author={Payne, Holly Lynn and Bregman, Mark Fielding and Vargas, Bogart and Solorio, Thamar and Maharjan, Suraj and Kar, Sudipta},
      year={2019},
      month=dec # " 10",
      group="patent",
      publisher={Google Patents},
      note={US Patent 10,503,829}

A method includes generating style values and experiential language tags (ELTs) for a plurality of books based on retrieved book content and reader reviews, respectively. The method further includes generating an ELT prediction model based on the style values and the ELTs. The ELT prediction model is configured to receive a set of style values for a new book and output a set of predicted ELTs for the new book, the set of predicted ELTs indicating predicted reader experiences with the new book. The method further includes receiving user-submitted book content from a remote user device, determining style values for the user-submitted book content, and determining a list of predicted ELTs for the user-submitted book content using the style values for the user-submitted book content and the ELT prediction model. Additionally, the method includes transmitting, to the user device, the list of predicted ELTs for the user-submitted book content.

Papers

Integrating Summarization and Retrieval for Enhanced Personalization via Large Language Models

Chris Richardson and Yao Zhang and Kellen Gillespie and Sudipta Kar and Arshdeep Singh and Zeynab Raeesy and Omar Zia Khan and Abhinav Sethy
@inproceedings{richardson2023integrating, title={Integrating Summarization and Retrieval for Enhanced Personalization via Large Language Models}, author={Chris Richardson and Yao Zhang and Kellen Gillespie and Sudipta Kar and Arshdeep Singh and Zeynab Raeesy and Omar Zia Khan and Abhinav Sethy}, year={2023}, booktitle={Personalized Generative AI @ CIKM 2023}, url="https://arxiv.org/abs/2310.20081", group="paper" }
MultiCoNER v2: a Large Multilingual dataset for Fine-grained and Noisy Named Entity Recognition

Fetahu, Besnik and Chen, Zhiyu and Kar, Sudipta and Rokhlenko, Oleg and Malmasi, Shervin
@inproceedings{fetahu-etal-2023-multiconer, title = MultiCoNER v2: a Large Multilingual dataset for Fine-grained and Noisy Named Entity Recognition", author = "Fetahu, Besnik and Chen, Zhiyu and Kar, Sudipta and Rokhlenko, Oleg and Malmasi, Shervin", booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2023", month = dec, year = {2023}, address = "Singapore", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2023.findings-emnlp.134", doi = "10.18653/v1/2023.findings-emnlp.134", pages = "2027–2051" }
SemEval-2023 Task 2: Fine-grained Multilingual Named Entity Recognition (MultiCoNER 2)

Fetahu, Besnik and Kar, Sudipta and Chen, Zhiyu and Rokhlenko, Oleg and Malmasi, Shervin
@inproceedings{fetahu-etal-2023-semeval, title = SemEval-2023 Task 2: Fine-grained Multilingual Named Entity Recognition (MultiCoNER 2)", author = "Fetahu, Besnik and Kar, Sudipta and Chen, Zhiyu and Rokhlenko, Oleg and Malmasi, Shervin", booktitle = "Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023)", month = jul, year = {2023}", address = "Toronto, Canada", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2023.semeval-1.310", doi = "10.18653/v1/2023.semeval-1.310", pages = "2247–2265" }
Learning to Retrieve Engaging Follow-Up Queries

Christopher Richardson, Sudipta Kar, Anjishnu Kumar, Anand Ramachandran, Omar Zia Khan, Zeynab Raeesy, Abhinav Sethy
EACL 2023
@inproceedings{richardson2023eacl, title=Learning to Retrieve Engaging Follow-Up Queries, author=Christopher Richardson and Sudipta Kar and Anjishnu Kumar and Anand Ramachandran and Omar Zia Khan and Zeynab Raeesy and Abhinav Sethy, booktitle=Findings of EACL 2023, year={2023} }
Preventing Catastrophic Forgetting in Continual Learning of New Natural Language Tasks

Sudipta Kar, Giuseppe Castellucci, Simone Filice, Shervin Malmasi, Oleg Rokhlenko
KDD 2022
@inproceedings{kar2022kdd, title=Preventing Catastrophic Forgetting in Continual Learning of New Natural Language Tasks, author=Kar, Sudipta, Castellucci, Giuseppe and Filice, Simone and Malmasi, Shervin and Rokhlenko, Oleg, booktitle=KDD 2022, year={2022} }
Survey on Aspect Category Detection

