Search

Variable selection is one of the most important contents of machine learning research and a large number of variable selection methods have been proposed at present. From deep learning perspective, we propose a new variable selection method based on multi-layer artificial neural networks in this paper. The method propagates errors gained form a trained network model through the network from output layer to input layer and the values gained by input layer give information about the relative contribution. The variables with big contribution value are selected, which we call neural inverse propagation (NIP). Specifically, after training a neural network, the gained errors between the actual network outputs and the desired outputs go through the network from top to down and arrive at the input layer. The absolute values gained by input units provide information about the relative importance (contribution) of input units, which is that the larger the value is, the more important the corresponding input variable is. We remove k variables corresponding to the first k minimum absolute value gained by input units to the predictive variables, to perform variable selection. The efficiency of the proposed method is illustrated through its application to classification and regression tasks on a number of synthetic datasets and a real-world image classification dataset. [ABSTRACT FROM AUTHOR]

Nowadays, improving a crop yield ( C Y ) is an emerging and essential task to reduce food scarcity. Factors impacting C Y improvement include rising population, water shortage, fertilizer use, climate change and unprecedented insect attacks. To resolve these issues, a smart agriculture technique is proposed in this work. Internet of Things (IoT) sensor devices are used to collect data from farms, following which the fuzzy association rule-based classification technique classifies the data into two, valuable and nonvaluable. An improved artificial neural network (IANN) algorithm is applied to identify and analyze the factors involved in monitoring C Y 's. Thereafter, all valuable data pertaining to the type of seed, fertilizer and crop cost is stored in blocks to secure data and communication between members of the farming community. Finally, an edge computing device is used to store the blocks and transfer information. The valuable data collected is classified using the fuzzy association rule and analyzed using the IANN technique, both of which facilitate a comparison with the historical data so as to enable better decision making in terms of seed and fertilizer selection. Similarly, crop price is predicted through a comparison of present and historical yields. To overcome breaches in security, a blockchain technique is employed in this work to secure communication between farmers, investors and merchants. The investor dispatches instructions on the selection of the seed and fertilizer, as well as the crop cost, through the blockchain to the farmer and the merchant. Such secure communication bypasses third-party involvement and inconsistencies in the data. When compared to the traditional method, the proposed technique offers better accuracy and profits, right from seed selection to trading. The proposed IANN technique produced a higher yield than the traditional method with a profit of 51%, 35% and 20% for rice, bananas and flowers, respectively. Similarly, the IANN technique provides 99.15% prediction accuracy in terms of a profit analysis. The blockchain and edge computing-based transactions improve security and reduce transactional latency. The proposed system ensures sustainability and traceability in agriculture. [ABSTRACT FROM AUTHOR]

Artificial neural networks, consisting of many levels of nonlinearities, have been widely used to deal with various supervised learning tasks. At present, the most popular and effective training method is back-propagation algorithm (BP). Inspired by manifold regularization framework, we introduce a novel regularization framework, which aims at preserving the inter-object-distance of the data. Then a refined BP algorithm (IOD-BP) is proposed by imposing the proposed regularization framework into the objective function of BP algorithm. Comparative experiments on various benchmark classification tasks show that the new regularization BP method significantly improves the performance of BP algorithm in terms of classification accuracy. [ABSTRACT FROM AUTHOR]