Open Access. Powered by Scholars. Published by Universities.®

Digital Commons Network™

Open Access. Powered by Scholars. Published by Universities.®

Singapore Management University

Discipline
Keyword
Publication Year
Publication
Publication Type
File Type

Articles 9271 - 9300 of 10474

Full-Text Articles in Entire DC Network

Innovation Engine, Arcot Desai Narasimhalu Jun 2007

Innovation Engine, Arcot Desai Narasimhalu

Research Collection School Of Computing and Information Systems

This paper describes a meta-model for innovation using an automobile engine as a metaphor. This innovation meta-model is used to manage a collection of innovation models. We develop an algorithm to identify innovations with potential for success using this meta-model. This meta-model can be used by corporations and individuals to identify plausible innovations at any given point in time.

Go to article

Usage-Based Versus Flat Pricing For E-Business Services With Differentiated Qos, Zhen Liu, Laura Wynter, Cathy Xia Jun 2007

Usage-Based Versus Flat Pricing For E-Business Services With Differentiated Qos, Zhen Liu, Laura Wynter, Cathy Xia

Research Collection School Of Computing and Information Systems

Design of e-commerce services that are competitive, in such a quickly responding market, requires the analyses of prices and price structures. We present a general model of an e-commerce market that allows us to analyze optimal price structures, both flat and usage-based. Based on the price structure of a major web hosting provider, we consider both single-tier and two-tier (burst-rate) pricing, and our result suggests that the more complex two-tier structure may not be worth the marketing effort, as the firm?s equilibrium profits will not increase through the use of this structure. An essential feature of our approach is that …

Go to article

Teaching "Global Project Management" With Distributed Team Projects, Randy Weinberg, Benjamin Gan Jun 2007

Teaching "Global Project Management" With Distributed Team Projects, Randy Weinberg, Benjamin Gan

Research Collection School Of Computing and Information Systems

The education of rising professionals must keep pace with changing forces of globalization and the realities of distributed work. Students who understand the basics of global project management, teamwork and collaboration are likely to find themselves at a competitive advantage over those who do not. This article describes the experiences in an undergraduate course called Global Project Management offered concurrently at two universities, one in the U.S. and one in Singapore, and incorporating collaborative student projects.

Go to article

An Efficient Identity-Based Key Exchange Protocol With Kgs Forward Secrecy For Low-Power Devices, Robert W. Zhu, Guomin Yang, Duncan S. Wong Jun 2007

An Efficient Identity-Based Key Exchange Protocol With Kgs Forward Secrecy For Low-Power Devices, Robert W. Zhu, Guomin Yang, Duncan S. Wong

Research Collection School Of Computing and Information Systems

For an ID-based key exchange (KE) protocol, KGS forward secrecy is about the protection of previously established session keys after the master secret key of the Key Generation Server (KGS) is compromised. This is the strongest notion of forward secrecy that one can provide for an ID-based KE protocol. Among all the comparable protocols, there are only a few of them that provide this level of forward secrecy, and all of these protocols require expensive bilinear pairing operations and map-to-point hash operations that may not be suitable for implementation on low-power devices such as sensors. In this paper, we propose …

Go to article

Simplifying Cyber Foraging For Mobile Devices, Rajesh Krishna Balan, Darren Gergle, Mahadev Satyanarayanan, James Herbsleb Jun 2007

Simplifying Cyber Foraging For Mobile Devices, Rajesh Krishna Balan, Darren Gergle, Mahadev Satyanarayanan, James Herbsleb

Research Collection School Of Computing and Information Systems

Cyber foraging is the transient and opportunistic use of compute servers by mobile devices. The short market life of such devices makes rapid modification of applications for remote execution an important problem. We describe a solution that combines a "little language" for cyber foraging with an adaptive runtime system. We report results from a user study showing that even novice developers are able to successfully modify large, unfamiliar applications in just a few hours. We also show that the quality of novice-modified and expert-modified applications are comparable in most cases.