Siva Uday Sampreeth Chebolu, Paolo Rosso, Sudipta Kar, Thamar Solorio
ACM Computing Surveys
@inproceedings{chebolu2022, title=Survey on Aspect Category Detection, author=Chebolu, Siva Uday Sampreeth and Rosso, Paolo and Kar, Sudipta and Solorio, Thamar, booktitle=ACM Computing Surveys, year={2022} }
SemEval-2022 Task 11: Multilingual Complex Named Entity Recognition (MultiCoNER)

Shervin Malmasi, Anjie Fang, Besnik Fetahu, Sudipta Kar, Oleg Rokhlenko
NAACL 2022
@inproceedings{multiconer-report, title=SemEval-2022 Task 11: Multilingual Complex Named Entity Recognition (MultiCoNER), author=Malmasi, Shervin and Fang, Anjie and Fetahu, Besnik and Kar, Sudipta and Rokhlenko, Oleg, booktitle=Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022), year=2022, publisher=Association for Computational Linguistics, }
MultiCoNER: a Large-scale Multilingual dataset for Complex Named Entity Recognition

Shervin Malmasi, Anjie Fang, Besnik Fetahu, Sudipta Kar, Oleg Rokhlenko
COLING 2022
@article{multiconer-data, title=MultiCoNER: a Large-scale Multilingual dataset for Complex Named Entity Recognition, author=Malmasi, Shervin and Fang, Anjie and Fetahu, Besnik and Kar, Sudipta and Rokhlenko, Oleg, booktitle = "Proceedings of the 29th International Conference on Computational Linguistics", month = oct, year = "2022", address = "Gyeongju, Republic of Korea", publisher = "International Committee on Computational Linguistics", url = "https://aclanthology.org/2022.coling-1.334", pages = "3798–3809", }

Syntax and Themes: How Context Free Grammar Rules and Semantic Word Association Influence Book Success

Henry Gorelick, Biddut Sarker Bijoy, Syeda Jannatus Saba, Sudipta Kar, Md Saiful Islam, Mohammad Ruhul Amin

RANLP 2021

@inproceedingsgorelick-etal-2021-syntax,
    title = "Syntax and Themes: How Context Free Grammar Rules and Semantic Word Association Influence Book Success",
    author = "Gorelick, Henry  and
      Bijoy, Biddut Sarker  and
      Jannatus Saba, Syeda  and
      Kar, Sudipta  and
      Islam, Md Saiful  and
      Amin, Mohammad Ruhul",
    booktitle = "Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2021)",
    month = sep,
    year = "2021",
    address = "Held Online",
    publisher = "INCOMA Ltd.",
    url = "https://aclanthology.org/2021.ranlp-main.53",
    pages = "463–474",
    abstract = "In this paper, we attempt to improve upon the state-of-the-art in predicting a novel’s success by modeling the lexical semantic relationships of its contents. We created the largest dataset used in such a project containing lexical data from 17,962 books from Project Gutenberg. We utilized domain specific feature reduction techniques to implement the most accurate models to date for predicting book success, with our best model achieving an average accuracy of 94.0%. By analyzing the model parameters, we extracted the successful semantic relationships from books of 12 different genres. We finally mapped those semantic relations to a set of themes, as defined in Roget’s Thesaurus and discovered the themes that successful books of a given genre prioritize. At the end of the paper, we further showed that our model demonstrate similar performance for book success prediction even when Goodreads rating was used instead of download count to measure success."

In this paper, we attempt to improve upon the state-of-the-art in predicting a novel’s success by modeling the lexical semantic relationships of its contents. We created the largest dataset used in such a project containing lexical data from 17,962 books from Project Gutenberg. We utilized domain specific feature reduction techniques to implement the most accurate models to date for predicting book success, with our best model achieving an average accuracy of 94.0%. By analyzing the model parameters, we extracted the successful semantic relationships from books of 12 different genres. We finally mapped those semantic relations to a set of themes, as defined in Roget’s Thesaurus and discovered the themes that successful books of a given genre prioritize. At the end of the paper, we further showed that our model demonstrate similar performance for book success prediction even when Goodreads rating was used instead of download count to measure success.

