The application of adaptive partitioned random search in feature selection problem

Feature selection is one of important and frequently used techniques in data preprocessing. It can improve the efficiency and the effectiveness of data mining by reducing the dimensions of feature space and removing the irrelevant and redundant information. Feature selection can be viewed as a global optimization problem of finding a minimum set of M relevant features that describes the dataset as well as the original N attributes. In this paper, we apply the adaptive partitioned random search strategy into our feature selection algorithm. Under this search strategy, the partition structure and evaluation function is proposed for feature selection problem. This algorithm ensures the global optimal solution in theory and avoids complete randomness in search direction. The good property of our algorithm is shown through the theoretical analysis.

Will Bangladesh's economic growth solve its environmental problems?

Bangladesh has experienced rising GDP and rising per capita incomes now for at least three decades. This article considers whether its continuing economic growth is likely to solve its environmental problems. In doing so, it critically considers the application to Bangladesh of Environmental Kuznets Curve relationships and applies other macro-methods of assessing the relationship between economic growth and the environment to Bangladesh’s situation. The consequences of Bangladesh's economic reforms for the economic welfare of Bangladeshis and the state of Bangladesh's environment are also examined. Particular attention is given to environmental change in agriculture in the light of Bangladesh economic growth, reforms and proposed growth strategy. Doubts are expressed about the environmental benefits claimed by the Bangladeshi Government for its agricultural development strategy. Indeed, it may exacerbate many existing environmental problems, such as depletion of soil fertility and water supplies, already present.

The Environment and the Selection of Aquaculture Species and Systems: An Economic Analysis

Environmental conditions play a significant role in the economic success of aquaculture. This article classifies environmental factors in a way that facilitates economic analysis of their implications for the selection of aquaculture species and systems. The implication of on-farm as on-site environmental conditions for this selection are considered first using profit-possibility frontiers and taking into account the biological law of environmental tolerance. However, in selecting, recommending and developing aquaculture species and systems, it is often unrealistic to assume the degree of managerial efficiency implied by the profit-possibility function. It is appropriate to take account of the degree of managerial inefficiency that actually exists, not all of which may be capable of being eliminated. Furthermore, experimental R&D should be geared to on-farm conditions, and the variability of these conditions needs to be taken into account. Particularly in shared water bodies, environmental spillovers between aquaculturalists can be important and as shown theoretically, can influence the socially optimal selection of aquaculture species and systems. Similarly, aquaculture can have environmental consequences for the rest of the community. The social economic implications of this for the selection of aquaculture species and systems are analyzed. Some paradoxical results are obtained. For example, if the quality of social governance of aquaculture is poor, aquaculture species and systems that cause a slow rate of environmental deterioration may be socially less satisfactory than those that cause a rapid rate of such deterioration. Socially optimal choice of aquaculture species and systems depends not only on their biophysical characteristics and market conditions but also on the prevailing state of governance of aquaculture. Failure to consider the last aspect can result in the introduction of new aquaculture species (and systems) doing more social harm than good.