Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.
Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
A fuzzy logic framework for integrating multiple learned models
Abstract
The Artificial Intelligence field of Integrating Multiple Learned Models (IMLM) explores ways to combine results from sets of trained programs. Aroclor Interpretation is an ill-conditioned problem in which trained programs must operate in scenarios outside their training ranges because it is intractable to train them completely. Consequently, they fail in ways related to the scenarios. We developed a general-purpose IMLM solution, the Combiner, and applied it to Aroclor Interpretation. The Combiner's first step, Scenario Identification (SI), learns rules from very sparse, synthetic training data consisting of results from a suite of trained programs called Methods. SI produces fuzzy belief weights for each scenario by approximately matching the rules. The Combiner's second step, Aroclor Presence Detection (AP), classifies each of three Aroclors as present or absent in a sample. The third step, Aroclor Quantification (AQ), produces quantitative values for the concentration of each Aroclor in a sample. AP and AQ use automatically learned empirical biases for each of the Methods in each scenario. Through fuzzy logic, AP and AQ combine scenario weights, automatically learned biases for each of the Methods in each scenario, and Methods' results to determine results for a sample. The Combiner is a general purpose IMLM solution. Its unique scenario-based architecture provides a clean method of including application-specific knowledge from human experts. Once scenario elements are defined, all rule sets are automatically learned. The Combiner is a powerful logic tool for evidence accrual. It solves the "chicken-and-egg" dilemma caused when relying on scenario-dependent programs to determine the scenario for a sample. It handles heterogeneous inputs and delivers heterogeneous outputs. Because of its straightforward approach, the rules and reasons the Combiner reaches its conclusions are easily available for accountability and tracability. Scenario weights are never explicitly needed by an operator, but are available and deliver a wealth of additional information about the real-world conditions affecting a sample. This dissertation documents the Combiner and its superior performance against Multicategory Classification, Dempster-Shafer, and the best Method in the Aroclor Interpretation suite.
Subject Area
Computer science
Recommended Citation
Den Hartog, Bobi Kai, "A fuzzy logic framework for integrating multiple learned models" (1998). ETD collection for University of Nebraska-Lincoln. AAI9902952.
https://digitalcommons.unl.edu/dissertations/AAI9902952