SentNoB: A Dataset for Analysing Sentiment on Noisy Bangla Texts

Khondoker Ittehadul Islam, Md Saiful Islam, Sudipta Kar, Mohammad Ruhul Amin

EMNLP 2021

Source Code

@inproceedings{sentnob-2021,
    title = "SentNoB: A Dataset for Analysing Sentiment on Noisy Bangla Texts",
    author = "Khondoker Ittehadul Islam,
    Md Saiful Islam, 
    Sudipta Kar, 
    Mohammad Ruhul Amin",
    booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
    month = Nov,
    year = "2021",
    publisher = "Association for Computational Linguistics",
    abstract = "In this paper, we propose an annotated sentiment analysis dataset made of informally written Bangla texts. This dataset comprises public comments on news and videos collected from social media covering 13 different domains, including politics, education, and agriculture. These comments are labeled with one of the polarity labels, namely positive, negative, and neutral. One significant characteristic of the dataset is that each of the comments is noisy in terms of the mix of dialects and grammatical incorrectness. Our experiments to develop a benchmark classification system show that hand-crafted lexical features provide superior performance than neural network and pretrained language models. We have made the dataset and accompanying models presented in this paper publicly available at https://git.io/JuuNB.",
}

In this paper, we propose an annotated sentiment analysis dataset made of informally written Bangla texts. This dataset comprises public comments on news and videos collected from social media covering 13 different domains, including politics, education, and agriculture. These comments are labeled with one of the polarity labels, namely positive, negative, and neutral. One significant characteristic of the dataset is that each of the comments is noisy in terms of the mix of dialects and grammatical incorrectness. Our experiments to develop a benchmark classification system show that hand-crafted lexical features provide superior performance than neural network and pretrained language models. We have made the dataset and accompanying models presented in this paper publicly available at https://git.io/JuuNB.

Multi-view Story Characterization from Movie Plot Synopses and Reviews

Sudipta Kar, Gustavo Aguilar, Mirella Lapata, Thamar Solorio

EMNLP 2020

Live Demo Slides

@inproceedings{kar-etal-2020-multiview,
    title = "Multi-view Story Characterization from Movie Plot Synopses and Reviews",
    author = "Kar, Sudipta  and
      Aguilar, Gustavo  and
      Lapata, Mirella and
      Solorio, Thamar.",
    booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
    month = Nov,
    year = "2020",
    publisher = "Association for Computational Linguistics",
    abstract = "This paper considers the problem of characterizing stories by inferring properties such as theme and style using written synopses and reviews of movies. We experiment with a multi-label dataset of movie synopses and a tagset representing various attributes of stories (e.g., genre, type of events). Our proposed multi-view model encodes the synopses and reviews using hierarchical attention and shows improvement over methods that only use synopses. Finally, we demonstrate how can we take advantage of such a model to extract a complementary set of story-attributes from reviews without direct supervision. We have made our dataset and source code publicly available at https://ritual.uh.edu/ multiview-tag-2020.",
}

This paper considers the problem of characterizing stories by inferring properties such as theme and style using written synopses and reviews of movies. We experiment with a multi-label dataset of movie synopses and a tagset representing various attributes of stories (e.g., genre, type of events). Our proposed multi-view model encodes the synopses and reviews using hierarchical attention and shows improvement over methods that only use synopses. Finally, we demonstrate how can we take advantage of such a model to extract a complementary set of story-attributes from reviews without direct supervision. We have made our dataset and source code publicly available at https://ritual.uh.edu/ multiview-tag-2020.

Attending the Emotions to Detect Online Abusive Language

Niloofar Safi Samghabadi, Afsheen Hatami, Mahsa Shafaei, Sudipta Kar and Thamar Solorio

WOAH4 @ EMNLP 2020

@inproceedings{safi-samghabadi-etal-2019-attending,
    title = "Attending the Emotions to Detect Online Abusive Language",
    author = "Safi Samghabadi, Niloofar and Hatami, Afsheen and Shafaei, Mahsa and Kar, Sudipta and  Solorio, Thamar",
    booktitle = "Proceedings of the Fourth Workshop on Online Abuse and Harms",
    month = nov,
    year = "2020",
    publisher = "Association for Computational Linguistics",
    abstract = "In recent years, abusive behavior has become a serious issue in online social networks. In this paper, we present a new corpus for the task of abusive language detection that is collected from a semi-anonymous online platform, and unlike the majority of other available resources, is not created based on a specific list of bad words. We also develop computational models to incorporate emotions into textual cues to improve aggression identification. We evaluate our proposed methods on different corpora and show promising results with respect to abusive language detection.",
}

In recent years, abusive behavior has become a serious issue in online social networks. In this paper, we present a new corpus for the task of abusive language detection that is collected from a semi-anonymous online platform, and unlike the majority of other available resources, is not created based on a specific list of bad words. We also develop computational models to incorporate emotions into textual cues to improve aggression identification. We evaluate our proposed methods on different corpora and show promising results with respect to abusive language detection.

