Sentiment analysis in Twitter

Martínez‐Cámara, Eugenio; Martín–Valdivia, María Teresa; Ureña‐López, L. Alfonso; Montejo-Ráez, Arturo

doi:10.1017/s1351324912000332

Cited by 199 publications

(94 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Tables 13 and 14 describe the performance of the Ngram models for spam detection, the performance gain for unigrams and bigrams is marginal while it decreases for rest of the higher order N-grams when taking the priors into account. This is consistent with the general trend for Twitter as shown in [16]. The decrease in accuracy (for N >= 3) is in tandem with the dip in the precision, recall, specificity and f1-score.Unigrams perform uniquely while including the priors, there is an increase in accuracy of 5.1% while the precision decreases by 2.36% at the same time.…”

Section: Results For Twittersupporting

confidence: 90%

Fashioning Data - A Social Media Perspective on Fast Fashion Brands

Chakraborty

Kundu²,

Agarwal³

2016

Proceedings of the 7th Workshop on Computational Approaches To Subjectivity, Sentiment and Social Media Analysis

View full text Add to dashboard Cite

In this paper, we study the performance of N-gram language models on classification tasks such as sentiment analysis and spam detection and evaluate the effect of prior probability estimates on the results. Our data is in the form of public online posts pertaining to fast fashion brands, from different social media channels (Twitter and Facebook). We propose a novel ensemble model based on the combination of different N-grams in order to deal with the heteroskedastic nature of data collected from these social media channels. This has been further extended to increase the efficacy of the classification results.

show abstract

Section: Results For Twittersupporting

confidence: 90%

Fashioning Data - A Social Media Perspective on Fast Fashion Brands

Chakraborty

Kundu²,

Agarwal³

2016

Proceedings of the 7th Workshop on Computational Approaches To Subjectivity, Sentiment and Social Media Analysis

View full text Add to dashboard Cite

show abstract

“…Location information is crucially important information for analyses such as disaster analysis (Sakaki et al, 2010), disease analysis (Culotta, 2010), and political analysis (Tumasjan et al, 2010). Such information is also useful for analyses such as sentiment analysis (Martínez-Cámara et al, 2014) and user attribute analysis (Rao et al, 2010) to undertake detailed region-specific analyses.…”

Section: Introductionmentioning

confidence: 99%

Unifying Text, Metadata, and User Network Representations with a Neural Network for Geolocation Prediction

Miura¹,

Taniguchi²,

Taniguchi³

et al. 2017

Proceedings of the 55th Annual Meeting of the Association For Computational Linguistics (Volume 1: Long Papers)

View full text Add to dashboard Cite

We propose a novel geolocation prediction model using a complex neural network. Our model unifies text, metadata, and user network representations with an attention mechanism to overcome previous ensemble approaches. In an evaluation using two open datasets, the proposed model exhibited a maximum 3.8% increase in accuracy and a maximum of 6.6% increase in accuracy@161 against previous models. We further analyzed several intermediate layers of our model, which revealed that their states capture some statistical characteristics of the datasets.

show abstract

“…However, analyses of the emotional tendencies of learners are usually coarse-grained at present, and the effect of sentiment change on graduation rate has not been analyzed in depth [48]. Therefore, it is necessary to propose a more fine-grained semantic analysis model to analyze sentiments expressed by learners in forums of courses.…”

Section: Related Workmentioning

confidence: 99%

Semantic Analysis of Learners’ Emotional Tendencies on Online MOOC Education

Wang

Zhou

2018

Sustainability

View full text Add to dashboard Cite

Abstract:As a new education product in the information age, Massive Open Online Courses (MOOCs) command momentous public attention for their unexpected rise and flexible application. However, the striking contrast between the high rate of registration and the low rate of completion has put their development into a bottleneck. In this paper, we present a semantic analysis model (SMA) to track the emotional tendencies of learners in order to analyze the acceptance of the courses based on big data from homework completion, comments, forums and other real-time update information on the MOOC platforms. Through emotional quantification and machine learning calculations, graduation probability can be predicted for different stages of learning in real time. Especially for learners with emotional tendencies, customized instruction could be made in order to improve completion and graduation rates. Furthermore, we classified the learners into four categories according to course participation time series and emotional states. In the experiments, we made a comprehensive evaluation of the students' overall learning status by kinds of learners and emotional tendencies. Our proposed method can effectively recognize learners' emotional tendencies by semantic analysis, providing an effective solution for MOOC personalized teaching, which can help achieve education for sustainable development.

show abstract

Sentiment analysis in Twitter

Cited by 199 publications

References 41 publications

Fashioning Data - A Social Media Perspective on Fast Fashion Brands

Fashioning Data - A Social Media Perspective on Fast Fashion Brands

Unifying Text, Metadata, and User Network Representations with a Neural Network for Geolocation Prediction

Semantic Analysis of Learners’ Emotional Tendencies on Online MOOC Education

Contact Info

Product

Resources

About