Go to article

Intelligence Through Interaction: Towards A Unified Theory For Learning, Ah-Hwee Tan, Gail A. Carpenter, Stephen Grossberg Jun 2007

Intelligence Through Interaction: Towards A Unified Theory For Learning, Ah-Hwee Tan, Gail A. Carpenter, Stephen Grossberg

Research Collection School Of Computing and Information Systems

Machine learning, a cornerstone of intelligent systems, has typically been studied in the context of specific tasks, including clustering (unsupervised learning), classification (supervised learning), and control (reinforcement learning). This paper presents a learning architecture within which a universal adaptation mechanism unifies a rich set of traditionally distinct learning paradigms, including learning by matching, learning by association, learning by instruction, and learning by reinforcement. In accordance with the notion of embodied intelligence, such a learning theory provides a computational account of how an autonomous agent may acquire the knowledge of its environment in a real-time, incremental, and continuous manner. Through a …

Go to article

Learning Nonparametric Kernel Matrices From Pairwise Constraints, Steven C. H. Hoi, Rong Jin, Michael R. Lyu Jun 2007

Learning Nonparametric Kernel Matrices From Pairwise Constraints, Steven C. H. Hoi, Rong Jin, Michael R. Lyu

Research Collection School Of Computing and Information Systems

Many kernel learning methods have to assume parametric forms for the target kernel functions, which significantly limits the capability of kernels in fitting diverse patterns. Some kernel learning methods assume the target kernel matrix to be a linear combination of parametric kernel matrices. This assumption again importantly limits the flexibility of the target kernel matrices. The key challenge with nonparametric kernel learning arises from the difficulty in linking the nonparametric kernels to the input patterns. In this paper, we resolve this problem by introducing the graph Laplacian of the observed data as a regularizer when optimizing the kernel matrix with …

Go to article

A Structuralist Model Of The Small Open Economy In The Short, Medium And Long Run, Hian Teck Hoon, Edmund S. Phelps Jun 2007

A Structuralist Model Of The Small Open Economy In The Short, Medium And Long Run, Hian Teck Hoon, Edmund S. Phelps

Research Collection School Of Economics

Open-economy macroeconomics contains a monetary model in the Keynesian tradition that is deemed serviceable for analyzing the short run and a nonmonetary neoclassical model thought capable of handling the long run. But do the Keynesian and neoclassical models meet the challenges thrown out by the main events of the past few decades—the ’80s shock to Europe from the sharp increase of external real interest rates; the kind of speculative shock experienced in the U.S. and parts of northern Europe in the second half of the ’90s: the prospect of new industries emerging in the future with needs for new capital; …

Go to article

A Structuralist Model Of The Small Open Economy In The Short, Medium And Long Run, Hian Teck Hoon, Edmund S. Phelps Jun 2007

A Structuralist Model Of The Small Open Economy In The Short, Medium And Long Run, Hian Teck Hoon, Edmund S. Phelps

Research Collection School Of Economics

Open-economy macroeconomics contains a monetary model in the Keynesian tradition that is deemed serviceable for analyzing the short run and a nonmonetary neoclassical model thought capable of handling the long run. But do the Keynesian and neoclassical models meet the challenges thrown out by the main events of the past few decades? We first indicate that the effects of these shocks on the open economy are not well captured by either the standard Keynesian model or the standard neoclassical theory. Next we provide a careful development of a nonmonetary model of the equilibrium path of the real exchange rate, share …

Go to article

Time Capsule Signature: Efficient And Provably Secure Constructions, Bessie C. Hu, Duncan S. Wong, Qiong Huang, Guomin Yang, Xiaotie Deng Jun 2007

Time Capsule Signature: Efficient And Provably Secure Constructions, Bessie C. Hu, Duncan S. Wong, Qiong Huang, Guomin Yang, Xiaotie Deng

Research Collection School Of Computing and Information Systems

Time Capsule Signature, first formalized by Dodis and Yum in Financial Cryptography 2005, is a digital signature scheme which allows a signature to bear a (future) time t so that the signature will only be valid at time t or later, when a trusted third party called time server releases time-dependent information for checking the validity of a time capsule signature. Also, the actual signer of a time capsule signature has the privilege to make the signature valid before time t.In this paper, we provide a new security model of time capsule signature such that time server is not required …

Go to article

A More Natural Way To Construct Identity-Based Identification Schemes, Guomin Yang, Jing Chen, Duncan S. Wong, Xiaotie Deng, Dongsheng Wang Jun 2007

A More Natural Way To Construct Identity-Based Identification Schemes, Guomin Yang, Jing Chen, Duncan S. Wong, Xiaotie Deng, Dongsheng Wang

