Several recent studies on dyadic humanhuman interactions have been done on conversations without specific business objectives. However, many companies might benefit from studies dedicated to more precise environments such as after sales services or customer satisfaction surveys. In this work, we place ourselves in the scope of a live chat customer service in which we want to detect emotions and their evolution in the conversation flow. This context leads to multiple challenges that range from exploiting restricted, small and mostly unlabeled datasets to finding and adapting methods for such context. We tackle these challenges by using Few-Shot Learning while making the hypothesis it can serve conversational emotion classification for different languages and sparse labels. We contribute by proposing a variation of Prototypical Networks for sequence labeling in conversation that we name ProtoSeq. We test this method on two datasets with different languages: daily conversations in English and customer service chat conversations in French. When applied to emotion classification in conversations, our method proved to be competitive even when compared to other ones. The code for Proto-Seq is available at https://github.com/ gguibon/ProtoSeq.
Emoji usage drastically increased recently, they are becoming some of the most common ways to convey emotions and sentiments in social messaging applications. Several research works automatically recommend emojis, so users do not have to go through a library of thousands of emojis. In order to improve emoji recommendation, we present and distribute two useful resources: an emoji embedding model from real usage, and emoji clustering based on these embeddings to automatically identify groups of emojis. Assuming that emojis are part of written natural language and can be considered as words, we only used unsupervised learning methods to extract patterns and knowledge from real emoji usage in tweets. Thereby, emotion categories of face emojis were obtained directly from text in a fully reproductible way. These resources and methodology have multiple usages; for example, they could be used to improve our understanding of emojis or enhance emoji recommendation.
Emojis are some of the most common ways to convey emotions and sentiments in social messaging applications. In order to help the user choose emojis among a vast range of possibilities, we aim at developing an automatic recommendation system based on user message analysis and real emoji usage, which goes beyond the simple dictionnary lookup that is done in the industry (mainly Android and iOS). For this purpose, we present a novel automatic emoji prediction model trained and tested on real data and based on sentiment-related features. Such a model differ from the ones learnt from tweets and can predict emojis with a 84.48% f1-score and a 95.49% high precision, using MultiLabel-RandomForest algorithm on real private instant message corpus. We want to determine the best discriminative features for this task.
In this paper we present the system submitted to the SemEval2018 task2 : Multilingual Emoji Prediction. Our system approaches both languages as being equal by first; considering word embeddings associated to automatically computed features of different types, then by applying bagging algorithm RandomForest to predict the emoji of a tweet.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.