CiteULike: neteler's algorithms

Novel methods improve prediction of species distributions from occurrence data

Jane Elith — 2006-04-13T15:45:22-00:00

Ecography, Vol. 29, No. 2. (April 2006), pp. 129-151.

Prediction of species' distributions is central to diverse applications in ecology, evolution and conservation science. There is increasing electronic access to vast sets of occurrence records in museums and herbaria, yet little effective guidance on how best to use this information in the context of numerous approaches for modelling distributions. To meet this need, we compared 16 modelling methods over 226 species from 6 regions of the world, creating the most comprehensive set of model comparisons to date. We used presence-only data to fit models, and independent presence-absence data to evaluate the predictions. Along with well-established modelling methods such as generalised additive models and GARP and BIOCLIM, we explored methods that either have been developed recently or have rarely been applied to modelling species' distributions. These include machine-learning methods and community models, both of which have features that may make them particularly well suited to noisy or sparse information, as is typical of species' occurrence data. Presence-only data were effective for modelling species' distributions for many species and regions. The novel methods consistently outperformed more established methods. The results of our analysis are promising for the use of data from museums and herbaria, especially as methods suited to the noise inherent in such data improve.

Elements of Machine Learning

Pat Langley — 2006-12-20T09:49:28-00:00

(01 September 1995)

Recent years have seen an explosion of work on machine learning, the computational study of algorithms that improve performance based on experience. Research on rule induction, neural networks, genetic algorithms, case-based reasoning, and probabilistic inference has produced a variety of robust methods for inducing knowledge from training data. This book covers the main induction algorithms explored in the literature and presents them within a coherent theoretical framework that moves beyond traditional paradigm boundaries. 'Elements of Machine Learning' provides a comprehensive introduction to the fundamental concepts and problems in the field. The book illustrates a variety of basic algorithms for inducing simple concepts from experience, presents alternatives for organizing learned concepts into large-scale structures, and discusses adaptations of the learning methods to more complex problem-solving tasks. The chapters describe these computational techniques in detail and give examples of their operation, along with exercises and references to the literature. This text is suitable for use in graduate courses on machine learning. Researchers and students in artificial intelligence, cognitive science, and statistics will find it a useful and informative addition to their libraries.

A language-independent software renovation framework

M Di Penta — 2005-07-15T12:47:51-00:00

Journal of Systems and Software, Vol. 77, No. 3. (September 2005), pp. 225-240.

One of the undesired effects of software evolution is the proliferation of unused components, which are not used by any application. As a consequence, the size of binaries and libraries tends to grow and system maintainability tends to decrease. At the same time, a major trend of today's software market is the porting of applications on hand-held devices or, in general, on devices which have a limited amount of available resources. Refactoring and, in particular, the miniaturization of libraries and applications are therefore necessary.We propose a Software Renovation Framework (SRF) and a toolkit covering several aspects of software renovation, such as removing unused objects and code clones, and refactoring existing libraries into smaller more cohesive ones. Refactoring has been implemented in the SRF using a hybrid approach based on hierarchical clustering, on genetic algorithms and hill climbing, also taking into account the developers' feedback. The SRF aims to monitor software system quality in terms of the identified affecting factors, and to perform renovation activities when necessary. Most of the framework activities are language-independent, do not require any kind of source code parsing, and rely on object module analysis.The SRF has been applied to GRASS, which is a large open source Geographical Information System of about one million LOCs in size. It has significantly improved the software organization, has reduced by about 50% the average number of objects linked by each application, and has consequently also reduced the applications' memory requirements.

Moving to smaller libraries via clustering and genetic algorithms

G Antoniol — 2005-04-27T19:40:06-00:00

(2003)

There may be several reasons to reduce a software system to its bare bone removing the extra fat introduced during development or evolution. Porting the software system on embedded devices or palmtops are just two examples.