CAS in Modernen Methoden der Informatik

Transkript

1 CAS in Modernen Methoden der Informatik Kurskatalog

2 Inhaltsverzeichnis CAS in Modernen Methoden der Informatik Inhaltsverzeichnis A Kursaufbau CAS in Modernen Methoden der Informatik Vorbemerkung Beschreibung Zielpublikum Kursdauer Zulassung zum Studiengang Kursprogramm Kursablauf und Bewertung Abschluss B Kurse 1 Modul Big Data Data Mining Cluster Computing, MapReduce, and the Future Heterogeneous Data Processing on the Web of Data / Information Visualization Virtualization and Clouds Computational Perception Multilingual Text Analysis and Information Extraction Social Computing Network-based Business Intelligence Coaching Day 15 2 Modul Engineering Solutions Requirements Engineering Software Quality Analysis Software Quality Assurance with Continuous Integration Development 2.0: Human-Centered Software Engineering / Open Innovation for Software Products & Processes 2.5 Human Computer Interaction Engineering Electronic Markets Digital Innovation and Social Networking for Companies Sustainability and Green IT Coaching Day 25 C Dozierende Seite 2 Universität Zürich, Institut für Informatik,

3 A Kursaufbau CAS in Modernen Methoden der Informatik Vorbemerkung Änderungen vorbehalten. Das Kursreglement ist Bestandteil dieses Kurskatalogs. Beschreibung Die Informatik ist einem schnellen Wandel unterworfen, und Wissen hat eine kurze Halbwertszeit. Die stetigen Neuerungen beeinflussen in immer stärkerem Masse den Berufsalltag, insbesondere die Art und Weise, wie Menschen und Computer interagieren. Weil sowohl der Stellenwert der Informatik in der Gesellschaft und Wirtschaft themenübergreifend immer wichtiger wird, als auch der zunehmende Einfluss der Sozialwissenschaften auf die Informationsverarbeitung erkennbar wird, ist der Kurs einerseits auf Big Data und andererseits auf den modernen Software-Entwurf fokussiert. Somit wird gewährleistet, dass die Teilnehmenden die neusten Entwicklungen der modernen Informatik kennenlernen, deren Auswirkungen einschätzen und auf die anstehenden Veränderungen angemessen reagieren können. Zielpublikum Informatikerinnen und Informatiker im Beruf mit Freude am Life-long Learning, die sich für die Berufspraxis der Zukunft mit den aktuellen Methoden in der Informatik und relevanten Forschungserkenntnissen einen Wissensvorsprung verschaffen wollen. Kursdauer Februar bis Juni 2016 Kurstage: Freitag und Samstag Zulassung zum Studiengang Hochschulabschluss auf Masterstufe oder gleichwertige Qualifikation sowie Berufserfahrung im Informatikbereich. Bei einem Bachelorabschluss oder anderen Gleichwertigkeitsfragen entscheidet die Kursdirektion «sur dossier» und abschliessend. Sie kann für Studienbewerberinnen und -bewerber, welche ausnahmsweise aufgrund vergleichbarer Qualifikationen zugelassen werden sollen, die Zulassung von einem erfolgreichen Aufnahmegespräch abhängig machen. Es besteht kein Anspruch auf Zulassung. Einzelne Module oder Teile davon können einem weiteren Personenkreis der universitären und ausseruniversitären Öffentlichkeit zugänglich gemacht werden. Der Besuch einzelner Module führt nicht zu einem Abschluss. Seite 3 Universität Zürich, Institut für Informatik,

4 Kursprogramm Das Kursangebot ist in zwei Module, Big Data und Engineering Solutions aufgeteilt. Um den CAS erfolgreich zu absolvieren, sind total 15.0 ECTS-Kreditpunkte nötig. Das Modul Big Data ergibt 8 ECTS-Kreditpunkte, das Modul Engineering Solutions ergibt 7 ECTS-Kreditpunkte. Ein ECTS-Kreditpunkt entspricht einem Arbeitsaufwand von ca. 30 Stunden. Dieser setzt sich zusammen aus Präsenzzeiten während den Kurstagen sowie Vor- und Nachbereitung. Die Module bestehen aus den angebotenen Kurstagen und den beiden Coaching Days. Teilnehmende, die den CAS nicht absolvieren, wählen ihre Kurstage frei aus dem Programm aus. Modul / Kurstage Unterrichtssprache* Modul 1: Big Data Data Mining Cluster Computing, MapReduce, and the Future Heterogeneous Data Processing on the Web of Data / Information Visualization Virtualization and Clouds Computational Perception Multilingual Text Analysis and Information Extraction Social Computing Network-based Business Intelligence Deutsch Deutsch Deutsch Englisch Englisch Deutsch Englisch Englisch Coaching Day Modul 2: Engineering Solutions Requirements Engineering 2.0 Software Quality Analysis Software Quality Assurance with Continuous Integration Development 2.0: Human-Centered Software Engineering / Open Innovation for Software Products & Processes Human Computer Interaction Engineering Electronic Markets Digital Innovation and Social Networking for Companies Sustainability and Green IT Deutsch Deutsch Deutsch Deutsch Englisch Englisch Deutsch Deutsch Coaching Day Seite 4 Universität Zürich, Institut für Informatik,