Research Collection School Of Computing and Information Systems

Constructing identification schemes is one of the fundamental problems in cryptography, and is very useful in practice. An identity-based identification (IBI) scheme allows a prover to identify itself to a public verifier who knows only the claimed identity of the prover and some common information. In this paper, we propose a simple and efficient framework for constructing IBI schemes. Unlike some related framework which constructs IBI schemes from some standard identification schemes, our framework is based on some more fundamental assumptions on intractable problems. Depending on the features of the underlying intractable problems presumed in our framework, we can derive …

Go to article

A Hybrid Of Plot-Based And Character-Based Interactive Storytelling, Yundong Cai, Chunyan Miao, Ah-Hwee Tan, Zhiqi Shen Jun 2007

A Hybrid Of Plot-Based And Character-Based Interactive Storytelling, Yundong Cai, Chunyan Miao, Ah-Hwee Tan, Zhiqi Shen

Research Collection School Of Computing and Information Systems

Interactive storytelling in the virtual environment attracts a lot of research interests in recent years. Story plot and character are two most important elements of a story. Based on these two elements, currently there are two research directions: plot-based and character-based interactive storytelling. However, plot-based approach lacks the refinement of character behaviors as character-based approach. On the other side, character-based approach does not follow a well organized story plot so that the moral of the story might be distorted. Therefore, there is a need to develop an integrated framework to achieve the balance between conveying story moral and enhancing the …

Go to article

Instance Weighting For Domain Adaptation In Nlp, Jing Jiang, Chengxiang Zhai Jun 2007

Instance Weighting For Domain Adaptation In Nlp, Jing Jiang, Chengxiang Zhai

Research Collection School Of Computing and Information Systems

Domain adaptation is an important problem in natural language processing (NLP) due to the lack of labeled data in novel domains. In this paper, we study the domain adaptation problem from the instance weighting per- spective. We formally analyze and charac- terize the domain adaptation problem from a distributional view, and show that there are two distinct needs for adaptation, cor- responding to the different distributions of instances and classification functions in the source and the target domains. We then propose a general instance weighting frame- work for domain adaptation. Our empir- ical results on three NLP tasks show that …

Go to article

Mobile G-Portal Supporting Collaborative Sharing And Learning In Geography Fieldwork: An Empirical Study, Yin-Leng Theng, Kuah-Li Tan, Ee Peng Lim, Jun Zhang, Dion Hoe-Lian Goh, Kalyani Chatterjea, Chew-Hung Chang, Aixin Sun, Han Yu, Nam Hai Dang, Yuanyuan Li, Minh Chanh Vo Jun 2007

Mobile G-Portal Supporting Collaborative Sharing And Learning In Geography Fieldwork: An Empirical Study, Yin-Leng Theng, Kuah-Li Tan, Ee Peng Lim, Jun Zhang, Dion Hoe-Lian Goh, Kalyani Chatterjea, Chew-Hung Chang, Aixin Sun, Han Yu, Nam Hai Dang, Yuanyuan Li, Minh Chanh Vo

Research Collection School Of Computing and Information Systems

Integrated with G-Portal, a Web-based geospatial digital library of geography resources, this paper describes the implementation of Mobile G-Portal, a group of mobile devices as learning assistant tools supporting collaborative sharing and learning for geography fieldwork. Based on a modified Technology Acceptance Model and a Task-Technology Fit model, an initial study with Mobile G-Portal was conducted involving 39 students in a local secondary school. The findings suggested positive indication of acceptance of Mobile G-Portal for geography fieldwork. The paper concludes with a discussion on technological challenges, recommendations for refinement of Mobile G-Portal, and design implications in general for digital libraries …

Go to article

Maximizing Broadcast And Multicast Traffic Load Through Link-Rate Diversity In Wireless Mesh Networks, Chun Tung Chou, Bao Hua Liu, Archan Misra Jun 2007

Maximizing Broadcast And Multicast Traffic Load Through Link-Rate Diversity In Wireless Mesh Networks, Chun Tung Chou, Bao Hua Liu, Archan Misra

Research Collection School Of Computing and Information Systems

This paper studies some of the fundamental challenges and opportunities associated with the network-layer broadcast and multicast in a multihop multirate wireless mesh network (WMN). In particular, we focus on exploiting the ability of nodes to perform link-layer broadcasts at different rates (with correspondingly different coverage areas). We first show how, in the broadcast wireless medium, the available capacity at a mesh node for a multicast transmission is not just a function of the aggregate pre-existing traffic load of other interfering nodes, but intricately coupled to the actual (sender, receiver) set and the link-layer rate of each individual transmission. We …

