Open Access. Powered by Scholars. Published by Universities.®
Physical Sciences and Mathematics Commons™
Open Access. Powered by Scholars. Published by Universities.®
- Discipline
-
- Computer Sciences (2498)
- Chemistry (1318)
- Physics (1078)
- Engineering (1076)
- Earth Sciences (875)
-
- Geology (669)
- Mathematics (649)
- Environmental Sciences (545)
- Statistics and Probability (545)
- Life Sciences (496)
- Oceanography and Atmospheric Sciences and Meteorology (410)
- Electrical and Computer Engineering (308)
- Applied Mathematics (276)
- Biochemistry, Biophysics, and Structural Biology (273)
- Biochemistry (245)
- Optics (212)
- Astrophysics and Astronomy (210)
- Information Security (200)
- Atmospheric Sciences (173)
- Operations Research, Systems Engineering and Industrial Engineering (153)
- Artificial Intelligence and Robotics (145)
- Social and Behavioral Sciences (143)
- Biostatistics (141)
- Computer Engineering (133)
- Software Engineering (129)
- Meteorology (126)
- Atomic, Molecular and Optical Physics (122)
- Medicine and Health Sciences (122)
- Civil and Environmental Engineering (116)
- Institution
-
- Air Force Institute of Technology (2216)
- Brigham Young University (1837)
- University of Wisconsin Milwaukee (872)
- University of South Carolina (805)
- Virginia Commonwealth University (691)
-
- University for Business and Technology in Kosovo (399)
- University of Texas Rio Grande Valley (364)
- University of North Dakota (347)
- Florida Institute of Technology (340)
- Illinois State University (109)
- Mississippi State University (103)
- Columbus State University (99)
- Rowan University (72)
- American University in Cairo (38)
- City University of New York (CUNY) (31)
- St. John's University (6)
- Nova Southeastern University (5)
- Pepperdine University (3)
- Utah State University (3)
- Texas A&M International University (1)
- Keyword
-
- Machine learning (152)
- Physical Sciences and Mathematics, Chemistry (84)
- #antcenter (80)
- Geology (70)
- Deep learning (65)
-
- Pure sciences (60)
- Machine Learning (58)
- Software engineering (49)
- Computer networks--Security measures (47)
- Synthesis (46)
- Chemistry (45)
- Computer security (44)
- Algorithms (43)
- Nanoparticles (41)
- Artificial intelligence (40)
- Simulation (38)
- Optimization (37)
- Mass spectrometry (36)
- Computer vision (35)
- Adaptive optics (34)
- Applied sciences (32)
- Image processing (32)
- Catalysis (31)
- Remote sensing (31)
- Neural networks (Computer science) (30)
- Bayesian (29)
- Physical Sciences and Mathematics, Mathematics (29)
- Classification (28)
- Security (27)
- Object-oriented programming (Computer science) (26)
- Publication Year
- File Type
Articles 7441 - 7470 of 8341
Full-Text Articles in Physical Sciences and Mathematics
An Object-Oriented, Formal Methods Approach To Organizational Process Modeling, Vincent S. Hibdon
An Object-Oriented, Formal Methods Approach To Organizational Process Modeling, Vincent S. Hibdon
Theses and Dissertations
This document presents a methodology for developing an organizational process model which is based on the principles of object oriented design and formal software engineering methods. The methodology begins with the development of an object oriented Rumbaugh model (27). The Rumbaugh model is then formally specified in Z (Zed) schemas. Finally, the Z specifications are translated into an executable model in the Software Refinery Environment(TM). This model is described based on the AF wing domain and developed in this domain. The proposed methodology is then shown to produce a very general model which is extendable across almost any domain. The …
Temporal Influence On Awareness, Don E. Hill
Temporal Influence On Awareness, Don E. Hill
Theses and Dissertations
Grossberg's Motion Oriented Contrast Filter (MOC) was extensively analyzed (7). The output from the filter's "global motion" neuronal layer was compared to a noncausal post-processing filter developed by AFIT. Both filters were shown to incorporate a weighted, noncausal temporal range of input data in processed output. The global motion framework was then implemented using a physiologically motivated pulsed neural model - the Pulse Coupled Neural Network (PCNN). By incorporating both spatial and temporal data, the PCNN was shown to exhibit a common visual illusion, apparent motion. The existence of a physiological temporal processing range was further investigated through implementation of …
Effects Of Ammonium Perchlorate On The Thyroid Hormone Levels Of The Sprague-Dawley Rat, James H. King Jr.
