Multi-label Classification of User Reactions in Online News

Curi, Zacarias; Britto, Alceu de Souza; Paraíso, Emerson Cabrera

doi:10.48550/arxiv.1809.02811

Cited by 1 publication

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DSL Roman-Urdu Sentiments corpus consists of 3241 mobile related sentiments manually annotated against positive, negative and neutral intents. Entire dataset is crawled from mobile review website namely WhatMobile 17 . Pre-processing of the corpus is performed in a same manner as applied for other corpus used for the generation of neural word embeddings (discussed in section 4.1)…”

Section: Benchmark Dataset: Dsl Roman-urdu Sentimentsmentioning

confidence: 99%

“…The availability of such rich resources has largely aided the researchers to perform a comparative analysis of diverse machine and deep learning methodologies and to assess the effectiveness of enhanced novel methodologies. Evidently, this progress has led the emergence of jaw-dropping applications for these rich resourced languages which are capable to perform sentiment classification in real-time such as Nexmo 7 , intent detection like LiveIntent 8 , emotion identification [13], emotion classification [14], constructing user interests profile [15] [16], and user reaction categorization [17].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Precisely Xtreme-Multi Channel Hybrid Approach for Roman Urdu Sentiment Analysis

et al. 2020

View full text Add to dashboard Cite

In order to accelerate the performance of various Natural Language Processing tasks for Roman Urdu, this paper for the very first time provides 3 neural word embeddings prepared using most widely used approaches namely Word2vec, FastText, and Glove. The integrity of generated neural word embeddings is evaluated using intrinsic and extrinsic evaluation approaches. Considering the lack of publicly available benchmark datasets, it provides a first-ever Roman Urdu public dataset which consists of 3241 sentiments annotated against positive, negative, and neutral classes. To provide benchmark baseline performance over the presented dataset for Roman Urdu sentiment analysis, we adapt diverse machine learning (Support Vector Machine, Logistic Regression, Naive Bayes), deep learning (convolutional neural network, recurrent neural network), and hybrid deep learning approaches. Performance impact of generated neural word embeddings based representation is compared with other most widely used bag of words based feature representation approaches using diverse machine and deep learning classifiers. In order to improve the performance of Roman Urdu sentiment analysis, it proposes a novel precisely extreme multi-channel hybrid methodology which makes use of convolutional and recurrent neural networks along with pre-trained neural word embeddings. The proposed hybrid approach outperforms adapted machine learning approaches by the significant figure of 9% and deep learning approaches by the figure of 4% in terms of F1-score. INDEX TERMS Fast-Text, Glove, Pretrain word embeddings for Roman Urdu, Roman Urdu Sentiment Analysis, Word2Vec,

show abstract

Section: Benchmark Dataset: Dsl Roman-urdu Sentimentsmentioning

confidence: 99%