Character-based Neural Embeddings for Tweet Clustering

Vakulenko, Svitlana; Nixon, Lyndon; Lupu, Mihai

doi:10.18653/v1/w17-1105

Cited by 10 publications

(8 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Related tasks include topic detection on Twitter task (Papadopoulos et al, 2014;Litvak et al, 2016;Zhou et al, 2016;Vakulenko et al, 2017), binary classification of Tweets (Rosenthal et al, 2017), classification of news related or political stance Tweets (Ghelani et al, 2017;Johnson and Goldwasser, 2016;Volkova et al, 2017), classification of news related articles (Ribeiro et al, 2017), and other classifications in NLP. Binary classification of texts and classification of news related texts and articles are most closely related to this task.…”

Section: Related Workmentioning

confidence: 99%

“…However, choosing important information for news reports from Twitter is very tough, because Twitter contains a vast amount of posts. For this reason, many researchers have studied how to extract important posts for each purpose (Papadopoulos et al, 2014;Litvak et al, 2016;Zhou et al, 2016;Vakulenko et al, 2017). A system using Neural Networks (NNs) has been developed by using models that are trained by extracting Tweets in factual TV program production, and these systems extract "Event-describing Tweets (EVENT)" which include incidents or accidents information for news reports from a large amount of Tweets (Miyazaki et al, 2017).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Classification of Tweets about Reported Events using Neural Networks

Makino¹,

Takei²,

Miyazaki³

et al. 2018

Proceedings of the 2018 EMNLP Workshop W-Nut: The 4th Workshop on Noisy User-Generated Text

View full text Add to dashboard Cite

We developed a system that automatically extracts "Event-describing Tweets" which include incidents or accidents information for creating news reports. Event-describing Tweets can be classified into "Reportedevent Tweets" and "New-information Tweets." Reported-event Tweets cite news agencies or user generated content sites, and Newinformation Tweets are other Event-describing Tweets. A system is needed to classify them so that creators of factual TV programs can use them in their productions. Proposing this Tweet classification task is one of the contributions of this paper, because no prior papers have used the same task even though program creators and other events information collectors have to do it to extract required information from social networking sites. To classify Tweets in this task, this paper proposes a method to input and concatenate character and word sequences in Japanese Tweets by using convolutional neural networks. This proposed method is another contribution of this paper. For comparison, character or word input methods and other neural networks are also used. Results show that a system using the proposed method and architectures can classify Tweets with an F1 score of 88 %.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Classification of Tweets about Reported Events using Neural Networks

Makino¹,

Takei²,

Miyazaki³

et al. 2018

Proceedings of the 2018 EMNLP Workshop W-Nut: The 4th Workshop on Noisy User-Generated Text

View full text Add to dashboard Cite

show abstract

“…OLDA+HC. Additionally, we also evaluate a tweet clustering approach [35] that uses character-based tweet embeddings (i.e. Tweet2Vec [8]) and outperforms the winner [13] of the 2014 SNOW breaking news detection competition 2,3 which was defined as a topic detection task.…”

Section: Datasetsmentioning

confidence: 99%

“…This method was named as Tweet2Vec+HC. All document clustering baselines employ a hierarchical agglomerative clustering algorithm as it is proven to be effective in [35].…”

Section: Datasetsmentioning

confidence: 99%

See 1 more Smart Citation

A Hierarchical Topic Modelling Approach for Tweet Clustering

Wang

Liakata

Zubiaga

et al. 2017

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. While social media platforms such as Twitter can provide rich and up-to-date information for a wide range of applications, manually digesting such large volumes of data is difficult and costly. Therefore it is important to automatically infer coherent and discriminative topics from tweets. Conventional topic models and document clustering approaches fail to achieve good results due to the noisy and sparse nature of tweets. In this paper, we explore various ways of tackling this challenge and finally propose a two-stage hierarchical topic modelling system that is efficient and effective in alleviating the data sparsity problem. We present an extensive evaluation on two datasets, and report our proposed system achieving the best performance in both document clustering performance and topic coherence.

show abstract