SemEval-2020 Task 9: Overview of Sentiment Analysis of Code-Mixed Tweets

Parth Patwa, Gustavo Aguilar, Sudipta Kar, Suraj Pandey, Srinivas PYKL, Björn Gambäck, Tanmoy Chakraborty, Thamar Solorio, Amitava Das

SemEval @ COLING 2020

@inproceedingspatwa2020sentimix,
      title={SemEval-2020 Task 9: Overview of Sentiment Analysis of Code-Mixed Tweets},
          author={Patwa, Parth and
                  Aguilar, Gustavo and
                  Kar, Sudipta and
                  Pandey, Suraj and
                  PYKL, Srinivas and
                  Gamb{ä}ck, Bj{ö}rn and
                  Chakraborty, Tanmoy and
                  Solorio, Thamar and  
                  Das, Amitava},
          booktitle = "Proceedings of the 14th International Workshop on Semantic Evaluation ({S}em{E}val-2020)",
          year = "2020",
          month = "December",
          address = "Barcelona, Spain",
          publisher = "Association for Computational Linguistics"

In this paper, we present the results of the SemEval-2020 Task 9 on Sentiment Analysis of Code-Mixed Tweets (SentiMix 2020). We also release and describe our Hinglish (Hindi-English) and Spanglish (Spanish-English) corpora annotated with word-level language identification and sentence-level sentiment labels. These corpora are comprised of 20K and 19K examples, respectively. The sentiment labels are - Positive, Negative, and Neutral. SentiMix attracted 89 submissions in total including 61 teams that participated in the Hinglish contest and 28 submitted systems to the Spanglish competition. The best performance achieved was 75.0% F1 score for Hinglish and 80.6% F1 for Spanglish. We observe that BERT-like models and ensemble methods are the most common and successful approaches among the participants.

LinCE: A Centralized Benchmark for Linguistic Code-switching Evaluation

Gustavo Aguilar, Sudipta Kar, Thamar Solorio

LREC, 2020

LinCE benchmark

@inproceedings{aguilar-etal-2020-lince,
        title = "{L}in{CE}: A Centralized Benchmark for Linguistic Code-switching Evaluation",
        author = "Aguilar, Gustavo  and
          Kar, Sudipta  and
          Solorio, Thamar",
        booktitle = "Proceedings of The 12th Language Resources and Evaluation Conference",
        month = may,
        year = "2020",
        group = "paper",
        address = "Marseille, France",
        publisher = "European Language Resources Association",
        url = "https://www.aclweb.org/anthology/2020.lrec-1.223",
        pages = "1803–1813",
        language = "English",
        ISBN = "979-10-95546-34-4",    
        }

Recent trends in NLP research have raised an interest in linguistic code-switching (CS); modern approaches have been proposed to solve a wide range of NLP tasks on multiple language pairs. Unfortunately, these proposed methods are hardly generalizable to different code-switched languages. In addition, it is unclear whether a model architecture is applicable for a different task while still being compatible with the code-switching setting. This is mainly because of the lack of a centralized benchmark and the sparse corpora that researchers employ based on their specific needs and interests. To facilitate research in this direction, we propose a centralized benchmark for Linguistic Code-switching Evaluation (LinCE) that combines eleven corpora covering four different code-switched language pairs (i.e., Spanish-English, Nepali-English, Hindi-English, and Modern Standard Arabic-Egyptian Arabic) and four tasks (i.e., language identification, named entity recognition, part-of-speech tagging, and sentiment analysis). As part of the benchmark centralization effort, we provide an online platform where researchers can submit their results while comparing with others in real-time. In addition, we provide the scores of different popular models, including LSTM, ELMo, and multilingual BERT so that the NLP community can compare against state-of-the-art systems. LinCE is a continuous effort, and we will expand it with more low-resource languages and tasks.

