Digital Commons Network^™

Two-View Transductive Support Vector Machines, Guangxia Li, Steven C. H. Hoi, Kuiyu Chang

Research Collection School Of Computing and Information Systems

Obtaining high-quality and up-to-date labeled data can be difficult in many real-world machine learning applications, especially for Internet classification tasks like review spam detection, which changes at a very brisk pace. For some problems, there may exist multiple perspectives, so called views, of each data sample. For example, in text classification, the typical view contains a large number of raw content features such as term frequency, while a second view may contain a small but highly-informative number of domain specific features. We thus propose a novel two-view transductive SVM that takes advantage of both the abundant amount of unlabeled data …

Perception Of The Risks And Benefits Of Bt Eggplant By Indian Farmers, Mark Chong

Research Collection Lee Kong Chian School Of Business

Several researchers most notably Lennart Sjoberg and his colleagues have proposed that the moral aspects of risk provide a better explanation of risk perception than the psychometric paradigm or Cultural Theory, neither of which accounts for moral concerns. This study is possibly the first to assess empirically the perception of the risks and benefits of a transgenic food crop transgenic Bt ( Bacillus thuringiensis ) eggplant by farmers in a developing country such as India. It also aims to assess if the moral aspects of risk figure in Indian farmers' perception of Bt eggplant and if economic benefits outweigh perceived …

Exclusive Lasso For Multi-Task Feature Selection, Yang Zhou, Rong Jin, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

We propose a novel group regularization which we call exclusive lasso. Unlike the group lasso regularizer that assumes co-varying variables in groups, the proposed exclusive lasso regularizer models the scenario when variables in the same group compete with each other. Analysis is presented to illustrate the properties of the proposed regularizer. We present a framework of kernel-based multi-task feature selection algorithm based on the proposed exclusive lasso regularizer. An efficient algorithm is derived to solve the related optimization problem. Experiments with document categorization show that our approach outperforms state-of-the-art algorithms for multi-task feature selection.

Applying Knowledge Management In University Research, Benjamin Loh, Ai-Chee Tang, Thomas Menkhoff, Hans-Dieter Evers

Research Collection Lee Kong Chian School Of Business

In the knowledge-based economy, universities are encountering dramatic changes. Their missions and functions are ‘pragmatized’ because of emerging new players and competing markets for knowledge production, the availability of higher education to a wider range of social classes and age groups, as well as the assimilation of information technology into the university environment. The dynamics and conduct of university research, in particular, has correspondingly become more sensitive to industry collaboration opportunities, commercial exploitation, and is increasingly transdisciplinary. This paper argues that knowledge management (KM) practices and tools can support universities in addressing these demands. Institutions of higher education can benefit …

Portfolio Selection Under Distributional Uncertainty: A Relative Robust Cvar In Portfolio Management, Dashan Huang, Shushang Zhu, Frank Fabozzi, Masao Fukushima

Research Collection Lee Kong Chian School Of Business

Robust optimization, one of the most popular topics in the field of optimization and control since the late 1990s, deals with an optimization problem involving uncertain parameters. In this paper, we consider the relative robust conditional value-at-risk portfolio selection problem where the underlying probability distribution of portfolio return is only known to belong to a certain set. Our approach not only takes into account the worst-case scenarios of the uncertain distribution, but also pays attention to the best possible decision with respect to each realization of the distribution. We also illustrate how to construct a robust portfolio with multiple experts …

Open Innovation In Platform Competition, Mei Lin

Research Collection School Of Computing and Information Systems

We examine the competition between a proprietary platform and an open platform,where each platform holds a two-sided market consisted of app developers and users.The open platform cultivates an innovative environment by inviting public efforts todevelop the platform itself and permitting distribution of apps outside of its own appmarket; the proprietary platform restricts apps sales solely within its app market. Weuse a game theoretic model to capture this competitive phenomenon and analyze theimpact of growth of the open source community on the platform competition. We foundthat growth of the open community mitigates the platform rivalry, and balances the developernetwork sizes on …

A Discriminative Model Approach For Accurate Duplicate Bug Report Retrieval, Chengnian Sun, David Lo, Xiaoyin Wang, Siau-Cheng Khoo

Research Collection School Of Computing and Information Systems

Bug repositories are usually maintained in software projects. Testers or users submit bug reports to identify various issues with systems. Sometimes two or more bug reports correspond to the same defect. To address the problem with duplicate bug reports, a person called a triager needs to manually label these bug reports as duplicates, and link them to their "master" reports for subsequent maintenance work. However, in practice there are considerable duplicate bug reports sent daily; requesting triagers to manually label these bugs could be highly time consuming. To address this issue, recently, several techniques have be proposed using various similarity …

