Running Wisdom Tests for DIKWP Artificial Consciousness System

Yucong Duan

International Standardization Committee of Networked DIKWP for Artificial Intelligence Evaluation(DIKWP-SC)

World Artificial Consciousness CIC(WAC)

World Conference on Artificial Consciousness(WCAC)

(Email: duanyucong@hotmail.com)

Introduction

Building upon the Knowledge Network (K_subj) constructed in the Knowledge Module, we'll now proceed to the Wisdom Module (W). In this module, we'll apply reasoning and decision-making processes using the knowledge network, taking into account the subjective differences and confidence levels. The goal is to make informed decisions that handle uncertainty effectively and align with the system's overarching purpose.

Wisdom Module (W)Goals of the Wisdom Module

1. Apply Reasoning: Use the knowledge network to derive insights and make decisions.

2. Handle Uncertainty: Factor in confidence levels to weigh decisions appropriately.

3. Align with Objectives: Ensure decisions are in line with the system's goals and purposes.

Step-by-Step Implementation of the Wisdom Module1. Define Decision-Making Criteria

We need to establish criteria for making decisions based on the knowledge network. These criteria should consider:

• Differences between data points.

• Confidence levels of the relationships.

• The system's overall goals.

Example Decision Criteria:

• Grouping Similar Items: Group data points that are similar in certain attributes with high confidence.

• Flagging Uncertain Relationships: Identify relationships with low confidence for further analysis.

2. Implement the Decision-Making FunctionpythonCopy codedef wisdom_decision_making(K_subj, similarity_threshold, confidence_threshold):     network = K_subj.get_network()     nodes = network['nodes']     edges = network['edges']     decisions = []    # Iterate over the edges to make decisions     for edge in edges:         node_i = nodes[edge['node_i']]         node_j = nodes[edge['node_j']]         attribute = edge['attribute']         difference = edge['difference']         confidence = edge['confidence']        # Determine if nodes are similar based on the attribute         if attribute == 'color_category':            # For categorical attributes, difference is 0 (same) or 1 (different)             similar = difference == 0         else:            # For numerical attributes, use the similarity threshold             similar = difference <= similarity_threshold        # Make decisions considering confidence levels         if confidence >= confidence_threshold:            if similar:                 decision = {                    'action': 'Group',                    'nodes': (node_i['id'], node_j['id']),                    'attribute': attribute,                    'reason': 'High confidence similarity',                    'confidence': confidence                 }            else:                 decision = {                    'action': 'Do not group',                    'nodes': (node_i['id'], node_j['id']),                    'attribute': attribute,                    'reason': 'High confidence difference',                    'confidence': confidence                 }        else:            # Low confidence decisions             decision = {                'action': 'Review',                'nodes': (node_i['id'], node_j['id']),                'attribute': attribute,                'reason': 'Low confidence relationship',                'confidence': confidence             }         decisions.append(decision)    return decisions3. Apply the Decision-Making Function

Set thresholds for similarity and confidence based on the system's requirements.

pythonCopy codesimilarity_threshold = 0.1  # For numerical attributes like 'size'confidence_threshold = 0.9  # Minimum acceptable confidence level# Execute the decision-making processdecisions = wisdom_decision_making(K_subj, similarity_threshold, confidence_threshold)4. Resulting Decisions

Example Output:

pythonCopy code[     {        'action': 'Group',        'nodes': (1, 2),        'attribute': 'color_category',        'reason': 'High confidence similarity',        'confidence': 1.0     },     {        'action': 'Do not group',        'nodes': (1, 3),        'attribute': 'color_category',        'reason': 'High confidence difference',        'confidence': 1.0     },     {        'action': 'Do not group',        'nodes': (2, 3),        'attribute': 'color_category',        'reason': 'High confidence difference',        'confidence': 1.0     },     {        'action': 'Group',        'nodes': (1, 2),        'attribute': 'size',        'reason': 'High confidence similarity',        'confidence': 1.0     },     {        'action': 'Group',        'nodes': (1, 3),        'attribute': 'size',        'reason': 'High confidence similarity',        'confidence': 1.0     },     {        'action': 'Group',        'nodes': (2, 3),        'attribute': 'size',        'reason': 'High confidence similarity',        'confidence': 1.0     }, ]5. Handling Uncertainty in Decisions

If any relationships had confidence levels below the threshold, the action would be 'Review', indicating that these cases need further analysis.

pythonCopy code# Example of a low confidence decision{    'action': 'Review',    'nodes': (1, 2),    'attribute': 'size',    'reason': 'Low confidence relationship',    'confidence': 0.85}6. Prioritizing Decisions

We can prioritize decisions based on confidence and the system's goals.

• High Confidence Decisions: Act upon immediately.

• Low Confidence Decisions: Queue for further review or collect more data.

Integrating with the System's Goals

While the Wisdom Module focuses on making decisions based on the knowledge network, it's crucial to ensure these decisions align with the system's overarching purpose. This integration is handled in the Purpose Module (P), which we'll proceed to next.

Conclusion of the Wisdom Module

We've successfully implemented the Wisdom Module, applying reasoning to make informed decisions based on the knowledge network. The module accounts for uncertainty by factoring in confidence levels and provides a set of actions that are ready to be aligned with the system's goals in the Purpose Module.

References for Further Reading

1. International Standardization Committee of Networked DIKWP for Artificial Intelligence Evaluation (DIKWP-SC),World Association of Artificial Consciousness(WAC),World Conference on Artificial Consciousness(WCAC). Standardization of DIKWP Semantic Mathematics of International Test and Evaluation Standards for Artificial Intelligence based on Networked Data-Information-Knowledge-Wisdom-Purpose (DIKWP ) Model. October 2024 DOI: 10.13140/RG.2.2.26233.89445 .  https://www.researchgate.net/publication/384637381_Standardization_of_DIKWP_Semantic_Mathematics_of_International_Test_and_Evaluation_Standards_for_Artificial_Intelligence_based_on_Networked_Data-Information-Knowledge-Wisdom-Purpose_DIKWP_Model

2. Duan, Y. (2023). The Paradox of Mathematics in AI Semantics. Proposed by Prof. Yucong Duan:" As Prof. Yucong Duan proposed the Paradox of Mathematics as that current mathematics will not reach the goal of supporting real AI development since it goes with the routine of based on abstraction of real semantics but want to reach the reality of semantics. ".

转载本文请联系原作者获取授权，同时请注明本文来自段玉聪科学网博客。
链接地址：https://blog.sciencenet.cn/blog-3429562-1456778.html

上一篇：Running Knowledge Test for DIKWP Artificial Conscious（初学者版）
下一篇：Running Purpose Test forDIKWP Artificial Consciousness（初学者版）