Symbolic artificial intelligence

Symbolic Artificial Intelligence (Symbolic AI), also known as good old-fashioned artificial intelligence (GOFAI), is a paradigm in Artificial Intelligence research that represents knowledge explicitly in terms of symbols and rules that can be directly understood by human users. This approach to AI research and development emphasizes the use of human-readable symbols to represent problems, objects, and their relationships within a domain of knowledge, and it employs rules for manipulating these symbols in order to solve problems or derive new information.

Overview

Symbolic AI was the dominant form of AI research from the 1950s into the late 1980s. It is based on the premise that all human thought can be understood as symbol manipulation, a hypothesis known as the Physical Symbol System Hypothesis. According to this view, solving a problem means searching through a space of possible solutions and applying rules to move from one state to another until a goal state is reached.

The key components of Symbolic AI include:

• Knowledge Base: A large database of knowledge about the world, which includes facts and rules about how those facts relate to each other.
• Inference Engine: A program that applies logical rules to the knowledge base to derive new information or make decisions.

Applications

Symbolic AI has been successfully applied in various domains, including:

• Expert Systems: Computer systems that emulate the decision-making ability of a human expert.
• Natural Language Processing (NLP): The ability of a computer program to understand human language as it is spoken or written.
• Computer Vision: Although primarily the domain of Connectionism and neural networks in recent years, early computer vision systems attempted to use symbolic representations to recognize objects.

Advantages and Limitations

The main advantages of Symbolic AI include its transparency and explainability. Since knowledge is represented in human-readable form, it is easier for users to understand how the AI system arrived at a particular decision. However, Symbolic AI also has limitations, including:

• Difficulty in handling uncertain or incomplete information.
• The need for extensive domain knowledge to be manually encoded into the system.
• Scalability issues, as the complexity of the knowledge base and the rules can grow exponentially with the complexity of the domain.

Recent Developments

With the rise of Machine Learning and Deep Learning, the focus of AI research has shifted towards statistical methods that learn from data. However, there is a growing interest in combining symbolic AI with these approaches to leverage the strengths of both. This hybrid approach, sometimes referred to as Neuro-Symbolic AI or Explainable AI (XAI), aims to create AI systems that are both powerful and interpretable.

See Also

• Artificial Intelligence
• Expert Systems
• Natural Language Processing
• Computer Vision
• Machine Learning
• Deep Learning
• Neuro-Symbolic AI
• Explainable AI

‎

---

Translate: - East Asian 中文, 日本, 한국어, South Asian हिन्दी, தமிழ், తెలుగు, Urdu, ಕನ್ನಡ, Southeast Asian Indonesian, Vietnamese, Thai, မြန်မာဘာသာ, বাংলা
European español, Deutsch, français, Greek, português do Brasil, polski, română, русский, Nederlands, norsk, svenska, suomi, Italian
Middle Eastern & African عربى, Turkish, Persian, Hebrew, Afrikaans, isiZulu, Kiswahili,
Other Bulgarian, Hungarian, Czech, Swedish, മലയാളം, मराठी, ਪੰਜਾਬੀ, ગુજરાતી, Portuguese, Ukrainian