Probabilistic Time-Series Classifier Using Fuzzy Logic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data mining techniques, particularly genetic algorithms, face challenges in extracting useful knowledge from large datasets as they rely on binary logic, which does not account for partial satisfaction of conditions, leading to incomplete decision-making in complex environments.
Innovation Solution
The implementation of probabilistic extensions to rule-based logic using fuzzy logic, allowing for the determination of how close input data is to satisfying conditions, enabling the output of probability values for decision-making rather than binary actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If binary logic is used in rule-based classification, then the system is simple and fast to compute, but the measurement precision and decision accuracy deteriorate because partial satisfaction of conditions cannot be accounted for
Solution Approach 1:
The patent transforms the binary logic parameter (true/false, 0/1) into a continuous probability parameter (0 to 1) that represents the degree of condition satisfaction. This allows the system to capture partial satisfaction of rules and provides more nuanced classification decisions while maintaining computational efficiency through probabilistic reasoning.
Solution Approach 2:
The system dynamically adjusts the classification outcome based on the aggregated probability values from multiple rules. Instead of static binary decisions, the system can adapt its decision threshold and interpret probability levels contextually, allowing flexible and accurate classification even when conditions are partially met.
2Measurement precision
If probabilistic extensions with fuzzy logic are implemented, then the measurement precision and decision accuracy improve by accounting for partial condition satisfaction, but the device complexity increases
Solution Approach 1:
The patent segments the classification process into distinct probabilistic rule evaluations, where each rule independently calculates a probability of condition satisfaction. These segmented probability assessments are then aggregated to form the final classification decision, allowing complex fuzzy logic operations to be broken down into manageable, computationally efficient steps.
3Loss of information
If binary decision-making is used, then the ease of operation is maintained, but the loss of information occurs because nuances in condition satisfaction are discarded
Solution Approach 1:
The patent introduces probability values as an intermediary between binary condition evaluation and final classification decisions. This intermediary layer preserves the nuanced information from partial condition satisfaction while still enabling straightforward operational interpretation through probability thresholds and aggregation, bridging the gap between information completeness and ease of operation.
Data Source
AI summary
In many environments, rules are trained on historical data to predict an outcome likely to be associated with new data. Described is a ruleset which predicts the probability of a particular outcome. Roughly described, an individual identifies a ruleset, where each of the rules has a plurality of conditions and also indicates a rule-level probability of a predetermined classification. The conditions indicate a relationship (e.g., ‘<’ or ‘!<’) between an input feature and a corresponding value. The rules are evaluated against input data to derive a certainty for each condition, and aggregated to a rule-level certainty. The rule probabilities are combined using the rule-level certainty values to derive a probability output for the ruleset, which can be used to provide a basis for decisions. In an embodiment, the per-condition certainty values are fuzzy values aggregated by fuzzy logic. A novel genetic algorithm can be used to derive the ruleset.