K-Means Clustering

##Load Packages needed to run the R code
library(readxl)
library(tidyverse)

##Import Data
file_path <- ##"file path of the source Excel file/Predictive Analytics with R in Power BI.xlsx"

data_set <- read_excel(file_path) %>% 
  select(id, total_points, now_cost, selected_by_percent, minutes, goals_scored, assists, clean_sheets, pos)

data_set$pos <- factor(data_set$pos, levels = c("GKP", "DEF", "MID", "FWD"), ordered = TRUE)

##Manipulate Data
model_data <- data_set %>% 
  filter(minutes > 180) %>% 
  mutate(points_per_90 = total_points / minutes * 90) %>%
  select(-minutes, -total_points)

##Create Clusters
set.seed(237)

fantasy_clusters <- kmeans(model_data[,2:6,8], 3)

model_data$cluster <- as.factor(fantasy_clusters$cluster)