NRC-Canada-2014: Detecting Aspects and Sentiment in Customer Reviews

By Svetlana Kiritchenko, Xiao-Dan Zhu, Colin Cherry, Saif M. Mohammad. 2014

673 citations recorded by the source corpus.

Reviews depict sentiments of customers towards various aspects of a product or service. Some of these aspects can be grouped into coarser aspect categories. SemEval-2014 had a shared task (Task 4) on aspect-level sentiment analysis, with over 30 teams participated. In this paper, we describe our submissions, which stood first in detecting aspect categories, first in detecting sentiment towards aspect categories, third in detecting aspect terms, and first and second in detecting sentiment towards aspect terms in the laptop and restaurant domains, respectively.

Subfields: Artificial Intelligence

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

SHINE: Signed Heterogeneous Information Network Embedding for Sentiment Link Prediction (2017, background): “Previous work [6, 11, 15] mainly focuses on sentiment classification based on the concrete content posted by users.”
Utilizing BERT for Aspect-Based Sentiment Analysis via Constructing Auxiliary Sentence (2019, methodology): “The earliest work on (T)ABSA relied heavily on feature engineering (Wagner et al., 2014; Kiritchenko et al., 2014), and subsequent neural network-based methods (Nguyen and Shirai, 2015; Wang et al., 2016; Tang et al., 2015, 2016; Wang et al., 2017) achieved higher accuracy.”
Lifelong Learning Memory Networks for Aspect Sentiment Classification (2018, background): “[6], [7], [19], where manual feature engineering and external resources are required.”
How Translation Alters Sentiment (2016, methodology): “In the process of developing these experiments to study how translation impacts sentiment, we created a new dialectal Arabic sentiment lexicon and a state-of-the-art Arabic sentiment analysis system by porting NRC-Canada’s competition winning system (Mohammad, Kiritchenko, & Zhu, 2013; Kiritchenko, ”
Attentional Encoder Network for Targeted Sentiment Classification (2019, methodology): “Non-RNN based baselines: • Feature-based SVM (Kiritchenko et al., 2014) is a traditional support vector machine based model with extensive feature engineering.”
Challenges in Sentiment Analysis (2017, methodology): “Bag-of-word models such as the NRC-Canada system (Mohammad et al., 2013a; Kir-itchenko, Zhu, Cherry, & Mohammad, 2014a; Kiritchenko et al., 2014b) and Unitn (Severyn & Moschitti, 2015) have been very successful in recent shared task competitions on determining sentiment of whole tweets, SMS…”
Multi-grained Attention Network for Aspect-Level Sentiment Classification (2018, methodology): “Traditional approaches (Jiang et al., 2011; Kiritchenko et al., 2014; Vo and Zhang, 2015) regard this task as the text classification problem and design effective features, which are utilized in statistical learning algorithms for training a classifier.”
Syntax-Aware Aspect Level Sentiment Classification with Graph Attention Networks (2019, methodology): “Kiritchenko et al. (2014) also propose a similar method and they show that adding parse context features could improve the prediction accuracy by more than one percent.”
Aspect-based Sentiment Classification with Aspect-specific Graph Convolutional Networks (2019, methodology): “In order to comprehensively evaluate the two variants of our model, namely, ASGCN-DG and ASGCN-DT, we compare them with a range of baselines and state-of-the-art models, as listed below: • SVM (Kiritchenko et al., 2014) is the model which has won SemEval 2014 task 4 with conventional feature extract”
Deep Learning for Aspect-Based Sentiment Analysis: A Comparative Review (2019, neutral): “5 Kiritchenko et al. (2014) SVM + n-grams + parse trees, + sentiment lexica Acc: 80.”
Targeted Aspect-Based Sentiment Analysis via Embedding Commonsense Knowledge into an Attentive LSTM (2018, background): “Early works on ABSA have mainly relied on feature-engineering to characterize sentences (Wagner et al. 2014; Kiritchenko et al. 2014).”
Sentiment Analysis of Short Informal Texts (2014, methodology): “In (Kiritchenko et al., 2014) we show how our approach can be adapted to identify the sentiment for a speciﬁed target.”
Revisiting Word Embedding for Contrasting Meaning (2015, background): “(2014) learned sentiment-specific embedding and applied it to sentiment analysis of tweets, which was often solved with more conventional methods (Zhu et al., 2014b; Kiritchenko et al., 2014a; Kiritchenko et al., 2014b).”
Current State of Text Sentiment Analysis from Opinion to Emotion Mining (2017, neutral): “For more elaboration on each of them, one can refer to Svetlana Kiritchenko and Mohammad [2014], Mohammad et al. [2013], Zhu et al. [2014], and Kiritchenko et al. [2014].”
Dual Graph Convolutional Networks for Aspect-based Sentiment Analysis (2021, background): “Earlier meth-ods (Titov and McDonald, 2008; Jiang et al., 2011; Kiritchenko et al., 2014; Vo and Zhang, 2015) are usually based on handcrafted features and fail to model the dependency between the given aspect and its context.”
Target-Dependent Sentiment Classification With BERT (2019, methodology): “DCU and NRC-Canada ranked the top two at SemEval-2014 task 4 subtask 2 (aspect term polarities) [28].”
Stance and Sentiment in Tweets (2016, methodology): “8 Positive and negative language tend to have sufﬁcient amount of commonality regardless of topic of discussion, and hence sentiment analysis systems traditionally learn a single model from all of the training data [Liu 2015; Kiritchenko et al. 2014a; Rosenthal et al. 2015].”
Aspect Based Sentiment Analysis with Gated Convolutional Networks (2018, methodology): “Without the large amount of sentiment lexicons, SVM perform worse than neural meth-ods.”