<p> Modern Internet services are multi-tiered and are typically hosted in virtualized shared platforms. While facilitating flexible service deployment, multi-tier architecture introduces significant challenges for Quality of Service (QoS) provisioning in hosted Internet services. Complex inter-tier dependencies and dynamic bottleneck tier shift are challenges inherent to tiered architectures. Hard-to-predict and bursty session-based Internet workloads further magnify this complexity. Virtualization of shared platforms adds yet another layer of complication in managing the hosted multi-tier Internet services. </p><p> We consider three critical aspects of Internet service management for improved performance and quality of service provisioning : admission control, dynamic resource provisioning and service differentiation. This thesis concentrates on statistical learning based approaches for multi-tier Internet service management to achieve efficient, balanced and scalable services. Statistical learning techniques are capable of solving complex dynamic problems through learning and adaptation with no <i>priori</i> domain-specific knowledge. We explore the effectiveness of supervised and unsupervised learning in managing multi-tier Internet services. </p><p> First, we develop a session based admission control strategy to improve session throughput of multi- tier Internet services. Using a supervised bayesian network, it achieves coordination among multiple tiers resulting in a balanced service. Second, we promote session-slowdown, a novel session-oriented metric for user perceived performance. We develop a regression based dynamic resource provisioning strategy, which utilizes a combination of offline training and online monitoring, for session slowdown guarantees in multi-tier systems. Third, we develop a reinforcement learning based coordinated combination of admission control and adaptive resource management for multi-tier Internet service differentiation and performance improvement in a shared virtualized platform. It addresses limitations of supervised learning by integrating model-independence of reinforcement learning and self-learning of neural networks for system scalability and agility. Finally, we develop an user interface based Monitoring and Management Console, intended for an administrator to monitor and fine tune the performance of hosted multi-tier Internet services. </p><p> We evaluate the developed management approaches using an e-commerce simulator and an implementation testbed on a virtualized blade server system hosting multi-tier RUBiS benchmark applications. Results demonstrate the effectiveness and efficiency of statistical learning approaches for QoS provisioning and performance improvement in virtualized multi-tier Internet services.</p>

<p> Communication now is easier than ever before. One consequence of this is the emergence of virtual communities, unconstrained by physical proximity. We perform two investigations into changing social trends. We study a corpus of 100 years of newspaper articles to see if we can find evidence to support the popular intuition that as news cycles have sped up, the public's attention span has gotten shorter. We find no such evidence: to the contrary, we find that the typical length of time that a person's name stays in the news has not changed over time, and celebrities now stay in the news for longer than ever before. We also investigate a new kind of community on Twitter called a group chat, where members have regular meetings to discuss a broad range of topics, from medical conditions to hobbies. We find that the phenomenon is growing over time, and paint a broad picture of the topics which one could find a group chat to discuss. With a view to helping connect new participants to group chats they may not have been able to find or might not have been aware of, we design an algorithm to rank group chats in the context of a topic given as a query.</p>

<p> Digital media have made it possible for people with disabilities to have better access to information and mainstream publications. New and improved guidelines by the World Wide Web Consortium (W3C) have helped millions of users with various types of disabilities to utilize the web to its full potential. The Web Content Accessibility Guideline (WCAG) 2.0 has pushed for several reforms to ensure that web pages are equally accessible for people with disabilities. How- ever, there are some limitations with the WCAG 2.0 when it comes to accommodating people with Low Vision. The W3C agrees and is currently working to draft a new set of guidelines that will address this specific problem. In the push to accommodate people with low vision, a team at California State University Long Beach, led by Dr. Wayne E Dick, proposed a software so- lution called Typometric Prescriptions or TRx which generates a user based custom stylesheet. The purpose of this thesis was to build on top of the working framework of TRx and shape it to be a completely functional stylesheet generator for people with low vision. The research and technical work put into this study have led to the development of a keyboard accessible color picker that makes it possible to pick a color from the possible 16 million choices, with less than 48 keystrokes.</p><p>

