الفهرس 
Title: A Machine Learning Technique for Detecting Water Quality Based on Algal Growth
AbstractOnly 14 pages are availabe for public view
 |  | from | 137 |  |  |
| | | from | 137 | | |
Abstract
Algae are receiving attention in the drinking water industry as a result of the continuing eutrophication of surface water supplies. Algae and algal metabolites greatly impact the treatment of potable water by (1) clogging intake screens, (2) increasing coagulant demand, (3) shortening filter runs, (4) increasing filter backwash water requirements, (5) increasing chlorine demand and disinfection byproduct formation, (6) producing unpleasant tastes and odors, (7) producing toxins, and (8) increasing the microbial regrowth potential in distribution systems. The proposed research model compiles data collected from 14 stations in Cairo from January to December from year 2012 to 2015. The stations are El Tebeen, El- Kafr El-Elwy, North Helwan, El- Maadi, El -Fustat, El Roda, Rod El-Farag, El- Ameria, Moustorod, Shubra El-Khima, El-Obour, El-Marg, El-Asher (1) and El Asher (2). The data of each station has 20 attributes and 1 attribute for diatoms (class label). After analyzing the data of algae, the results demonstrated the most three types of spread algae in Egypt: green, blue green algae and diatoms which is the most common of the three in Cairo. The diatoms have the highest percentage to be focused on toestablish a prediction model for diatoms. The data sets were categorized into three types: physical, chemical and biological for forecasting the movement and growth of algae in river systems as particularly important for operational managers responsible for the distribution and supply of potable water. Algae affect the taste and smell of water and pose considerable filtration problems at water treatment plants. The final results in the proposed model referred to Meta learning techniques that used were the best from using regression techniques