Dataset Macro & Micro IOT Anfis

The dataset used in this study contains real-time nutrient measurements of sugarcane plants collected through an IoT sensor system. The data includes variables such as temperature, soil moisture, pH, electrical conductivity, as well as macronutrient (N, P, K) and micronutrient (Zn, Mn, Fe) concentrations. This dataset is provided in a CSV format, which is machine-readable and ready for analysis. The source code for the Adaptive Neuro-Fuzzy Inference System (ANFIS) model is hosted on GitHub and is used to estimate nutrient requirements for sugarcane plants. Written in MATLAB, this code involves regression based on ANFIS to model the nonlinear relationships between soil conditions, environmental factors, and nutrient demands. The code also includes preprocessing of raw data, training the ANFIS model, and evaluating the prediction results using performance metrics such as MAE, RMSE, and R². The uploaded files consist of the raw dataset (dataset.csv) and the MATLAB code for the ANFIS model (dataset.m), which is available via the GitHub URL. This dataset and code are intended for precision agriculture research aimed at optimizing fertilization strategies based on sensor data. The source code can be used to replicate the model and conduct further analysis on nutrient data for crops using an ANFIS-based AI approach.

本研究所采用的数据集，为通过物联网（IoT）传感器系统采集的甘蔗植株实时养分监测数据。数据包含温度、土壤含水量、pH值、电导率，以及大量元素（N、P、K）与微量元素（Zn、Mn、Fe）的浓度等变量。该数据集以逗号分隔值（CSV）格式存储，具备机器可读性，可直接用于分析工作。自适应神经模糊推理系统（Adaptive Neuro-Fuzzy Inference System, ANFIS）模型的源代码托管于GitHub平台，该模型用于估算甘蔗植株的养分需求。此代码基于MATLAB编写，通过基于ANFIS的回归建模，拟合土壤条件、环境因子与养分需求之间的非线性关联。代码还涵盖原始数据预处理、ANFIS模型训练，以及采用平均绝对误差（Mean Absolute Error, MAE）、均方根误差（Root Mean Squared Error, RMSE）和决定系数（Coefficient of Determination, R²）等性能指标对预测结果进行评估的完整流程。本次上传的文件包含原始数据集（dataset.csv）与ANFIS模型的MATLAB代码（dataset.m），相关代码可通过指定的GitHub链接获取。本数据集与代码旨在服务于精准农业研究，通过传感器数据优化施肥策略。该源代码可用于复现该模型，并依托基于ANFIS的人工智能方法，针对作物养分数据开展进一步分析研究。