My Start Page

Acoustics, Speech, and Signal Processing, 2005. Proceedings. (ICASSP '05). IEEE International Conference on, Vol. 1 (2005), pp. 589-592.

Vol. 2 (2003), pp. 7-9.

Sources of training data suitable for language modeling of conversational speech are limited. In this paper, we show how training data can be supplemented with text from the web filtered to match the style and/or topic of the target recognition task, but also that it is possible to get bigger performance gains from the data by using class-dependent interpolation of N-grams. 1

Computers & Education, Vol. In Press, Corrected Proof

This paper describes an academic experience of podcasting, which involved a group of students of a course on multimedia communication and human-computer interaction. These students acted both as users of the university's podcasting service, and as creators of podcasted lessons. A comprehensive analysis based on the evaluation of the effects on student performance, on data from student satisfaction surveys, from interviews and from instructors' observations provided encouraging results: Full-time students co-involved in lessons' podcasting outperformed colleagues of the previous years and achieved higher levels of what we define as competitive agency, that led them to better understand the theoretical issues of the course and to more effective practical skills.

EDUCAUSE Review, Vol. 40, No. 6. (0 2005), pp. 33-34,.

Podcasting, by definition, combines "iPod" with "broadcasting," and has the advantages of ease of publication, ease of subscription, and ease of use across multiple environments. This article discusses how those in higher education can determine the potential uses and value of podcasting, which, done well, can reveal to students, faculty, staff, and communities the essential humanity at the heart of higher education. (Contains 13 notes.)

Human Interface and the Management of Information. Interacting in Information Environments (2007), pp. 404-412.

This paper presents an experience of educational podcasting set up at the University of Bergamo (Italy), and derives from that experience some remarks upon the quality of podcasting services, in order to promote the definition of guidelines on podcasting quality. We discuss three main attributes of a podcasting environment: quality of the production environment (recording and editing), quality of the product (content and communication style), quality of the distribution environment (paratext and management).

Organization Studies, Vol. 10, No. 1. (1 January 1989), pp. 23-44.

A general framework of salient variables is put forward in this paper to be used in the analysis and explanation of the organizational phenomena around technical change, particularly the introduction of 'new technology'. Such variables are grouped into output, workflow, organization structure, skills and knowledge, and societal and business context factors. Their interrelationships are discussed on the basis of the organization research tradition and more recent findings. Hypotheses are derived which may be used to explain further findings, establish a better link between organizational theory and research on recent technical change, and benefit from synergy between organization, industrial sociology and business administration research. 10.1177/017084068901000102

Trends in Biotechnology, Vol. In Press, Corrected Proof

During the current genomics revolution, the genomes of a large number of living organisms have been fully sequenced. However, with the advent of new sequencing technologies, genomics research is now at the threshold of a second revolution. Several second-generation sequencing platforms became available in 2007, but a further revolution in DNA resequencing technologies is being witnessed in 2008, with the launch of the first single-molecule DNA sequencer (Helicos Biosciences), which has already been used to resequence the genome of the M13 virus. This review discusses several single-molecule sequencing technologies that are expected to become available during the next few years and explains how they might impact on genomics research.

(26 Aug 2008)

"Cognizing" (i.e., thinking, understanding, knowing, and having the capacity to do what cognizers can do) is a mental state. Systems without mental states, such as cognitive technology, can sometimes also do some of what cognizers can do, but that does not make them cognizers. Cognitive technology allows cognizers to offload some of the functions they would otherwise have had to execute with their own brains and bodies alone; it also extends cognizers' performance powers beyond those of brains and bodies alone. Language itself is a form of cognitive technology that allows cognizers to offload some of their brain functions onto the brains of other cognizers. Language also extends cognizers' individual and joint performance powers, distributing the load through interactive and collaborative cognition. Reading, writing, print, telecommunications and computing further extend cognizers' capacities. And now the web, with its distributed network of cognizers, digital databases and sofware agents, has become the Cognitive Commons in which cognizers and cognitive technology can interact globally with a speed, scope and degree of interactivity that yield performance powers inconceivable with unaided individual cognition alone.

TechTrends, Vol. 52, No. 4. (1 July 2008), pp. 66-70.