5 * Die Unterrichtssprache ist die Sprache, welche während des Kurstages gesprochen wird. Die Kursunterlagen sind generell auf Englisch abgefasst. Kursablauf und Bewertung Pro Modul muss ein Leistungsnachweis erbracht werden. Dieser besteht aus einem schriftlichen Test über alle Themen des Moduls sowie in einer schriftlichen Arbeit zu einem gewählten Thema. Die CAS-Absolvierenden entscheiden sich im Laufe des Moduls für ein Thema nach Absprache mit einem Dozierenden. Die Arbeit soll einen Umfang von bis zu 20 Seiten haben und ist in Deutsch oder Englisch zu verfassen. Am Ende des jeweiligen Moduls findet der Coaching Day statt. An diesem Tag bereiten sich die CAS- Absolvierenden anhand von Anleitungen und Beispielen auf die schriftliche Arbeit vor. Sie können Fragen diskutieren sowie das gewählte Thema konkretisieren oder vertiefen. Der Abgabetermin der schriftlichen Arbeit ist jeweils 30 Tage nach dem Coaching Day des entsprechenden Moduls. Die schriftliche Arbeit wird vom Dozierenden benotet, mit den Noten 1 bis 6. Halbe Noten sind zulässig. Noten unter 4 sind ungenügend. Ein ungenügender Leistungsnachweis kann einmal am nächstmöglichen Termin, spätestens nach drei Monaten nach Kenntnis des Nichtbestehens, wiederholt werden. Andernfalls gilt er als definitiv nicht bestanden. Abschluss Certificate of Advanced Studies UZH in Modernen Methoden der Informatik (CAS UZH in Modernen Methoden der Informatik) Das Zertifikat wird verliehen, wenn 15.0 ECTS Kreditpunkte erworben worden sind und zwei schriftliche Arbeiten (Leistungsnachweise) mit einer Note 4.0 bewertet worden sind. Studierende, denen der Abschluss nicht verliehen wird, erhalten einen Nachweis über die erbrachten Leistungen. Seite 5 Universität Zürich, Institut für Informatik,

6 B Kurse 1 Modul Big Data 1.1 Data Mining Computers increasingly need to operate in environments that do not lend themselves to be modeled in the black and white certainties associated with binary system. Indeed, the world is mostly an uncertain place and Computer Science has largely decided to ignore the fact almost since its inception. Aggravating the issue, we collect ever more data about our endeavors that could be exploited if we would have consistent means to deal with the uncertainties that are usually associated with data of varying quality. This course will show how those limitations of the binary system can be overcome by the coherent usage of methods from probability and information theory. Based on these foundations it will introduce sound methods for reasoning about data that includes uncertainties. In addition, it will introduce the major inference techniques (i.e., predictive analysis methods oftentimes called data mining or business intelligence algorithms) that are used to improve computer systems performance in almost all sectors of their application. The course will exemplify its approaches using an industrial strength open source data mining tool that the participants are going to be using in a number of exercises. Course content From Binary to Probabilities o Shortcomings of binary computing o Uncertainty and probability o Probabilistic Reasoning o Bayes Rule Data Mining o The Main Data Mining Tasks: Prediction, association rules, and clustering o Exemplary Methods for each of the tasks Instructor Weka RapidMiner Prof. Abraham Bernstein, Ph.D. German Seite 6 Universität Zürich, Institut für Informatik,

7 1.2 Cluster Computing, MapReduce, and the Future Driven by the need for low-latency, large-scale data processing a numerous Internet companies and open source projects have embraced the principles of heterogonous cluster computing to process their data. These approaches veer away from using large centralized systems for processing and data storage but rely on large, low-cost commodity clusters to store and process their data. Through the systematic use of powerful abstractions these approaches achieve almost seamless parallelization, distribution, continuous availability, fail-over safety, and instantaneous deployment of new features. This course will start by introducing the complexities of distributed computing to establish the problems that need to be addressed by possible solutions. Based on these foundational problems it will introduce the powerful abstractions developed by the above-mentioned community and explain how they address the problems mentioned. These will expemlify how companies such as Google, facebook, Yahoo!, and others deal with todays data deluge. Next, the course will provide a revue of tools (chiefly among them Hadoop, HBase, Akka, and Signal/Collect) available freely to harness the powerful abstractions mentioned. The course will contain plenty of practical example algorithms as well as the possibility to develop first simple programs in class. Course content Distributed computing problems & large-scale data processing; Distributed computing in heterogeneous clusters Bulk Synchronous Processing abstractions for cluster computing: MapReduce and BigTable for harnessing heterogeneous clusters: Hadoop and HBase Beyond Bulk Synchronous Processing: Signal/Collect and Akka Instructor Hadoop HBase Akka Signal/Collect Prof. Abraham Bernstein, Ph.D. German Seite 7 Universität Zürich, Institut für Informatik,

8 1.3 Heterogeneous Data Processing on the Web of Data / Information Visualization Heterogeneous Data Processing on the Web of Data The World Wide Web has changed the way we all work. Its principles of decentralized control everybody can add information and everybody can link to anything have lead to an enormous explosion of information available at our fingertips today. Data, in contrast, is today mostly held imprisoned in data silos and difficult to process in a decentralized fashion. The advent of the Web of Data (sometimes also called the Semantic Web) has changed the equation. Based on the same principles simple addition of additional data points and linking of data to other data the web of data has been growing at an astonishing speed. Indeed, the publication of government data in the US (data.gov) and the UK (data.gov.uk) has stirred a number of worldwide initiatives to follow suite (in Switzerland the city of Zurich is expected to open a site in the first half of this year). The currently proposed approaches allow the schema-agnostic, robust processing of data from heterogeneous sources and have a number of interesting implications for both intra- end interorganizational data processing. This course will introduce the most relevant techniques for large-scale, schema-agnostic, heterogeneous data processing. Besides introducing the techniques it will also engage in a number of smaller practical exercises that exemplify the power of the proposed approaches. Course content Heterogeneous data; Semi-structured data; Data on the web: principles and approaches; Standards: RDF, RDFa, SPARQL, OWL, schema.org, microformats; The Linked Open Data Cloud; Processing on the web of data; Distributed data integration Instructor Jena CKAN Prof. Abraham Bernstein, Ph.D. German Seite 8 Universität Zürich, Institut für Informatik,

