Lexical TF-IDF: An n-gram Feature Space for Cross-Domain Classification of Sentiment Reviews

Dey, Atanu; Jenamani, Mamata; Thakkar, Jitesh J.

doi:10.1007/978-3-319-69900-4_48

Cited by 27 publications

(11 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Most of the works in the field of NLP were focused on feature engineering and extraction, such as bag-of-words, TF-IDF, and N-grams techniques [10]. Formerly, those techniques have been used in combination with conventional machine learning.…”

Section: B Nlp and ML Methodsmentioning

confidence: 99%

Explainable Artificial Intelligence (XAI) towards Model Personality in NLP task

Adi

Nurdin

2021

IJOE

View full text Add to dashboard Cite

In recent years, the development of Deep Learning in the field of Natural Language Processing, especially in sentiment analysis, has achieved significant progress and success. It is because of the availability of large amounts of text data and the ability of deep learning techniques to produce sophisticated predictive results from various data features. However, the sophisticated predictions that are not accompanied by sufficient information on what is happening in the model will be a major setback. Therefore, the significant development of the Deep Learning model must be accompanied by the development of the XAI method, which helps provide information about what drives the model to get predictable results. Simple Bidirectional LSTM and complex Bi-GRU-LSTM-CNN model for Sentiment Analysis were proposed in the present research. Both models were analyzed further using three different XAI methods (LIME, SHAP, and Anchor) in which they were used and compared to two proposed models, proving that XAI is not limited to giving information about what happens in the model but can also help us to understand and distinguish models' personality and behaviour. Keywords⎯ Deep learning, Explainable artificial intelligence, Natural language processing, Sentiment analysis I.

show abstract

Section: B Nlp and ML Methodsmentioning

confidence: 99%

Explainable Artificial Intelligence (XAI) towards Model Personality in NLP task

Adi

Nurdin

2021

IJOE

View full text Add to dashboard Cite

show abstract

“…The two substantial tasks are indexing and weighting; TF-IDF manages the weighting. It determines the weight of t in a document D. TF-IDF is derived from TF and IDF as follows [18]: Finally, in order to validate the introduced model, we introduced a comparative study that showed the efficiency of the introduced model over other methods in the related literature.…”

Section: Tf-idf Vectorizermentioning

confidence: 99%

A New Model for a Secure Social Media Application

Riad

Elhoseny

2022

Applied Sciences

View full text Add to dashboard Cite

The progress of data technology and wireless networks is generated by open online communication channels. Unfortunately, trolls are abusing the technology for executing cyberattacks and threats. An automated cybersecurity solution is vital for avoiding the threats and security issues from social media. This can be a requirement for tackling and considering cyberbullying in various aspects including prevention of such incidents and automated detection. This study introduces a novel Artificial Fish Swarm Algorithm with Weighted Extreme Learning Machine (AFSA-WELM) model for cybersecurity on social media. The proposed model is mostly intended to detect the existence of cyberbullying on social media. The proposed model starts by processing the dataset and making it ready for the next stages of the model. It then uses the TF-IDF vectorizer for word embedding. After that, it uses the WELM model for the identification and classification of cyberbullying. Finally, the optimal tunning parameters used in the WELM model are derived for the AFSA model. The experimental analysis has shown that the proposed model achieves maximum accuracy compared with existing algorithms. Moreover, our proposed model achieves maximum precision–recall performance with various datasets.

show abstract

“…• Aumento do alcance dos transformadores de valência: a estratégia usada era baseada no 3-gram [Dey et al 2017], onde 3 era o número de palavras alcançadas por negações, intensificadores ou redutores de valência. O seu alcance foi estendido até o sinal de pontuação mais próximo, abrangendo e influenciando todas as outras palavras no caminho.…”

Section: Sentimento (Rede De Frames)unclassified

Identificação do Perfil Emocional do Aluno Através de Análise de Sentimento: Combatendo a Evasão Escolar

Bóbó¹,

Campos²,

Ströele³

et al. 2019

Anais Do XXX Simpósio Brasileiro De Informática Na Educação (SBIE 2019)

View full text Add to dashboard Cite

This article presents the SASys architecture, which main goal is to identify the student's emotional profile through the use of FrameNet Brasil to establish the author's feeling in texts. A recommendation system, based on the student's emotional profile and learning style, sends motivational messages to mitigate school dropout. The proposal was evaluated by a case study with students of the Methodology of Scientific and Educational Research class of a learning distance course. The results point to the feasibility of the proposal for the assertiveness of the student's emotional profile during the class and detection of students' risk of dropout.Resumo. Este artigo apresenta a arquitetura SASys, cujo objetivo principal é identificar o estado emocional do aluno, através do uso da FrameNet Brasil para determinar o sentimento do autor em textos. Um sistema de recomendação, baseado no estado emocional do aluno e do seu estilo de aprendizagem, envia mensagens motivacionais para mitigar a evasão. A proposta foi avaliada com a condução de um estudo de caso com alunos da disciplina de Metodologia de Pesquisa Científica e Educacional de um curso à Distância. Os resultados apontam para a viabilidade da proposta para a assertividade do estado emocional do aluno ao longo da disciplina e detecção de alunos com risco de evasão.

show abstract

Lexical TF-IDF: An n-gram Feature Space for Cross-Domain Classification of Sentiment Reviews

Cited by 27 publications

References 7 publications

Explainable Artificial Intelligence (XAI) towards Model Personality in NLP task

Explainable Artificial Intelligence (XAI) towards Model Personality in NLP task

A New Model for a Secure Social Media Application

Identificação do Perfil Emocional do Aluno Através de Análise de Sentimento: Combatendo a Evasão Escolar

Contact Info

Product

Resources

About