Dealing with temporal variation in patent categorization

In this paper, we quantify the existence of concept drift in patent data, and examine its impact on classification accuracy. When developing algorithms for classifying incoming patent applications with respect to their category in the International Patent Classification (IPC) hierarchy, a temporal mismatch between training data and incoming documents may deteriorate classification results. We measure the effect of this temporal mismatch and aim to tackle it by optimal selection of training data. To illustrate the various aspects of concept drift on IPC class level, we first perform quantitative analyses on a subset of English abstracts extracted from patent documents in the CLEF-IP 2011 patent corpus. In a series of classification experiments, we then show the impact of temporal variation on the classification accuracy of incoming applications. We further examine what training data selection method, combined with our classification approach yields the best classifier; and how combining different text representations may improve patent classification. We found that using the most recent data is a better strategy than static sampling but that extending a set of recent training data with older documents does not harm classification performance. In addition, we confirm previous findings that using 2-skip-2-grams on top of the bag of unigrams structurally improves patent classification. Our work is an important contribution to the research into concept drift for text classification, and to the practice of classifying incoming patent applications.     

Book Review

This book is a good example of the contribution a clear conceptual analysis can make to highlighting and clarifying problematic thinking about real issues in education. In the literature there are different claims of varying strength and plausibility which require very careful sorting. David Carr demonstrates such careful sorting in the difficult field of ethics and maps some paths for fruitful thinking about professionalism in teaching. ‘The distinctions of Peters and others between education and teaching, theory and practice, and intrinsic and instrumental value, stand in need of more careful mapping and refinement’ which could enable us to steer clear of ‘educationally pernicious conclusions’ (p. 179). Carr succeeds in doing just this.     

Green Doesn't Always Mean ‘Go’: possible tensions in the desirability and implementation of environmental education

Although there are different interpretations of environmental education, they all share two assumptions, i.e. that environmental education is a desirable objective, and that it needs to be implemented as widely and as urgently as possible. This paper argues that these two assumptions cannot be taken for granted in South Africa, with its particular history of apartheid and developmental needs. I examine four possible responses to the dominant Western/Northern interpretation of environmental education and argue that in South Africa the fourth response may prove to be the most fruitful one to pursue, for it moves beyond tensions generated by race and class issues.     

Identifying the ‘subnormal’ child in an age of expansion of special education and child science in the Netherlands (c.1945–1965)

Between c.1945 and 1965 across the West special education has grown and differentiated substantially. In the Netherlands this expansion ran parallel to the academic recognition and rapid development of the study of learning disabilities. How are these two processes related? This article argues that in this country child science and special education have mutually stimulated each other’s growth and development. The creation of new categories of special-needs children brought about a climate in which the study of learning disabilities and their treatment could flourish. This, in turn, produced further differentiation between children with learning difficulties. Soon problems of identification and categorisation of mentally ‘subnormal’ children proved too complicated to rely on intelligence testing and medical-psychological diagnosis alone. Educational prognosis, based on long-term observation and all kinds of testing, became the key to a child’s future at school and educationists instead of psychologists became the foremost keyholders.     

The dangers of schooling: the introduction of school medical inspection in the Netherlands (c.1900)

In this article the authors address the question of why school medical inspection in the Netherlands developed not only considerably slower than the British service but did so also on a more modest scale in terms of the impact on children’s lives. In the Netherlands school doctors were not allowed to treat children’s illnesses and therefore never opened clinics of their own. Likewise, they did not become part of the school system or even join forces in a national school medical service. To explain this the authors discuss the history of school hygienic concern since the mid‐nineteenth century, the respective roles played and arguments provided by the medical and teaching professions in the process of claiming the necessity for school medical inspection, and the conditions that have determined the appointment and instructions of school doctors since 1904. They draw the conclusion that the rapidly growing power of religious groups has limited the expansion of the medical regime. In the Netherlands schoolchildren did not become part of the ‘social body’ because of national fitness, but because society no longer accepted the inadequacy of the protection of pupils’ health against the dangers of schooling, especially after compulsory education was introduced.