Hello,
I'm new to machine learning and need some help. Perhaps someone here knows of similar examples.
I have a dataset of 900 geographical objects. For these objects, I have annual values for 11 years. I want to create an algorithm that finds dependencies from objects for which data is known at a lower temporal resolution, and upon inputting 1-2 elements, fills in the remaining "squares" (likely referring to missing data points or future predictions) with corresponding values. I can at least add parameters such as population density, land use type, elevation, and slope to each "square". However, I don't understand how to make the model learn to find patterns from the values at the stations, cause they are similar for every object in a year selection.
As i analysed a liiterature it is more conviniet to use RF or Generate Spatial Weights Matrix.
Thank you!