<p> The internet's original design, guided by the end-to-end design principle, pushed all application-specific logic and complexity to the edges of the network and kept the core of the network focused on the simple task of delivering data. The original end-to-end principle, however, did not explicitly account for trust and security. There are several central points of trust and failure on the traditional internet. These include root servers for the Domain Name System (DNS) and public-key infrastructure like Certificate Authorities (CAs) that publish security certificates. Further, the success of cloud hosted services in the last decade means that most user data is stored on remote servers and end-users need to trust these remote servers for correct execution of their applications. </p><p> In this thesis, we present a new internet architecture that explicitly follows the trust-to-trust design principle, i.e., end-users don't need to trust the core of the network for anything, and end-users can use applications and services in a fully decentralized way. We make the observation that cryptocurrency blockchains, like Bitcoin, can be used to bootstrap trust for new nodes joining a network. We identify the various limitations, like high latency and limited bandwidth, of contemporary blockchains and discuss how our architecture can scale by moving most operations outside of the blockchain layer. </p><p> We detail our experience of running a large production system on top of a cryptocurrency blockchain and how that experience guided our design. We present the implementation of a new decentralized internet, called Blockstack, that takes the trust-to-trust architecture from a theoretical concept to a production system. Deploying new systems by modifying production blockchains is hard because it requires coordination and agreement from several parties. We introduce virtualchains, a virtual blockchain constructed by processing data from underlying blockchains, to enable the seamless introduction of new functionality on top of blockchains without requiring any consensus-breaking changes. Blockstack is already powering several fully decentralized applications, like OpenBazaar; it's released as open-source software and, to date, more than 70,000 domains have been registered on it.</p><p>

<p> In this research, we focus on identifying false identities in social networks. We performed a detailed study on different string matching techniques to identify user profiles with real or fake identity. In this thesis, we focus on a specific case study on sex offenders. Sex offenders are not supposed to be online on social networking sites in few states. To identify the existence of offenders in social networks, we ran experiments to compare datasets downloaded from Facebook and offender registries. To identify the most suitable string matching technique to solve this particular problem, we performed experiments on various methods and utilized the most appropriate technique, the Jaro-Winkler algorithm. The major contribution of our research is a weight based scoring function that is capable of identifying user records with full or partial data revealed in social networks. Based on our data samples created using metadata information of Facebook, we were able to identify the sex offender profiles with real identity and seventy percent of the sex offenders with partial information.</p>

<p> Every year a plethora of students come to pursue their careers at California State University, Long Beach (CSULB). The university supports an arena of web portals for enrollment, courses, department, student recreation and wellness amongst others.</p><p> Although there are enrollment services available, they involve mandatory check-ins and paperwork. GoSchoolPro portal eases the process of enrollment for the students by providing convenient services. This portal will not only help in facilitating the process for the incoming students but will also give voice to the past experiences of the alumni. It will be a student democrat: For the students and by the students.</p><p> The current project presents GoSchoolPro as a web application, which leverages the Model-View-Controller (MVC) architecture as its foundation and implements the Active Server Pages using Dot NET (ASP.NET) framework. GoSchoolPro uses the backbone of the Object-Oriented Analysis and Design (OOAD) methodology and other Software Engineering (SE) best practices.</p>

<p> User experience changes over time as a consequence of prolonged engagement with many products or services. Interaction evolves as we learn, adopt, and shift from old use patterns to new ones. Scholarly work in HCI has been successful in its investigation of interactive experiences situated in current places and times. This dissertation aims to provide a complementary perspective by paying attention to user experiences that happen in the past. In particular, it explores visual thinking as a method of describing and analyzing interactive experiences over time. </p><p> In addition, this dissertation has a pragmatic purpose: to develop applicable research approaches for professional practice. My professional experience as a designer yielded the insight that many practitioners collect user stories as references for envisioning possible futures. However, the collection and application of user stories can be improved. In many cases, the analysis of user stories is fragmented and lacking a systematic approach. Scholars are in a unique position to support practitioners by developing cohesive, systematic methods. I believe that academically developed methods should aim to support professional practice. </p><p> The overall aim of this dissertation is to produce methods for studying users and artifacts, the two essential components of a user experience. </p><p> The research commenced with a preliminary study of the use of visual diagrams as interview aids for recalling daily email usage. The research continued with a review study comparing different existing approaches to using visual support in user research as well as an analytic study of how to examine the history of interaction artifacts using "visual annotations." Finally, the research concluded with an exploratory study wherein timeline annotations of "retrospective interaction histories," were applied in workshops. Following these four studies, I examined how visual methods in a retrospective study might help capture and represent heterogeneous individual user experiences. This examination led to my proposal of a theory of temporal anchors. Temporal anchors capture the idea that the measurement of user experience over time and the history of interactive artifacts might serve as anchor points in users' retrospection. I conclude the dissertation with a discussion of potential future research directions.</p>