TechTrends, Vol. 52, No. 4. (1 July 2008), pp. 34-41.

Internet Mathematics, Vol. 1, No. 1. (2003)

Computers & Education, Vol. 50, No. 2. (February 2008), pp. 491-498.

In this paper we describe a study of the effectiveness of mobile learning (m-learning) in the form of podcasting, for teaching undergraduate students in Higher Education. Podcasting involves downloading a series of audio or video broadcasts (files) onto a digital media player, via a computer, over a period of weeks. These can then be watched or listened to when, where and as often as students choose. The use of digital media players, popularised by Apple's iPod(TM), is widespread amongst undergraduate students. A pilot survey of Business and Management students indicated that over 74% owned some form of digital media player, with a further 7% indicating that they intended to purchase one in the next six months. Whilst podcasting is being utilized as a teaching tool by some educators in the secondary sector, its use in higher education, and its effectiveness as a learning tool for adults, remains to be established. In our study, a separate group of just under 200 first-level students were given a series of revision podcasts after completing a course in Information and Communications Technology (and prior to their examination). As part of the subscription process, they had to complete an online questionnaire about their experience. The questionnaire utilized a five-point Likert scale comparing their attitudes to lectures, podcasts, notes, textbooks and multimedia e-learning systems. Statistical analysis of the results of the study indicates that students believe that podcasts are more effective revision tools than their textbooks and they are more efficient than their own notes in helping them to learn. They also indicate that they are more receptive to the learning material in the form of a podcast than a traditional lecture or textbook. The study suggests that the use of podcasts as a revision tool has clear benefits as perceived by undergraduate students in terms of the time they take to revise and how much they feel they can learn. Coupled with the advantages of flexibility in when, where and how it is used, podcasting appears to have significant potential as an innovative learning tool for adult learners in Higher Education.

Health Policy Plan., Vol. 19, No. 5. (1 September 2004), pp. 302-309.

The high burden of disease in developing countries often makes it difficult for health systems in these countries to attain the same level of specialist skills as industrialized countries. Technology transfer is one way to improve specialist skills whilst at the same time reducing the burden of disease. This paper describes the use of teleophthalmology, a form of telemedicine, as a mode of technology transfer between the United Kingdom and South Africa. As the burden of eye disease in South Africa is high, the country cannot afford the level of ophthalmic specialization achieved in the UK. The paper estimates the cost-effectiveness of the technology transfer project in terms of a cost per Disability Adjusted Life Year (DALY) averted. We found the technology transfer project to be cost-effective in reducing the burden of eye disease, and that practitioners in South Africa also learned novel procedures that could help future patients and improve cost-effectiveness. Technology transfer using telemedicine is a cost-effective method that richer countries can employ to aid capacity building in the health care systems of poorer countries. 10.1093/heapol/czh035

Journal of the American Society for Information Science, Vol. 46, No. 8. (1995), pp. 581-589.

Business organizations have become interested in recognizing gatekeepers who may improve business prospects through informal external communication. Past research has identified gatekeepers within organizations, but no technique existed for identifying gatekeepers among groups of people working in an industry, profession, or other external group. This paper develops and tests a simple rule for identifying gatekeepers among members of a profession with a common group of potential interpersonal communication channels. The rule is used to classify gatekeepers among information technology professionals, based on self-reported information dissemination behavior. The rule passes five tests that compare the characteristics and information-seeking strategies of individuals classified using this rule with other members of the profession. The results of the study confirm that self-reported information dissemination behavior can be used to identify gatekeepers among individuals with diverse information-gathering behaviors, but a common group of potential interpersonal communication channels. © 1995 John Wiley & Sons, Inc.

(2002), pp. 133-142.

This paper proposes a new method for evaluating the quality of retrieval functions. Unlike traditional methods that require relevance judgements by experts or explicit user feedback, it is based entirely on clickthrough data. This is a key advantage, since clickthrough data can be collected at very low cost and without overhead for the user. Taking an approach from experiment design, the paper proposes an experiment setup that generates unbiased feedback about the relative quality of two search results without explicit user feedback. A theoretical analysis shows that the method gives the same results as evaluation with traditional relevance judgements under mild statistical assumptions. An empirical analysis veries that the assumptions are indeed justied and that the new method leads to conclusive results in a WWW retrieval study. 1

Science (14 August 2008), 1160379.