9 Information Visualization Increasing amounts of data, often in abstract or plain numerical form, are gathered and stored in many applications, business software and data processing systems. Besides using systematic analysis of these datasets, e.g. using statistical and data mining approaches, visual data exploration and presentation can significantly support the intuitive understanding of the relations which are implicitly present in the complex data collections. Patterns, clusters, trends and relative proportions within data collections are often only easily apprehensible if given in a visual form, thus exploiting the immense power of the human visual system to detect spatial relations and configurations. Unlike in most 3D graphics and scientific visualization domains, abstract (business) data often lacks a direct mapping to a visual representation. In this course we will look at the established basic principles and best practices in data visualization. The contents will cover the representation of quantities, use of color and visual perception, as well as many algorithms and methods for spatial mapping and visual display of data. Many principles discussed in this course are applicable to both (business) presentations as well as interactive data visualization and analysis systems. Course content Principles of information visualization Representation of quantitative values Color theory and perception Spatial mapping Methods and techniques Data visualization systems Visual analytics Instructors Various examples from research and practice Prof. Dr. Renato Pajarola German Seite 9 Universität Zürich, Institut für Informatik,

10 1.4 Virtualization and Clouds Significant advances in the domain of Information and Communications Technology (ICT) have been seen in the last thirty years. As such the usage of processing cycles may be considered to be a new utility. Computing services are highly essential to support daily demands in society, economics, and production sectors. One of the newer paradigms, following the cluster computing, high-performance computing, and grid computing waves is termed cloud computing, which utilizes the powerful concept of virtualization to provide such computing demands as suitable as possible. Thus, based on the definition of cloud computing and respective architectures a suitable resource allocation overview via virtualization concepts will be discussed. While a number of alternatives on virtualization will be distinguished, such as full virtualization, OS virtualization, hardware virtualization, storage virtualization, network virtualization, virtual machines, and hypervisors, the general functions needed for successful operations do include (a) real-time monitoring of virtualized workloads, (b) resource allocation, (c) configuration change tracking, and (d) reporting feedback for validating SLAs and assessing ROI for IT spending. As such a system-independent (as far as possible) view and introduction will be presented, of course, by relying on existing theory and available examples, such as Xen, VMWare Server, VMWare ESX Server, Virtual Iron, Dell VM, or Windows Virtual Server. Finally, the operational aspects of virtualization in selected cases will be taken up for further detailed study. Course content Definitions of cloud computing and virtualization Introduction into various virtualization approaches Insights into relevant functionality Discussions of selected operational aspects of virtualization Instructors To be decided, likely Xen and VMWare Server Prof. Dr. Burkhard Stiller English Seite 10 Universität Zürich, Institut für Informatik,

11 1.5 Computational Perception Computational perception (or computer vision) is a field that includes methods for acquiring, processing, analyzing, and understanding images and, in general, image datasets in order to produce numerical or symbolic information. Its goal is the extraction of semantic and/or geometrical information from images. Computer vision is used in a plethora of heterogeneous applications, such as image photography, augmenter reality, object recognition, security and surveillance, industrial inspection, 3D measurements, medical imaging, and so on. If one only considers the smartphone market, we can count already several thousands of apps using computer vision algorithms, and their number is doubling every 6 months. On the Internet, the availability of large image collections, such as Google Images or Flick.com, makes it necessary to develop tools to organize all this information so that it can be managed by humans in practical ways. In this course, we will introduce methods for processing digital images and extracting salient information. Topics will include camera image formation, filtering, edge detection, feature extraction, object recognition, image retrieval, and structure from motion. Practical example applications will be provided throughout the course. Course content Instructor Camera image formation Filtering Edge detection Feature extraction Object recognition Image retrieval Structure from motion TBD Prof. Dr. Davide Scaramuzza English Seite 11 Universität Zürich, Institut für Informatik,

12 1.6 Multilingual Text Analysis and Information Extraction Natural language is the main medium for human communication and for storing information. The internet has resulted in vast amounts of natural language texts in electronic form. The processing and interpretation of these texts for information extraction is of increasing importance. Texts can occur in a multitude of languages which requires automatic language identification, specific processing of various languages, machine translation of queries or texts, and the fusion of information gathered from various language sources. If the same text is available in multiple languages, this provides interesting clues for interpreting words and serves as a central resource for building machine translation systems. In this course we will introduce the methods for processing texts in multiple languages including word class tagging and grammatical analysis. We will present the challenges in analyzing complex words and phrases and in identifying different types of names (persons, geographical entities, organizations). We show how information extracted from texts in various languages can be merged via multilingual ontologies. We introduce the technology of modern machine translation systems and demonstrate how we have employed them profitably in practical applications. Course content Multilingual Text Analysis Name Recognition and Classification Machine Translation Information Extraction Multilingual Ontologies Instructor Word Analysis System Gertwol Dependency Parser ParZu Google Translate Prof. Dr. Martin Volk German Seite 12 Universität Zürich, Institut für Informatik,

