HUMIR at IEST-2018: Lexicon-Sensitive and Left-Right Context-Sensitive BiLSTM for Implicit Emotion Recognition

Abstract

This paper describes the approaches used in HUMIR system for the WASSA-2018 shared task on the implicit emotion recognition. The objective of this task is to predict the emotion expressed by the target word that has been excluded from the given tweet. We suppose this task as a word sense disambiguation in which the target word is considered as a synthetic word that can express 6 emotions depending on the context. To predict the correct emotion, we propose a deep neural network model that uses two BiLSTM networks to represent the contexts in the left and right sides of the target word. The BiLSTM outputs achieved from the left and right contexts are considered as context-sensitive features. These features are used in a feed-forward neural network to predict the target word emotion. Besides this approach, we also combine the BiL-STM model with lexicon-based and emotionbased features. Finally, we employ all models in the final system using Bagging ensemble method. We achieved macro F-measure value of 68.8 on the official test set and ranked sixth out of 30 participants.

Knowledge‐enriched joint‐learning model for implicit emotion cause extraction

“…A few studies have been carried out in the implicit emotion analysis task [1,[27][28][29][30][31][32]. Due to the lack of emotion and context information, one of the research hotspots of emotion-related tasks is how to effectively introduce external knowledge to assist their proposed models in understanding emotions in the text [20,[33][34][35][36][37].…”

Türkçe Metinlerde Sözlük Tabanli Yaklaşimla Duygu Analizi Ve Görselleştirme

“…Bu veriseti 91309 yorum içermektedir [24]. Üçüncü verisetinde ise 10.000 olumlu 10.000 olumsuz metin yer almaktadır [25].…”

Türkçe bilgisayarlı dil bilimi çalışmalarında his analizi

“…Hisler karşılıklı konuşma esnasında ses tonu, jest ve mimiklerden belirgin şekilde fark edilebilirken yazılı iletişimde yazarın hangi hislerle metni yazdığını belirlemek çoğu zaman bir insan için dahi zordur. Çünkü metinler bağlama göre anlamı ve yansıttığı duygusu farklı olan ifadelerden oluşmaktadır (Naderalvojoud, Ucan ve Akcapinar Sezer, 2018). Bu kelime ve ifadelerin yansıttığı hisler belirlenebilirse cümlenin, paragrafın ve bunlara bağlı olarak metnin yansıttığı hisler oransal olarak bulunabilir.…”