Search

Hexavalent chromium is one of the most toxic and carcinogenic species known and can be released into the environment from several sources. In Sarigkiol basin (N Greece) the presence of Cr(VI) in soil, sediments and groundwater may originate from both natural (ophiolitic rocks and their weathering products) and anthropogenic (dispersed fly ash produced from lignite power plants) sources. In this study, the distribution of contents and origin of environmentally available Cr(VI) in soils, sediments, regoliths and fly ash of Sarigkiol basin is presented. Detailed geochemical and mineralogical studies were performed on soil samples (up to 1 m) and regoliths, while leaching tests were also applied to fresh and old fly ash samples. Leachable chromium from soil and sediment samples generally increased with depth and the highest concentrations were observed near to the power plant of Agios Dimitrios. The speciation of chromium in leachates revealed that Cr(VI) concentrations accounted for more than 96% of total Cr. Leaching tests of regoliths established that the natural contribution of Cr(VI) is up to 14 μg kg−1. Therefore, the measurement of higher concentrations (up to 80 μg kg−1) of environmentally available Cr(VI) in soils and sediments can be attributed to the impact/presence of dispersed fly ash in the soils and sediments of the same area. This was also supported by the low correlation recorded between environmentally available chromium and Cr-bearing minerals (mainly serpentine and talc). The influenced zone is located in the eastern part of the basin near the local power plant and surrounds an open conveyor belt that transfers fly ash to an open temporary storage pit. This zone overlies an unconfined porous aquifer thus explaining the elevated concentrations of Cr(VI) in groundwater (up to 120 μg L−1) previously reported in this area.

We propose a knowledge framework for garment recommendations, which is based on two pillars. The first pillar incorporates knowledge about aspects of fashion, such as materials, garments, colours, body types, facial features, social occasion etc., as well as their interrelations, with the purpose of providing personalised recommendations. The said knowledge is encoded in the form of an owl ontology, the origin of which is attributed to fashion experts. Moreover, in commercial fashion sites, customers purchase garments of various types. Because of that, interesting patterns in their purchase behaviour can be sought, and thus groups of garments that tend to be purchased together can be discovered. This forms the second pillar, that can be used to enhance the first pillar with community based garment recommendations. This paper is the description and integration of the aforementioned pillars in a knowledge framework.

The article introduces a method, which is based on reinforcement learning, for extracting rules of the form if-then-else from a labeled data-set. The constituent parts of a rule are the input dimensions of the labeled data-set, each accompanied by an appropriate interval of activation, and a label which stands for class membership. Initially, the input space is partitioned using tiles. The algorithm tries to compose the largest possible orthogonal intervals out of tiles. After the creation of intervals for each dimension the rule receives credit for its classification ability. This credit with the aid of reinforcement will be used to improve its constituent parts. The effectiveness of the proposed method has been tested on five different classification problems: the Iris data set, the Concentric data, the 4 Gaussians, the Pima-Indians set and the Image Segmentation data set.

We propose a neural network method for the generation of symbolic expressions using reinforcement learning. Usually, the symbolic form expressed in terms of a calculus (propositional, first-order, lambda, etc.) is deemed comprehensible by humans and it is necessary as far as the acceptance of neural networks is concerned. According to the proposed method, a human decides on the kind and number of primitive functions which, with the appropriate composition (in the mathematical sense), can represent a mapping between two domains. The appropriate composition is achieved by an agent which tries many compositions and receives a reward depending on the quality of the composed function. Naturally, the learning agent (which in our case is a recurrent neural net) must perform the credit assignment task. Results are encouraging concerning the derivation of simple arithmetic expressions.