Multimedia Lab $@$ ACL WNUT NER Shared Task: Named Entity Recognition for Twitter Microposts using Distributed Word Representations

We take a multi-task learning approach to the shared Task 1 at SemEval-2018. The general idea concerning the model structure is to use as little external data as possible in order to preserve the task relatedness and reduce complexity. We employ multi-task learning with hard parameter sharing to exploit the relatedness between sub-tasks. As a base model, we use a standard recurrent neural network for both the classification and regression subtasks. Our system ranks 32nd out of 48 participants with a Pearson score of 0.557 in the first subtask, and 20th out of 35 in the fifth subtask with an accuracy score of 0.464.

show abstract

“…These embeddings were obtained from Godin et al (2015). Words which are out-of-vocabulary are replaced with UNK.…”

Section: Preprocessing Of Text and Textual Representationmentioning

confidence: 99%

KU-MTL at SemEval-2018 Task 1: Multi-task Identification of Affect in Tweets

Nyegaard-Signori¹,

Helms²,

Bjerva

et al. 2018

Proceedings of the 12th International Workshop on Semantic Evaluation

View full text Add to dashboard Cite

show abstract

“…(b) Extrinsic methods evaluate word vectors by measuring their performance when used for downstream NLP tasks e.g., dependency parsing, named entity recognition (Passos et al, 2014;Godin et al, 2015).…”

Section: Embedding Evaluationmentioning

confidence: 99%

ESTEEM: A Novel Framework for Qualitatively Evaluating and Visualizing Spatiotemporal Embeddings in Social Media

Arendt¹,

Volkova²

2017

Proceedings of ACL 2017, System Demonstrations

View full text Add to dashboard Cite

Analyzing and visualizing large amounts of social media communications and contrasting short-term conversation changes over time and geolocations is extremely important for commercial and government applications. Earlier approaches for largescale text stream summarization used dynamic topic models and trending words. Instead, we rely on text embeddingslow-dimensional word representations in a continuous vector space where similar words are embedded nearby each other. This paper presents ESTEEM, 1 a novel tool for visualizing and evaluating spatiotemporal embeddings learned from streaming social media texts. Our tool allows users to monitor and analyze query words and their closest neighbors with an interactive interface. We used stateof-the-art techniques to learn embeddings and developed a visualization to represent dynamically changing relations between words in social media over time and other dimensions. This is the first interactive visualization of streaming text representations learned from social media texts that also allows users to contrast differences across multiple dimensions of the data.

show abstract

“…We perform a grid-search using cross-validation on our training set for parameter tuning, and report results on our test set. For each of the models, we establish a baseline with W2V features (Google News-trained Word2Vec size 300 for the debate forums, and Twitter-trained Word2Vec size 400 (Godin et al, 2015), for the tweets). We experiment with different embedding representations, finding that we achieve best results by averaging the word embeddings for each input when using SVM, and creating an embedding matrix (number of words by embedding size for each in- Figure 1: LSTM Network Architecture put) as input to an embedding layer when using LSTM.…”

Section: Rqs Vs Information-seeking Qsmentioning

confidence: 99%

Proceedings of the 18th Annual SIGdial Meeting on Discourse and Dialogue

2017

View full text Add to dashboard Cite

We are excited to welcome you to this year's SIGdial Conference, the 18th Annual Meeting of the Special Interest Group on Discourse and Dialogue. We are pleased to hold the conference in Saarbrücken on August 15-17th, co-located with SemDial 2017, and in close proximity to both INTERSPEECH 2017 and YRRSDS 2017, the Young Researchers' Roundtable on Spoken Dialog Systems.The SIGdial conference remains a premier publication venue for research in discourse and dialogue. This year, the program includes oral presentations, poster sessions, and one demo session. SIGdial 2017 also hosts three special sessions, two joint with SemDial 2017.We received 113 submissions this year. In only one previous year has there been a greater number of submissions to SIGdial. All long and short papers received at least 3 reviews. We carefully considered both the numeric ratings and the tenor of the comments in making our selections for the program. Overall, the members of the Program Committee did an excellent job in reviewing the submitted papers. We thank them for their important role in selecting the accepted papers and for helping to come up with a high quality program for the conference. We also thank Pierre Lison, Mentoring Chair for SIGdial 2017, for his dedicated work on the mentoring process. The goal of mentoring is to assist authors of papers that contain important ideas but lack clarity. In line with the SIGdial tradition, our aim has been to create a balanced program that accommodates as many favorably rated papers as possible. We accepted 29 long papers, 11 short papers and 7 demo presentations. These numbers give an overall acceptance rate of 41.6%. The rates separately for types of papers are 42% for long papers, 30% for short and 87% for demo submissions. Of the long papers, 18 were presented as oral presentations. The remaining 11 long papers and all the short papers were presented as posters, split into two poster sessions.This year SIGdial has three special sessions on topics of growing interest. The chosen sessions were (i) the Special Session on Negotiation Dialog organized by Amanda Stent, Aasish Pappu, Diane Litman and Marilyn Walker; (ii) the Second WOCHAT Special Session on Chatbots and Conversational Agents organized by Ryuichiro Higashinaka, Ron Artstein, Rafael E. Banchs, and Wolfgang Minker; and (iii) Special Session on Natural Language Generation for Dialog Systems organized by Marilyn Walker, Verena Rieser, Vera Demberg, Dietrich Klakow, Dilek Hakkani-Tur, David M. Howcroft and Shereen Oraby. These specialized topics brought diverse paper submissions to our technical program. At the conference, the special sessions also featured panel discussions and position talks, allowing for active engagement of the conference participants. This year, two of these special sessions, the one on negotiation and on conversational agents, are part of the joint SIGdial/SemDial program at the conference bringing both communities to participate in them. This year's SIGdial conference runs 3 full days compared to previous year...

show abstract

Multimedia Lab $@$ ACL WNUT NER Shared Task: Named Entity Recognition for Twitter Microposts using Distributed Word Representations

Cited by 160 publications

References 5 publications

KU-MTL at SemEval-2018 Task 1: Multi-task Identification of Affect in Tweets

KU-MTL at SemEval-2018 Task 1: Multi-task Identification of Affect in Tweets

ESTEEM: A Novel Framework for Qualitatively Evaluating and Visualizing Spatiotemporal Embeddings in Social Media

Proceedings of the 18th Annual SIGdial Meeting on Discourse and Dialogue

Contact Info

Product

Resources

About