Rank Aggregation for Pattern Classifier Selection in Remote Sensing Images

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Model selection for discriminative restricted boltzmann machines through meta-heuristic techniques

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Isolation, selection and characterization of root-associated growth promoting bacteria in Brazil Pine (Araucaria angustifolia)

Araucaria angustifolia, a unique species of this genus that occurs naturally in Brazil, has a high socio-economic and environmental value and is critically endangered of extinction, since it has been submitted to intense predatory exploitation during the last century. Root-associated bacteria from A. angustifolia were isolated, selected and characterized for their biotechnological potential of growth promotion and biocontrol of plant pathogenic fungi. Ninety-seven strains were isolated and subjected to chemical tests. All isolates presented at least one positive feature, characterizing them as potential PGPR. Eighteen isolates produced indole-3-acetic acid (IAA), 27 were able to solubilize inorganic phosphate, 21 isolates were presumable diazotrophs, with pellicle formation in nitrogen-free culture medium, 83 were phosphatases producers, 37 were positive for siderophores and 45 endospore-forming isolates were antagonistic to Fusarium oxysporum, a pathogen of conifers. We also observed the presence of bacterial strains with multiple beneficial mechanisms of action. Analyzing the fatty acid methyl ester (FAME) and partial sequencing of the 16S rRNA gene of these isolates, it was possible to characterize the most effective isolates as belonging to Bacillaceae (9 isolates), Enterobacteriaceae (11) and Pseudomonadaceae (1). As far as we know, this is the first study to include the species Ewingella americana as a PGPR. (C) 2011 Elsevier GmbH. All rights reserved.

Three-feature model to reproduce the topology of citation networks and the effects from authors' visibility on their h-index

Various factors are believed to govern the selection of references in citation networks, but a precise, quantitative determination of their importance has remained elusive. In this paper, we show that three factors can account for the referencing pattern of citation networks for two topics, namely "graphenes" and "complex networks", thus allowing one to reproduce the topological features of the networks built with papers being the nodes and the edges established by citations. The most relevant factor was content similarity, while the other two - in-degree (i.e. citation counts) and age of publication - had varying importance depending on the topic studied. This dependence indicates that additional factors could play a role. Indeed, by intuition one should expect the reputation (or visibility) of authors and/or institutions to affect the referencing pattern, and this is only indirectly considered via the in-degree that should correlate with such reputation. Because information on reputation is not readily available, we simulated its effect on artificial citation networks considering two communities with distinct fitness (visibility) parameters. One community was assumed to have twice the fitness value of the other, which amounts to a double probability for a paper being cited. While the h-index for authors in the community with larger fitness evolved with time with slightly higher values than for the control network (no fitness considered), a drastic effect was noted for the community with smaller fitness. (C) 2012 Elsevier Ltd. All rights reserved.

Radiomics of NSCLC: Quantitative CT Image Feature Characterization and Tumor Shrinkage Prediction

Radiomics is the high-throughput extraction and analysis of quantitative image features. For non-small cell lung cancer (NSCLC) patients, radiomics can be applied to standard of care computed tomography (CT) images to improve tumor diagnosis, staging, and response assessment. The first objective of this work was to show that CT image features extracted from pre-treatment NSCLC tumors could be used to predict tumor shrinkage in response to therapy. This is important since tumor shrinkage is an important cancer treatment endpoint that is correlated with probability of disease progression and overall survival. Accurate prediction of tumor shrinkage could also lead to individually customized treatment plans. To accomplish this objective, 64 stage NSCLC patients with similar treatments were all imaged using the same CT scanner and protocol. Quantitative image features were extracted and principal component regression with simulated annealing subset selection was used to predict shrinkage. Cross validation and permutation tests were used to validate the results. The optimal model gave a strong correlation between the observed and predicted shrinkages with . The second objective of this work was to identify sets of NSCLC CT image features that are reproducible, non-redundant, and informative across multiple machines. Feature sets with these qualities are needed for NSCLC radiomics models to be robust to machine variation and spurious correlation. To accomplish this objective, test-retest CT image pairs were obtained from 56 NSCLC patients imaged on three CT machines from two institutions. For each machine, quantitative image features with concordance correlation coefficient values greater than 0.90 were considered reproducible. Multi-machine reproducible feature sets were created by taking the intersection of individual machine reproducible feature sets. Redundant features were removed through hierarchical clustering. The findings showed that image feature reproducibility and redundancy depended on both the CT machine and the CT image type (average cine 4D-CT imaging vs. end-exhale cine 4D-CT imaging vs. helical inspiratory breath-hold 3D CT). For each image type, a set of cross-machine reproducible, non-redundant, and informative image features was identified. Compared to end-exhale 4D-CT and breath-hold 3D-CT, average 4D-CT derived image features showed superior multi-machine reproducibility and are the best candidates for clinical correlation.