Age Suitability Rating: Predicting the MPAA Rating Based on Movie Dialogues

Mahsa Shafaei, Niloofar Safi Samghabadi, Sudipta Kar, Thamar Solorio

LREC 2020

@inproceedings{shafaei-etal-2020-age,
            title = "Age Suitability Rating: Predicting the {MPAA} Rating Based on Movie Dialogues",
            author = "Shafaei, Mahsa  and
              Safi Samghabadi, Niloofar  and
              Kar, Sudipta  and
              Solorio, Thamar",
            booktitle = "Proceedings of The 12th Language Resources and Evaluation Conference",
            month = may,
            year = "2020",
            group = "paper",
            address = "Marseille, France",
            publisher = "European Language Resources Association",
            url = "https://www.aclweb.org/anthology/2020.lrec-1.166",
            pages = "1327–1335",
            language = "English",
            ISBN = "979-10-95546-34-4",

Movies help us learn and inspire societal change. But they can also contain objectionable content that negatively affects viewers’ behaviour, especially children. In this paper, our goal is to predict the suitability of movie content for children and young adults based on scripts. The criterion that we use to measure suitability is the MPAA rating that is specifically designed for this purpose. We create a corpus for movie MPAA ratings and propose an RNN based architecture with attention that jointly models the genre and the emotions in the script to predict the MPAA rating. We achieve 81% weighted F1-score for the classification model that outperforms the traditional machine learning method by 7%.

BanFakeNews: A Dataset for Detecting Fake News in Bangla

Md Zobaer Hossain, Md Ashraful Rahman, Md Saiful Islam, Sudipta Kar

LREC 2020

Dataset

@inproceedings{hossain-etal-2020-banfakenews,
            title = "{B}an{F}ake{N}ews: A Dataset for Detecting Fake News in Bangla",
            author = "Hossain, Md Zobaer  and
              Rahman, Md Ashraful  and
              Islam, Md Saiful  and
              Kar, Sudipta",
            booktitle = "Proceedings of The 12th Language Resources and Evaluation Conference",
            month = may,
            year = "2020",
            group = "paper",
            address = "Marseille, France",
            publisher = "European Language Resources Association",
            url = "https://www.aclweb.org/anthology/2020.lrec-1.349",
            pages = "2862–2871",
            language = "English",
            ISBN = "979-10-95546-34-4",

Observing the damages that can be done by the rapid propagation of fake news in various sectors like politics and finance, automatic identification of fake news using linguistic analysis has drawn the attention of the research community. However, such methods are largely being developed for English where low resource languages remain out of the focus. But the risks spawned by fake and manipulative news are not confined by languages. In this work, we propose an annotated dataset of \mbox≈ 50K news that can be used for building automated fake news detection systems for a low resource language like Bangla. Additionally, we provide an analysis of the dataset and develop a benchmark system with state of the art NLP techniques to identify Bangla fake news. To create this system, we explore traditional linguistic features and neural network based methods. We expect this dataset will be a valuable resource for building technologies to prevent the spreading of fake news and contribute in research with low resource languages. The dataset and source code are publicly available at https://github.com/Rowan1697/FakeNews.

Multimodal Identification of Medical Conditions from Chest X-ray Images and Radiology Reports

Sudipta Kar, Hsi-Ming Chang, Ratna Saripalli, Gopal Avinash

GE Data Science Symposium, 2019

Folksonomication: Predicting Tags for Movies from Plot Synopses using Emotion Flow Encoded Neural Network

Sudipta Kar, Suraj Maharjan, Thamar Solorio

COLING 2018

Live Demo Code

@InProceedings{C18-1244,
          author = 	"Kar, Sudipta
                and Maharjan, Suraj
                and Solorio, Thamar",
          title = 	"Folksonomication: Predicting Tags for Movies from Plot Synopses using Emotion Flow Encoded Neural Network",
          booktitle = 	"Proceedings of the 27th International Conference on Computational Linguistics",
          year = 	"2018",
          group = "paper",
          publisher = 	"Association for Computational Linguistics",
          pages = 	"2879–2891",
          location = 	"Santa Fe, New Mexico, USA",
          url = 	"http://aclweb.org/anthology/C18-1244"
        }

Folksonomy of movies covers a wide range of heterogeneous information about movies, like the genre, plot structure, visual experiences, soundtracks, metadata, and emotional experiences from watching a movie. Being able to automatically generate or predict tags for movies can help recommendation engines improve retrieval of similar movies, and help viewers know what to expect from a movie in advance. In this work, we explore the problem of creating tags for movies from plot synopses. We propose a novel neural network model that merges information from synopses and emotion flows throughout the plots to predict a set of tags for movies. We compare our system with multiple baselines and found that the addition of emotion flows boosts the performance of the network by learning ≈18% more tags than a traditional machine learning system.

