Detection in Time Series Data Mining

For many data mining applications, finding the outliers is more interesting than finding the common patterns of the data. Outliers are frequently adapted in time series data mining analysis. The main objective of this paper, outliers on forecasting in agricultural production is analyzed. Outliers in time series data was carried out by Fox (1972). Outlier detection has been used for detect and, where appropriate, remove inconsistent observations from data. The original outlier detection methods were arbitrary but new, Principled and systematic techniques are used, drawn from the full scope of computer science and statistics. In agricultural production outliers are initially detected and then forecast using ARIMA model. Predictions made after detecting outliers are compared with numerically and graphically the predictions made before detecting outliers.