Multi-criteria selection of structural adhesives to bond ABS parts obtained by rapid prototyping

One of the most used methods in rapidprototyping is Fused Deposition Modeling (FDM), which provides components with a reasonable strength in plastic materials such as ABS and has a low environmental impact. However, the FDM process exhibits low levels of surface finishing, difficulty in getting complex and/or small geometries and low consistency in “slim” elements of the parts. Furthermore, “cantilever” elements need large material structures to be supported. The solution of these deficiencies requires a comprehensive review of the three-dimensional part design to enhance advantages and performances of FDM and reduce their constraints. As a key feature of this redesign a novel method of construction by assembling parts with structuraladhesive joints is proposed. These adhesive joints should be designed specifically to fit the plastic substrate and the FDM manufacturing technology. To achieve this, the most suitable structuraladhesiveselection is firstly required. Therefore, the present work analyzes five different families of adhesives (cyanoacrylate, polyurethane, epoxy, acrylic and silicone), and, by means of the application of technical multi-criteria decision analysis based on the analytic hierarchy process (AHP), to select the structuraladhesive that better conjugates mechanical benefits and adaptation to the FDM manufacturing process

Comparison of metaheuristic strategies for peakbin selection in proteomic mass spectrometry data

Mass spectrometry (MS) data provide a promising strategy for biomarker discovery. For this purpose, the detection of relevant peakbins in MS data is currently under intense research. Data from mass spectrometry are challenging to analyze because of their high dimensionality and the generally low number of samples available. To tackle this problem, the scientific community is becoming increasingly interested in applying feature subset selection techniques based on specialized machine learning algorithms. In this paper, we present a performance comparison of some metaheuristics: best first (BF), genetic algorithm (GA), scatter search (SS) and variable neighborhood search (VNS). Up to now, all the algorithms, except for GA, have been first applied to detect relevant peakbins in MS data. All these metaheuristic searches are embedded in two different filter and wrapper schemes coupled with Naive Bayes and SVM classifiers.

Consciousness, Action Selection, Meaning and Phenomenic Anticipation

Phenomenal states are generally considered the ultimate sources of intrinsic motivation for autonomous biological agents. In this article, we will address the issue of the necessity of exploiting these states for the design and implementation of robust goal-directed artificial systems. We will provide an analysis of consciousness in terms of a precise definition of how an agent "understands" the informational flows entering the agent and its very own action possibilities. This abstract model of consciousness and understanding will be based in the analysis and evaluation of phenomenal states along potential future trajectories in the state space of the agents. This implies that a potential strategy to follow in order to build autonomous but still customer-useful systems is to embed them with the particular, ad hoc phenomenality that captures the system-external requirements that define the system usefulness from a customer-based, requirements-strict engineering viewpoint.

Regularized logistic regression and multi-objective variable selection for classifying MEG data

This paper addresses the question of maximizing classifier accuracy for classifying task-related mental activity from Magnetoencelophalography (MEG) data. We propose the use of different sources of information and introduce an automatic channel selection procedure. To determine an informative set of channels, our approach combines a variety of machine learning algorithms: feature subset selection methods, classifiers based on regularized logistic regression, information fusion, and multiobjective optimization based on probabilistic modeling of the search space. The experimental results show that our proposal is able to improve classification accuracy compared to approaches whose classifiers use only one type of MEG information or for which the set of channels is fixed a priori.

Collaborative data stream mining in ubiquitous environments using dynamic classifier selection

In ubiquitous data stream mining applications, different devices often aim to learn concepts that are similar to some extent. In these applications, such as spam filtering or news recommendation, the data stream underlying concept (e.g., interesting mail/news) is likely to change over time. Therefore, the resultant model must be continuously adapted to such changes. This paper presents a novel Collaborative Data Stream Mining (Coll-Stream) approach that explores the similarities in the knowledge available from other devices to improve local classification accuracy. Coll-Stream integrates the community knowledge using an ensemble method where the classifiers are selected and weighted based on their local accuracy for different partitions of the feature space. We evaluate Coll-Stream classification accuracy in situations with concept drift, noise, partition granularity and concept similarity in relation to the local underlying concept. The experimental results show that Coll-Stream resultant model achieves stability and accuracy in a variety of situations using both synthetic and real world datasets.