Point-Based Backup For Decentralized Pompds: Complexity And New Algorithms, Akshat Kumar, Shlomo Zilberstein

Research Collection School Of Computing and Information Systems

Decentralized POMDPs provide an expressive framework for sequential multi-agent decision making. Despite their high complexity, there has been significant progress in scaling up existing algorithms, largely due to the use of point-based methods. Performing point-based backup is a fundamental operation in state-of-the-art algorithms. We show that even a single backup step in the multi-agent setting is NP-Complete. Despite this negative worst-case result, we present an efficient and scalable optimal algorithm as well as a principled approximation scheme. The optimal algorithm exploits recent advances in the weighted CSP literature to overcome the complexity of the backup operation. The polytime approximation scheme …

Creative Industries Debate: Unfinished Business: Labour, Management, And The Creative Industries, Stefano Harney

Research Collection Lee Kong Chian School Of Business

In what follows I am going to argue that the rise of the creative industries has in general been understood too narrowly. This narrow understanding has had implications for the way that a politics of management and labour in the creative industries has been framed and contained, and it has held back an analysis of class struggle in the creative industries. To elaborate an understanding of labour in the creative industries I am going to revisit some insights related to the development of British cultural studies, and try to link these insights to what Stuart Hall calls the conditions of …

Learning User Profiles For Personalized Information Dissemination, Ah-Hwee Tan, Christine Teo

Research Collection School Of Computing and Information Systems

Personalized information systems represent the recent effort of delivering information to users more effectively in the modern electronic age. This paper illustrates how a supervised Adaptive Resonance Theory (ART) system, known as fuzzy ARAM, can be used to learn user profiles for personalized information dissemination. ARAM learning is on-line, fast, and incremental. Acquisition of new knowledge does not require re-training on previously learned cases. ARAM integrates both user-defined and system-learned knowledge in a single framework. Therefore inconsistency between the two knowledge sources will not arise. ARAM has been used to develop a personalized news system known as PIN. Preliminary experiments …

Awareness 2.0: Staying Aware Of Projects, Developers And Tasks Using Dashboards And Feeds, Christoph Treude, Margaret-Anne Storey

Research Collection School Of Computing and Information Systems

Software development teams need to maintain awareness of various different aspects ranging from overall project status and process bottlenecks to current tasks and incoming artifacts. Currently, there is a lack of theoretical foundations to guide tool selection and tool design to best support awareness tasks. In this paper, we explore how the combination of highly configurable project, team and contributor dashboards along with individual event feeds is used to accomplish extensive awareness. Our results stem from an empirical study of several large development teams, with a detailed study of a team of 150 developers and additional data from another four …

Bridging Lightweight And Heavyweight Task Organization: The Role Of Tags In Adopting New Task Categories, Christoph Treude, Margaret-Anne Storey

Research Collection School Of Computing and Information Systems

In collaborative software development projects, tasks are often used as a mechanism to coordinate and track shared development work. Modern development environments provide explicit support for task management where tasks are typically organized and managed through predefined categories. Although there have been many studies that analyze data available from task management systems, there has been relatively little work on the design of task management tools. In this paper we explore how tagging with freely assigned keywords provides developers with a lightweight mechanism to further categorize and annotate development tasks. We investigate how tags that are frequently used over a long …

The Role Of Emergent Knowledge Structures In Collaborative Software Development, Christoph Treude

Research Collection School Of Computing and Information Systems

Many collaboration features in software development tools draw on lightweight technologies such as tagging and wikis. We propose to study the role of emergent knowledge structures created through these features. Using a mixed-methods approach, we investigate which processes emergent knowledge structures support and how tool support can leverage them.

Web2se: First Workshop On Web 2.0 For Software Engineering, Christoph Treude, Margaret-Anne Storey, Kate Ehrlich, Arie Van Deursen

Research Collection School Of Computing and Information Systems

Social software is built around an "architecture of participation" where user data is aggregated as a side-effect of using Web 2.0 applications. Web 2.0 implies that processes and tools are socially open, and that content can be used in several different contexts. Web 2.0 tools and technologies support interactive information sharing, data interoperability and user centered design. For instance, wikis, blogs, tags and feeds help us organize, manage and categorize content in an informal and collaborative way. One goal of this workshop is to investigate how these technologies can improve software development practices. Some of these technologies have made their …

Towards Finding Robust Execution Strategies For Rcpsp/Max With Durational Uncertainty, Na Fu, Pradeep Varakantham, Hoong Chuin Lau

