Open Access Open Access  Restricted Access Subscription or Fee Access

Analysing K-Nearest Neighbor technique for Classification of Agricultural land Soils

R. Beulah, M. Ravichandran

Abstract


Soil is a significant input aspect of cultivation. The main intention of the effort work is to predict soil type using data mining classification techniques.  Soil kind is predicted using data mining classification techniques such as KNN. This classifier algorithm is functional to take out the knowledge from soil data and the soil types. In this paper, Data Mining and agricultural Data Mining are epigrammatic. The KNN model can produce more reliable results of this data and the RMSE, RSquared, MAE values. For solute the problems in Big Data, proficient methods can be formed that exploit Data Mining to develop the meticulousness of classification of huge top soil data sets.


Keywords


KNN, RMSE, RSquared, MAE

Full Text:

PDF

References


Jaganathan, P, Kuppuchamy, R, Rajkumar, N.(2010).“ A hybrid Feature Selection Method Based on Information Gain Combined with Correlation for Medical Dataset”, National Conference on Knowledge Discovery in Data mining and Data Warehousing,

http://avinashilingamkvk.org/Dist%20Profile.pdf

Ian H. Witten & Eibe Frank. (2005). ”Data Mining: Practical machine learning tools and techniques”, 2nd Edition, Morgan Kauf-mann, San Francisco.

Jaganathan, P, Kuppuchamy, R, Rajkumar, N.(2010).“ Information Gain Based Feature Selection Applied in KNN Classification”, UGC Sponsored National Conference on Emerging Trends in Computational Science and its Applications

Willmott, C. and Matsuura, K.: Advantages of the Mean Absolute Error (MAE) over the Root Mean Square Error (RMSE)


Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.