The Application of Data Mining in Analyzing Factors Affecting the Classification of Technical Efficiency of Wheat Farmers: A Study in Ahar County

نویسندگانVahedi, J. - Ghahremanzadeh, M.- Dashti, G. - Pakrooh, P
نشریهJournal of Agricultural Economics & Development
نوع مقالهFull Paper
تاریخ انتشار۲۰۲۶.۰۳.۱۵
رتبه نشریهISI
نوع نشریهچاپی
کشور محل چاپایران

چکیده مقاله

Ahar County, as one of the main agricultural production areas, produces a considerable share of rainfed wheat in East Azerbaijan Province. Improving technical efficiency (TE) in this region can have a significant impact on farmers' productivity and economic sustainability. Therefore, the present study was conducted with the aim of applying data mining to analyze the efficiency of rainfed wheat farmers in Ahar County. To this end, farmers' TE was calculated using Data Envelopment Analysis (DEA), and those with efficiency scores above the regional average (0.59) were classified as the high-efficiency group, while the rest were categorized as the low-efficiency group. Subsequently, t-tests and chi-square tests were employed to identify variables likely to influence TE. Machine learning algorithms, including logistic regression, support vector machines (SVM), k-nearest neighbors (k-NN), and random forest (RF), were then applied for classification and analysis of TE. The results indicated that logistic regression outperformed the other algorithms. The output of this algorithm revealed that factors such as herbicides, weed control, manure, land rental value, nitrate fertilizers, number of farm plots, pesticides, farmer's age, combined harvesting method, experience, household members with university education, seed procurement from the Agricultural Organization, and residence in rural areas have a positive effect on TE. Conversely, factors including mixed landownership (personal-rental), seed procurement from personal sources, and non-agricultural income exerted a negative influence on efficiency. Based on the findings, it is recommended that farmers receive necessary training on the optimal management of influential agricultural inputs, including herbicides, manure, nitrate fertilizers, and pesticides. Furthermore, policymakers are advised to enhance the motivation of farmers operating on rented lands by providing financial incentives and advisory services. The development of supportive programs for the supply of quality seeds and agricultural inputs through the Agricultural Jihad Organization is also proposed.