Effects Of Ammonium Perchlorate On The Thyroid Hormone Levels Of The Sprague-Dawley Rat, James H. King Jr.
Theses and Dissertations
The purpose of this research was to determine the threshold dose for ammonium perchlorate (AP) in the Sprague-Dawley rat. No dose response data exist for AP and the EPA has studied literature on the subject of perchlorates to determine a provisional reference dose. The Perchiorate Group, a consortium of DoD and industry representatives, believes this provisional reference dose is too conservative. This experiment was executed to provide dose response data on which to base a more accurate reference dose. The study consisted of eight groups of 12 Sprague-Dawley rats, six male and six female, which were exposed to incremental doses …
A Comparison Of Remediation Priorities Developed By The Defense Priority Model, The Relative Risk Evaluation Method, And A Quantitative Risk Assessment Approach, David M. Hunter
Theses and Dissertations
The Superfund, established by the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) of 1980, seriously underestimated both the number of severely contaminated sites and the associated cleanup cost. The magnitude of projected cleanup costs, coupled with shrinking federal budgets, necessitated the development and use of risk-based prioritization models among some federal agencies. Among these tools, the DOD prioritization models are meant to give priority to sites posing the greatest threat to human health. Their failure to properly rank sites could incorrectly shift the focus from those that pose substantial risk to sites of lesser risk. The DOD site ranking …
Dynamic Transfer Of Control Between Manned And Unmanned Simulation Actors, Neal W. Schneider
Dynamic Transfer Of Control Between Manned And Unmanned Simulation Actors, Neal W. Schneider
Theses and Dissertations
This thesis continues the ongoing research at the Air Force Institute of Technology's Virtual Environments Laboratory in the area of distributed simulation. As the relevance and interest of interactive simulation as a training medium continues to grow, there is a pressing need to provide more realistic and numerous intelligent autonomous agents for simulations. As those autonomous agents mature and become more realistic, the need exists to be able to handle individual agents by taking control of them and operating them as manned agents at certain points within the simulation. The author started with a protocol proposed in a working draft …
An Assessment Of The Impact Of Fuel Jettisoning Events Using Simulation And Impact Models, Jeffrey M. Todd
An Assessment Of The Impact Of Fuel Jettisoning Events Using Simulation And Impact Models, Jeffrey M. Todd
Theses and Dissertations
Work has been accomplished to determine the impact of jettisoned fuel when it reaches the surface. While previous work indicates that jettisoning JP-4 jet fuel results in a negligible ground fall impact, the impact of jettisoning lower volatile JP-8 jet fuel has not been thoroughly characterized. Several efforts have been made to mathematically model the evaporation, advection, and dispersion of the plume of fuel as it travels to the surface. The AFIT Fuel Jettisoning Model, the Fuel Jettisoning Simulation Model, and Fuel-Dumping Impact Assessment Model were evaluated and compared to assess the impact of jettisoned JP-8 jet fuel. Additionally, the …
Experimental Investigation And Computer Modeling Of Optical Switching In Distributed Bragg Reflector And Vertical Cavity Surface Emitting Laser Structures, Richard J. Bagnell
Experimental Investigation And Computer Modeling Of Optical Switching In Distributed Bragg Reflector And Vertical Cavity Surface Emitting Laser Structures, Richard J. Bagnell
Theses and Dissertations
The optical switching capabilities of Distributed Bragg Reflector (DBR) structures, including Vertical Cavity Surface Emitting Lasers (VCSELs) are examined. Reflectivity switching is demonstrated using both thermal and carrier generated effects to alter the DBR/VCSEL layers' refractive indices. Optical bistability is demonstrated at room temperature, under CW photopumped excitation. The optical bistability hysteresis is controllable by spectral location of the pump on the stop band edge. In the VCSEL, reflective bistability is also evidenced; additionally, this bistability is accompanied by a bistability in the VCSEL lasing output intensity, spot size, and wavelength. Modeling of the DBR/VCSEL thermally induced bistability was accomplished …
Analysis And Interpretation Of Ion Data Associated With Neutral Gas Releases In The Earth's Ionosphere, Timothy M. Shadid
Analysis And Interpretation Of Ion Data Associated With Neutral Gas Releases In The Earth's Ionosphere, Timothy M. Shadid
Theses and Dissertations
Barium and strontium release experiments were conducted throughout 1991 from the Combined Release and Radiation Effects Satellite (CRRES) to study both natural and man-made disturbances in the earth's ionosphere. A mass spectrometer on the spacecraft counted the Ba and Sr ions as the cloud expanded. In this study, data from the G-1 (in sunlight) and G-11b (in darkness) releases were modeled to understand the source of the ion signals. The model reproduced the Ba+ sun data well assuming photoionization (τ = 28 s) was the primary ionization mechanism. However, it was not able to account for the remaining ion …