The future of manufacturing

Anyone who looks at the title of this special issue will agree that the intent behind the preparation of this volume was ambitious: to predict and discuss “The Future of Manufacturing”. Will manufacturing be important in the future? Even though some sceptics might say not, and put on the table some old familiar arguments, we would strongly disagree. To bring subsidies for the argument we issued the call-for-papers for this special issue of Journal of Manufacturing Technology Management, fully aware of the size of the challenge in our hands. But we strongly believed that the enterprise would be worthwhile. The point of departure is the ongoing debate concerning the meaning and content of manufacturing. The easily visualised internal activity of using tangible resources to make physical products in factories is no longer a viable way to characterise manufacturing. It is now a more loosely defined concept concerning the organisation and management of open, interdependent, systems for delivering goods and services, tangible and intangible, to diverse types of markets. Interestingly, Wickham Skinner is the most cited author in this special issue of JMTM. He provides the departure point of several articles because his vision and insights have guided and inspired researchers in production and operations management from the late 1960s until today. However, the picture that we draw after looking at the contributions in this special issue is intrinsically distinct, much more dynamic, and complex. Seven articles address the following research themes: 1.new patterns of organisation, where the boundaries of firms become blurred and the role of the firm in the production system as well as that of manufacturing within the firm become contingent; 2.new approaches to strategic decision-making in markets characterised by turbulence and weak signals at the customer interface; 3.new challenges in strategic and operational decisions due to changes in the profile of the workforce; 4.new global players, especially China, modifying the manufacturing landscape; and 5.new techniques, methods and tools that are being made feasible through progress in new technological domains. Of course, many other important dimensions could be studied, but these themes are representative of current changes and future challenges. Three articles look at the first theme: organisational evolution of production and operations in firms and networks. Karlsson's and Skold's article represent one further step in their efforts to characterise “the extraprise”. In the article, they advance the construction of a new framework, based on “the network perspective” by defining the formal elements which compose it and exploring the meaning of different types of relationships. The way in which “actors, resources and activities” are conceptualised extends the existing boundaries of analytical thinking in operations management and open new avenues for research, teaching and practice. The higher level of abstraction, an intrinsic feature of the framework, is associated to the increasing degree of complexity that characterises decisions related to strategy and implementation in the manufacturing and operations area, a feature that is expected to become more and more pervasive as time proceeds. Riis, Johansen, Englyst and Sorensen have also based their article on their previous work, which in this case is on “the interactive firm”. They advance new propositions on strategic roles of manufacturing and discuss why the configuration of strategic manufacturing roles, at the level of the network, will become a key issue and how the indirect strategic roles of manufacturing will become increasingly important. Additionally, by considering that value chains will become value webs, they predict that shifts in strategic manufacturing roles will look like a sequence of moves similar to a game of chess. Then, lastly under the first theme, Fleury and Fleury develop a conceptual framework for the study of production systems in general derived from field research in the telecommunications industry, here considered a prototype of the coming information society and knowledge economy. They propose a new typology of firms which, on certain dimensions, complements the propositions found in the other two articles. Their telecoms-based framework (TbF) comprises six types of companies characterised by distinct profiles of organisational competences, which interact according to specific patterns of relationships, thus creating distinct configurations of production networks. The second theme is addressed by Kyläheiko and SandstroÍm in their article “Strategic options based framework for management of dynamic capabilities in manufacturing firms”. They propose a new approach to strategic decision-making in markets characterised by turbulence and weak signals at the customer interface. Their framework for a manufacturing firm in the digital age leads to active asset selection (strategic investments in both tangible and intangible assets) and efficient orchestrating of the global value net in “thin” intangible asset markets. The framework consists of five steps based on Porter's five-forces model, the resources-based view, complemented by means of the concepts of strategic options and related flexibility issues. Thun, GroÍssler and Miczka's contribution to the third theme brings the human dimension to the debate regarding the future of manufacturing. Their article focuses on the challenges brought to management by the ageing of workers in Germany but, in the arguments that are raised, the future challenges associated to workers and work organisation in every production system become visible and relevant. An interesting point in the approach adopted by the authors is that not only the factual problems and solutions are taken into account but the perception of the managers is brought into the picture. China cannot be absent in the discussion of the future of manufacturing. Therefore, within the fourth theme, Vaidya, Bennett and Liu provide the evidence of the gradual improvement of Chinese companies in the medium and high-tech sectors, by using the revealed comparative advantage (RCA) analysis. The Chinese evolution is shown to be based on capabilities developed through combining international technology transfer and indigenous learning. The main implication for the Western companies is the need to take account of the accelerated rhythm of capability development in China. For other developing countries China's case provides lessons of great importance. Finally, under the fifth theme, Kuehnle's article: “Post mass production paradigm (PMPP) trajectories” provides a futuristic scenario of what is already around us and might become prevalent in the future. It takes a very intensive look at a whole set of dimensions that are affecting manufacturing now, and will influence manufacturing in the future, ranging from the application of ICT to the need for social transparency. In summary, this special issue of JMTM presents a brief, but undisputable, demonstration of the possible richness of manufacturing in the future. Indeed, we could even say that manufacturing has no future if we only stick to the past perspectives. Embracing the new is not easy. The new configurations of production systems, the distributed and complementary roles to be performed by distinct types of companies in diversified networked structures, leveraged by the new emergent technologies and associated the new challenges for managing people, are all themes that are carriers of the future. The Guest Editors of this special issue on the future of manufacturing are strongly convinced that their undertaking has been worthwhile.

