An automated two-step NDVI-based method for the production of low-cost historical burned area map records over large areas

The recognition and understanding of long-term fire-related processes and patterns, such as the possible connection between the increased frequency of wildfires and global warming, requires the study of historical data records. In this study, a methodology was proposed for the automated production of long historical burned area map records over large-scale regions. The methodology was based on remotely sensed, high temporal resolution, normalized difference vegetation index (NDVI) data that could be easily acquired at medium or low spatial resolution. The proposed methodology was used to map the burned areas of the wildfires that occurred over the Peloponnese peninsula, Greece, during the summer of 2007. The method was built upon the NDVI data collected by the Moderate Resolution Imaging Spectroradiometer (MODIS) and Système Pour l’Observation de la Terre (SPOT)-VEGETATION. The higher spatial resolution data of MODIS resulted in higher burned area user accuracy (91.10%) and kappa (0.85) values than the respective ones for VEGETATION (79.29% and 0.77). The majority of classification errors were located along the perimeter of the burned areas and were mainly attributed to spatial resolution limitations of the remotely sensed data. The commission errors located away from the fire perimeter were primarily attributed to topographically shaded areas and land-cover types spectrally similar to burned areas. The omission errors resulted primarily from the small size and elongated shape of remote burned areas. Owing to their geometry, they have a high proportion of mixed pixels, whose spectral properties failed to meet the strict set of criteria for core fire pixels. The benefits of the proposed methodology are maximized when applied to data of the highest available spatial resolution, such as those collected by MODIS and the Project for On-Board Autonomy – Vegetation (PROBA-V) and when land-cover types spectrally similar to burned areas are masked prior to its application.     

Strength-based learning classifier systems revisited: Effective rule evolution in supervised classification tasks

Learning classifier systems (LCS) are machine learning systems designed to work for both multi-step and single-step decision tasks. The latter case presents an interesting challenge for such algorithms, especially when they are applied to real-world data mining (DM) problems. The present investigation departs from the popular approach of applying accuracy-based LCS to single-step classification and aims to uncover the potential of strength-based LCS in such tasks. Although the latter family of algorithms have often been associated with poor generalization and performance, we aim at alleviating these problems by defining appropriate extensions to the traditional strength-based LCS framework. These extensions are detailed and their effect on system performance is studied through the application of the proposed algorithm on a set of artificial problems, designed to challenge its scalability and generalization abilities. The comparison of the proposed algorithm with UCS, its state-of-the-art accuracy-based counterpart, emphasizes the effects of our extended strength-based approach and validates its competitiveness in multi-class problems with various class distributions. Overall, our work presents an investigation of strength-based LCS in the domain of supervised classification. Our extensive analysis of the learning dynamics involved in these systems provides proof of their potential as real-world DM tools, inducing tractable rule-based classification models, even in the presence of severe class imbalances.     

An Engineered E. coli Strain for Direct in Vivo Fluorination

Selectively fluorinated compounds are found frequently in pharmaceutical and agrochemical products where currently 25–30 % of optimised compounds emerge from development containing at least one fluorine atom. There are many methods for the site-specific introduction of fluorine, but all are chemical and they often use environmentally challenging reagents. Biochemical processes for C−F bond formation are attractive, but they are extremely rare. In this work, the fluorinase enzyme, originally identified from the actinomycete bacterium Streptomyces cattleya, is engineered into Escherichia coli in such a manner that the organism is able to produce 5′-fluorodeoxyadenosine (5′-FDA) from S-adenosyl-l -methionine (SAM) and fluoride in live E. coli cells. Success required the introduction of a SAM transporter and deletion of the endogenous fluoride efflux capacity in order to generate an E. coli host that has the potential for future engineering of more elaborate fluorometabolites.     

