Apparently fraud is a growth industry. The monetary losses from Internet fraud have increased every year since first officially reported by the Internet Crime Complaint Center (IC3) in 2000. Prior research studies and third-party reports of fraud show rates substantially higher than eBay’s reported negative feedback rate of less than 1%. The conclusion is most buyers are withholding reports of negative feedback. Researchers Nikitov and Stone in a forensic case study of a single opportunistic eBay seller found buyers sometimes embedded negative comments in positive feedback as a means of avoiding retaliation from sellers and damage to their reputation. This category of positive feedback was described as “negative-positive” feedback. An example of negative-positive type feedback is “Good product, but slow shipping.” This research study investigated the concept of using negative-positive type feedback as a signature to identify potential opportunistic sellers in an online auction population. As experienced by prior researchers using data extracted from the eBay web site, the magnitude of data to be analyzed in the proposed study was massive. The nature of the analysis required - judgment of seller behavior and contextual analysis of buyer feedback comments – could not be automated. The traditional method of using multiple dedicated human raters would have taken months of labor with a correspondingly high labor cost. Instead, crowdsourcing in the form of Amazon Mechanical Turk was used to reduce the analysis time to a few days and at a fraction of the traditional labor cost. The research’s results found that the presence of subtle buyer behavior in the form of negative-positive type feedback comments are an inter-buyer signal indicating that a seller was behaving fraudulently. Sellers with negative-positive type feedback were 1.82 times more likely to be fraudulent. A correlation exists between an increasing number of negative-positive type feedback comments and an increasing probability that a seller was acting fraudulently. For every one unit increase in the number of negative-positive type feedback comments a seller was 4% more likely to be fraudulent.

No description available.

No description available.

No description available.

No description available.

While the relational model for data storage is sufficient for the modelling and manipulation of a large number of application domains, a growing class of application domains are either difficult or impossible for the relational model to deal with efficiently. The realisation of this fact has led to a proliferation of data models that attempt to increase the complexity and semantic capture of the domains that they can model - the development of object-oriented databases and the various semantic data models are a result of this. The idea of using logic to define, manipulate and constrain data has given rise to large numbers of systems that interface - not always successfully - a database system and a logic processing system. Most such systems are based on Prolog or its derivations. This thesis describes the development and use of an object-oriented and semantically rich form of logic - conceptual graph theory - as a system for the definition, manipulation, and constraint of data. It describes a theoretical correspondence between conceptual graph theory and the relational model, and proceeds to develop an augmented, hybrid theory that is formally more expressive and as rigorous as those languages based on the relational algebra or calculus. This thesis also describes the design and implementation of a hybrid relational database - conceptual graph system, that has a cleaner and more principled system of semantic capture than other (for example, Prolog-based) systems, and that is also adaptive in nature - it automatically modifies its underlying storage structures in accordance with modifications made to the structures of the application domain over time. This completely shields the user from any responsibility for database design and maintenance, and so the user need only be concerned with application domain knowledge. Although the implementation described is incomplete, it can be extended to produce a deductive, object-oriented database system based on conceptual graphs.

The objective of this thesis is to examine the returns to mobility of civilian personnel within the Department of Defense (DoD). This study employs panel data provided by the Defense Manpower Data Center (DMDC) and drawn from the Department of Defense Civilian Personnel Data Files. The dataset consisted of 21,143 personnel who were new hires in years 1994-1995. Between 1994-1995 and 2003, 3,267 (15.4%) employees were interstate migrants. The data were set up as an unbalanced panel with a total of 132,068 observations. This study uses ordinary least squares (OLS), probit and Heckman selection-correction techniques to explore two returns to mobility measures: compensation and promotion. Multivariate models were specified and estimated for each performance measure. The results indicated workers who migrate are more likely subsequently to be promoted. Migration is a strategic move for workers to advance and maximize their personal utility since migrants earn higher salaries than non-migrants. Females present no evidence of tied-mover effects, and pursue promotion and salary opportunities like males. Women promote faster than men, and women migrants increase their promotion rates even more. Females, however, earn lower salaries than males. The models also reveal that veterans earn lower salaries than non-veterans and have no significant advantages in promotion over their counterparts.

