Shale brittleness index prediction method based on improved KNN algorithm

The invention discloses a shale brittleness index prediction method based on an improved KNN algorithm. The method comprises the following steps: determining a training well and a test well; performing correlation analysis and comparison on each logging parameter in the training well and the brittleness index, selecting three logging parameters with first three absolute values of correlation coefficients as independent variables, and taking the brittleness index as a dependent variable; constructing a training sample and a training database by using data in the training well; iteratively optimizing the training database by using a KNN algorithm to obtain an optimal training database; taking the data in the optimal training database as training data, taking the corresponding data in the test well and the training well as test data, and obtaining an optimal K value by adopting a cross validation method; and establishing a KNN model by using the optimal training database and the optimal K value to obtain a predi