Micromirror Array Control Of A Phase-Locked Laser Diode Array, Carl J. Christensen
Micromirror Array Control Of A Phase-Locked Laser Diode Array, Carl J. Christensen
Theses and Dissertations
A ten element micromirror array has been designed, fabricated, and employed to control the far field irradiance pattern of a phase locked laser diode array. The laser array used in this experiment was a ten element, gain guided array lasing at a nominal wavelength of 828 nm and operating in the it out of phase supermode. The laser's near field irradiance was imaged onto a micromirror array, where the it phase differences between adjacent laser elements were corrected. This was accomplished by moving the micromirrors with individually applied voltages. The result was the desirable single lobed far field pattern, placing …
Objectism 3.0: A Software Architecture For The Development Of Portable Visual Simulation Applications, Shawn M. Hannan
Objectism 3.0: A Software Architecture For The Development Of Portable Visual Simulation Applications, Shawn M. Hannan
Theses and Dissertations
A visual simulation software architecture is a reusable design for visual simulation applications. This thesis effort was the third stage in an ongoing refinement of such an architecture, named ObjectSim. The primary goals of this stage were to improve the architecture by eliminating its dependence on two platform-specific graphics libraries (named GL and Performer, from Silicon Graphics, Inc.), and to examine the potential for expanding the architecture to accommodate distributed simulations. The effort resulted in a new version of the architecture which allows the development of visual simulation applications which take full advantage of the aforementioned libraries without calling those …
Transforming Algebraically-Based Object Models Into A Canonical Form For Design Refinement, Charles G. Beem
Transforming Algebraically-Based Object Models Into A Canonical Form For Design Refinement, Charles G. Beem
Theses and Dissertations
The understandability of object-oriented design techniques and the rigor of formal methods have improved the state of software development; however, both ideas have limitations. Object-oriented techniques, which are semi-formal, can still result in incorrect designs, while formal methods are complex and require an extensive mathematical background. The two approaches can be coupled, however, to produce designs that are both understandable and verifiable, and to produce executable code. This research proposes an approach where object-oriented models are first represented algebraically in a formal specification language such as LARCH and then transformed into a canonical form suitable for design refinement. In the …
The Application Of Hybridized Genetic Algorithms To The Protein Folding Problem, Robert L. Gaulke
The Application Of Hybridized Genetic Algorithms To The Protein Folding Problem, Robert L. Gaulke
Theses and Dissertations
The protein folding problem consists of attempting to determine the native conformation of a protein given its primary structure. This study examines various methods of hybridizing a genetic algorithm implementation in order to minimize an energy function and predict the conformation (structure) of Met-enkephalin. Genetic Algorithms are semi-optimal algorithms designed to explore and exploit a search space. The genetic algorithm uses selection, recombination, and mutation operators on populations of strings which represent possible solutions to the given problem. One step in solving the protein folding problem is the design of efficient energy minimization techniques. A conjugate gradient minimization technique is …
Neural Network Analysis Of Chemical Compounds In Nonrebreathing Fisher-344 Rat Breath, Robert E. Sackett Jr.
Neural Network Analysis Of Chemical Compounds In Nonrebreathing Fisher-344 Rat Breath, Robert E. Sackett Jr.
Theses and Dissertations
This research applies statistical and artificial neural network analysis to data obtained from measurement of organic compounds in the breath of a Fisher-344 rat. The Research Triangle Institute (RTI) developed a breath collection system for use with rats in order to collect and determine volatile organic compounds (VOCs) exhaled. The RTI study tested the hypothesis that VOCs, including endogenous compounds, in breath can serve as markers to exposure to various chemical compounds such as drugs, pesticides, or carcinogens normally foreign to living organisms. From a comparative analysis of chromatograms, it was concluded that the administration of carbon tetrachloride dramatically altered …
Determining The Effects Of Waste Coal Ash On Landfill Radon Levels, Richard S. Krysiak Jr.
Determining The Effects Of Waste Coal Ash On Landfill Radon Levels, Richard S. Krysiak Jr.
