Entity-Based Opinion Mining from Text and Multimedia

Maynard, Diana; Hare, Jonathon

doi:10.1007/978-3-319-18458-6_4

Cited by 9 publications

(6 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Challenges of sentiment analysis in Twitter: Sentiment analysis on Twitter is particularly challenging, due to abbreviations, the terseness of the tweets, lack of context and the ambiguity which results. This has already been documented (Maynard and Bontcheva, 2016;Maynard and Hare, 2015), with others developing specially adapted NLP tools for the domain (Bontcheva et al, 2013). In addition, there is the difficulty inherent in detecting the sarcasm present in many tweets, as investigated by Maynard and Greenwood (2014).…”

Section: Related Workmentioning

confidence: 83%

“…Testing on a dataset for targetdependent twitter sentiment classification, they show that by incorporating context words for the target into an LSTM they improve over all baselines. The challenges of entity-based opinion mining, analysing the sentiment of a tweet towards a particular entity contained within it, has already been studied by Maynard and Hare (2015). In addition to addressing the extraction of opinions on crucial events in society for the purposes of archiving, they also expanded their research to cover the integration of multimedia through the extraction of sentiment evidence from images.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Regional Sentiment Bias in Social Media Reporting During Crises

et al. 2018

View full text Add to dashboard Cite

Crisis events such as terrorist attacks are extensively commented upon on social media platforms such as Twitter. For this reason, social media content posted during emergency events is increasingly being used by news media and in social studies to characterize the public's reaction to those events. This is typically achieved by having journalists select 'representative' tweets to show, or a classifier trained on prior human-annotated tweets is used to provide a sentiment/emotion breakdown for the event. However, social media users, journalists and annotators do not exist in isolation, they each have their own context and world view. In this paper, we ask the question, 'to what extent do local and international biases affect the sentiments expressed on social media and the way that social media content is interpreted by annotators'. In particular, we perform a multilingual study spanning two events and three languages. We show that there are marked disparities between the emotions expressed by users in different languages for an event. For instance, during the 2016 Paris attack, there was 16% more negative comments written in the English than written in French, even though the event originated on French soil. Furthermore, we observed that sentiment biases also affect annotators from those regions, which can negatively impact the accuracy of social media labelling efforts. This highlights the need to consider the sentiment biases

show abstract

Section: Related Workmentioning

confidence: 83%

Section: Related Workmentioning

confidence: 99%

Regional Sentiment Bias in Social Media Reporting During Crises

et al. 2018

View full text Add to dashboard Cite

show abstract

“…The paper by Vlachidis and Tudhope (2016) presented a semantic annotation system, OPTIMA, which performs the tasks of NER and relation extraction (RE) using handcrafted rules and ontological and domain vocabulary resources. Significant effort has been put into exploring the Web and its assistance in entity annotation, semantic indexing and search of different collections, both by handcrafted rules and data-driven or machine learning (ML) approaches, for example, in the domain of television and radio news, as well as semantic annotation of Web content itself, opinion mining from textual and multimedia content or an IE system for microblog content (Bontcheva et al, 2013;Dowman et al, 2005;Maynard and Hare, 2015). More platforms for automatic semantic annotations that use a rule-based or ML approach are MultimediaN, AeroDAML, Armadillo, KIM, MnM, MUSE, Ont-O-Mat, SemTag and ONTEA (Laclavik et al, 2012;Reeve and Han, 2005;Schreiber et al, 2008).…”

Section: Related Workmentioning

confidence: 99%

HerCulB: content-based information extraction and retrieval for cultural heritage of the Balkans

