Abstract
Nowadays there is a visible shift from product-oriented economy towards service-oriented economy. This shift relates to a huge pressure on innovation in services and new offerings to the market. Concurrently with this shift, capabilities of IT/software tools are increasing rapidly. Such movements are opening new space in the dynamic market and are encouraging new ventures to be founded. A challenge for managers and leaders is to design processes and set/manage an internal environment that inherently supports creativity and cooperation between team members in new ventures. In this paper, we characterize a management model named INnovation&COoperation (INCO) suitable for innovative software start-ups. We propose that the management processes and leadership of such enterprises have to have six distinctive qualities to support innovation and cooperation of team members. Proposed qualities are as follows: (1) network-hierarchical organizational structure; (2) assessment function of added value; (3) context awareness; (4) equal opportunities; (5) portfolio project management and uniformity; (6) embedded feedback. We believe that these qualities are essential for setting up a dynamic and flexible enterprise in the long-term perspective. A case study of a spin-off from Masaryk University, Czech Republic, is used to illustrate and validate this proposal in practice. This IT/software SME follows a network-hierarchical organizational structure, project team members are assigned dynamic roles and continuous feedback is provided to control and enhance collaboration and creativity.

Abstract
This paper explores the evolution of productivity of the artisanal dredge fleet that operates in the south coast of Portugal. This fleet is considered to be one of the most significant in the sector, essentially due to the volume and value of its catches. In this context, the study carried out sought in first place to determine variation in productivity over a time window of 10 years (between 1995 and 2004). Secondly, it sought to distinguish the performance of local and coastal vessels comprising the chosen fleet, The performance of the five homeports in the Algarve coast was also compared. We used Malmquist indexes to measure productivity change and explored the impact of changes in stock conditions and in regulatory policies on productivity levels. During the time period analysed, the measures defined by regulatory entities included allowing the use of a new type of dredge and the establishment of fishing quotas per vessel and per species. Finally, the fishing quotas allowed for each vessel were confronted with the captures officially declared.

Abstract
This paper develops a method based on data envelopment analysis (DEA) for efficiency assessments taking into account the effect of non-discretionary factors. A typology that classifies the non-discretionary factors into two groups is proposed: the factors that characterize the external conditions where the decision making units (DMUs) operate (external factors), and the factors that are internal to the production process but cannot be controlled by the decision makers (internal factors). This paper proposes an enhanced DEA model that accommodates non-discretionary inputs and outputs and treats them differently depending on their classification as internal or external to the production process. This generalized model integrates the previous approaches for dealing with non-discretionary variables described in the DEA literature. The model defines the efficient frontier based exclusively on the discretionary variables and internal non-discretionary factors, but the potential peers of each DMU are restricted to other units facing comparable external conditions (represented by the external non-discretionary factors). The peer selection criteria implemented in the DEA model is informed by decision makers' opinion, The applicability of the model developed is illustrated with a real-world assessment of retailing stores.

Abstract

Abstract

Abstract
Integration methods for ensemble learning can use two different approaches: combination or selection. The combination approach (also called fusion) consists on the combination of the predictions obtained by different models in the ensemble to obtain the final ensemble predication. The selection approach selects one (or more) models from the ensemble according to the prediction performance of these models on similar data from the validation set. Usually, the method to select similar data is the k-nearest neighbors with the Euclidean distance. In this paper we discuss other approaches to obtain similar data for the regression problem. We show that using similarity measures according to the target values improves results. We also show that selecting dynamically several models for the prediction task increases prediction accuracy comparing to the selection of just one model.