Aquaculture - Water Quality Dataset

This dataset is useful to train and test deep learning models that developed to assess the quality of the water in fish ponds based on the parameters like Temperature, Turbidity, Dissolved Oxygen, Biochemical oxygen demand (BOD), $CO_{2}$, pH, Alkalinity, Hardness, Calcium, Ammonia, Nitrite, Phosphorus, $H_{2}S$ and Plankton. The quality of water in fish ponds is classified in three categories like Excellent, Good and Poor quality. This dataset is prepared based on the threshold values of each input feature for acceptable range, desirable range and stress range for the growth of fishes in ponds. This AWD dataset consists of three different quality samples. They are excellent represented with 0, good quality is labelled with 1 and poor quality id labelled with 2. The developed dataset consists total 1500 poor quality water samples, 1400 excellent quality water samples and 1400 good quality water samples. The developed dataset consists of total 4300 samples with 14 input feature and one output label column.

本数据集可用于训练与测试旨在通过温度、浊度、溶解氧、生化需氧量（Biochemical Oxygen Demand, BOD）、二氧化碳、pH值、碱度、硬度、钙、氨、亚硝酸盐、磷、硫化氢及浮游生物等参数评估鱼塘水质的深度学习模型。鱼塘水质被划分为优秀、良好、差劣三个等级。本数据集依据鱼类池塘养殖生长所需的各输入特征的可接受范围、适宜范围及胁迫范围的阈值构建而成。该AWD数据集包含三类水质样本：优秀水质以0标记，良好水质以1标记，差劣水质以2标记。本数据集总计包含4300条样本，其中差劣水质样本1500条、优秀水质样本1400条、良好水质样本1400条，共涵盖14个输入特征列与1个输出标签列。