Theses and Dissertations
Coal contains trace amounts of the primary radionuclides 40K, and elements of the 4n (232Th), 4n+2 (238U), and 4n+3 (235U) series including 220Rn and 222Rn. Combustion of coal by electric power and heat plants result in concentration of noncombustible mineral matter, including most of the radionuclides, in the coal ash. The increased radiation due to the concentration of radionuclides is known as technologically enhanced natural radiation. The purpose of this research was to determine the effects of landfilled coal ash on one specific aspect of technologically enhanced natural radiation, radon levels. …
The Effect Of Meteorological Data Averaging Times On Plume Concentrations From Explosive Ordnance Disposal Open Burning Operations, Ida L. Widmann
The Effect Of Meteorological Data Averaging Times On Plume Concentrations From Explosive Ordnance Disposal Open Burning Operations, Ida L. Widmann
Theses and Dissertations
Explosive Ordnance Disposal (EOD) Open Burning (OB) operations are performed to treat and dispose of unserviceable munitions in the Department of Defense (DOD) inventory. Air pollution modeling of OB operations is an Environmental Protection Agency requirement for permit issuance at OB sites. Specific OB regulation is still in its infancy; therefore, establishment of OB modeling techniques is still in the early stages. This thesis effort sought to develop a computer model, based upon the Gaussian Puff Equation. The model varies from standard plume modeling practices by not making the assumption that the wind direction, wind speed and turbulence are uniform …
Model Of Chlorocarbon (Cfc-12) Chemisorption On Solid Rocket Motor Alumina Exhaust Particles, Gary E. Lund
Model Of Chlorocarbon (Cfc-12) Chemisorption On Solid Rocket Motor Alumina Exhaust Particles, Gary E. Lund
Theses and Dissertations
Solid Rocket (SRMs) that power Titan IV rockets and Space Shuttles, exhaust large quantities of potentially ozone damaging pollutants directly into the stratosphere, while in powered flight. In the past, studies on potential stratospheric impact of the exhaust products from aluminum/ammonium perchlorate based SRMs have focused on the effect of gaseous HCl from SRMs on the stratosphere. Until recently, the impact of heterogeneous chemistry on stratospheric ozone was believed to be relatively insignificant. This research investigates the potential heterogeneous process of CFC-12 dissociative chemisorption on alumina surfaces and the release of reactive halogen species known to destroy ozone. Through a …
Comparison Of Gravel Substrate Vs Soil Substrate For The Construction Of An Experimental Fen, Carolyn S. Langley
Comparison Of Gravel Substrate Vs Soil Substrate For The Construction Of An Experimental Fen, Carolyn S. Langley
Theses and Dissertations
Under the Clean Water Act Section 404 of 1972 and 33 CFR 320-330 and 40 CFR 230 moderate the destruction of wetlands by the Air Force to make way for other uses. To obtain a permit for a design or construction project which affects a wetland, the Air Force must agree to create new wetlands, or replace lost wetland acreage through wetland creation or restoration. The Air Force is interested in building 'successful' wetlands as inexpensively as possible. It has been common practice to use hydric soil, which often had to be hauled in, as the substrate at the restored …
Design And Analysis Of Parallel Hierarchical Battlefield Simulation, Conrad P. Masshardt
Design And Analysis Of Parallel Hierarchical Battlefield Simulation, Conrad P. Masshardt
Theses and Dissertations
The purpose of this research is to determine if hierarchically partitioning a discrete event battlefield simulation reduces runtime and, if reduction exists, to characterize the run time reduction given any particular partition configuration. A hierarchical discrete event simulation of a main battle tank was constructed. Implementations were built for both a single processor and a multiprocessing machine. The implementations used the Message Passing Interface to increase portability to other parallel and distributed configurations. Three test cases were generated and run on three parallel and distributed environments, a network of Sun SparcStation 20's, a Silicon Graphics Power Challenge, and a Paragon …
A Comparison Of Loose And Tight Gps/Ins Integration Using Real Ins And Gps Data, Warren H. Nuibe
A Comparison Of Loose And Tight Gps/Ins Integration Using Real Ins And Gps Data, Warren H. Nuibe
Theses and Dissertations
An extended Kalman filter (EKE) is used to combine the information obtained from a Global Positioning System (GPS) receiver and an Inertial Navigation System (INS) to provide a navigation solution. This research compares the results of a tightly-coupled GPS/INS integrated system with a loosely-coupled integrated system, using real world data. A fair comparison is accomplished by using the same sets of data, and keeping the integration structures as close as possible. Both integrations are feedforward and have the same error states in the navigation Kalman filters. Differences between the two, such as navigation solutions and tuning values, are shown in …