Research Collection School Of Computing and Information Systems

Resource Constrained Project Scheduling Problems with minimum and maximum time lags (RCPSP/max) have been studied extensively in the literature. However, the more realistic RCPSP/max problems — ones where durations of activities are not known with certainty – have received scant interest and hence are the main focus of the paper. Towards addressing the significant computational complexity involved in tackling RCPSP/max with durational uncertainty, we employ a local search mechanism to generate robust schedules. In this regard, we make two key contributions: (a) Introducing and studying the key properties of a new decision rule to specify start times of activities with …

Oral History Interview With Janice Bellace: Conceptualising Smu, Janice Bellace

Oral History Collection

The interview with Janet Bellace covered: first involvement with SMU, collaboration with Wharton School, roles and responsibilities, first building, risks and opportunities, founding team, faculty recruitment, board of trustees, government relationship, groundbreaking for city campus, convocation, student interaction, student feedback, graduate placement, milestones, future of SMU.

Biography:

Board of Trustees, SMU, 2001–present
Founding President, SMU, 1999–2001

“Thank you, Janice!” read the front page tribute article of the October 2001 smuhub newsletter. It went on to describe some of Professor Janice Bellace’s contributions as consultant and then founding president of SMU. The article noted that when Janice arrived as president, “…there …

Watch The Vote: Avoiding Undesirable Zero-Sum Outcomes, Knowledge@Smu

Knowledge@SMU

Luxury watches often invoke a lot of debate amongst its aficionados. After the great lengths to which people go to, to collect, study, appraise and critique these mobile mechanical clocks, it is no surprise that they feel very strongly about their picks. Decisions made by prestigious bodies, such as the Grand Prix d’Horlogerie, are hence quite delicate. People's money, ego, and pride are at stake. So picking the winner that will keep the majority satisfied is crucial, yet quite difficult to do. Drawing comparisons to political election outcomes, SMU Economics professor Massimiliano Landi explores a new voting system to address …

To Invent Future Leaders, General Electric Redefines Tough Love, Knowledge@Smu

Knowledge@SMU

It has been almost a decade since Jack Welch retired as GE’s chairman and CEO, but the legacy and methods of one of the biggest names in corporate America is still widely talked about, even revered. During his two-decade tenure, the conglomerate grew its profits by nearly 30 times – an astounding feat for an old, sizeable multinational corporation. However, times have changed and so have employee attitudes. Will Welch's confrontational, no-nonsense, outcome-oriented methods, that have become so synonymous with GE, go down well its new generations of workers?

What Are Mobile Friendly Library Sites Offering? A Survey, Chee Hsien Aaron (Zheng Zhixian) Tay

Research Collection Library

Everyone knows mobile is hot now, and many libraries are scrambling to add mobile friendly pages. Currently themost complete listing of libraries with mobile sites can be found at the Library Success Wiki site , which lists over40 such sites.I thought it would be helpful to look at these 40+ sites to see what are the common services that have beingoffered to mobile users. To that end, I have created a google doc of the 40 sites comparing their common features.I grouped each feature below according to whether it exists in a mobile site and then choose representativeexamples with images …

What Are Mobile Friendly Library Sites Offering? A Survey, Chee Hsien Aaron (Zheng Zhixian) Tay

Research Collection Library

Everyone knows mobile is hot now, and many libraries are scrambling to add mobile friendly pages. Currently themost complete listing of libraries with mobile sites can be found at the Library Success Wiki site , which lists over40 such sites.I thought it would be helpful to look at these 40+ sites to see what are the common services that have beingoffered to mobile users. To that end, I have created a google doc of the 40 sites comparing their common features.I grouped each feature below according to whether it exists in a mobile site and then choose representativeexamples with images …

Changing Our Service Interface To Support Learning: Information Literacy In The Virtual Environment At The City University Of Hong Kong Library, Bethany Wilkes, Hsianghoo Steve Ching, Shun-Wa Lee, Joanna Pong, Anna Ryan

Research Collection Library

No abstract provided.

From Post-Its To Face-Bras: How 3m Puts Innovation Into Practice, Knowledge@Smu

Knowledge@SMU

Most people are familiar with the genesis of the Post-it, 3M's signature yellow sticky pad: an accidental outcome that followed the discovery of a non-permanent adhesive. While innovations as such might be viewed as a fluke – an invention that occurred by chance – it is helpful to note that certain environments and conditions can render an organisation more 'accident'-prone. Geoff Nicholson, the "father" of 3M's line of Post-it Note products and an advocate of "risky business" behaviours, shared his insights and observations at SMU recently.

Playing With Recognizers: A Call For An Extensible Editor, Richard Christopher Davis

Research Collection School of Computing and Information Systems

The sketch recognition interface community has not produced a ―killer‖ application, because access to sketch recognition technology has been too restricted. If recognition technologies were more freely available for experimentation, powerful new applications would evolve. This paper proposes a rough architecture for an extensible graphical editor that facilitates collaboration between recognition technology developers, user interface designers, and early adopters of sketch recognition interfaces. Only by serving all three communities will we reach the critical mass necessary for killer applications to emerge.