Go to article

Editors' Introduction To Regular Research Section, Jae Kyu Lee, Patrick Y. K. Chau, Robert J. Kauffman, Norman M. Sadeh Jun 2007

Editors' Introduction To Regular Research Section, Jae Kyu Lee, Patrick Y. K. Chau, Robert J. Kauffman, Norman M. Sadeh

Research Collection School Of Computing and Information Systems

The second section of this issue consists of four regular submissions.In “A New Mobile Payment Scheme for Roaming Services”, Ren-Junn Hwang, Sheng-Hua Shiau and Ding-Far Jan present a new mobile payment scheme that supports anonymity while allowing consumers to roam outside of their home operator’s network.Vasudeva Akula and Daniel A. Menascé’s article on “Two-level Workload Characterization of Online Auctions” analyzes the workload of an online auction site. The analysis reveals the presence of heavy tailed distributions and looks at bidding activity during the closing minutes of auctions. The article suggests that results from this analysis could be used to devise …

Go to article

Are Radio Sources And Gamma Ray Bursts Luminal Booms?, Manoj Thulasidas Jun 2007

Are Radio Sources And Gamma Ray Bursts Luminal Booms?, Manoj Thulasidas

Research Collection School Of Computing and Information Systems

The softening of a Gamma Ray Burst (GRB) afterglow bears remarkable similarities to the frequency evolution in a sonic boom. Atthe front end of the sonic boom cone, the frequency is infinite, much like a GRB. Inside the cone, the frequencyrapidly decreases to infrasonic ranges and the sound source appears at two places at the same time,mimicking the double-lobed radio sources. Although a “luminal” boom violates the Lorentz invarianceand is therefore forbidden, it is tempting to work out the details and compare them with existing data. This temptation is further enhanced by the observed superluminality in the celestial objects associated …

Go to article

Similarity Beyond Distance Measurement, Feng Kang, Rong Jin, Steven C. H. Hoi Jun 2007

Similarity Beyond Distance Measurement, Feng Kang, Rong Jin, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

One of the keys issues to content-based image retrieval is the similarity measurement of images. Images are represented as points in the space of low-level visual features and most similarity measures are based on certain distance measurement between these features. Given a distance metric, two images with shorter distance are deemed to more similar than images that are far away. The well-known problem with these similarity measures is the semantic gap, namely two images separated by large distance could share the same semantic content. In this paper, we propose a novel similarity measure of images that goes beyond the distance …

Go to article

Making The ‘Most’ Out Of Rfid Technology: A Research Agenda For The Study Of The Adoption, Use And Impact Of Rfid, John Curtin, Robert J. Kauffman, Frederick J. Riggins Jun 2007

Making The ‘Most’ Out Of Rfid Technology: A Research Agenda For The Study Of The Adoption, Use And Impact Of Rfid, John Curtin, Robert J. Kauffman, Frederick J. Riggins

Research Collection School Of Computing and Information Systems

Radio frequency identification (RFID) technology dramatically increases the ability of the organization to acquire a vast array of data about the location and properties of any entity that can be physically tagged and wirelessly scanned within certain technical limitations. RFID can be applied to a variety of tasks, structures, work systems and contexts along the value chain, including business-to-business logistics, internal operations, business-to-consumer marketing, and after-sales service applications. As industry adoption of RFID increases there is an emerging interest by academic researchers to engage in scholarly investigation to understand how RFID relates to mobility, organizational and systems technologies (MOST). In …

Go to article

Vertically Coupled Microring Laser Devices Based On Inp Using Bcb Wafer Bonding, M. Hamacher, H. Heidrich, U. Troppenz, D. Syvridis, D. Alexandropoulos, S. Mikroulis, A. Kapsalis, Chyng Wen Tee, K. A. Williams, V. Dragoi, M. Alexe, D. Cristea, M. Kusko Jun 2007

Vertically Coupled Microring Laser Devices Based On Inp Using Bcb Wafer Bonding, M. Hamacher, H. Heidrich, U. Troppenz, D. Syvridis, D. Alexandropoulos, S. Mikroulis, A. Kapsalis, Chyng Wen Tee, K. A. Williams, V. Dragoi, M. Alexe, D. Cristea, M. Kusko

Research Collection Lee Kong Chian School Of Business

In this paper, we present our approach on the realization of vertical coupling technology that has been verified for microlaser fabrication as well as for the realization of high Q resonators with 'loss-less' cavities based on InP. The concept is based on full wafer technology and provides the integration of an optical transparent passive bus waveguide vertically coupled to the active ring laser.

