Data Mining in Agriculture

By: Paola A.

Every year, technology changes and new developments help many economic sectors discover new ways to improve, forecast a change, etc. For example, one economic sector that is benefitting from using new developments in technology is agriculture. In agriculture, a way to discover this type of changes is through data mining. But what exactly is data mining and how is agriculture benefitting from this. “Data mining involves the process of finding large quantity of previously unknown data, and then their use in important business decision making” (Milovic & Radojevic, 2015). For instance, in developing countries such as India, using data mining for “price prediction helps the farmers and also Government to make effective decision[s]” (Hemageetha & Nasira, 2012). Furthermore, data mining is something that could also help consumers by preparing them in case of any change in prices. Finally, this method of collecting data is something that can definitely improve and benefit the way farmers, government, and consumers make better decisions in the future with different applications, but it also has challenges and limitations.
Using data mining in agriculture benefits farmers, government, etc. in many ways. One of the ways that using data mining is beneficial in agriculture is “possibility to study hidden patterns in datasets in agricultural domain. These patterns can be used for diagnosing crop condition, prognosing market development, monitoring customer solvency” (Milovic & Radojevic, 2015). In other words, it can help farmers tell around the time that their crop would flourish, make predictions of when their product is more likely to sell, and what product customers are buying more. Also, another benefit is that “Agricultural institutions use data mining technique and applications for different areas, for instance agronomists use patterns measuring growth indicators of plants, crop quality indicators, success of taken agro technical measures and managers of agricultural organizations pay attention on user satisfaction and economically optimal decisions” (Milovic & Radojevic, 2015).

Data mining uses many applications to forecast several things that could happen with produce, crops, etc. For instance, farmers use K – means cluster algorithm in order to determine which apples could be sold for a lower price and which apples could be sold at a normal price based on how much damaged they have. Moreover, during k – means cluster, the apples are divided and based on those apples that are bad, they are divided again to make sure which ones go for sale and which ones don’t. Furthermore, in order to get which crops are used more and which ones are the closest to that crop in case that is missing, we use Support Vector Machine (SVM).

Even though data mining has a lot of benefits through the usage of applications, there are still challenges and limitations that need improvement. For example, one of the challenges and limitations that data mining in agriculture has is “missing, invalid, inconsistent or nonstandard data like parts of information recorded in different formats from different data sources create a large obstacle for successful data mining” (Milovic & Radojevic, 2015). In other words, if you don’t have enough information or correct information, it could lead to results that may not be true and could potentially lead to an economic loss.
The usage of data mining in agriculture is an innovative way that has helped and benefitted many farmers, managers, etc., and even countries. Also, based on applications, farmers can discover what decisions to make in order to improve their crop, sales, and what is the closest crop they have if they run out of the one they want. Even though data mining has it challenges and limitations, I believe that in the future those challenges will be solved to better the usage of data mining in agriculture.

Milovic, B., & Radojevic, V. (2015). Application of Data Mining in Agriculture. Retrieved January 10, 2017, from

Patel, H., & Patel, D. (2014, June). A Brief survey of Data Mining Techniques Applied to Agricultural Data. Retrieved January 10, 2017, from

Support Vector Machines for Crop Classification Using Hyperspectral Data
Gustavo Camps-Valls-Luis Gómez-Chova-Javier Calpe-Maravilla-Emilio Soria-Olivas-José Martín-Guerrero-José Moreno –