Aspect Clustering Methods for Sentiment Analysis

Vargas, Francielle Alves; Pardo, Thiago Alexandre Salgueiro

doi:10.1007/978-3-319-99722-3_37

Cited by 13 publications

(12 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In this work, we implement and use this as a baseline classifier, along with another baseline classifier built from word embedding representations of review texts. Others such as (Titov & McDonald, 2008;Huang, Rogers, & Joo, 2014;Zhou, Wan, & Xiao, 2015;Vargas & Pardo, 2018) have used topic modelling techniques such as LDA or PLSA to identify the aspects that users discuss in restaurant reviews, however they did not study its effectiveness for review rating prediction. The review rating prediction task is different from work on recommender systems (Resnick & Varian, 1997;Sarwar, Karypis, Konstan, & Riedl, 2000;Z.…”

Section: Related Workmentioning

confidence: 99%

Leveraging aspect phrase embeddings for cross-domain review rating prediction

Jiang

Zubiaga

2019

PeerJ Computer Science

View full text Add to dashboard Cite

Online review platforms are a popular way for users to post reviews by expressing their opinions towards a product or service, as well as they are valuable for other users and companies to find out the overall opinions of customers. These reviews tend to be accompanied by a rating, where the star rating has become the most common approach for users to give their feedback in a quantitative way, generally as a likert scale of 1-5 stars. In other social media platforms like Facebook or Twitter, an automated review rating prediction system can be useful to determine the rating that a user would have given to the product or service. Existing work on review rating prediction focuses on specific domains, such as restaurants or hotels. This, however, ignores the fact that some review domains which are less frequently rated, such as dentists, lack sufficient data to build a reliable prediction model. In this paper, we experiment on 12 datasets pertaining to 12 different review domains of varying level of popularity to assess the performance of predictions across different domains. We introduce a model that leverages aspect phrase embeddings extracted from the reviews, which enables the development of both in-domain and cross-domain review rating prediction systems. Our experiments show that both of our review rating prediction systems outperform all other baselines. The cross-domain review rating prediction system is particularly significant for the least popular review domains, where leveraging training data from other domains leads to remarkable improvements in performance. The in-domain review rating prediction system is instead more suitable for popular review domains, provided that a model built from training data pertaining to the target domain is more suitable when this data is abundant.

show abstract

Section: Related Workmentioning

confidence: 99%

Leveraging aspect phrase embeddings for cross-domain review rating prediction

Jiang

Zubiaga

2019

PeerJ Computer Science

View full text Add to dashboard Cite

show abstract

“…Technical literature presents few related work, we found the following works about WE applied to Portuguese texts: [10] , [19], and [23]. Hartmann et al (2007) [10] trained 31 WE models using the Word2Vec, Fast-Text, Wang2Vec and GloVe algorithms.…”

Section: Related Workmentioning

confidence: 99%

“…Nowadays, one trend in NLP is to use vectors of words whose syntactic similarities correlate with semantic similarities. These vectors are used to calculate similarities between terms [10,19,23]. Bengio et al (2003) [2] were among the first to introduce the term Word Embedding, where words or phrases are mapped to vectors of real numbers.…”

Section: Introductionmentioning

confidence: 99%

“…Likewise other NLP tasks, research about WE applied to Portuguese texts is still in its first steps [23]. [10] made syntactic, semantic, intrinsic and extrinsic analysis of different trained approaches of WE.…”

Section: Introductionmentioning

confidence: 99%

“…Vargas and Pardo [23] used WE to aspect (explicit and implicit) identification and clustering that are critical tasks for sentiment analysis.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Word Embedding Analysis towards Ontology Enrichment

Poetsch¹,

Corrêa²,

Freitas³

2019

RCS

View full text Add to dashboard Cite

Word Embedding is a set of language modeling and feature learning techniques in Natural Language Processing where words or phrases are mapped to vectors of real numbers. This approach could be used in many tasks of Natural Language Processing, such as Text Classification, Part-Of-Speech Tagging, Named Entity Recognition, Sentiment Analysis, and others. In this paper we created different Word Embedding models, using TripAdvisor's hotel reviews. The corpus was pre-processed, in order to reduce noise, and then submitted to four Word Embedding algorithms: Word2Vec, FastText, Wang2Vec, and GloVe. Finally, HOntology concepts and relations are compared with the outputs of models created aiming to improve it, enriching this domain ontology.

show abstract

Identifying Fine-Grained Opinion and Classifying Polarity on Coronavirus Pandemic

Vargas

Santos

Rocha

2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

In this paper, we explore the fine-grained opinion identification and polarity classification tasks using twitter data on the COVID-19 pandemic in Brazilian Portuguese. We trained machine learning-based classifiers using a few different methods and tested how well they performed different tasks. For polarity classification, we tested a crossdomain strategy in order to measure the performance of the classifiers among different domains. For fine-grained opinion identification, we provide a taxonomy of opinion aspects and employed them in conjunction with machine learning methods. Based on the obtained results, we found that the cross-domain data improved the results of the polarity classification. For fine-grained opinion identification, the use of a domain taxonomy presented competitive results for the Portuguese language.

show abstract

Aspect Clustering Methods for Sentiment Analysis

Cited by 13 publications

References 15 publications

Leveraging aspect phrase embeddings for cross-domain review rating prediction

Leveraging aspect phrase embeddings for cross-domain review rating prediction

A Word Embedding Analysis towards Ontology Enrichment

Identifying Fine-Grained Opinion and Classifying Polarity on Coronavirus Pandemic

Contact Info

Product

Resources

About