Реализация алгоритмов кластеризации и прогнозирования с R

3 000 руб. за проект

07 июня 2020, 23:10 • 1 отклик • 26 просмотров

1st Objective (partitioning clustering)

Find the ideal number of clusters – justify it by showing all necessary steps/methods,
K-means with the best two clusters,
Find the mean of each attribute for the winner cluster,
Check consistency of your results against 19th column,
Check for any pre-processing tasks (scaling, outliers)

2nd Objective (MLP)

Discuss the input selection problem for time series prediction and propose various input

configurations
Perform any pre-processing steps (such as normalisation) before training
Implement a number of MLPs, using various structures (layers/nodes) / input parameters / network

parameters and show in a table their performances comparison (based on testing data) through the provided stat. indices.
Provide your best results both graphically (prediction output vs desired output) and via performance indices

Файлы