الفهرس 
Soft computing approaches for knowledge discovery from big data
Abstract EnglishOnly 14 pages are availabe for public view
 |  | from | 161 |  |  |
| | | from | 161 | | |
Abstract
Commonly, Big Data is considered one of the most important signs of the recent decades Because everything around us has become a source of data. So that incase of the ability to analyze such huge amount of data it will has a big return on the organization that own such data. Although the Big Data is considered a treasure of valuable knowledge, it is not easy to mine the hidden knowledge for such big complex and imprecise treasure using traditional knowledge discovery techniques. Actually, such challenges are returned to the characteristics of Big Data.The first characteristics is the huge volume of datasets. Secondly, the velocity that such data required for processing. Thirdly, the variety of data types that may be structured, non-structured and semi-structured. Fourthly, the veracity of data and uncertainty that the datasets contain. Fifthly, the value that is hide inside the hills of data. Accordingly, Nowadays, the data scientists do not spare any effort for developing new techniques that help in discovering the valuable knowledge from Big Data. This thesis proposes two soft computing approaches for knowledge discovery from Big Data.The first approach is a fuzzy approach based on Hadoop for discovering crops plantation knowledge from the Agro-climatic historical database of the years from 1901 to 2016 of Egypt. Generally, the proposed approach offers a list of scenarios for crops plantation dates with a suitability degree for the suggested scenarios. Also, it helps handling the whole crops plantation life cycle from all other aspects with water requirements follow up depending on the data streamed from weather station.The proposed approach has been tested on a set of crops with cooperation of researchers from Cairo university and Climate Change Information and Renewable Energy and Expert Systems Center from agricultural research center. The results of testing the propped approach show the added value of it against other related works because it not only provides a set of planation date but also the expected diseases, the harvesting dates and the expected crops water requirements. For example, the proposed approach discovered a set of plantation dates with suitability degrees reaches to 96% for Wheat, 97% for Soybean and 95% for Bean