Master of Science / Department of Computing and Information Sciences / William H. Hsu / Stock trading is a complex process which is subject to distinct events both inside and outside an organization. An increase in revenue is a direct influence which causes the stock price to move upwards; likewise, the price of a stock fluctuates due to indirect influences. For example, a stock’s price may move upwards due to a firm arriving at beneficial deals or adding eminent professionals to its board of directors; a rise in correlated stock prices. The stock prices are affected to a great extent by statements of the finance minister and other related officials. In addition, subjective judgments and emotions of traders can also influence the variation of indices and stock prices in the market. The efficient market hypothesis proposes that stock price is unpredictable, assuming all past information has been influenced on current price and therefore it is not useful for the prediction of future price. Nevertheless, there are opposing theories which state that stock prices are predictable through the identification of trends and price patterns based upon past data such as price and volume quotes, balance sheets, and income statements. In the stock market, naive traders (or investors) assume risks due to the above uncertainties, but still have opportunities to make profits through proper, in-depth analysis on sufficient quantities of past data. There are many indicators that are accessible and can help predict the direction of future prices or index values using fundamental and technical data. Fundamental data, derived from the balance sheets and income statements, is preferred for mid-term and long-term investors but not suitable for short-term investors; meanwhile, technical data can be used for short-term investors as well. Technical data is preferred for short-term investments but it can also be used for long-term investments by choosing a specific window of time to look back in determining the indicators for long periods. The current trading model for stocks and indices was developed using an accuracy-based learning classifier system (XCS), which combines reinforcement learning, genetic algorithms, and other heuristics to form an adaptive system whose purpose is to execute stock trades for profit. A test bed developed for experimenting with this system consists of technical data, with candidate features chosen as the most popular indicators.

Master of Science / Department of Computing and Information Sciences / William H. Hsu / In decision-making applications, the Skyline query is used to find a set of non-dominated data points (called Skyline points) in a multi-dimensional dataset. A data point dominates another data point if it is at least as good as the other data point in all dimensions and better in at least one dimension. The skyline consists of data points not dominated by any other data point. Computing the skyline points of a dataset is essential for applications that involve multi-criteria decision making. Skyline queries filter out the interesting tuples from a potentially large dataset. No matter how we weigh our preferences along the attributes, only those tuples which score best under a monotone scoring function are part of the skyline. In other words, the skyline does not contain tuples which are nobody's favorite. With a growing number of real-world applications involving multi-criteria decision making over multiple dimensions, skyline queries can be used to answer those problems accurately and efficiently. This report mainly focuses on various skyline computing algorithms which can be used for online processing efficiently and are suitable to present multi-criteria decision making scenario. I implemented the Branch-and-Bound skyline Algorithm on two different data sets; one is a synthetic dataset and the other is a real dataset. My aim is to explore various subspaces of a given dataset and compute skylines over them, especially those subspace skylines which contain the least number of the skyline points.

Master of Science / Department of Computing and Information Sciences / Daniel Andresen / The objective of this project is to develop an iPhone application for the Konza Prairie LTER and play a significant role in the development of their website. The data of the Konza Prairie LTER is vast and includes the spatial datasets, weather reports, text archives, information about the flora and fauna at Konza Prairie Natural Area, integrated project areas, Dataset Codes, LTER Core Areas, Related projects, Konza Documents and Permits, and Publications. The module which was developed by me provides users information like publications, personnel information and datasets. The titles of the publications along with their PIs will be retrieved from the database and formatted onto the webpage. The 'Personnel' page has links directing to Primary Contacts, Faculty and Staff and Graduate Students. The attractive feature of this personnel page is that each personnel in the list will have a '+' link which enables the user more information about that particular personnel like their field of interests, mailing address and a link to view their profile. As the website already has all the required data and information in detail, the mobile app only briefs about them. People who are working on-site can make of this app efficiently as its features includes list of upcoming events, map of Konza with GPS feature, Post Card feature, Primary personnel to be contacted and other features. The website is developed using Visual Studio 2010 and SQL Server 2008 database. The iPhone app enhanced my knowledge and provided me with real-time exposure to iPhone SDK tools, X-code, Interface builder and its development environment.

