Publication

Abstract : Computer technologies have recently been introduced into the field of agriculture, leading to significant advancements in spatial data mining. Spatial data refers to information associated with locations on the Earth's surface. A spatial co-location pattern consists of multiple groups that correlate spatial features or events frequently occurring within the same geographical region. Existing systems employ probabilistic prevalent co-location mining to identify likely prevalent co-location patterns; however, there is scope for improving efficiency. This paper focuses on spatial co-location mining in agriculture by proposing a novel multi-resolution pruning technique to address the challenge of mining co-location data patterns containing rare spatial features. The proposed combinatorial spatial co-location mining algorithm identifies localities, provides soil information, and recommends suitable crop varieties, thereby supporting improved agricultural decision-making.