No abstract

This paper presents a dragonfly optimization (DFO) based ANN model for predicting India's primary fuel demand. It involves socio-economic indicators such as population and per capita GDP and uses two ANNs, which are trained through DFO algorithm. The method optimizes the connection weights of ANN models through effectively searching the problem space in finding the global best solution. Primary fuel demands during the years 1990-2012 forms the data for training and validating the model. The proposed model requires an input, the year of the forecast, and predicts the primary fuels' demand. The forecasts up to the year 2025 are compared with that of the RM with a view to illustrate the accuracy. General Termsforecasting, primary energy fuels Keywords artificial neural network, dragonfly optimization, regression model.

Prediction of stock market return is an important issue in finance. Fuzzy and Artificial neural networks have been used in stock market prediction during the last decade. Studies were performed for the forecast of stock index values as well as daily direction of change in the index. This work compares fuzzy and arrangement of ANN model and makes these models to train with the past 5 years stock price datasets of various companies like (TCS, HCL) and the prediction of future stock price of company has been found. Membership functions (LOW, MEDIUM, HIGH) based fuzzy model will give recommendation for investor which says the current situation of the stock market. The Root Mean Square error (RMSE), Mean Absolute Performance (MAPE) metrics calculates the error rate value of each model. The proposed Hybrid Network model has expecting to given high performance.

This paper presents Neuro-Fuzzy approach for forecasting analysis in power load. Forecasting the power load is a difficult task for a country and both positive and negative load forecasting makes a big problem for the country. An approach that Neuro-Fuzzy model is proposed for forecast power load in this paper. The proposed model a fuzzy back propagation network is constructed and then a fuzzy intersection is applied and after that de-fuzzify the result to generate a crisp value by using Radial Basis Function network (RBF). The proposed model improves the accuracy of power load forecasting. The forecasted results obtained by neuro-fuzzy method were compared with the Artificial Neural Network by using Mean Absolute Percentage Error (MAPE) to measure accuracy of the result. The experimental result shows that the neuro-fuzzy implementations have more accuracy.