data_processing.Rmd

---
title: "Data Preprocessing"
author: "Mohanapriya Singaravelu"
date: "03/09/2021"
output: html_document
---

```{r}
library(dplyr)
data <- read.csv("data/premier_league_matches.csv")
head(data)

```
```{r}
nrow(data)
#first 380 - 2019-2020 season
#next 286 - 2020-2021 season
```
Splitting the dataset into two seasons based on the date:
i) 2019-2020 season
```{r}
s1 <- data %>% 
  filter(date <= '2020-07-26') %>% arrange(match_id)
print(s1)

```

ii) 2020-2021 season: 
```{r}
s2 <- data %>% 
  filter(date > '2020-07-26') %>% arrange(match_id)
print(s2)

```

Removing the matches in 2019-2020 season with no fans in attendance:
```{r}
fans_data <- s1 %>% 
  filter(date < '2020-06-17') %>% arrange(match_id) 

fans_data['fans_present'] <- 'Y'
fans_data
```

Removing the matches where the home teams are :
1. AFC Bournemouth
2. Watford
3. Norwich City
```{r}
fans_data<- filter(fans_data,home_team!="AFC Bournemouth" & home_team!="Watford" & home_team!="Norwich City")
fans_data
```

Considering only the first 13 home games of the remaining 17 teams
```{r}
#groupby home_team, count the no of matches, drop the matches >13th 
fans_data = fans_data %>% arrange(home_team) %>% group_by(home_team) %>% slice(1:13)
fans_data
```

```{r}
fans_data = arrange(fans_data,match_id)
fans_data
write.csv(fans_data,'with_fans.csv')
```

Repeating the process for 2020-21 season

Removing the matches with fans
```{r}
limited_fans_matches <- c(59005, 58999,  59003, 59000, 58997, 59008, 59014, 59007, 59009, 59006, 59024, 59033, 59030, 59026, 59041)
no_fans_data <- s2 %>% subset(!match_id %in% limited_fans_matches)
no_fans_data
```

Removing the matches where the home teams are: 
1. Leeds United
2. Fulham
3. West Bromwich Albion 

```{r}
no_fans_data<- filter(no_fans_data,home_team!="Leeds United" & home_team!="Fulham" & home_team!="West Bromwich Albion")
no_fans_data
```
```{r}
no_fans_data = no_fans_data %>% arrange(home_team) %>% group_by(home_team) %>% slice(1:13)
no_fans_data


```


```{r}
no_fans_data = arrange(no_fans_data,match_id)
no_fans_data
write.csv(no_fans_data,'wo_fans.csv')
```