13 1.7 Social Computing This course provides an introduction to social computing, a topic at the intersection of computer science and economics. At its core, this course is based on the realization that people are motivate in many different ways (not just by money), to work for projects like Wikipedia or Linux for example. The Internet has enabled many new forms of production that open up new ways to think about how tasks are best being solved in online domains. We begin by studying peer production, which describes decentralized collaborations among individuals that result in successful large-scale projects, without the use of monetary incentives. Next, we introduce human computation, which solves problems that currently cannot be solved by computers alone (e.g., translation, image tagging) by combining small units of human work with an automated algorithm to organize the work. Third, we introduce crowdsourcing, i.e., the act of outsourcing tasks to an undefined, large group of people or community (a crowd) through an open call (e.g., via MTurk). In the second half of the course, we study systems that rely on the wisdom of the crowds idea. We show how to optimally design contests (e.g., between programmers on TopCoder) such that the best solution to a problem is determined via a competition. Next, we study how to optimally design badges that reward contributors on social platforms, such that the number and quality of the contributions is maximized. Finally, we discuss prediction markets, which are markets specifically designed to predict the outcome of certain events (e.g., the launch of a new product). Throughout the course we will leave time for discussions on how these new social computing concepts could be applied in the participants work domains. Course content 1. Peer Production 2. Human Computation Systems 3. Crowdsourcing Markets 4. Contest Design 5. Badge Design 6. Prediction Markets Instructor MTurk Turkit Inkling prediction markets Prof. Abraham Bernstein, Ph.D. Prof. Sven Seuken, Ph.D. English Seite 13 Universität Zürich, Institut für Informatik,

14 1.8 Network-based Business Intelligence Course content Instructors In the current business environment, individuals, organizations, and systems have been increasingly interacting and collaborating in the form of networks through computer-based techniques and technologies. This trend has generated a huge amount of network (relational) data in various business domains such as Finance and Marketing. How to effectively model, analyze, and utilize such network data through computing technologies to support business decision making becomes a major challenge for nowadays business intelligence (BI) practices in large organizations. This course will help you get a solid start at understanding what (business) networks are, what they can do, and how you can develop and employ network-based BI techniques and applications in organizations. Moreover, this course will introduce main research directions and works in Network Science and Business Intelligence. The basic concepts of networks (organizational networks and social networks) Social network analysis methods Network-based BI techniques Networks in Finance and Marketing tba Prof. Daning Hu, Ph.D. English Seite 14 Universität Zürich, Institut für Informatik,

15 1.9 Coaching Day à Vorbereitung auf das Verfassen der schriftlichen Arbeit Seite 15 Universität Zürich, Institut für Informatik,

16 2 Modul Engineering Solutions 2.1 Requirements Engineering 2.0 Requirements Engineering (RE) is a key activity in any software development endeavor. If requirements are wrong, neglected, or misunderstood, the resulting software system will not satisfy its stakeholders expectations and needs, which means that the system will eventually fail, regardless how well it is designed, coded, and tested. Modern Requirements Engineering is no longer about writing voluminous requirements documents. It is about creating a shared vision for a system-to-be and then specifying this vision in a degree of detail that ensures a successful implementation. The course will start by introducing some general principles and facts about Requirements Engineering. Based on these foundations, we will then present practices, techniques and practical advice about how to elicit, document and validate requirements in various contexts and application domains. Course content Successfully capturing, negotiating, and communicating user needs Understanding requirements in context Employing social networks, spontaneous feedback and computer supported cooperative work for capturing and validating requirements Creating requirements for innovative systems Embedding Requirements Engineering both in traditional and in agile project settings Instructor Selected RE tools will be demonstrated Prof. Martin Glinz, Dr. rer. nat. German Seite 16 Universität Zürich, Institut für Informatik,

17 2.2 Software Quality Analysis In the development of today s software systems, in particular for agile development processes, developers need to continuously adapt and evolve the system to reflect the frequently changing stakeholders needs. This need for continuous adaptation of the software requires a high quality of the evolving software throughout the whole development process and developers to frequently assess the software quality. Thereby, the quality of a software system comprises many aspects of a system and its development process and can be measured in various ways. For example, one can meter inner code structures through a coupling and cohesion analysis or one can analyze the development process by looking at the evolution of changes. Given the size and complexity of many of today s software systems, it is also important to provide adequate abstraction and visualization for analyzing a software system. Therefore, we will also introduce software visualization and exploration as means to better understand and evolve large software systems. In this course, we will address software quality analysis techniques, software evolution tools and techniques, as well as software visualization mechanisms and tools that can address the needs of stakeholders such as software developers, software architects, or quality assurance managers. Course content Software evolution analysis (history analysis, hotspot identification, design erosion) System of systems analysis (addressing software ecosystems) Mining software archives (correlations between changes, people, teams, and distribution/outsourcing Bug prediction models and quality Visualization of quality aspects Human effects on quality (empirical findings) Instructors Software evolution tools such as the Evolizer tool suite Change analysis tools such as ChangeDistiller Software Analysis as a Service platform SOFAS Software exploration tools such as CocoViz or Cod Bug pattern identification tools such as Findbugs Prof. Dr. Harald Gall German Seite 17 Universität Zürich, Institut für Informatik,