RiTUAL-UH at TRAC 2018 Shared Task: Aggression Identification

Niloofar Safi Samghabadi, Deepthi Mave, Sudipta Kar, Thamar Solorio

TRAC-I @ COLING 2018

@inproceedings{safi-samghabadi-etal-2018-ritual,
        title = "{R}i{TUAL}-{UH} at {TRAC} 2018 Shared Task: Aggression Identification",
        author = "Safi Samghabadi, Niloofar  and
          Mave, Deepthi  and
          Kar, Sudipta  and
          Solorio, Thamar",
        booktitle = "Proceedings of the First Workshop on Trolling, Aggression and Cyberbullying ({TRAC}-2018)",
        month = aug,
        year = "2018",
        address = "Santa Fe, New Mexico, USA",
        publisher = "Association for Computational Linguistics",
        url = "https://www.aclweb.org/anthology/W18-4402",
        pages = "12–18",
        group = "paper",
        abstract = "This paper presents our system for {“}TRAC 2018 Shared Task on Aggression Identification{”}. Our best systems for the English dataset use a combination of lexical and semantic features. However, for Hindi data using only lexical features gave us the best results. We obtained weighted F1-measures of 0.5921 for the English Facebook task (ranked 12th), 0.5663 for the English Social Media task (ranked 6th), 0.6292 for the Hindi Facebook task (ranked 1st), and 0.4853 for the Hindi Social Media task (ranked 2nd).",
    }

This paper presents our system for “TRAC 2018 Shared Task on Aggression Identification”. Our best systems for the English dataset use a combination of lexical and semantic features. However, for Hindi data using only lexical features gave us the best results. We obtained weighted F1-measures of 0.5921 for the English Facebook task (ranked 12th), 0.5663 for the English Social Media task (ranked 6th), 0.6292 for the Hindi Facebook task (ranked 1st), and 0.4853 for the Hindi Social Media task (ranked 2nd).

Letting Emotions Flow: Success Prediction by Modeling the Flow of Emotions in Books

Suraj Maharjan, Sudipta Kar, Manuel Montes, Fabio A. Gonzalez, Thamar Solorio

NAACL 2018

@InProceedings{maharjan-EtAl:2018:N18-2,
      author    = {Maharjan, Suraj  and  Kar, Sudipta  and  Montes, Manuel  and  Gonzalez, Fabio A.  and  Solorio, Thamar},
      title     = {Letting Emotions Flow: Success Prediction by Modeling the Flow of Emotions in Books},
      booktitle = {Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers)},
      month     = {June},
      year      = {2018},
      group="paper",
      address   = {New Orleans, Louisiana},
      publisher = {Association for Computational Linguistics},
      pages     = {259–265},
      abstract  = {Books have the power to make us feel happiness, sadness, pain, surprise, or sorrow. An author’s dexterity in the use of these emotions captivates readers and makes it difficult for them to put the book down. In this paper, we model the flow of emotions over a book using recurrent neural networks and quantify its usefulness in predicting success in books. We obtained the best weighted F1-score of 69% for predicting books’ success in a multitask setting (simultaneously predicting success and genre of books).},
      url       = {http://www.aclweb.org/anthology/N18-2042}
    }

Books have the power to make us feel happiness, sadness, pain, surprise, or sorrow. An author’s dexterity in the use of these emotions captivates readers and makes it difficult for them to put the book down. In this paper, we model the flow of emotions over a book using recurrent neural networks and quantify its usefulness in predicting success in books. We obtained the best weighted F1-score of 69% for predicting books’ success in a multitask setting (simultaneously predicting success and genre of books).

