Data Mining the Smartphone Manipulation Skills in a Coffee Farming Community: A Step for Risk Analysis

Abstract

The Philippines is largely an agricultural country, and the importance of coffee in the Philippines cannot be undervalued. However, the coffee plantations are generally confronted with various insect pests and diseases. This is the reason why authorities continue to look for solutions through technological applications for coffee farming. Smartphones are becoming a functional tool in agriculture because its mobility served as an advantage to agriculture. However, challenges regarding the level of ICT, particularly of smartphones technology usage among the rural community is low due to limited knowledge and skills. Thus, this study has the primary objective to apply data mining to the smartphone manipulation skills of possible users' dataset in the province of Quirino, Philippines. Specifically, it sought to determine the optimal number of the types of potential users and to identify the different types of possible users and their skills that emerged from the clustering. The result shows that k=4 is the best choice for the dataset. The four clusters formed to represent the four groups of possible users are the good users with 25 instances, skilled users with 35 instances, the users with limited skills with 83 instances and finally, the expert users with 32 instances. Each of the groups possesses their distinct skills which emerged from the clustering technique implemented.