Exploiting Optimal Self-Attention Deep Learning-based Recognition of Textual Emotions for Disabled Persons DOI
Haya Mesfer Alshahrani, Ishfaq Yaseen,

Suhanda Drar

и другие.

Fractals, Год журнала: 2024, Номер 32(09n10)

Опубликована: Сен. 24, 2024

A disability is a significant issue that has posed and continues to pose challenge. Disability basis of frustration because it can be observed as mental, constraint, cognitive, physical handicap inhibits the individual’s growth involvement. Consequently, effort been put into removing these kinds restrictions. These initiatives address trouble disabled people encounter. People with disabilities often need rely on others meet their requirements. Machine learning (ML) excelling in producing smart cities offering secure environment for individuals. Emotional detection an important research domain expose many appreciated inputs. Emotion expressed differently through speech facial expressions, gestures, written text. text document fundamentally content-based classification task, utilizing models from deep (DL), complex systems natural language processing (NLP). This paper presents Optimal Self-Attention DL-based Recognition Textual Emotions (OSADL-RTE) technique Disabled Persons. The presented OSADL-RTE focuses identifying distinct types emotions textual data. As primary preprocessing step, comprises different phases transform input useful way. For word embedding, bag words (BoWs) approach exploited. derives self-attention long short-term memory (SA-LSTM) identify emotions. Lastly, arithmetic fractals optimization algorithm (AOA) correctly tunes hyperparameter selection SA-LSTM approach. experimental study occurs emotion database. investigational outcome portrayed superior accuracy 99.59% over existing methods.

Язык: Английский

Leveraging Corpus Linguistics and Data-Driven Deep Learning for Textual Emotion Analysis DOI
Somia Asklany,

Najla I. Al-shathry,

Abdulkhaleq Q. A. Hassan

и другие.

Fractals, Год журнала: 2024, Номер 32(09n10)

Опубликована: Сен. 27, 2024

Emotions have played a major part in the conversation, as they express context to conversation. Text or words conversation contain contextual and lexical meanings. In recent times, obtaining emotion from text has been an attractive area of research. With emergence machine learning (ML) algorithms hardware aid ML method, identifying with provides significant promising solutions. The main objective Textual Emotion Analysis (TEA) is analyze extract user’s emotional states text. Many different Complex Systems Deep Learning (DL) fast-paced developed proved their effectiveness several fields including audio, image, natural language processing (NLP). This moved researchers away classical DL for academic research work. study develops new Corpus Linguistics Data-Driven (CLD3L-TEA) technique. CLD3L-TEA technique mainly investigates distinct types emotions that endure social media model, raw data can be pre-processed ways. Next, multi-weighted TF–IDF model used generate feature vectors. For identification emotions, applied gated recurrent unit (GRU). At last, hyperparameter range GRU executed by Fractal Harris Hawks Optimization (HHO) model. experimental validation on benchmark dataset illustrates supremacy this over approaches.

Язык: Английский

Процитировано

1
Unraveling the World of Artificial Emotional Intelligence DOI
Belghachi Mohammed

Advances in psychology, mental health, and behavioral studies (APMHBS) book series, Год журнала: 2024, Номер unknown, С. 17 - 51

Опубликована: Фев. 26, 2024

Artificial emotional intelligence (AEI) is a burgeoning field at the intersection of technology and human emotion, seeking to imbue machines with capacity perceive, understand, respond emotions. This chapter encapsulates comprehensive exploration AEI, encompassing its foundations, technical intricacies, applications, user perspectives, ethical considerations, challenges, real-world case studies. The foundations AEI involve unraveling complexities emotions, from facial expressions voice tones physiological signals. Understanding aspects including data acquisition, feature extraction, machine learning models, multi-modal fusion, provides insights into sophisticated mechanisms driving emotionally intelligent systems. AEI's applications span diverse domains, virtual health assistants providing mental well-being support aware educational platforms adapting students' needs.

Язык: Английский

Процитировано

0
Exploiting Optimal Self-Attention Deep Learning-based Recognition of Textual Emotions for Disabled Persons DOI
Haya Mesfer Alshahrani, Ishfaq Yaseen,

Suhanda Drar

и другие.

Fractals, Год журнала: 2024, Номер 32(09n10)

Опубликована: Сен. 24, 2024

A disability is a significant issue that has posed and continues to pose challenge. Disability basis of frustration because it can be observed as mental, constraint, cognitive, physical handicap inhibits the individual’s growth involvement. Consequently, effort been put into removing these kinds restrictions. These initiatives address trouble disabled people encounter. People with disabilities often need rely on others meet their requirements. Machine learning (ML) excelling in producing smart cities offering secure environment for individuals. Emotional detection an important research domain expose many appreciated inputs. Emotion expressed differently through speech facial expressions, gestures, written text. text document fundamentally content-based classification task, utilizing models from deep (DL), complex systems natural language processing (NLP). This paper presents Optimal Self-Attention DL-based Recognition Textual Emotions (OSADL-RTE) technique Disabled Persons. The presented OSADL-RTE focuses identifying distinct types emotions textual data. As primary preprocessing step, comprises different phases transform input useful way. For word embedding, bag words (BoWs) approach exploited. derives self-attention long short-term memory (SA-LSTM) identify emotions. Lastly, arithmetic fractals optimization algorithm (AOA) correctly tunes hyperparameter selection SA-LSTM approach. experimental study occurs emotion database. investigational outcome portrayed superior accuracy 99.59% over existing methods.

Язык: Английский

Процитировано

0