Master of Science / Department of Computing and Information Sciences / Mitchell Neilsen / E-Learning System offers online training for the employees of the company. Every company uses different technologies based on the type of work they do, products they develop and services they offer. Fresher's who are just recruited many not be familiar with the technologies the company is using or the employees already working in the company may not have required level of efficiency in those technologies. It is necessary to train all the employees of the company according to their profiles with different technologies the company uses. The main aim of the project is to create web interface for E-learning System for the employees of the company. In order to use this system the employees needs to register as member and can start using services. The registered users have more access to system and can make use of training system more efficiently then unregistered users. The registered users will have access to master library where all the materials and videos of different courses are available. The users can enroll into different levels of different courses and get trained by downloading and studying the materials available .The users can also schedule for a test for the courses and levels they are registered into. Records are created for the level and course they re passed so that admin they view them. Creating a web interface will help the employees to study easily, manage their courses efficiently and can also check their performances. It also aims for developing reports for each users in the E-Learning System so that the users can see their performances and the admin can also recommend him/her to appropriate project's based on the performance of the employee in the technologies he/she is registered into.

Master of Science / Department of Computing and Information Sciences / Daniel Andresen / Mobile applications can be one of the best ways to keep consumers engaged with a brand as they are on the move. With the increase in demand for smartphones and efficiency of wireless networks, the demand for mobile applications has increased incredibly. Android is one of the most popular open source platforms that offers the developers full access to the framework API’s so as to build innovative applications. The main aim of this project is to build an Android application that helps the users to find a Restaurant in a specified location and according to the specified tastes. The main features provided by the Restaurant Finder application are as follows: 1) Basic Search where the user can search for a particular restaurant based on any keyword and Advanced Search where the user can specify the category, rating and the distance range for the restaurants. 2) Google Maps that shows the top 5 restaurants in the city of the current location and the routes to a particular restaurant. 3) The users can write a review, see the reviews and invite a friend/colleague to meet at a particular restaurant. 4) Google Calendar where the user can mark an event.

Pen-based computers bring the promise of tapping into people’s expressiveness with pen and paper and producing a platform that feels familiar while providing new functionalities only possible within an electronic medium. To this day, pen computers’ success is marginal because their interfaces mainly replicate keyboard and mouse ones. Maximizing the potential of pen computers requires redesigning their interfaces so that they are sensitive to the pen’s input modalities and expressiveness. In particular, pressure is an important and expressive, yet underutilized, pen input modality. This dissertation advances our knowledge about pressure-aware, pen-based interactions and how people use these techniques. We systematically explore their design by first investigating how pressure can affect pen interactions. We propose novel techniques that take advantage of the pressure modality of a pen to control, link, and annotate digital video. We then study people’s performance using pressure to navigate through a set of elements and find that they can discriminate a minimum of six different pressure regions. We introduce the concept of Pressure Widgets and suggest visual and interaction properties for their design. We later explore pressure’s use to enhance the adjustment of continuous parameters and propose Zliding, a technique in which users vary pressure to adjust the scale of the parameter space, while sliding their pen to perform parameter manipulations. We study Zliding and find it a viable technique, which is capable of enabling arbitrarily precise parameter adjustments. We finally present a novel interaction technique defined by the concurrent variation in pressure applied while dragging a pen. We study these pressure marks and find that they are a compact, orientation-independent, full interaction phrase that can be 30% faster than a stateof-the-art selection-action interaction phrase. This dissertation also makes a number of key contributions throughout the design and study of novel interaction techniques: -It identifies important design issues for the development of pressure-sensitive, pen operated widgets and interactions, -It provides design guidelines for interaction techniques and interface elements utilizing pressure-enabled input devices, -It presents empirical data on people’s ability to control pressure, and -It charts a visual design space of pressure-sensitive, pen-based interactions.

For almost 50 years, researchers have been exploring the use of stereoscopic displays for visualizing and interacting with three-dimensional (3D) data. Unfortunately, a number of unfavorable qualitative properties have impeded the wide-spread adoption of traditional 3D displays. The volumetric display, a more recent class of 3D display to emerge, possesses unique features which potentially makes it more suitable for integration into workplace, classroom, and even home environments. In this dissertation we investigate volumetric displays as an interactive platform for 3D applications. We identify the inherent affordances unique to volumetric displays, such as their true 3D display volume, 360° viewing angle, and enclosing surface. Identifying these properties exposes human factor issues which we investigate and interaction issues which we address. First, we evaluate the user’s ability perceive imagery displayed by a volumetric display. In a formal experiment, we show that depth perception can be improved, in comparison to more traditional platforms. We then perform an experiment which evaluates users’ ability to read text under 3D rotations, and present a new algorithm which optimizes text rotation when viewed my multiple users. Next, we investigate the user’s ability to select 3D imagery within the display. Results show that the dimension defining the depth of the object can constrain user performance as much as or more than the other two dimensions of the target. This leads us to explore alternative methods of selection which are less constraining to the user. We define a suite of new selection techniques, of which several are found to have significant benefits in comparison to techniques traditionally used in 3D user interfaces. Next, we describe our development of the first working interactive application, where a volumetric display is the sole device for input and display. The application presents a first glance at what the equivalent of today’s graphical user interface might be on a volumetric display. We then develop a prototype application which allows multiple users to simultaneously interact with the volumetric display. We discuss and address the core issues related to providing such a collaborative user interface, and report feedback obtained from usage sessions and expert interviews.