18 2.3 Software Quality Assurance with Continuous Integration Course content Instructors The customer s high demand of up-to-date, high quality software puts a challenging task to the software industry. We need to be able to deliver a new release at any moment, thus always have a running-system at hand. To cope with these requirements we have to enrich our software development process with the necessary measures and tools. We need a thorough test-base for quality assurance and very short feedback cycles for developers during their daily work. Software testing on different levels of granularity and continuous integration with multi-stage builds provides the means to ensure high quality and release-readiness of software products at the same time. Moreover, we can leverage the tools provided to configure continuous integration for different software development processes. In this course, we address software quality assurance by means of software testing on different levels of granularity (unit, integration, system, end-to-end, and load) in conjunction with multi-stage builds in a continuous integration environment. We show how we can use continuous integration for releasingsoftware in different development processes and how we can get towards continuous delivery. In addition, we present the challenges of continuous integration with respect to infrastructure and acceptance among different stakeholders. The course uses state-of-the-art open-source tools and drives the content with examples. The target audience of the course includes people in technical positions, e.g., software engineering professionals, software architects/technical project leaders, and quality assurance engineers. Software testing on different levels of granularity (unit, integration, system, end-to-end, and load) Multi-stage builds with continuous integration Deployment/delivery with continuous integration Quality analysis with continuous integration Continuous integration people and infrastructure challenges Continuous integration platform Jenkins Common development, build, and test infrastructre: Java EE, JUnit, Maven Integration testing with Arquillian Web testing framework Selenium/WebDriver Load testing platform JMeter Quality analysis platform Sonar Dr. Beat Fluri German Seite 18 Universität Zürich, Institut für Informatik,

19 2.4 Development 2.0: Human-Centered Software Engineering / Open Innovation for Software Products & Processes Development 2.0: Human-Centered Software Engineering Software is built by humans. Current development environments are centered around models of the artifacts used in development, rather than of the people who perform the work, making it difficult and sometimes infeasible for the developer to satisfy her information needs. For instance, a central model in a development environment is an abstract syntax tree, which provides an abstraction of source code to facilitate feedback about the syntax of code being written, facilitate navigation in terms of code structure and facilitate code transformations. By providing feedback about artifacts, these models benefit the developers using the environment. However, these models fall short for developers by supporting only a small fraction of their information needs and tasks when they are performing work on the system. This course will introduce relevant current approaches and technologies to support a better human-centric software development and to satisfy information needs of software engineers. Besides introducing the techniques it will also engage in a number of smaller practical and interactive exercises. Course content Information needs and fragments Information filtering and tailoring (to stakeholder needs and tasks) New interfaces and devices for the software development process (e.g., interactive touch displays, tabletops); Collaborative software development and awareness Recommender systems Social networks in software development (tentative list) Instructors Collaboration tools such as Mylyn New IDEs such as Code Bubbles and Debugger Canvas Development support tools such as Crystal NUI tools such as SmellTagger Prof. Thomas Fritz, PhD German Seite 19 Universität Zürich, Institut für Informatik,

20 Open Innovation for Software Products & Processes Open Innovation (OI) in software engineering often is seen in the context of open source software projects. But both open and closed source projects offer a great deal of expertise, experience, and knowledge to enable technological leadership for future software innovations. One of the basic questions for OI is the level of quality of a software product that should be improved, enhanced, refurbished or even freshly developed. This question is essential as the product functions as a key enabler for an idea to become an innovation: is the point of time right, is the quality level appropriate, are the feature sets defining a good foundation for the next steps towards customers? We will center the course around the definition of OSS Watch: Processes and tools that encourage innovation by boosting internal, and harnessing external, creativity, and by bringing the innovation results to market through both internal and external channels. (OSS Watch, 2011) Course content Open Innovation Processes (Intrapreneurship); Open development vs. closed development; OI platforms and models for enabling innovation incubation; OI and software quality aspects; OI for software design and development Instructor OI platforms (socio-technical platforms) Software analysis platforms Expertise finder tools Social media for OI Prof. Dr. Harald Gall German Seite 20 Universität Zürich, Institut für Informatik,

21 2.5 Human Computer Interaction Advances in technology have a profound impact on human activities, shaping the way that we work, communicate, and use information. At the same time, technology must address and respond to human needs and capabilities in order to be successful. As evidenced by the way that the modern web browser suddenly made the internet accessible to the general population after years of niche usage, or by the way the Apple ipad and similar recent devices have made the decades-old tablet computing paradigm commonplace, advanced technology needs to fit with human practices to unlock its real value and achieve broad adoption. The right interface or design can be critical for transforming technology from obscure science to supportive, useful, and effective tools. At the same time, identifying or creating these necessary innovations is a challenge. The field of Human-Computer Interaction focuses on the co-evolution of technology and human practices. It focuses on understanding human needs and activities, and designing technologies that are informed by this understanding. It offers methods and approaches for addressing human questions, translating findings into technology design, and evaluating technologies for both usability and usefulness. This course will provide participants with a valuable and versatile toolbox of methods and skills essential for user research and design. In particular, it will focus on the iterative design process as a practical and costefficient way of developing solutions, as well as simple but effective lowresource techniques for collecting user data and evaluating technology solutions. Course content The HCI iterative design process Usability criteria and principles and their application to design Practical techniques for gathering and analyzing user data Graduated prototyping techniques Low-resource methods for evaluating solutions with end users Instructors Balsamiq Prof. Elaine M. Huang, Ph.D. English Seite 21 Universität Zürich, Institut für Informatik,

