IMPROVING ON DEMAND COLD CHAIN FORECASTING MODEL BASED ON DEEP LEARNING USING OPTIMIZED SUSTAINABLE NEURAL NETWORK
European Journal of Molecular & Clinical Medicine,
2020, Volume 7, Issue 11, Pages 2101-2122
AbstractBig data processing is a crucial one because, huge volume of data resource to increasing the processing cost. Cold chain is a developing logistics approach relatively on supply chain management which is storing, transporting economical needs of products maintenance in various levels. By maintaining the large number of information in big data processing leads more complex to predict the data, especially agricultural information process depends the cold stock product in supply chain management. By the fact increasing features to analyses the data is more complex to produce good classification results to make future prediction. The prediction and classification is a major issue that remains to be solved based on available data. So we need a reducing framework to classify the agriculture oriented information processing. Now a day’s Deep learning is a tremendous data analysis techniques to improve the prediction support to clod chain management. To resolve such a type of problems,to propose a Subset Reduct Core Spider Optimization Model (SRCSO) with optimized with social spider which is used to improve the feature selection which is for classification performance. To deploy a sustainable neural classifier for an effective classification using Optimized Cuckoo Genetic (OCG) search features to recommend the modified artificial neural network. The performance accuracy for effective categorization using marginal relevance weightage by spectral classification to improve the classification accuracy. The selected features are trained as recurrent neurons which is intended as sigmoid activation function. This selects the optimal values get closer to the neuronsas search with optimum weights. The target result reached the classifier produce the resultant to categorize the agro classes for transportation recommendation. The proposed system produce higher performance in sensitivity and specification result than any other previous system.
- Article View: 143
- PDF Download: 207