Intelligent advisory system

As explained in Knowledge Graphs and LLMs in Action

An advisory system is an intelligent system that supports human decision-making rather than acting independently.

An intelligent advisory system (IAS) supports human decision-making by providing information, insights, suggestions, and recommendations without executing actions independently. Unlike an autonomous system, it keeps the user responsible for decisions and benefits from interaction, contextual information, and explanatory feedback. section 2.2.2

An IAS typically provides:

• Decision support that helps users compare options and make informed choices. section 2.2.2
• Context awareness that adapts advice to user preferences and specific scenarios. section 2.2.2
• Interactive explanations that let users explore options, ask questions, and inspect recommendations. section 2.2.2

A hybrid IAS combines knowledge graphs (KGs) and large language models (LLMs). KGs contribute explicit, structured knowledge and interpretable reasoning; LLMs contribute natural-language understanding, contextual interpretation, and handling of ambiguity. Graph neural networks can add structural pattern detection, while LLMs explain resulting insights in accessible language. section 2.7 section 9.1

Healthcare advisory workflow

A healthcare IAS can detect entities in clinical language, retrieve information from knowledge sources, and disambiguate context-dependent mentions. For example, named entity disambiguation can distinguish ordinary Zika infection from congenital Zika infection before connecting each mention to a knowledge-base entity. In personalized medicine, graph models can process patient-similarity and biological-pathway graphs while an LLM synthesizes medical literature and clinical guidelines into treatment rationales. section 7.1 section 9.1

Figure 2.3 Differences between (a) an intelligent autonomous system and (b) an intelligent advisor system. The first acts on behalf of users, whereas the second only makes suggestions. In both cases, the intelligent system supports or helps the end user; it is not meant to replace the user.

Compare an IAS with an autonomous system: the autonomous system acts on behalf of the user, whereas the IAS only makes suggestions. section 2.2.2

How Knowledge Graphs and LLMs in Action covers it

How deep it goes

Defines advisory behavior and hybrid architecture but stops short of specifying a complete deployment or execution procedure.

Examples

Uses fraud detection, predictive policing, biomedical diagnosis, personalized medicine, and Zika entity disambiguation as application examples.

What you can do after

Evaluate/understand how KGs, LLMs, and GNNs can support explainable recommendations without replacing human decision-makers.

How it connects

Connects decision support with autonomous systems, knowledge graphs, large language models, graph neural networks, and named entity disambiguation.

liveBookmanning.com

Read Knowledge Graphs and LLMs in Action on liveBook Get the book at manning.com

Related concepts

• human-in-the-loopuses · Human-in-the-loop is a system pattern in which people provide input, judgment, review, or intervention within an automated process.
• intelligent systemkind of · An intelligent system is a system that uses information to reason, adapt its actions to context, and pursue meaningful objectives.
• data integrationuses · Data integration is the process of combining data from multiple sources into a coherent form for consistent access, processing, or analysis.
• knowledge graphused by · A knowledge graph is a graph-structured representation of entities and their semantically meaningful relationships, often including associated attributes or facts.
• graph neural networkproduced by · A graph neural network is a neural-network model that learns representations or predictions from graph-structured data by combining information from connected entities.
• language modelproduced by · A language model is a computational model that represents patterns in linguistic tokens to estimate sequence probabilities and predict or generate language from context.

This explainer was generated by Manning's knowledge graph from the text of Knowledge Graphs and LLMs in Action. The section links point to the parts of the book it draws on.