22 2.6 Engineering Electronic Markets Due to the Internet, electronic markets are becoming the most important means of trading goods and services in today s society. Millions of items change hands via ebay every day; Billions of online ads are priced automatically on Google every day; lots of companies are using automated markets to automatically source goods and services worth Billions of dollars every year. However, as the dot-com bubble and the recent financial crisis have shown, markets can also be broken and, in the worst case, may even collapse. This illustrates that special care must be taken when setting up a new electronic market, or when trying to fix a broken one. In this course, we discuss how to design (or engineer ) electronic markets, under economic and computational considerations. The first half of the course covers a variety of electronic auctions. We first discuss how to design auctions that can be used to sell individual items to customers, or to automatically price advertisements on the Internet. Then we move on to more complex, so-called combinatorial auctions, and show how they can be used for sourcing a large company s supplies, for optimizing server architectures, or for selling computational resources like bandwidth. The second half of the course takes a broader view on electronic markets, covering a number of common challenges that arise when starting a new online business and ways to solve them. First, we adopt the perspective of a business manager who wants to start or continue to grow an online business in a domain with network effects. Then we discuss how to identify good and bad users in a market domain, how to identify good and bad products, and how to recommend products or services to users they will like based on their subjective taste. Finally, we wrap up the course with a case study on how to monetize a social network. The case study is designed to illustrate many of the concepts covered in this course. Throughout the course, we use many real-world examples to illustrate the new concepts, and we devote enough time for extended discussions on the pros and cons of various approaches. Course content 1. Introduction to Game Theory and Auctions 2. Internet Auctions (ebay, Google, etc.) 3. Combinatorial Auctions 4. Electronic Markets with Network Effects 5. Reputation and Recommender Systems (Tripadvisor, Amazon, etc.) 6. Case Study: Monetizing a Social Network (e.g., Facebook, Twitter) Instructors No software or computer necessary New concepts will be practiced via case studies Prof. Dr. Sven Seuken English Seite 22 Universität Zürich, Institut für Informatik,

23 2.7 Digital Innovation and Social Networking for Companies Course content Instructors Over the past years, the increasing use of information and communication technologies has brought about a significant transformation of organizations. Alongside other technologies, social software, such as wikis, weblogs, and social network sites, have become a focal area for many organizations. The aim of the course is to give an overview over the potential of enterprise social software (ESS) and to provide you with methods to plan, implement and evaluate the success of ESS. We will discuss the underlying principles (e.g. transparency, malleability) and use cases (e.g. expert finding, knowledge exchange) of ESS as well as (firsthand account) best practices and motives of companies using ESS. You will learn how ESS contribute to changes in leadership and organizational culture in companies and discuss the challenges on the way to a networked organization. tba tba Dr. Alexander Richter German Seite 23 Universität Zürich, Institut für Informatik,

24 2.8 Sustainability and Green IT Course content Instructors Green IT is the study and practice of environmentally sustainable computing throughout the whole life cycle of IT systems. IT is responsible for roughly 10 % of our electricity demand and for 2-3 % of total energy consumption. Although it is the hardware that consumes energy and creates demand for scarce metals and other important resources, software products have decisive impact on how much hardware is installed, how it is utilized, how it can adapt to changing conditions (including the fluctuating availability of renewable energy) and when a device becomes obsolete. There have been many attempts to define the sustainability of IT systems from a software perspective and to derive recommendations for developers from conceptual frameworks, and to develop tools to measure indicators of sustainability, e.g., of Web applications. The course will provide an overview of the existing approaches to Green IT with a focus on software and the causal chains leading from software architecture to physical effects, such as power consumption of hardware, network traffic and its energy intensity, storage and energy intensity, and other sustainability-related issues. General overview and conceptual framework of Green IT Green IT in datacenters (state of the art) Green IT and communications (energy intensity of the Internet and different access networks) Green software engineering (software architecture and energy, introduction to the GREENSOFT model) Energy-aware systems Standardization (current state of Green IT standardization activities by ITU, ISO and Greenhouse Gas Protocol) none Prof. Dr. Lorenz Hilty German Seite 24 Universität Zürich, Institut für Informatik,

25 2.9 Coaching Day à Vorbereitung auf das Verfassen der schriftlichen Arbeit Seite 25 Universität Zürich, Institut für Informatik,