One of the most essential parts of any software requirements analysis effort is the exploration of alternative ways by which stakeholder problems can be solved. Systematic modeling and analysis of requirements variability allows better decision making during the early requirements phase and substantiates design choices pertaining to the configurability aspect of the system-to-be. This thesis proposes the use of goal models for capturing and reasoning about requirements variability. The goal models we adopt consist of AND/OR decompositions of stakeholder goals and express alternative ways by which stakeholders may wish to achieve them. By capturing goal variability using such models, we propose a shift of focus from variability of the software design, to variability of the problem that the design is intended to solve. This way, we ensure that every important variation of the problem is identified and analyzed before variations of the solution are specified. The thesis exploits opportunities that arise from this new viewpoint. Firstly, a variability-intensive goal decomposition process is proposed. The process is based on associating each high-level goal to a set of variability concerns that must be addressed through decomposition. We introduce a universal categorization of such concerns and also show how domain-specific variability concerns can be identified by annotating domain corpora. Concern-driven decomposition offers a structured way of thinking about problem variability, while systematizing its identification process. Further, an expressive LTL-based preference language is introduced to support leverage of large spaces of goal alternatives. The language allows the expression of preferences over behavioral and qualitative properties of solutions and a reasoning tool allows the identification of alternatives that satisfy these preferences. This way, individual stakeholders can get the solution that exactly fits their needs in a particular situation, through simply specifying desired high-level characteristics of these solutions. Finally, a framework for connecting alternatives at the goal level to alternative configurations of common desktop applications is presented. The framework shows how a vast number of configurations of a software application can be evaluated and ranked with respect to a small number of quality goals that are more intuitive to and comprehensible by end users.

A fundamental goal of computer vision is the ability to analyze motion. This can range from the simple task of locating or tracking a single rigid object as it moves across an image plane, to recovering the full pose parameters of a collection of nonrigid objects interacting in a scene. The current state of computer vision research, as with the preponderance of challenges that comprise "artificial intelligence", is that the abilities of humans can only be matched in very narrow domains by carefully and specifically engineered systems. The key to broadening the applicability of these successful systems is to imbue them with the flexibility to handle new inputs, and to adapt automatically without the manual intervention of human engineers. In this research we attempt to address this challenge by proposing solutions to motion analysis tasks that are based on machine learning. We begin by addressing the challenge of tracking a rigid object in video, presenting two complementary approaches. First we explore the problem of learning a particular choice of appearance model---principal components analysis (PCA)---from a very limited set of training data. However, PCA is far from the only appearance model available. This raises the question: given a new tracking task, how should one select the most-appropriate models of appearance and dynamics? Our second approach proposes a data-driven solution to this problem, allowing the choice of models, along with their parameters, to be learned from a labelled video sequence. Next we consider motion analysis at a higher-level of organization. Given a set of trajectories obtained by tracking various feature points, how can we discover the underlying non-rigid structure of the object or objects? We propose a solution that models the observed sequence in terms of probabilistic "stick figures", under the assumption that the relative joint angles between sticks can change over time, but their lengths and connectivities are fixed. We demonstrate the ability to recover the invariant structure and the pose of articulated objects from a number of challenging datasets.

Effective coordination and communication are essential to the success of software organizations, but their study to date has been impaired by theoretical confusion and fragmentation. I articulate a theory that argues that the members of software organizations face a constant struggle to share and negotiate an understanding of their goals, plans, status, and context. This struggle lies at the heart of their coordination and communication problems. The theory proposes an analysis of organizational strategies based on four attributes of interaction that foster the development of shared understanding: synchrony, proximity, proportionality, and maturity. Organizations that have values, structures, and practices which facilitate these qualities find it easier to coordinate and communicate effectively. This argument has serious implications for traditional concepts in our literature. Project lifecycle processes and documentation are poor substitutes for informal but unscalable coordination and communication mechanisms. Practices and tools are valuable to the extent that they enable the development of shared understanding across our criteria. Co-location and group cohesion take advantage of the four criteria and therefore have direct advantages for software teams. Finally, growth is detrimental to the effectiveness of the organization because it hinders the use of small-scale mechanisms and it leads to an undesirable formalization. The theory is supported with empirical evidence collected from five case studies of a wide variety of software organizations, and it has explanatory and predictive power. The thesis links this theory to other current research efforts and shows that it complements and enhances them by providing a more solid theoretical foundation and by reclaiming the relevance of synchronous, proximate, proportionate, and mature interactions in software organizations.