Go to article

Asymmetric Discounting In Intertemporal Choice: A Query Theory Account, E. U. Weber, E. J. Johnson, K. F. Milch, Hannah H. Chang, J. C. Brodscholl, D. G. Goldstein Jun 2007

Asymmetric Discounting In Intertemporal Choice: A Query Theory Account, E. U. Weber, E. J. Johnson, K. F. Milch, Hannah H. Chang, J. C. Brodscholl, D. G. Goldstein

Research Collection Lee Kong Chian School Of Business

People are impatient and discount future rewards more when they are asked to delay consumption than when they are offered the chance to accelerate consumption. The three experiments reported here provide a process-level account for this asymmetry, with implications for designing decision environments that promote less impulsivity. In Experiment 1, a thought-listing procedure showed that people decompose discount valuation into two queries. Whether one considers delayed or accelerated receipt of a gift certificate influences the order in which memory is queried to support immediate versus delayed consumption, and the order of queries affects the relative number of patient versus impatient …

Go to article

Forgery Attack To An Asymptotically Optimal Traitor Tracing Scheme, Yongdong Wu, Feng Bao, Robert H. Deng May 2007

Forgery Attack To An Asymptotically Optimal Traitor Tracing Scheme, Yongdong Wu, Feng Bao, Robert H. Deng

Research Collection School Of Computing and Information Systems

In this paper, we present a forgery attack to a black-box traitor tracing scheme [2] called as CPP scheme. CPP scheme has efficient transmission rate and allows the tracer to identify a traitor with just one invalid ciphertext. Since the original CPP scheme is vulnerable to the multi-key attack, we improved CPP to thwart the attack. However, CPP is vulnerable to a fatal forgery attack. In the forgery attack, two traitors can collude to forge all valid decryption keys. The forged keys appear as perfect genuine keys, can decrypt all protected content, but are untraceable by the tracer. Fortunately, we …

Go to article

Cognitive Evaluation Of Information Modeling Methods, Keng Siau, Yuan Wang May 2007

Cognitive Evaluation Of Information Modeling Methods, Keng Siau, Yuan Wang

Research Collection School Of Computing and Information Systems

In the field of information system engineering, information modeling method is a technique to capture user requirements and to understand system complexity. The importance of information modeling has been recognized by practitioners and researchers, but little has been explored to analyze the available information modeling methods or to evaluate them in terms of their strengths, weaknesses, and effectiveness. This research analyzes six information-modeling methods: use case diagram, rich picture diagram, entity-relationship diagram, Trochim’s concept mapping, repertory grid, and causal mapping. These information-modeling methods are analyzed from a cognitive perspective in order to better understand their nature, the assumptions, and the …

Go to article

Learning To Classify E-Mail, Irena Koprinska, Josiah Poon, James Clark, Jason Yuk Hin Chan May 2007

Learning To Classify E-Mail, Irena Koprinska, Josiah Poon, James Clark, Jason Yuk Hin Chan

Research Collection School Of Computing and Information Systems

In this paper we study supervised and semi-supervised classification of e-mails. We consider two tasks: filing e-mails into folders and spam e-mail filtering. Firstly, in a supervised learning setting, we investigate the use of random forest for automatic e-mail filing into folders and spam e-mail filtering. We show that random forest is a good choice for these tasks as it runs fast on large and high dimensional databases, is easy to tune and is highly accurate, outperforming popular algorithms such as decision trees, support vector machines and naive Bayes. We introduce a new accurate feature selector with linear time complexity. …

Go to article

Modeling Architectural Strategy Using Design Structure Networks, C. Jason Woodard May 2007

Modeling Architectural Strategy Using Design Structure Networks, C. Jason Woodard

Research Collection School Of Computing and Information Systems

System architects face the formidable task of purposefully shaping an evolving space of complex designs. Their task s further complicated when they lack full control of the design process, and therefore must anticipate the behavior of other stakeholders, including the designers of component products and competing systems. This paper presents a conceptual tool called a design structure network (DSN) to help architects and design scientists reason effectively about these situations. A DSN is a graphical representation of a system’s design space. DSNs improve on existing representation schemes by providing a compact and intuitive way to express design options—the ability to …