MPST: A Corpus of Movie Plot Synopses with Tags

Sudipta Kar, Suraj Maharjan, A. Pastor López-Monroy, Thamar Solorio

LREC 2018

Dataset

@InProceedings{KAR18.332,
     author = {Sudipta Kar and Suraj Maharjan and A. Pastor López-Monroy and Thamar Solorio},
     title = {\MPST}: A Corpus of Movie Plot Synopses with Tags},
     booktitle = {Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)},
     year = {2018},
     month = {May},
     date = {7-12},
     group="paper",
     location = {Miyazaki, Japan},
    editor = {Nicoletta Calzolari (Conference chair) and Khalid Choukri and Christopher Cieri and Thierry Declerck and Sara Goggi and Koiti Hasida and Hitoshi Isahara and Bente Maegaard and Joseph Mariani and Hélène Mazo and Asuncion Moreno and Jan Odijk and Stelios Piperidis and Takenobu Tokunaga},
     publisher = {European Language Resources Association (ELRA)},
     address = {Paris, France},
     isbn = {979-10-95546-00-9},
     language = {english}
     }

Social tagging of movies reveals a wide range of heterogeneous information about movies, like the genre, plot structure, soundtracks, metadata, visual and emotional experiences. Such information can be valuable in building automatic systems to create tags for movies. Automatic tagging systems can help recommendation engines to improve the retrieval of similar movies as well as help viewers to know what to expect from a movie in advance. In this paper, we set out to the task of collecting a corpus of movie plot synopses and tags. We describe a methodology that enabled us to build a fine-grained set of around 70 tags exposing heterogeneous characteristics of movie plots and the multi-label associations of these tags with some 14K movie plot synopses. We investigate how these tags correlate with movies and the flow of emotions throughout different types of movies. Finally, we use this corpus to explore the feasibility of inferring tags from plot synopses. We expect the corpus will be useful in other tasks where analysis of narratives is relevant.

RiTUAL-UH at SemEval-2017 Task 5: Sentiment Analysis on Financial Data Using Neural Networks

Sudipta Kar, Suraj Maharjan, Thamar Solorio

SemEval @ ACL 2017

Code

@inproceedings{kar-etal-2017-ritual,
        title = "{R}i{TUAL}-{UH} at {S}em{E}val-2017 Task 5: Sentiment Analysis on Financial Data Using Neural Networks",
        author = "Kar, Sudipta  and
          Maharjan, Suraj  and
          Solorio, Thamar",
        booktitle = "Proceedings of the 11th International Workshop on Semantic Evaluation ({S}em{E}val-2017)",
        month = aug,
        year = "2017",
        group = "paper",
        address = "Vancouver, Canada",
        publisher = "Association for Computational Linguistics",
        url = "https://www.aclweb.org/anthology/S17-2150",
        doi = "10.18653/v1/S17-2150",
        pages = "877–882",
    }

In this paper, we present our systems for the “SemEval-2017 Task-5 on Fine-Grained Sentiment Analysis on Financial Microblogs and News”. In our system, we combined hand-engineered lexical, sentiment and metadata features, the representations learned from Convolutional Neural Networks (CNN) and Bidirectional Gated Recurrent Unit (Bi-GRU) with Attention model applied on top. With this architecture we obtained weighted cosine similarity scores of 0.72 and 0.74 for subtask-1 and subtask-2, respectively. Using the official scoring system, our system ranked the second place for subtask-2 and eighth place for the subtask-1. It ranked first for both of the subtasks by the scores achieved by an alternate scoring system.

UH-PRHLT at SemEval-2016 Task 3: Combining Lexical and Semantic-based Features for Community Question Answering

Marc Franco-Salvador, Sudipta Kar, Thamar Solorio, Paolo Rosso

SemEval @ NAACL 2016

@InProceedings{semeval2016,
          author    = {Franco-Salvador, Marc  and  Kar, Sudipta  and  Solorio, Thamar  and  Rosso, Paolo},
          title     = {UH-PRHLT at SemEval-2016 Task 3: Combining Lexical and Semantic-based Features for Community Question Answering},
          booktitle = {Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval-2016)},
          month     = {June},
          year      = {2016},
          address   = {San Diego, California},
          publisher = {Association for Computational Linguistics},
          pages     = {814–821},
          url       = {http://www.aclweb.org/anthology/S16-1126}
        }

In this work we describe the system built for the three English subtasks of the Se- mEval 2016 Task 3 by the Department of Computer Science of the University of Houston (UH) and the Pattern Recog- nition and Human Language Technology (PRHLT) research center - Universitat Polit‘ecnica de Val‘encia: UH-PRHLT. Our system represents instances by using both lexical and semantic-based similarity mea- sures between text pairs. Our semantic features include the use of distributed rep- resentations of words, knowledge graphs generated with the BabelNet multilingual semantic network, and the FrameNet lexi- cal database. Experimental results outper- form the random and Google search en- gine baselines in the three English sub- tasks. Our approach obtained the highest results of subtask B compared to the other task participants.