Multilingual emotion classification using supervised learning: Comparative experiments

Becker, Karin; Moreira, Viviane Pereira; Santos, Aline Gondim

doi:10.1016/j.ipm.2016.12.008

Cited by 40 publications

(27 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These representations can arise from feature design and control the noise of data. In our case, the noise is likely to come from two sources, namely, incorrect translations or features that are not appropriate [4], [13], [14]. Thus it is crucial to distinguish between the drop in accuracy that caused by inappropriate feature representation from that might have occurred because of erroneous translation.…”

Section: Feature Engineeringmentioning

confidence: 98%

“…More recently, Balahur and Trurchi, [4] and Becker et al [14] investigated how a simple strategy can address the problem of sentiment analysis in multiple languages. Particularly, they analyze how the use of machine translation systems -such as Google Translate -can affect the performance of English Sentiment Analysis methods in nonEnglish datasets.…”

Section: Related Workmentioning

confidence: 99%

“…For this purpose, the present work adopted Google Translate (GT) to translate the corpora from English-to-Arabic as it offers API access and is considered the state-of-the-art machine translation system used today [4], [13], [14]. www.ijacsa.thesai.org With the aid of machine translation, we can translate the data in both directions to form target language to the source language (i.e., Arabic to English) or from source language to the target language (i.e.…”

Section: A Data Acquisitionmentioning

confidence: 99%

See 2 more Smart Citations

Cross-Lingual Sentiment Classification from English to Arabic using Machine Translation

Al-Shabi¹,

Adel²,

Omar³

et al. 2017

ijacsa

View full text Add to dashboard Cite

Abstract-Cross-lingual sentiment learning is becoming increasingly important due to the multilingual nature of usergenerated content on social media and the scarce resources for languages other than English. However, cross-lingual sentiment learning is a challenging task due to the different distribution between translated data and original data and due to the language gap, i.e. each language has its own ways to express sentiments. This work explores the adaptation of English resources for sentiment analysis to a new language, Arabic. The aim is to design a light model for cross-lingual sentiment classification from English to Arabic, without any manual annotation effort which, at the same time, is easy to build and does not require deep linguistic analysis. The ultimate goal is to find an optimal baseline model and to determine the relation between the noise in the translated data and the accuracy of sentiment classification. Different configurations of several factors are investigated including feature representation, feature reduction methods, and the learning algorithms to find the optimal baseline model. Experiments show that a good classification model can be obtained from translated data regardless of the artificial noise added by machine translation. The results also show a significant cost to automation, and thus the best path to future enhancement is through the inclusion of language-specific knowledge and resources.

show abstract

Section: Feature Engineeringmentioning

confidence: 98%

Section: Related Workmentioning

confidence: 99%

Section: A Data Acquisitionmentioning

confidence: 99%

See 1 more Smart Citation

Cross-Lingual Sentiment Classification from English to Arabic using Machine Translation

Al-Shabi¹,

Adel²,

Omar³

et al. 2017

ijacsa

View full text Add to dashboard Cite

show abstract

“…However, loss of polarity or the issue of labeling mismatching caused by the erroneous machine translation of source language training data severely deteriorates the analysis performance [ 3 ]. Previous work predominantly relies on machine translation engines or bilingual lexicons to project data from the source language to the target language [ 1 , 2 , 13 ]. Machine translation quality is still low and far from satisfactory [ 1 , 2 , 9 ].…”

Section: Related Studiesmentioning

confidence: 99%

“…First, translation-based methods are dependent on machine translation or bilingual dictionaries [ 17 , 18 ] to project the annotations from the source language into the target language, or vice versa. A classification model is then trained over the target data with projected annotations [ 2 , 13 ]. Earlier works utilize the bilingual dictionaries to transfer the sentiment features from one language to another.…”

Section: Related Studiesmentioning

confidence: 99%

Cross Lingual Sentiment Analysis: A Clustering-Based Bee Colony Instance Selection and Target-Based Feature Weighting Approach

Almansor

Zhang

Khan

et al. 2020

Sensors

View full text Add to dashboard Cite

The lack of sentiment resources in poor resource languages poses challenges for the sentiment analysis in which machine learning is involved. Cross-lingual and semi-supervised learning approaches have been deployed to represent the most common ways that can overcome this issue. However, performance of the existing methods degrades due to the poor quality of translated resources, data sparseness and more specifically, language divergence. An integrated learning model that uses a semi-supervised and an ensembled model while utilizing the available sentiment resources to tackle language divergence related issues is proposed. Additionally, to reduce the impact of translation errors and handle instance selection problem, we propose a clustering-based bee-colony-sample selection method for the optimal selection of most distinguishing features representing the target data. To evaluate the proposed model, various experiments are conducted employing an English-Arabic cross-lingual data set. Simulations results demonstrate that the proposed model outperforms the baseline approaches in terms of classification performances. Furthermore, the statistical outcomes indicate the advantages of the proposed training data sampling and target-based feature selection to reduce the negative effect of translation errors. These results highlight the fact that the proposed approach achieves a performance that is close to in-language supervised models.

show abstract

Emotion Detection from Text in Learning Environments: A Review

Bustos-López

Cruz-Ramírez

Guerra‐Hernández

et al. 2021

Studies in Computational Intelligence

View full text Add to dashboard Cite

Multilingual emotion classification using supervised learning: Comparative experiments

Cited by 40 publications

References 24 publications

Cross-Lingual Sentiment Classification from English to Arabic using Machine Translation

Cross-Lingual Sentiment Classification from English to Arabic using Machine Translation

Cross Lingual Sentiment Analysis: A Clustering-Based Bee Colony Instance Selection and Target-Based Feature Weighting Approach

Emotion Detection from Text in Learning Environments: A Review

Contact Info

Product

Resources

About