Efficient algorithms for ranking with SVMs

RankSVM (Herbrich et al. in Advances in large margin classifiers. MIT Press, Cambridge, MA, 2000; Joachims in Proceedings of the ACM conference on knowledge discovery and data mining (KDD), 2002) is a pairwise method for designing ranking models. SVMLight is the only publicly available software for RankSVM. It is slow and, due to incomplete training with it, previous evaluations show RankSVM to have inferior ranking performance. We propose new methods based on primal Newton method to speed up RankSVM training and show that they are 5 orders of magnitude faster than SVMLight. Evaluation on the Letor benchmark datasets after complete training using such methods shows that the performance of RankSVM is excellent.     

Information-analytical support of priority fields in science and technology: Branch and differentiated approaches

This paper considers the urgent problems of information-analytical support of Russian scientific technological and innovation policy, as well as modern tasks of information-analytical support of the innovation activities of scientific and educational organizations, and proposes methodical approaches that have potential for improving the Russian system of patent searches and analytics. It is shown to be expedient to build complex information-retrieval systems with maximum access to relevant arrays of branch information and multi-service analytical components, which make it possible to implement the differentiated approach to consumers and a diversity of scenarios of scientific information activities, which are particular to different groups of consumers.     

Tasks of information support of innovative economic development and the role of engineering

The determining factors of the pattern of innovative development of Russia’s economy are considered. The structural functional insufficiency of the current interactions between basic science and industry is shown. The steady upward trend in exports of software and IT services of Russian companies is analyzed. The role and potential of innovation engineering companies in the field of information technologies is emphasized. The current relevant tasks of information support of innovative research, development, and technology transfer are discussed. Prospective areas for improving the efficiency of the national information structure are presented.     

Aspects of Analytical Information Support of the Social Assessment of Technical and Technology Risks

The tendencies towards the increasing relevance of studying the negative effects of research, technology, and industrial development are analyzed. The methodological aspects of social assessment of technical and technology risks are considered. The prognostication, computer-aided modeling, and minimization of harmful effects of hi-tech industrial development and adoption of new technologies are discussed. The tasks of analytical information support of interdisciplinary studies aimed at technology assessment are considered from a system perspective. The relevance of developing the conceptual form of the national system of social assessment of equipment and new technologies and the need for the structural and substantial organization of this system are shown.