CID: an efficient complexity-invariant distance for time series

The ubiquity of time series data across almost all human endeavors has produced a great interest in time series data mining in the last decade. While dozens of classification algorithms have been applied to time series, recent empirical evidence strongly suggests that simple nearest neighbor classification is exceptionally difficult to beat. The choice of distance measure used by the nearest neighbor algorithm is important, and depends on the invariances required by the domain. For example, motion capture data typically requires invariance to warping, and cardiology data requires invariance to the baseline (the mean value). Similarly, recent work suggests that for time series clustering, the choice of clustering algorithm is much less important than the choice of distance measure used.In this work we make a somewhat surprising claim. There is an invariance that the community seems to have missed, complexity invariance. Intuitively, the problem is that in many domains the different classes may have different complexities, and pairs of complex objects, even those which subjectively may seem very similar to the human eye, tend to be further apart under current distance measures than pairs of simple objects. This fact introduces errors in nearest neighbor classification, where some complex objects may be incorrectly assigned to a simpler class. Similarly, for clustering this effect can introduce errors by “suggesting” to the clustering algorithm that subjectively similar, but complex objects belong in a sparser and larger diameter cluster than is truly warranted.We introduce the first complexity-invariant distance measure for time series, and show that it generally produces significant improvements in classification and clustering accuracy. We further show that this improvement does not compromise efficiency, since we can lower bound the measure and use a modification of triangular inequality, thus making use of most existing indexing and data mining algorithms. We evaluate our ideas with the largest and most comprehensive set of time series mining experiments ever attempted in a single work, and show that complexity-invariant distance measures can produce improvements in classification and clustering in the vast majority of cases.

A system for classification of time-series data from industrial non-destructive device

This work proposes a system for classification of industrial steel pieces by means of magnetic nondestructive device. The proposed classification system presents two main stages, online system stage and off-line system stage. In online stage, the system classifies inputs and saves misclassification information in order to perform posterior analyses. In the off-line optimization stage, the topology of a Probabilistic Neural Network is optimized by a Feature Selection algorithm combined with the Probabilistic Neural Network to increase the classification rate. The proposed Feature Selection algorithm searches for the signal spectrogram by combining three basic elements: a Sequential Forward Selection algorithm, a Feature Cluster Grow algorithm with classification rate gradient analysis and a Sequential Backward Selection. Also, a trash-data recycling algorithm is proposed to obtain the optimal feedback samples selected from the misclassified ones.

Structure-based drug design studies on a series of aldolase inhibitors

Human African trypanosomiasis, also known as sleeping sickness, is a major cause of death in Africa, and for which there are no safe and effective treatments available. The enzyme aldolase from Trypanosoma brucei is an attractive, validated target for drug development. A series of alkyl‑glycolamido and alkyl-monoglycolate derivatives was studied employing a combination of drug design approaches. Three-dimensional quantitative structure-activity relationships (3D QSAR) models were generated using the comparative molecular field analysis (CoMFA). Significant results were obtained for the best QSAR model (r2 = 0.95, non-cross-validated correlation coefficient, and q2 = 0.80, cross-validated correlation coefficient), indicating its predictive ability for untested compounds. The model was then used to predict values of the dependent variables (pKi) of an external test set,the predicted values were in good agreement with the experimental results. The integration of 3D QSAR, molecular docking and molecular dynamics simulations provided further insight into the structural basis for selective inhibition of the target enzyme.

Nutrientes y termoclinas en la Estación Europea de Series Temporales Oceánicas de Canarias

[ES]El mantenimiento de observaciones continuadas, durante largos períodos de tiempo en un mismo área, es una estrategia científica y logística de gran interés oceanográfico. El estudio detallado de las series así establecidas, nos permitirá conocer los cambios temporales que tienen lugar tanto en la fisica, como en la biogeoquímica oceánica, y los procesos que ejercen el control de los mismos. Dentro de esta escala temporal, la base de datos hidrográficos que resulta nos ayudará a establecer los ciclos estacionales y los ciclos interanuales del conjunto de parámetros hidrográficos y geoquímicos medidos. El Archipiélago Canario está considerado, dada la combinación de procesos de larga escala que se dan, como una zona óptima para estudios de las aguas oceánicas a escala global. Se encuentra rodeado de aguas profundas inmersas en el régimen de recirculación Este de la Corriente del Golfo, constituido por la Corriente de Canarias, influenciado por el afloramiento del Noroeste Africano y por la deposición de partículas eólicas desde el Sahara.El programa ESTOC supone una contribución a los proyectos internacionales y multidisciplinarios WOCE (World Ocean Circulation Experiment) y JGOFS (Joint Global Ocean Flux Study) que pretenden resolver el problema científico que supone la escasez de información sobre lo que está ocurriendo en los océanos a escala global.

Análisis del estado de los recursos pesqueros de Gran Canaria a partir del estudio de las series históricas de captura

Máster en Gestión Sostenible de Recursos Pesqueros

Comité de Expertos para combinación de fuentes de predicción. Aplicación en series temporales eólicas.

Máster Universitario en Sistemas Inteligentes y Aplicaciones Numéricas en Ingeniería (SIANI)

La institucionalización científica de la Arqueología en España a través de sus revistas y series periódicas

[ES] En nuestro trabajo abordamos las principales fases históricas del proceso de institucionalización e independencia científica de las publicaciones arqueológicas en España, en cuyos periodos de crecimiento y recesión es posible determinar el proceso por el que ha atravesado la investigación arqueológica en nuestro país, desde el siglo XVIII hasta el presente.