<p> Because data on alumnae/i is difficult to collect and change over time, obtaining up-to-date information on graduates &mdash; for the purposes of highlighting their professional achievements and facilitating networking &mdash; is a challenge faced by many higher education institutions. Social networks provide a means for alumnae/i to share successes, but viewing this information typically involves an account or a paid subscription. In this project, I address these issues by designing and implementing a free, easy-to-use, and updateable Mills Computer Science alumnae/i website backed by a database. The website has the potential to showcase the most recent or impressive alumnae/i achievements and prominently feature all graduates' accomplishments via their profiles. Additionally the database can store manually provided alumnae/i information and extract further data from LinkedIn. Evaluations of the project indicate that users and administrators find the website easy to search, navigate, and update and the information provided helpful. Because of its versatile design, this project could serve similar needs for other disciplines and institutions. </p>

<p> Interaction design is the definition of digital behavior, from desktop software and mobile applications to components of appliances, automobiles, and even biomedical devices. Where architects plan buildings, graphic designers make visual compositions, and industrial designers give form to three-dimensional objects, interaction designers define the digital components of products and services. These include websites, mobile applications, desktop software, automobiles, consumer electronics, and more. Interaction design is a relatively new but fast-growing discipline, emerging with the explosive growth of the World Wide Web. In a software-saturated world, every day, multiple times a day, billions of people encounter the work products of interaction design. </p><p> Given the reach of their profession, how interaction designers work is of paramount concern. In considering interaction design, this dissertation turns away from a longstanding question of design studies: <i>How does interaction design demonstrate a special form of human thought?</i> And towards a set of questions drawn from practice-oriented studies of science and technology: <i>What kinds of objects and subjects do interaction design practices make, and how do those practices produce them? </i> </p><p> Based on participant observation at three San Francisco interaction design consultancies and interviews with designers in California's Bay Area, this dissertation argues that performance practices organize interaction design work. By &ldquo;performance practices,&rdquo; I mean episodes of storytelling and narrative that take place before an audience of witnesses. These performances instantiate &mdash; make visible and tangibly felt &mdash; the human and machine behaviors that the static deliverables seem unable on their own to materialize. In doing so, performances of the project help produce and sustain alignment within teams and among designers, clients, and developers. </p><p> In this way, a focus on episodes of performance turns our concerns from cognition, in which artifacts assist design thinking, to one of enactment, in which documents, spaces, tools, and bodies actively participating in producing the identities, responsibilities, and capacities of project constituents. It turns our attention to questions of political representation, materiality and politics. From this perspective, it is not necessarily how designers <i> think</i> but how they stage and orchestrate performances of the project that makes accountable, authoritative decision-making on behalf of clients and prospective users possible.</p>

<p> The objective of this thesis was to explore the topic of scalable web development, and it answered the question, &ldquo;How do you scale a website to handle more traffic at peak times without wasting resources?&rdquo; This is important research to any web company that has issues with rising costs as demand for their website increases. It would be wise for every online business to be prepared for more web traffic, before it occurs, without spending the budget of a multi-million user web company in low traffic periods. The last thing you want is an error as your customer base starts to arrive, giving them a bad experience for their first impressions, which would result in lost revenue.</p><p> Scalable software development architectures, including microservices, big data, and Kubernetes were studied, in addition to similar web service companies including Facebook, Twitter, and Match.com. A scalable architecture was designed for a social media web service, MeAndYou, using the big data configuration with a shared Aurora database, which was configured using an auto-scaling group attached to a load balancer in Amazon Web Services (AWS). It was tested using a custom threaded Selenium-based Python script that applied simulated user load to the servers. As the load was applied, AWS added more Elastic Compute Cloud (EC2) instances running a virtual disk image of the web server. After the load was removed, the instances were terminated automatically by AWS to save costs.</p><p> Countless steps were taken to make the web service bigger and more scalable than it originally was, before testing, including adding more fields to user profiles, adding more search types, and separating the layers of code into different Hypertext Preprocessor (PHP) files in the front-end. A version control system was configured on the servers using GitHub and rsync. The systems architecture designed suggests the Match Engine should use a stream processing message queue, which would allow the system to factor searches one at a time as they are created, with horizontal scaling capabilities, rather than grabbing the entire database and storing it in memory. The backend Match Engine was also tested for accuracy using Structured Query Language (SQL) injection, which determined how the match algorithm should be improved in the future.</p><p>