26 Institut für Informatik C Dozierende Abraham Bernstein Spezialgebiete: Abraham Bernsteins Forschungsinteressen beinhalten das Semantische Web, Data-Mining, heterogene Datenintegration, verteilte Systeme, Verarbeitung von Graph-Daten sowie Datenströmen und das Wechselspiel zwischen sozialen und technischen Elementen der Informatik. Aktivitäten: Mitglied des Editorial Board mehrerer internationaler Fachzeitschriften. Mitglied verschiedener Führungsgremien und wissenschaftlicher Vereinigungen. Leiter mehrerer internationaler Forschungsprojekte. Vorstandsmitglied der Schweizer Informatik Gesellschaft (SI) und des ICTSwitzerland. Ordentlicher Professor für Informatik, Institut für Informatik, Universität Zürich Beat Fluri Senior Technical Project Leader, AdNovum Informatik AG Seite 26 Website: Dr. Beat Fluri hat an der ETH Zürich Informatik-Ing. studiert (Abschluss 2004) und bei Prof. Harald Gall an der Universität Zürich im Bereich Software Evolutions Analyse promoviert (Abschluss 2008). Nach der Dissertation arbeitete er ein Jahr als Senior Research Associate an der Universität Zürich. Von 2009 bis 2011 gründete und entwickelte er zusammen mit Freunden die unter TischtennisSpielern bekannte Sportplattform spood.me. Von Okt 2011 bis Mai 2013 war er Software Architekt und technischer Projektleiter bei der Entwicklung von Web- und Enterprise Java Applikationen bei der Comerge AG. In der Zeit von 2009 bis 2011 hielt er als externer Dozent die Master-Vorlesung "Software Testing" an der Universität Zürich. Seit Juni 2013 ist Beat Fluri als Senior Technical Project Leader bei der AdNovum Informatik AG tätig. Sein Fokus liegt in der server-seitigen Entwicklung von WebApplikationen mit JEE 6. Als Technischer Leiter von solchen anspruchsvollen, agilen Software-Projekten ist er verantwortlich, dass Kunden qualitativ hochstehende Produkte erhalten. Dazu setzt er in seinen Projekten Test-Driven Development, Continuous Integration, Continuous Deployment, Clean Code und TestAutomatisierung ein. Universität Zürich, Institut für Informatik,

27 Institut für Informatik Thomas Fritz Assistenzprofessor für Software Quality, Institut für Informatik, Universität Zürich Thomas Fritz ist Assistenzprofessor am Institut für Informatik der Universität Zürich. Er erhielt seinen PhD von der University of British Columbia, Kanada, in 2011 und sein Diplom von der Ludwig-Maximilians-Universität München, Deutschland, in Er hat Erfahrung mit verschiedensten Firmen und Forschungsgruppen gesammelt, so wie IBM in Ottawa und Zürich und die OBASCO Gruppe an der École des Mines de Nantes in Frankreich. Aktivitäten: In seiner Forschung interessiert er sich dafür, Stakeholdern im Softwareentwicklungsprozess zu helfen besser mit der Information und den Systemen, an denen sie arbeiten, umzugehen. Heutige Ansätze konzentrieren sich stark auf die Artifakte im Softwareentwicklungsprozess anstatt der Menschen, die diese Artifakte erstellen. Dies macht es oft schwer, die Informationsbedürfnisse der Stakeholder zu befriedigen. Thomas Fritz erforscht, wie man Modelle spezifisch für die Stakeholder erstellen kann um heutige Ansätze zu erweitern und die Informationsbedürfnisse der Stakeholder zu adressieren. Website: seal.ifi.uzh.ch/fritz Harald Gall Ordentlicher Professor für Software Engineering, Institut für Informatik, Universität Zürich Seite 27 Spezialgebiete: Harald Gall forscht und lehrt im Gebiet der Software Evolutionsanalyse und dem Mining von SoftwareArchiven. Seit mehr als zehn Jahren erarbeitet seine Forschergruppe neue Modelle, Techniken und Werkzeuge zur Untersuchung von Software-Archiven für die bessere Unterstützung des SoftwareEntwicklungsprozesses. Die Arbeitsbereiche beinhalten Software Qualitätsanalyse, Software Visualisierung, Software Architektur, Kollaborative Software Entwicklung sowie Service-zentrierte Software Systeme. Aktivitäten: Mitglied des Editorial Board mehrerer internationaler Fachzeitschriften sowie Mitglied verschiedener Führungsgremien und wissenschaftlicher Vereinigungen. Website: seal.ifi.uzh.ch/gall Universität Zürich, Institut für Informatik,

28 Institut für Informatik Martin Glinz Spezialgebiete: Requirements Engineering, Software Engineering, Software Qualität, Modellierung von Systemen. Aktivitäten: Direktor des Instituts für Informatik der Universität Zürich. Mitglied des Editorial Board mehrerer internationaler Fachzeitschriften sowie des Steering Committees internationaler Konferenzen. General Chair und Program Chair erstklassiger internationaler Konferenzen. Mitglied des International Requirements Engineering Board. Mitglied in den Führungsgremien verschiedener wissenschaftlicher Vereinigungen. Dr. rer. nat., Ordentlicher Professor für Informatik, Institut für Informatik, Universität Zürich Website: Daning Hu Specialties : Daning Hu s research interests include Business Intelligence, Network Analysis, Social Media, Web and Enterprise 2.0, data mining, financial intelligence Activities: Member of Association for Information Systems and INFORMS Website: Assistenzprofessor für Information Systems, Institut für Informatik, Universität Zürich Seite 28 Universität Zürich, Institut für Informatik,

