Data Association Methodology to Improve Spatial Predictions in Alternative Marketing Circuits in Ecuador

Area of Science:

• Agricultural Economics
• Spatial Statistics
• Data Mining

Background:

• Accurate commercialization forecasting is crucial for agricultural markets.
• Multivariate spatial prediction techniques like cokriging are used for estimations.
• Identifying product relationships can improve prediction accuracy.

Purpose of the Study:

• To develop a methodology for reducing future estimation errors in agricultural product commercialization.
• To integrate association rule mining with multivariate spatial prediction.
• To enhance the accuracy of spatial predictions by considering product interdependencies.

Main Methods:

• Utilizing the Apriori algorithm to discover association rules among agricultural product sales in local markets.
• Applying multivariate spatial prediction techniques (cokriging) informed by identified association rules.
• Comparing prediction accuracy with and without the use of association rules.

Main Results:

• The Apriori algorithm successfully identified significant association rules between agricultural products.
• Integrating these association rules into cokriging significantly improved spatial prediction accuracy.
• The methodology demonstrated a reduction in future estimation errors.

Conclusions:

• Association rule mining, specifically using the Apriori algorithm, is effective in enhancing multivariate spatial prediction for agricultural product commercialization.
• Considering product associations leads to more reliable future estimations in local markets.
• This integrated approach offers a valuable tool for agricultural market analysis and forecasting.