Customer value based entry decision in International markets the concept of International Added Customer Equity

Market entry decisions are some of a firm's most important long-term strategic choices. Still, the international marketing literature has not yet fully incorporated the idea of relationship marketing in general, and the customer value concept in particular, as a basis for market entry decisions. This article presents some conceptual ideas about a customer value based market selection model. The metric International Added Customer Equity (IACE), a straightforward decision criterion derived from the customer equity concept is presented as an additional decision criterion for export market selection and ultimately market entry.

Managing customer relationship management projects:the case of a large French telecommunication company

Customer relationship management (CRM) implementation projects reflect a growing conceptual shift from the traditional engineering view of projects. Such projects are complex and risky because they call for both organisational and technological changes. This requires effective project management across various phases of the implementation process. However, few empirical researches have dealt with these project management issues. The aim of this research is to investigate how a “project team” manages CRM implementation projects successfully, across the different phases of the implementation process. We conducted an in-depth case study of the “Firm-Clients Branch” of a large telecommunications company in France. The findings show that, to manage CRM implementation projects successfully, an integrated and balanced approach is required. This involves appropriate system selection, effective process re-engineering and further development of organizational structures. We highlight the need for a “technochange approach” to achieve successful organisational transition and effective CRM implementation. The study reveals that the project team plays a central role throughout the implementation phases. Furthermore the effectiveness of technochange depends on project team performance, technology efficiency and close coordination with stakeholders.

Dynamic decision-making based on NFR for managing software variability and configuration selection

Due to dynamic variability, identifying the specific conditions under which non-functional requirements (NFRs) are satisfied may be only possible at runtime. Therefore, it is necessary to consider the dynamic treatment of relevant information during the requirements specifications. The associated data can be gathered by monitoring the execution of the application and its underlying environment to support reasoning about how the current application configuration is fulfilling the established requirements. This paper presents a dynamic decision-making infrastructure to support both NFRs representation and monitoring, and to reason about the degree of satisfaction of NFRs during runtime. The infrastructure is composed of: (i) an extended feature model aligned with a domain-specific language for representing NFRs to be monitored at runtime; (ii) a monitoring infrastructure to continuously assess NFRs at runtime; and (iii) a exible decision-making process to select the best available configuration based on the satisfaction degree of the NRFs. The evaluation of the approach has shown that it is able to choose application configurations that well fit user NFRs based on runtime information. The evaluation also revealed that the proposed infrastructure provided consistent indicators regarding the best application configurations that fit user NFRs. Finally, a benefit of our approach is that it allows us to quantify the level of satisfaction with respect to NFRs specification.

Modelling question selection behaviour in online communities

Value of online Question Answering (QandA) communities is driven by the question-answering behaviour of its members. Finding the questions that members are willing to answer is therefore vital to the effcient operation of such communities. In this paper, we aim to identify the parameters that cor- relate with such behaviours. We train different models and construct effective predictions using various user, question and thread feature sets. We show that answering behaviour can be predicted with a high level of success.