29 Elaine M. Huang Specialties: Elaine M. Huang conducts research on Human- Computer Interaction (HCI), Computer-Supported Cooperative Work (CSCW), and Ubiquitous Computing. Her current interests include systems to support environmentally sustainable behavior, smart home technologies influenced by domestic routines, pervasive and multi-display environments, and tangible interfaces to support group work. Ausserordentliche Professorin für Human- Computer Interaction, Institut für Informatik Universität Zürich Activities: Member of organizing committees and program committees for several top-tier HCI and Ubiquitous Computing conferences. Forum editor for ACM Interactions magazine. Involved in activities geared the promotion and engagement of women in Computer Science. Actively involved in events aimed at raising the profile of HCI within Switzerland. Website: Lorenz Hilty Ausserordentlicher Professor für Informatics and Sustainability, Institut für Informatik, Universität Zürich Lorenz Hilty ist Professor am Institut für Informatik der Universität Zürich, Leiter der Forschungsgruppe Informatik und Nachhaltigkeit an der Eidgenössischen Materialprüfungs- und Forschungsanstalt Empa und Affiliate Professor am Royal Institute of Technology KTH in Stockholm. Er habilitierte sich an der Universität Hamburg, Deutschland, im Jahr 1997, und verfügt über langjährige Forschungserfahrung im Überschneidungsbereich von Informatik und Umweltforschung, unter anderem am Institut für Wirtschaft und Ökologie IWÖ der Universität St. Gallen, an der Abteilung Umweltinformationssysteme des Forschungsinstituts für Anwendungsorientierte Wissensverarbeitung FAW in Ulm und als Leiter der Abteilung Technologie und Gesellschaft der Empa. Aktivitäten: In seiner Forschung arbeitet Lorenz Hilty daran, die Möglichkeiten der Informatik für die Analyse und Lösung von Umweltproblemen, für die Einsparung von Energie und Material und zur Förderung einer insgesamt nachhaltigen Entwicklung einzusetzen. Seit einigen Jahren ist der Energie- und Materialverbrauch durch die IT selbst zu einem relevanten Thema geworden, so dass ein wachsender Teil der Projekte dem Bereich Green IT zuzuordnen sind. Hier stehen methodische Fragen der Messung von Material- und Energieverbrauchs über den gesamten Lebenszyklus von IT-Produkten und der Energiebedarf des Datenverkehrs im Internet und seinen Zugangsnetzen im Vordergrund. Website: Seite 29 Universität Zürich, Institut für Informatik,

30 Institut für Informatik Renato Pajarola Ordentlicher Professor für Informatik, Institut für Informatik, Universität Zürich Alexander Richter Oberassistent, Forschungsgruppe Innovation & Social Networking, Institut für Informatik, Universität Zürich Spezialgebiete: Renato Pajarolas Forschungsinteressen decken vor allem die Gebiete der interaktiven 3D Computergrafik (z.b. 3D Datenvisualisierung, Virtual Reality, 3D Games) und der wissenschaftlichen Visualisierung ab (z.b. Geo-Visualisierung, Biomedizinische Bildgebung). Im Fokus stehen vor allem die effiziente Verarbeitung und Darstellung von grossen mehrdimensionalen Daten mittels schneller Algorithmen, Datenstrukturen und verteilter paralleler Prozesse. Aktivitäten: Mitglied von Editorial Boards internationaler Forschungszeitschriften, mehreren wissenschaftlichen Komitees und Vereinigungen. Leiter mehrerer internationaler und nationaler Forschungsprojekte. Website: vmml.ifi.uzh.ch Dr. Alexander Richter ist seit Oberassistent am Institut für Informatik der Universität Zürich. Im Rahmen seiner Dissertation beschäftigte er sich von 2007 bis 2009 mit den Herausforderungen des Einsatzes von Social Networking Services im Unternehmenskontext. Danach war er mehrere Jahre lang als Bereichsleiter Social Business in der Forschungsgruppe Kooperationssysteme an der Universität der Bundeswehr München tätig. In seiner Rolle unterstützte er Unternehmen wie Allianz, Deutsche Bahn, Bayer, Bosch, Capgemini, EADS, Schott oder Siemens bei der Auswahl, Einführung und Erfolgsmessung von Social Software. Aktivitäten: Im Rahmen seiner Forschung untersucht er wie Social Software die Zusammenarbeit und das Wissensmanagement in einem Unternehmen unterstützen kann. Gleichzeitig möchte er dazu beitragen den Wandel zu einer vernetzten Organisation, der sich aktuell in vielen Unternehmen vollzieht, greifbar zu machen und Wege aufzeigen damit umzugehen. Website: Seite 30 Universität Zürich, Institut für Informatik,

31 Davide Scaramuzza Assistant Professor for Robotics, Department of Informatics, University of Zurich Davide Scaramuzza is Professor of Robotics at the University of Zurich. He is the founder and director of the Robotics and Perception Group ( He received his PhD (2008) in Robotics and Computer Vision at ETH Zurich. He was Postdoc at both ETH Zurich and the University of Pennsylvania, where he worked on autonomous navigation of micro aerial vehicles. From 2009 to 2012, he led the European project SFLY, which focused on autonomous navigation of micro helicopters in GPS-denied environments using vision as the main sensor modality. For his research, he was awarded the Robotdalen Scientific Awards (2009) and the European Young Researcher Award (2012), sponsored by the IEEE and the European Commission. He is coauthor of the 2nd edition of the book Introduction to Autonomous Mobile Robots (MIT Press). He is also author of the first open-source Omnidirectional Camera Calibration Toolbox for MATLAB, which, besides thousands of downloads worldwide, is also currently used at NASA, Philips, Bosch, and Daimler. Since 2009, he has been consultant for the company Dacuda, a startup from ETH Zurich, inventor of the world's first scanner mouse, currently sold by LG. This mouse uses robot SLAM technology to scan documents in real time. Finally, he is author of several topranked robotics and computer vision journals. Aktivitäten: His research interests are field and service robotics, intelligent vehicles, and computer vision. Specifically, he investigates the use of cameras as the main sensors for robot navigation, mapping, exploration, reasoning, and interpretation. His interests encompass both ground and flying vehicles. Website: Seite 31 Universität Zürich, Institut für Informatik,