Finding Influentials Based On The Temporal Order Of Information Adoption In Twitter, Changhyun Lee, Haewoon Kwak, Hosung Park, Sue Moon

Research Collection School Of Computing and Information Systems

Twitter offers an explicit mechanism to facilitate information diffusion and has emerged as a new medium for communication. Many approaches to find influentials have been proposed, but they do not consider the temporal order of information adoption. In this work, we propose a novel method to find influentials by considering both the link structure and the temporal order of information adoption in Twitter. Our method finds distinct influentials who are not discovered by other methods.

Cold War Origins Of The International Federation For Information Processing, Ksenia Tatarchenko

Research Collection College of Integrative Studies

The International Federation for Information Processing (IFIP) was born as a nongovernmental federation with the main goal of bringing together computer professionals from countries in the East and West. This article examines the Cold War context of the IFIP's origins and the mechanisms its founders used to reconcile computing and politics and to construct computing as an international discipline.

Mashup Environments In Software Engineering, Lars Grammel, Christoph Treude, Margaret-Anne Storey

Research Collection School Of Computing and Information Systems

Too often, software engineering (SE) tool research is focused on creating small, stand-alone tools that address rarely understood developer needs. We believe that research should instead provide developers with flexible environments and interoperable tools, and then study how developers appropriate and tailor these tools in practice. Although there has been some prior work on this, we feel that flexible tool environments for SE have not yet been fully explored. In particular, we propose adopting the Web 2.0 idea of mashups and mashup environments to support SE practitioners in analytic activities involving multiple information sources.

Efficient Skyline Maintenance For Streaming Data With Partially-Ordered Domains, Yuan Fang, Chee-Yong Chan

Research Collection School Of Computing and Information Systems

We address the problem of skyline query processing for a count-based window of continuous streaming data that involves both totally- and partially-ordered attribute domains. In this problem, a fixed-size buffer of the N most recent tuples is dynamically maintained and the key challenge is how to efficiently maintain the skyline of the sliding window of N tuples as new tuples arrive and old tuples expire. We identify the limitations of the state-of-the-art approach STARS, and propose two new approaches, STARS+ and SkyGrid, to address its drawbacks. STARS+ is an enhancement of STARS with three new optimization techniques, while SkyGrid is …

Policy-Driven Distributed And Collaborative Demand Response In Multi-Domain Commercial Buildings, Archan Misra, Henning Schulzrinne

Research Collection School Of Computing and Information Systems

Enabling a sophisticated Demand Response (DR) framework, whereby individual consumers adapt their electricity consumption in response to price variations, is a major objective of the emerging Smart Grid. We first point out why the current model, of EMS-based centralized control of a static repository of high load appliances, is inappropriate for supporting DR in future commercial buildings and campuses, where the consuming appliances are controlled by multiple users. To enable DR in such multi-domain environments, we envision a more collaborative and autonomous model, where a large set of heterogeneous smart electrical devices autonomously self-organize and negotiate their collective DR. Enabling …

A Verification System For Interval-Based Specification Languages, Chunqing Chen, Jin Song Dong, Jun Sun, Andrew P. Martin

Research Collection School Of Computing and Information Systems

Interval-based specification languages have been used to formally model and rigorously reason about real-time computing systems. This usually involves logical reasoning and mathematical computation with respect to continuous or discrete time. When these systems are complex, analyzing their models by hand becomes error-prone and difficult. In this article, we develop a verification system to facilitate the formal analysis of interval-based specification languages with machine-assisted proof support. The verification system is developed using a generic theorem prover, Prototype Verification System (PVS). Our system elaborately encodes a highly expressive set-based notation, Timed Interval Calculus (TIC), and can rigorously carry out the verification …

Adaptive Ensemble Classification In P2p Networks, Hock Hee Ang, Vivekanand Gopalkrishnan, Steven C. H. Hoi, Wee Keong Ng

Research Collection School Of Computing and Information Systems

Classification in P2P networks has become an important research problem in data mining due to the popularity of P2P computing environments. This is still an open difficult research problem due to a variety of challenges, such as non-i.i.d. data distribution, skewed or disjoint class distribution, scalability, peer dynamism and asynchronism. In this paper, we present a novel P2P Adaptive Classification Ensemble (PACE) framework to perform classification in P2P networks. Unlike regular ensemble classification approaches, our new framework adapts to the test data distribution and dynamically adjusts the voting scheme by combining a subset of classifiers/peers according to the test data …