CAPTCHAs (Completely Automated Public Turing test to tell Computers and Humans Apart) are widespread security measures in the World Wide Web that prevent automated programs from abusing online services. They do so by asking humans to perform a task that computers cannot yet perform, such as deciphering distorted characters. Our research explored whether such human effort can be channeled into a useful purpose: helping to digitize old printed material by asking users to decipher scanned words from books that computerized optical character recognition (OCR) failed to recognize. We showed that this method can transcribe text with word accuracy over 99%, matching the guarantee of professional human transcribers. Our apparatus is deployed in over 40,000 Web sites and has transcribed over 440 million words. 10.1126/science.1160379

(2008), pp. 652-655.

Social Epistemology, Vol. 22, No. 1. (2008), pp. 71-95.

This essay presents a theory of the role of technology in the distribution and exercise of social power. The paper studies how technical artefacts and systems are used to construct, maintain or strengthen power relations between agents, whether individuals or groups, and how their introduction and use in society differentially empowers and disempowers agents. The theory is developed in three steps. First, a definition of power is proposed, based on a careful discussion of opposing definitions of power, and it is argued that a theory of power should have two components: a theory of power relations and a theory of empowerment. Second, an analysis of power relations is presented, in which five basic types of power relations between agents are distinguished, and this analysis is applied to technology, resulting in an account of the possible roles of technical artefacts in power relations. Third, I analyse how technology can lead to or contribute to empowerment and disempowerment, and what resistance strategies are possible against disempowerment through technological means. The theory of technology and power presented in this paper is claimed to be an essential ingredient of a critical theory of technology, which is a theory that analyses and critiques the role of technology in the distribution and exercise of power in society. In the final section of this paper, it is argued that the theoretical analysis of power and technology presented in this paper provides an adequate basis for the further development of such a critical theory of technology. I study how it may, specifically, be used to develop strategies for the democratization of technology.

PLoS ONE, Vol. 3, No. 8. (2008)

BACKGROUND: Data, data everywhere. The diversity and magnitude of the data generated in the Life Sciences defies automated articulation among complementary efforts. The additional need in this field for managing property and access permissions compounds the difficulty very significantly. This is particularly the case when the integration involves multiple domains and disciplines, even more so when it includes clinical and high throughput molecular data. METHODOLOGY/PRINCIPAL FINDINGS: The emergence of Semantic Web technologies brings the promise of meaningful interoperation between data and analysis resources. In this report we identify a core model for biomedical Knowledge Engineering applications and demonstrate how this new technology can be used to weave a management model where multiple intertwined data structures can be hosted and managed by multiple authorities in a distributed management infrastructure. Specifically, the demonstration is performed by linking data sources associated with the Lung Cancer SPORE awarded to The University of Texas MDAnderson Cancer Center at Houston and the Southwestern Medical Center at Dallas. A software prototype, available with open source at www.s3db.org, was developed and its proposed design has been made publicly available as an open source instrument for shared, distributed data management. CONCLUSIONS/SIGNIFICANCE: The Semantic Web technologies have the potential to addresses the need for distributed and evolvable representations that are critical for systems Biology and translational biomedical research. As this technology is incorporated into application development we can expect that both general purpose productivity software and domain specific software installed on our personal computers will become increasingly integrated with the relevant remote resources. In this scenario, the acquisition of a new dataset should automatically trigger the delegation of its analysis.

Lit Linguist Computing, Vol. 19, No. 2. (1 June 2004), pp. 161-180.

Within the last twenty years historians of science and technology have asked how a recent history might be written, and within the last ten interest has significantly increased, culminating in an online project at MIT. Since humanities computing owes its existence to developments in recent technology, and needs to become historically self-aware to be fully of the humanities, work toward an historiography of recent things is deeply relevant. In this essay I draw on this work to highlight the difficulties and opportunities of such an historiography, in particular its ethnographic character and the tempting lure of prediction. I focus on the crucial question of tacit object-knowledge, concluding that it is gained by concernful action. I recommend that we awaken from a progress-and-democratization chronicle to a genuine history of scholarly technology. 10.1093/llc/19.2.161