Open Access. Powered by Scholars. Published by Universities.®
Physical Sciences and Mathematics Commons™
Open Access. Powered by Scholars. Published by Universities.®
- Discipline
Articles 1 - 6 of 6
Full-Text Articles in Physical Sciences and Mathematics
Augmenting Latent Dirichlet Allocation And Rank Threshold Detection With Ontologies, Laura A. Isaly
Augmenting Latent Dirichlet Allocation And Rank Threshold Detection With Ontologies, Laura A. Isaly
Theses and Dissertations
In an ever-increasing data rich environment, actionable information must be extracted, filtered, and correlated from massive amounts of disparate often free text sources. The usefulness of the retrieved information depends on how we accomplish these steps and present the most relevant information to the analyst. One method for extracting information from free text is Latent Dirichlet Allocation (LDA), a document categorization technique to classify documents into cohesive topics. Although LDA accounts for some implicit relationships such as synonymy (same meaning) it often ignores other semantic relationships such as polysemy (different meanings), hyponym (subordinate), meronym (part of), and troponomys (manner). To …
Assessing Information Trustability In A Secure Web Services Environment, Charles G. Penner
Assessing Information Trustability In A Secure Web Services Environment, Charles G. Penner
Theses and Dissertations
Decisions are made based on available information. A decision support system endeavors to provide information that is timely, accurate, and trustable. Information gathered from secure web service transactions has attributes that can be used to assess a level of trustability. The trust assessments enable a decision maker to determine a basis for confidence in the information presented from the web service. Existing trust assessment models do not provide a way to determine from a particular trust assessment what information attributes contributed to its computation. The present work creates trust values that retain and denote meaning, allowing a decision maker to …
Modeling Information Quality Expectation In Unmanned Aerial Vehicle Swarm Sensor Databases, Patrick D. Baldwin
Modeling Information Quality Expectation In Unmanned Aerial Vehicle Swarm Sensor Databases, Patrick D. Baldwin
Theses and Dissertations
Swarming Unmanned Aerial Vehicles (UAVs) are the future of Intelligence, Surveillance and Reconnaissance (ISR). Swarms of hundreds of these vehicles, each equipped with multiple sensors, will one day fill the skies over hostile areas. As the sensors collect hundreds of gigabytes of data, telemetry data links will be unable to transmit the complete data picture to the ground in real time. The collected data will be stored on board the UAVs and selectively downloaded through queries issued from analysts on the ground. Analysts expect to find relevant sensor data within the collection of acquired sensor data. This expectation is not …
Bubble World - A Novel Visual Information Retrieval Technique, Christopher L. Van Berendonck
Bubble World - A Novel Visual Information Retrieval Technique, Christopher L. Van Berendonck
Theses and Dissertations
With the tremendous growth of published electronic information sources in the last decade and the unprecedented reliance on this information to succeed in day-to-day operations, comes the expectation of finding the right information at the right time. Sentential interfaces are currently the only viable solution for searching through large infospheres of unstructured information, however, the simplistic nature of their interaction model and lack of cognitive amplification they can provide severely limit the performance of the interface. Visual information retrieval systems are emerging as possible candidate replacements for the more traditional interfaces, but many lack the cognitive framework to support the …
Application Of Information Retrieval Techniques To Heterogeneous Databases In The Virtual Distributed Laboratory, Rodney D. Lykins
Application Of Information Retrieval Techniques To Heterogeneous Databases In The Virtual Distributed Laboratory, Rodney D. Lykins
Theses and Dissertations
The Department of Defense (DoD) maintains thousands of Synthetic Aperture Radar (SAR), Infrared (IR), Hyper-Spectral intelligence imagery and Electro-Optical (EO) target signature data. These images are essential to evaluating and testing individual algorithm methodologies and development techniques within the Automatic Target Recognition (ATR) community. The Air Force Research Laboratory Sensors Directorate (AFRL/SN) has proposed the Virtual Distributed Laboratory (VDL) to maintain a central collection of the associated imagery metadata and a query mechanism to retrieve the desired imagery. All imagery metadata is stored in relational database format for access from agencies throughout the federal government and large civilian universities. Each …
Three-Dimensional Data Visualization Of Electronic Military Intelligence Using The Project Broadsword System, Michael L. Goeringer
Three-Dimensional Data Visualization Of Electronic Military Intelligence Using The Project Broadsword System, Michael L. Goeringer
Theses and Dissertations
Today's military electronic infrastructure solves many problems while creating others. Using computers, battlefield and global awareness is brought to bear through the near real-time linking of sensor platforms from around the globe. These intelligence networks produce vast amounts of data that must be parsed, interpreted, digested and stored by information gathering systems. As the amount of intelligence data continues to increase, these text-based systems become cumbersome and inadequate. To ensure vital information is not overlooked or discovered too late, other forms of intelligence product management and data navigation need to be investigated. This thesis explores procedures for enhancing the capabilities …