Maximum Likelihood Estimation Of Wave Front Slopes Using A Hartmann-Type Sensor, Scott A. Sallberg
Maximum Likelihood Estimation Of Wave Front Slopes Using A Hartmann-Type Sensor, Scott A. Sallberg
Theses and Dissertations
Current methods for estimating the wave front slope at the pupil of a telescope equipped with a Hartmann-type wave front sensor (H-WFS) are based on a simple centroid calculation of the intensity distributions (spots) recorded in each subaperture of the H-WFS. The centroid method does not include any knowledge concerning correlation properties of the slopes over the subapertures or the amount of light collected by the telescope and diverted to the H-WFS for wave front reconstruction purposes. This thesis devises a maximum likelihood (ML) estimation of the spot centroids by incorporating statistical knowledge of the spot shifts. The light level …
A Monte Carlo Analysis Of Computerized Tomography, Karyl J. Davis
A Monte Carlo Analysis Of Computerized Tomography, Karyl J. Davis
Theses and Dissertations
The purpose of this thesis was to model a computerized tomography device (CT) using Monte Carlo methods to determine the scattered radiation spectrum inside and the dose outside the suite at Wright-Patterson Medical Center and at a generic suite to non-occupational personnel. This was driven by the recent inclusion of the most recent NCRP recommended dose limits into 10CFR20 of 50 mSv, occupational; 1 mSv; non-occupational continuous exposure; and 5 mSv, infrequent exposure. The rotating fan beam of the CT was modeled for MCNP, a general-purpose Monte Carlo n-particle transport modeL The CT target was a standard human phantom defined …
Macroscale Diffusion-Limited Sorption Modeling--A Preliminary Modeling Exercise For A Dover Afb Site, Jason T. Herman
Macroscale Diffusion-Limited Sorption Modeling--A Preliminary Modeling Exercise For A Dover Afb Site, Jason T. Herman
Theses and Dissertations
A modification was made to the USGS SUTRA code which allowed the simulation of macro scale diffusion effects from specific layers. This modification utilized a split-operator finite element numerical technique to incorporate the macroscale diffusion. The code was applied to a conceptual site developed from a field site at Dover AFB, DL Simulations were done to compare the modified code to the unmodified code which clearly showed the modified code as a closer representation of reality. Simulations were also done to study the effects of pulsed and continuous pumping within the time frame of a field experiment at Dover. These …
Reducing Lag In Virtual Displays Using Multiple Model Adaptive Estimation, David W. Kyger
Reducing Lag In Virtual Displays Using Multiple Model Adaptive Estimation, David W. Kyger
Theses and Dissertations
Multiple Model Adaptive Estimation is an effective method for reducing lag in virtual environment displays. Lag in displays (the time from head motion to the appearance of the proper image on the display) is a significant detriment to realism in virtual environments. Increasing the speed of the computers which control the virtual display is not a final answer. No matter how fast the processors work, there will always be demands to do more. Predicting angular head positions (look-angles) can reduce the lag by allowing the computer to calculate the appropriate scene before it is needed on the display. Single predictors …
Computer-Aided Diagnosis Of Mammographic Masses, William E. Polakowski
Computer-Aided Diagnosis Of Mammographic Masses, William E. Polakowski
Theses and Dissertations
A new Model-Based Vision algorithm was developed to find possibly cancerous regions of interest (ROIs) in digitized mammograms and to correctly identify the malignant masses. This work has shown a sensitivity of 92 percent for locating malignant ROIs. The database contained 272 images (12 bit, 1OO microns) with 36 malignant and 53 benign mass images. Of the 53 biopsied benign cases, 74 percent were correctly classified. The Focus of Attention (segmentation) Module algorithm used a physiologically motivated Difference of Gaussians (DoG) filter to highlight mass-like regions in the mammogram. The Index Module labeled the regions by their hypothesized class: large …
The Role Of Frame Selection And Bispectrum Phase Reconstruction For Speckle Imaging Through Atmospheric Turbulence, Elizabeth A. Harpold
The Role Of Frame Selection And Bispectrum Phase Reconstruction For Speckle Imaging Through Atmospheric Turbulence, Elizabeth A. Harpold
Theses and Dissertations