A model for European Biogenic Volatile Organic Compound emissions: Software development and first validation

A grid-oriented Biogenic Emission Model (BEM) has been developed to calculate Non-Methane Volatile Organic Compound (NMVOC) emissions from vegetation in high spatial and temporal resolutions. The model allows the emissions calculation for any modeling domain covering Europe on the basis of: 1) the U.S. Geological Survey 1-km resolution land-use database, 2) a land-use specific, monthly isoprene, monoterpene and Other Volatile Organic Compound (OVOC) emission potentials and foliar biomass densities database, 3) temperature and solar radiation data provided by the mesoscale meteorological model MM5. The model was applied for Europe in 30-km spatial resolution for the year 2003. The European total emissions for 2003 consist of 33.0% isoprene, 25.5% monoterpenes and 41.5% OVOC. BEM results are compared with those from the well-documented global Model of Emissions of Gases and Aerosols from Nature (MEGAN). The BEM total emissions compare well with the MEGAN ones. In July 2003, the results of both models agree within a factor of 1.2 for total isoprene emissions and within a factor of 2 for total monoterpene emissions. The comparison of the spatial distributions of the July 2003 isoprene and monoterpene emissions calculated with BEM and MEGAN shows that, in the greater part of the study area, the differences are below the current uncertainty limit for the estimation of spatially-resolved biogenic VOC emissions in Europe being equal to about ±600 kg km^?2 month^?1. Differences that are above this limit are found mainly in the eastern European countries for isoprene and in the Mediterranean countries for monoterpenes.     

Agent Mining: The Synergy of Agents and Data Mining

Autonomous agents and multiagent systems (or agents) and data mining and knowledge discovery (or data mining) are two of the most active areas in information technology. Ongoing research has revealed a number of intrinsic challenges and problems facing each area, which can't be addressed solely within the confines of the respective discipline. A profound insight of bringing these two communities together has unveiled a tremendous potential for new opportunities and wider applications through the synergy of agents and data mining. With increasing interest in this synergy, agent mining is emerging as a new research field studying the interaction and integration of agents and data mining. In this paper, we give an overall perspective of the driving forces, theoretical underpinnings, main research issues, and application domains of this field, while addressing the state-of-the-art of agent mining research and development. Our review is divided into three key research topics: agent-driven data mining, data mining-driven agents, and joint issues in the synergy of agents and data mining. This new and promising field exhibits a great potential for groundbreaking work from foundational, technological and practical perspectives.     

Guest Editors' Introduction: Agents and Data Mining

On top of two active research streams, agents and data mining, a most recent and exciting trend is their interaction and integration. Agent mining has emerged as a very promising field due to its unique contributions to complementary and innovative methodologies, techniques, and applications for complex problem-solving. This editorial summarizes the structure of this special issue.     

An integrated infrastructure for monitoring and evaluating agent-based systems

Driven by the urging need to thoroughly identify and accentuate the merits of agent technology, we present in this paper, MEANDER, an integrated framework for evaluating the performance of agent-based systems. The proposed framework is based on the Agent Performance Evaluation (APE) methodology, which provides guidelines and representation tools for performance metrics, measurement collection and aggregation of measurements. MEANDER comprises a series of integrated software components that implement and automate various parts of the methodology and assist evaluators in their tasks. The main objective of MEANDER is to integrate performance evaluation processes into the entire development lifecycle, while clearly separating any evaluation-specific code from the application code at hand. In this paper, we describe in detail the architecture and functionality of the MEANDER components and test its applicability to an existing multi-agent system.     

The phytate and phytase of soybean tempeh

Soybeans were fermented into tempeh by Rhizopus oligosporus NRRL 2710. The phytic acid content of soybeans was reduced by about one-third as a result of this fermentation, while an equivalent amount of phosphate was released in the tempeh. The reduction of phytic acid was due to phytase elaborated by the mould of the fermentation. The pH optimum of this enzyme was 5.6 and the K_m of the phytase-phytate reaction was 0.28 × 10^?3 M.