Accuracy Analysis Mechanism for Agriculture Data Using the Ensemble Neural Network Method

Abstract

Abstract:With the rise and development of information technology (IT) services, the amount of data generated is rapidly increasing. Data from many different places are inconsistent. Data capture, storage and analysis have major challenges. Most data analysis methods are unable to handle such large amounts of data. Many studies employ neural networks, mostly specifying the number of hidden layers and neurons according to experience or formula. Different sets of network topologies have different results, and the best network model is selected. This investigation proposes a system based on the ensemble neural network (ENN). It creates multiple network models, each with different numbers of hidden layers and neurons. A model that does not achieve the accuracy rate is discarded. The proposed system derives the weighted average of all remaining network models to improve the accuracy of the prediction. This study applies the proposed method to generate agricultural yield predictions. The agricultural production process in Taiwan is more complex than those of manufacturing or other industries. The Council of Agriculture provides agricultural forecasting primarily based on the planted area and experience to predict the yield, but without consideration of the overall planting environment. This work applies the proposed data analysis method to agriculture. The method based on ENN has a much lower error rate than traditional back-propagation neural networks, while multiple regression analysis has an error rate of 12.4%. Experimental results reveal that the ENN method is better than traditional back-propagation neural networks and multiple regression analysis.

Weather and Crop Prediction Using Modified Self Organizing Map for Mysore Region

“…Chen, and P.Y. Tsai, [15] investigated a new system based on the ENN. This system derives the weighted average of all remaining network models to improve the accuracy of the prediction.…”

“…Here, the existing methods: KNN, SOM and E-NN [15] and the advanced scheme (W-SOM) were evaluated in three different combinations of testing and training percentage like 40% training and 60% testing, 70% training and 30% testing, 80% training and 20% testing of collected data, which is displayed in Table 3. Table 3 shows the performance evaluation of existing methodologies and the proposed method.…”

Insurance risk assessment in the face of climate change: Integrating data science and statistics

“…Pantazi, Moshou, Alexandridis, Whetton, and Mouazen () use supervised Kohonen networks, counter‐propagation artificial networks, and XY‐fusion networks to forecast winter wheat yield in 2013 on a small scale—22 ha in Bedfordshire, U.K., using NDVI and a number of soil parameters that are typically unavailable for insurance companies (soil moisture content, pH, and nutrient concentrations). Kung, Kuo, Chen, and Tsai () propose to combine several neural networks in an ensemble to improve predictions of tomato yields based on a number of meteorological factors. Overall, forecasting results that are obtained with machine learning for larger geographical areas and could be used by insurance companies are yet scarce (see Table combining information on these and the earlier mentioned studies).…”

Combining the Technology Acceptance Model and Uses and Gratifications Theory to examine the usage behavior of an Augmented Reality Tour-sharing Application

“…A query-answering server is proposed with the proposed novel random neural networks (RNNs) [44] to analyze the query from users and classify the query into question class for searching answers. The processes of query-answering server include: (1) receiving a query sentence; (2) using segmentation tool to retrieve words; (3) computing term frequency and inverse document frequency (TF-IDF); (4) using RNNs to analyze question and answer; and (5) replying the answer of user's query.…”