Go to article

Integrated 10th Order Fresnel Lens Design For Beam Quality Enhancement In Tapered Laser Diode, F. K. Lau, Chyng Wen Tee, C. H. Kwok, R. V. Penty, I. H. White, N. Michel, M. Krakowski May 2007

Integrated 10th Order Fresnel Lens Design For Beam Quality Enhancement In Tapered Laser Diode, F. K. Lau, Chyng Wen Tee, C. H. Kwok, R. V. Penty, I. H. White, N. Michel, M. Krakowski

Research Collection Lee Kong Chian School Of Business

An integrated 10th order Fresnel lens capable of improving the laser beam quality is reported. The far-field divergence is narrowed by an average of 1.9 deg (29%) and an overall M2- factor improvement of 15% is recorded.

Go to article

Vertically-Coupled Microring Laser Array For Dual-Wavelength Generation, Chyng Wen Tee, K. A. Williams, R. V. Penty, I. H. White, M. Hamacher, U. Troppenz, H. Heidrich May 2007

Vertically-Coupled Microring Laser Array For Dual-Wavelength Generation, Chyng Wen Tee, K. A. Williams, R. V. Penty, I. H. White, M. Hamacher, U. Troppenz, H. Heidrich

Research Collection Lee Kong Chian School Of Business

We report the first demonstration of continuous-wave operation of a tunable, compact microring laser array based on a vertical-coupling architecture, well suited to larger-scale integration. Wavelength separation tunability from 4.9 to 6.3nm is observed.

Go to article

Gprune: A Constraint Pushing Framework For Graph Pattern Mining, Feida Zhu, Xifeng Yan, Jiawei Han, Philip S. Yu May 2007

Gprune: A Constraint Pushing Framework For Graph Pattern Mining, Feida Zhu, Xifeng Yan, Jiawei Han, Philip S. Yu

Research Collection School Of Computing and Information Systems

In graph mining applications, there has been an increasingly strong urge for imposing user-specified constraints on the mining results. However, unlike most traditional itemset constraints, structural constraints, such as density and diameter of a graph, are very hard to be pushed deep into the mining process. In this paper, we give the first comprehensive study on the pruning properties of both traditional and structural constraints aiming to reduce not only the pattern search space but the data search space as well. A new general framework, called gPrune, is proposed to incorporate all the constraints in such a way that they …

Go to article

Analysis Of Topological Characteristics Of Huge Online Social Networking Services, Yong-Yeol Ahn, Seungyeop Han, Haewoon Kwak, Sue Moon, Hawoong Jeong May 2007

Analysis Of Topological Characteristics Of Huge Online Social Networking Services, Yong-Yeol Ahn, Seungyeop Han, Haewoon Kwak, Sue Moon, Hawoong Jeong

Research Collection School Of Computing and Information Systems

Social networking services are a fast-growing business in the Internet. However, it is unknown if online relationships and their growth patterns are the same as in real-life social networks. In this paper, we compare the structures of three online social networking services: Cyworld, MySpace, and orkut, each with more than 10 million users, respectively. We have access to complete data of Cyworld's ilchon (friend) relationships and analyze its degree distribution, clustering property, degree correlation, and evolution over time. We also use Cyworld data to evaluate the validity of snowball sampling method, which we use to crawl and obtain partial network …

Go to article

Enhancing The Performance Of Semi-Supervised Classification Algorithms With Bridging, Jason Yuk Hin Chan, Josiah Poon, Irena Koprinska May 2007

Enhancing The Performance Of Semi-Supervised Classification Algorithms With Bridging, Jason Yuk Hin Chan, Josiah Poon, Irena Koprinska

Research Collection School Of Computing and Information Systems

Traditional supervised classification algorithms require a large number of labelled examples to perform accurately. Semi-supervised classification algorithms attempt to overcome this major limitation by also using unlabelled examples. Unlabelled examples have also been used to improve nearest neighbour text classification in a method called bridging. In this paper, we propose the use of bridging in a semi-supervised setting. We introduce a new bridging algorithm that can be used as a base classifier in any supervised approach such as co-training or selflearning. We empirically show that classification performance increases by improving the semi-supervised algorithm’s ability to correctly assign labels to previouslyunlabelled …

Go to article

First Prev Next Last