Tanasijević

Pavlović-Lažetić

2020

View full text Add to dashboard Cite

Purpose The purpose of this paper is to provide a methodology for automatic annotation of a multimedia collection of intangible cultural heritage mostly in the form of interviews. Assigned annotations provide a way to search the collection. Design/methodology/approach Annotation is based on automatic extraction of metadata and is conducted by named entity and topic extraction from textual descriptions with a rule-based approach supported by vocabulary resources, a compiled domain-specific classification scheme and domain-oriented corpus analysis. Findings The proposed methodology for automatic annotation of a collection of intangible cultural heritage, applied on the cultural heritage of the Balkans, has very good results according to F measure, which is 0.87 for the named entity and 0.90 for topic annotation. The overall methodology enables encapsulating domain-specific and language-specific knowledge into collections of finite state transducers and allows further improvements. Originality/value Although cultural heritage has a significant role in the development of identity of a group or an individual, it is one of those specific domains that have not yet been fully explored in case of many languages. A methodology is proposed that can be used for incorporating natural language processing techniques into digital libraries of cultural heritage.

show abstract

“…No primeiro grupo, por exemplo, o uso de textos ao redor de imagensé utilizado por [Habibian et al 2015] para verificar correspondências entre as imagens e textos e assim encontrar os conjuntos inter-relacionados de termos e temas, ao contrário de simplesmente anotar textos já definidos. Já [Maynard and Hare 2015] apresentam uma abordagemútil quando não há texto associadoàs mídias. Os autores abordam como a análise das questões sociais nas mídias podem ajudar os editores a selecionar material relevante e como a mineração de mídias sociais pode contribuir para relacionar arquivos.…”

Section: Anotação Semântica Em Vídeosunclassified

Comparativo entre fontes de dados para anotação automática de videoaulas

Gomes

Souza

Barrére

2017

Anais Do XXVIII Simpósio Brasileiro De Informática Na Educação (SBIE 2017)

View full text Add to dashboard Cite

Abstract. The volume of the educational video repositories has rapidly increased as the creation of new digital resources is facilited. However, such repositories need better indexing and searching engines. Users find it difficult to find useful information when educational videos has poor quality metadata. The proccess of adding new descriptors to videos is called annontation. In this work, we will discuss how the information extracted from educational videos influence the automatic semantic annotation task. We will present an comparative analysis among these information sources and we will demonstrate which information sources are most useful for annontation task.Resumo. A facilidade de criação de conteúdos digitais tem permitido que repositórios de recursos educacionais possam crescer rapidamente, gerando necessidade de melhores mecanismos de indexação e busca. Para usuários, ainda há muita dificuldade de encontrar vídeos de interesse quando estes possuem metadados de baixa qualidade. O processo de atribuir novos descritores aos vídeosé chamado de anotação. Este trabalho tem como objetivo explorar as fontes de informação presentes em videoaulas que podem auxiliar no processo no anotação semântica. O artigo apresenta uma análise comparativa entre a qualidade dessas fontes de dados de forma a demonstrar quais fontes geram melhor resultado para o processo de anotação. IntroduçãoA facilidade de criação de conteúdos digitais tem permitido que repositórios de recursos educacionais possam crescer rapidamente, gerando necessidade de melhores mecanismos de indexação e busca desses conteúdos. De todas as mídias, o vídeoé a que tem o maior destaque quando o assuntoé educação, seja pela sua capacidade de adaptaçãoàs diversas plataformas [de Oliveira et al. 2010] ou pela sua atratividade natural, unindo fatores auditivos e visuais numaúnica mídia [Medeiros and Pansanato 2015]. Por outro lado, para usuários dos repositórios de videoaulas,é quase impossível encontrar vídeos de interesse quando se utiliza apenas pequenos trechos de texto associado ao vídeo, como título e alguns metadados disponíveis nos repositórios que, quase sempre, são curtos e de alto nível [Yang and Meinel 2014].O fato dos usuários ainda não conseguirem extrair o máximo potencial em suas buscas mostra a relevância desses estudos, principalmente quando estão relacionados a

show abstract

Entity-Based Opinion Mining from Text and Multimedia

Cited by 9 publications

References 37 publications

Regional Sentiment Bias in Social Media Reporting During Crises

Regional Sentiment Bias in Social Media Reporting During Crises

HerCulB: content-based information extraction and retrieval for cultural heritage of the Balkans

Comparativo entre fontes de dados para anotação automática de videoaulas

Contact Info

Product

Resources

About