Table 7: Wilk’s Lambda
Table 8: Classification Results*
* 98.7 % original grouped classes correctly classified
Figure 1: Classification Map of Dharmapuri Figure 2: Classification Map of Krishnagiri
Figure 3: Classification Map of Namakkal Figure 4: Classification Map of Salem
In Table 1 provides the groupings done by cluster analysis. Figure 1 to 4 shows the groupings of crops into 3 clusters for each districts of the study period in Table 1, we rated the members in the first cluster as High Yield District (HYD), and the second as Moderate Yield District (MYD) and the third as Low Yield District (LYD).Crops belonging to HYD category are the ones that perform better than those of MYD and LYD.Similarly the crops belonging to MYD category are superior to those of LYD, indicating the members in the category LYD are at a low report in terms of the crops yield considered in the present analysis. Table 7 shows that Wilks' lambda is a measure of how well each function separates cases into groups. Smaller values of Wilks’ lambda indicate greater discriminatory ability of the function. The associated chi-square statistic tests the hypothesis that the means of the functions listed are equal across groups. The small significance value indicates that the discriminant function does better than chance at separating the groups. In this study, Salem district is significantly different fromother districts. The classification matrix is reported in table 8. The pictorial representation of GIS mapping is drawn using the standardized discriminant functions evaluated at the group centroids. From the GIS maps in Figure 5, it is evident that the three groups of rated crops are very well separated and represented in the GIS maps for four districts. The Spatial pattern of districts (Figure 5) shows that Salem district get High Yield, Krishnagiri and Namakkal get Moderate Yield and Dharmapuri gets Low yield performance during the study period. The low yieldmay be due to shortage of rainfall, Irrigated land usage and variations in the cropping pattern during the study period.
Figure 5: Spatial Patterns of Districts Map
MYD LYD HYD
In this paper we identified meaningful groups of districts that are classified as best with respect to their total yield in terms of crops using data mining and GIS techniques. An attempt is made to analyze the agriculture productivity yield data relating to major crops of North-western agro-climatic zone over a period of ten years from 2003 to 2012.The present analysis shows that only 3 groups could be meaningfully formed for each district.This indicates that only 3 types of yields existed over a period of ten years.Further, the districts find themselves classified into High Yield District (HYD), Moderate Yield District (MYD)and Low Yield District (LYD) categories depending on the crops. Agriculture Analysts can make use of these techniques of classifying, and the districts can project the yield on the basis of crops that has been considered in this study. Geographical Information System (GIS) may be used for crops yield, evaluation, and tracking changes in crops, among other uses. A generalization of the results is under investigation to obtain a set of 3 groups of yields for any given year.