Skip to main content
SpringerLink
Log in
Springer
Cognitive Neurodynamics
Publishing model:
Hybrid

Cognitive Neurodynamics - Special Issue Call for Paper

Special issue title

Frontier Research on Artificial Intelligence for Sentiment Analysis of Intelligent Systems

Guest editors

Dr. Ahmed A. Abd El-Latif, Department of Mathematics and Computer Science, Menoufia University, Egypt
Dr. Edmond Shu-lim Ho, Department of Computer and Information Sciences, Northumbria University, UK
Dr. Jialiang Peng, School of Data Science and Technology, Heilongjiang University, China

About the special issue

Complex intelligent systems, such as Internet-of-Things (IoT), social network services (SNS), virtual reality (VR), augmented reality (AR), etc., are becoming an essential part of human life. The dynamics of these systems alter human behavior in many ways. Hence, it is crucial to understand the underlying system dynamics and what this entails for users' sentiments. Sentiment analysis is a very powerful and commonly used analysis method in data mining. It provides an excellent option to determine, evaluate, monitor, and understand the sentiment of consumers regarding a product or a service. They are useful for organizations to make decisions to improve their business processes. In the past few years, many researchers have worked on sentiment analysis problems and applied machine learning approaches to solve them, such as neural networks, support vector machines, and other methods. But there still exist some problems in current sentiment analysis that could affect the performance of classifiers in the future.

Sentiment analysis is a key element in solving complex, intelligent systems problems. Complex Intelligent Systems rely on the data that is generated and collected from users, which has various sources and formats. They are affected by many factors such as domain knowledge, sentiment orientation, sentiment word association, etc. Since the data contain complex multi-level structures with high-level semantics and deep relationships, conventional artificial intelligence methods that fail to convert data into machine knowledge become insufficient in responding to the complex problems of complex, intelligent systems. Neural networks are good at detecting low-level features but fail to deal with complicated semantic structures and problems. In contrast, symbolic artificial intelligence is able to deal with various challenges but with weak generalization ability due to the massive domain knowledge engineering involved. Neuro-symbolic artificial intelligence combines the strengths of both neural networks and symbolic artificial intelligence for solving challenges like those in complex, intelligent systems. It is a novel idea that presents forward-looking directions and new research frontiers in artificial intelligence.

Neuro-symbolic artificial intelligence is the combination of neural networks and symbolic artificial intelligence--a hybrid approach combining a "perception oriented" neural network and a "reasoning oriented" knowledge grounded system. Although neuro-symbolic integration has been developing for a long time, symbol manipulating with deep neural networks has become a subject of considerable attention in recent years. Neuro-symbolic hybrid techniques provide interesting possibilities for sentiment analysis in complex, intelligent systems due to their ability to explicitly represent semantic knowledge, flexibility, and robustness. Machine learning algorithms may work well when the size of labeled data is large, but they do not scale well to large vocabularies or to rare words or multiword expressions. It is possible to overcome these limitations by combining machine learning and symbolic reasoning, which paves the way for more advanced research in neuro-symbolic artificial intelligence for sentiment analysis in complex intelligent systems.

This special issue focuses on the challenges and opportunities related to sentiment analysis in complex, intelligent systems. It briefly examines several recent advances in neuro-symbolic artificial intelligence. The goal is to highlight the core concepts of operating with knowledge and neural computations, maps, and algorithms in neuro-symbolic artificial intelligence, which can render a robust capability for intelligent reasoning and decision making in complex, intelligent systems.

List of Topics (include, but not limited to the following):

  • Neuro-symbolic artificial intelligence for knowledge representation in complex intelligent systems
  • Sentiment and emotion analysis in complex intelligent systems with neuro-symbolic artificial intelligence
  • Critical assessment of sentiment analysis and emotion understanding in complex systems with neuro-symbolic artificial intelligence
  • Reasoning and decision making in complex intelligent systems with neuro-symbolic artificial intelligence
  • Effective ways of information integration and knowledge processing in complex intelligent systems with neuro-symbolic artificial intelligence
  • Neuro-symbolic artificial intelligence for explainable sentiment and emotion prediction in complex intelligent systems
  • Aspect-based sentiment analysis with neuro-symbolic artificial intelligence for complex intelligent systems
  • Neuro-symbolic knowledge representation and analysis for complex intelligent systems
  • Commonsense reasoning in complex intelligent systems with neuro-symbolic artificial intelligence
  • Multimodal and multilingual aspects of sentiment analysis in complex intelligent systems with neuro-symbolic artificial intelligence

Important Dates

- Submissions Deadline: 01 November 2022
- Notification to Authors: 13 January 2023
- Deadline for revision submissions: 28 March 2023  
- Notification of final decisions: 05 June 2023

Search

Navigation