A Database Management System (DBMS) is system software for managing a large amount of data in secondary memory. The standard DBMS used today in both industry and the military is the Relational DBMS (RDBMS). The RDBMS is based upon the relational paradigm, whereas modern software development technologies that interact with the RDBMS are based upon the object-oriented paradigm. This difference in paradigms presents a conceptual mismatch which greatly reduces programmer and developer productivity. Additionally, wireless handheld devices have become ubiquitous both in the military and in the community at large. These handheld devices provide a convenient means of information access. To date, the military has failed to capitalize on the use of handheld devices as a convenient means of information access with respect to the large amounts of information stored in its databases. This thesis investigates various database application architectures and proposes an architecture that will not only overcome the conceptual mismatch between the relational and object-oriented paradigms, but also allows handheld device access to the database. A proof-of-concept prototype database application that provides handheld device access to a military personnel database is built to show the viability of the proposed architecture.

A considerable amount of research has been recently conducted on face recognition tasks, due to increasing demands for security and authentication applications. Recent technological developments in uncooled IR imagery technology have boosted IR face recognition research applications. Our study is part of an on-going research initiated at the Naval Postgraduate School that considers an uncooled low-resolution and low-cost IR camera used for face recognition applications. This work investigates a recent approach which approximates nonlinear kernel-based methods at a significantly reduced computational cost. Our research was applied to an IR database. Results show that this scheme may perform sufficiently close to its â kernelizedâ version considered in a previous study, at a fraction of the computational cost, provided that the associated parameters are well tuned. The thesis considers a relative comparison between the two algorithms, based on identification and verification experiments and considers a statistical test to investigate whether classification performance differences may be considered statistically significant. Results show that, from a cost perspective, a low-resolution uncooled IR camera in conjunction with a low computational-cost classification scheme can be embedded in a robust face recognition system to efficiently address the issue of authentication in security-related tasks.

The vision of the future Internet of Things is posing new challenges due to gigabytes of data being generated everyday by millions of sensors, actuators, RFID tags, and other devices. As the volume of data is growing dramatically, so is the demand for performance enhancement. When it comes to this big data problem, much attention has been given to cloud computing and virtualization for their almost unlimited resource capacity, flexible resource allocation and management, and distributed processing ability that promise high scalability and availability. On the other hand, the variety of types and nature of data is continuously increasing. Almost without exception, data centers supporting cloud based services are monitored for performance and security and the resulting monitoring data needs to be stored somewhere. Similarly, billions of sensors that are scattered throughout the world are pumping out huge amount of data, which is handled by a database. Typically, the monitoring data consists time series, that is numbers indexed by time. To handle this type of time series data a distributed time series database is needed.   Nowadays, many database systems are available but it is difficult to use them for storing and managing large volumes of time series data. Monitoring large amounts of periodic data would be better done using a database optimized for storing time series data. The traditional and dominant relational database systems have been questioned whether they can still be the best choice for current systems with all the new requirements. Choosing an appropriate database for storing huge amounts of time series data is not trivial as one must take into account different aspects such as manageability, scalability and extensibility. During the last years NoSQL databases have been developed to address the needs of tremendous performance, reliability and horizontal scalability. NoSQL time series databases (TSDBs) have risen to combine valuable NoSQL properties with characteristics of time series data from a variety of use-cases.   In the same way that performance has been central to systems evaluation, energy-efficiency is quickly growing in importance for minimizing IT costs. In this thesis, we compared the performance of two NoSQL distributed time series databases, OpenTSDB and InfluxDB, based on the energy consumed by them in different scenarios, using the same set of machines and the same data. We evaluated the amount of energy consumed by each database on single host and multiple hosts, as the databases compared are distributed time series databases. Individual analysis and comparative analysis is done between the databases. In this report we present the results of this study and the performance of these databases based on energy consumption.