Frame selection using quality sharpness metrics have been shown in previous AFIT theses, to be effective in improving the final product of images obtained using adaptive optics. This thesis extends this idea to noncompensated speckle image data. Speckle image reconstruction is simulated with and without frame selection. Speckle images require the processing of hundreds of data frames. Frame selection is a method of reducing the amount of data required to reconstruct the image. A collection of short exposure image data frames of a single object are sorted based on sharpness metrics. Only the highest quality frames are retained and processed …
Photoluminescence And Electroluminescence Of Erbium And Neodymium Implanted Semiconductors, James R. Hunter
Photoluminescence And Electroluminescence Of Erbium And Neodymium Implanted Semiconductors, James R. Hunter
Theses and Dissertations
Low temperature photoluminescence (PL) and electroluminescence (EL) measurements were used to study the excitation of erbium- and neodymium-implanted GaAs and AlxGa1-xAs (x=0. 1, 0.3) pn-junctions. The rare-earth (RE) emissions were investigated as a function of ion dose, aluminum mole fraction, laser excitation power, and applied forward bias voltage for the implanted samples. Low temperature PL was also measured from Er doped silicon grown by the metalorganic chemical vapor-phase deposition (MOCVD) method using various growth parameters.. The MOCVD-grown Si samples were studied as a function of metalorganic source temperature, silane (SiH4) flow, growth time, and …
A Comparison Of Methods For Estimating Rms Error: A `Brute Force" Approach Versus A Mathematically-Elegant Approach, As Applied To The Calculation Of A Specific Retrieval Error For A Limb-Scanning Microwave Radiometer-Spectrometer, Larry L. Johnson
Theses and Dissertations
The Millimeter-wave Atmospheric Sounder (MAS) is a high resolution limb- scanning microwave radiometer-spectrometer which has been flown aboard the Space Shuttle. The instrument was designed to sense the microwave emission from several upper atmospheric constituents, including water vapor and ozone. The resonant frequencies of water vapor and ozone are extremely close (183 and 184 GHz, respectively), so that the high-frequency wing of the water vapor spectrum overlaps the low frequency wing of the ozone spectrum. Consequently, the measured ozone spectrum incorporates a slight water vapor contribution; therefore, the retrieved ozone profile is dependent on the true water vapor profile. The …
A System Dynamics Approach To Modelling The Degradation Of Biochemical Oxygen Demand In A Constructed Wetland Receiving Stormwater Runoff, Leslie A. Mudgett
A System Dynamics Approach To Modelling The Degradation Of Biochemical Oxygen Demand In A Constructed Wetland Receiving Stormwater Runoff, Leslie A. Mudgett
Theses and Dissertations
The objective of this research was to develop a tool to aid the Air Force Environmental Manager in the identification of the design parameters of a constructed wetland system that may be optimized to provide a desired biochemical oxygen demand (BOD) removal efficiency during the treatment of Air Force stormwater runoff. The objective is achieved through the development and use of a system dynamics model which simulates the hydrological functions of a constructed wetland as well as the processes within the wetland responsible for degradation of BOD. Based on literature review, the primary mechanism responsible for the degradation of BOD …
Semantic Interpretation Of An Artificial Neural Network, Stanley D. Kinderknecht
Semantic Interpretation Of An Artificial Neural Network, Stanley D. Kinderknecht
Theses and Dissertations
Recent advances in machine learning theory have opened the door for applications to many difficult problem domains. One area that has achieved great success for stock market analysis/prediction is artificial neural networks. However, knowledge embedded in the neural network is not easily translated into symbolic form. Recent research, exploring the viability of merging artificial neural networks with traditional rule-based expert systems, has achieved limited success. In particular, extracting production (IF.. THEN) rules from a trained neural net based on connection weights provides a valid set of rules only when neuron outputs are close to 0 or 1 (e.g. the output …
Deriving Optimal Solutions From Incomplete Knowledge Bases, Shawn A. Northrop
Deriving Optimal Solutions From Incomplete Knowledge Bases, Shawn A. Northrop
Theses and Dissertations
Many real world domains cannot be represented using Bayesian Networks due to the need for complete probability tables and acyclic knowledge. However, Bayesian Knowledge Bases (BKBs) are a viable method for representing these incomplete domains, but very little research has been performed on inferencing with them. This paper presents three inference engines for extracting optimal solutions from three distinct BKB subclasses: singly- connected, multiply-connected with mutually exclusive cycles, and cyclic. The singly-connected inference engine has a worst case polynomial run time. Performance improvement techniques for increasing inference engine speed are discussed, in addition to a new tool for measuring incompleteness …