INTRODUCTION OF PROJECT:

FIFA 20 is a football simulation video game published by Electronic Arts as part of the FIFA series. It is the 27th installment in the FIFA series, and was released on 27 September 2019 for Microsoft Windows, PlayStation 4, Xbox One, and Nintendo Switch.

BUISNESS CASE: WITH THE FIFA20 DATASET WE NEED TO CLUSTER THE PLAYER BY THEIR SKILL INTO CERTAIN GROUP

TASK: CLUSTERING

DOMAIN ANALYSIS OF FEATURE:

• SOFIFA-ID: Id of fifa20 player

• PLAYER URL:A URL (Uniform Resource Locator) is a unique identifier used to locate a resource on the Internet. It is also referred to as a web address.

• SHORT NAME:Short or nick name, Name of player

• LONG NAME:Full name of player

• AGE:Age in years

• DOB:Date of birth of player

• HEIGHT CM:Player height in centimeter

• WEIGHT KG:Player body weight

• NATIONALITY:Nationality is the legal status, which represents the country from which an individual belongs.

• CLUB:A group of people formed for the purpose of playing sports.

• OVERALL:The OVR (Overall) is made up of all the player's in game stats.

• POTENTIAL:The potential rating is the rating they will achieve for sure during the FIFA 20 career model.

• VALUE EUR:The value of player in euro.

• WAGE EUR:A living wage is the amount of income needed to provide an player with a basic but socially acceptable standard of living.

• PLAYER POSITIONS:Positions indicate approximately where each player should be lined up prior to the start of a play.

• PREFEERED FOOT:Footedness may refer to the foot a player uses to kick with the greatest force and skill.

• INTERNATIONAL REPUTATION:International reputation is rated with 1-5 stars.

• WEAK FOOT:A player's foot (left or right) that is weaker than their preferred foot. A player's attribute rated between 1 to 5 which specifies the shot power and ball control for the other foot of that player than his preferred foot's.

• SKILL MOVES:Skill moves, it is assigned a number between 1 (least complex) and 5 (most complex). Players with higher skills are able to perform more complex moves.

• WORK RATE:Work rates are how hard the player works on each end of the field.

• BODY-TYPE:A custom body type and character of football player.

• REAL FACE:Face scan of real football player if it 'yes' then the character is look super realistic.

• RELEASE CLAUSE EUR:A release clause is the minimum amount of money you're willing to accept for that player.

• PLAYER TAGS:The tags of player is identified the player skills, like [Dribbler, Distance shooter, Crosser, Speedster]

• TEAM POSITION:The player’s positioning on the field

• TEAM JERSEY NUMBER:The number on the back of a football player's jersey is to identify a specific player.

• LONED FROM:It's a temporary restriction placed on player due to their club being short of players.

• JOINED:Joining date of club

• CONTRACT VALID UNTIL:The end year of contract of players.

• NATION POSITION:Nation position is similar to the team position of players.

• NATION JERSEY NUMBER:Jersey number of player.

• PACE:Fastest pace of player, striker or winger

• SHOOTING:Shooting is hitting the ball in an attempt to score a goal. It is usually done using the feet or head.

• PASSING:The purpose of passing is to keep possession of the ball by maneuvering it on the ground between different players with the objective of advancing it up the playing field.

• DRIBLING:Dribbling Is Maneuvering A Ball By One Player While Moving In A Given Direction, Avoiding Problems’ Attempts To Intercept The Ball.

• DEFENDING:Defending is most relate to the defender to stop the attack

• PHYSIC:Physic in player means intensity and their effort.

• GK DIVING:Goal keeper diving to save a shot.

• GK HANDLING:As a goalkeeper, you are allowed to handle the ball with your hands within your penalty area.

• GK KICKING:The goalkeeper is the player, who usually kicks the ball to resume play, and they either kick it short to a teammate or it is kicked extremely hard to travel a long distance.

• GK REFLEX:Ability to react quickly.

• GK SPEED:Speed of goal keeper.

• GK POSITIONING:Position of goal keeper is around only goal posts.

• PLAYER TRAITS:A player’s traits are actions that he will be more likely to perform in matches.

• ATTACKING CROSSING:The aim of a cross is to create the easiest possible scenario for the attacking player to score a goal.

• ATTACKING FINISHING:Create a attack on field to score a goal

• ATTACKING HEADING ACCURACY:One of the fundamental principles of the attacking header is to head the ball downwards.

• ATTACKING SHORT PASSING:Short passes of ball to score a goal

• ATTACKING VOLLEYS:A volley is an air-borne strike in association football, where a player's foot meets and directs the ball in an angled direction before it has time to reach the ground.

• SKILL DRIBBLING:The score of dribbling skill with football

• SKILL CURVE:The curve of football at the time of goal

• SKILL FK ACCURACY:It stands for field goal. It is when a player kicks a ball through the uprights from a set distance.

• SKILL LONG PASSING:The long passing skill score of the player

• SKILL BALL CONTROL:The ball control of the player at playing time

• MOVEMENT ACCERLATION:how quickly player can increase your velocity towards maximum speed.

• MOVEMENT SPRINT SPPED:Sprinting speed score of the player.

• MOVEMENT AGILITY:Agility is the ability to move quickly and easily, including moving in different directions without losing balance

• MOVEMENT REACTION:The score of the quick relation of player.

• MOVEMENT BALANCE:Balance is the ability to control your body in its own space.

• SHOT POWER:Power generates to score a goal.

• POWER JUMPING:The jumping ability of the player on field.

• POWER STAMINA:Stamina of player on filed.

• POWER STRENGTH:Strength of player on field.

• POWER LONG SHOT:Total no of power long shot of football.

• MENTAILITY AGGRESSION:Aggression is a characteristic that can have many negative as well as positive effects on performance of player.

• MENTALITY INTERCEPTION:Interception is thrown the possession of the ball is given to the other team.

• MENTALITY POSITIONING:The mentality positioning of player determines how defensive or attacking your players set out to play.

• MENTALITY VISION:The vision/mindset of player on filed.

• MENTALITIES PENALTIES:When a player commits an infraction against the rules of the game a foul will be called and a penalty will be assessed.

• MENTALITY COMPOSURE:Composure stands for: Recognize–Regroup–Refocus. The first step is to Recognize that you are dwelling on the mistake, which limits your ability focus on the next play.

• DEFENDING MAKING:When your team is defending, you want to make the field small and the options of the other team limited.

• DEFENDING STANDING TACKLE:Defending standing Tackle of the player to get a ball on filed.

• DEFENDING SLIDING TACKLE:Defending sliding tackle of player to get a ball.

• GOALKEEPING DIVING:Goal keeper diving to save a goal.

• GOALKEEPING HANDLING:Handle the ball with your hands within your penalty area.

• GOALKEEPING KICKING:A kick taken by the goalkeeper when the attacking team has kicked the ball over the goal line.

• GOALKEEPING POSITIONING:A Positioning of player to save the goal.

• GOALKEEPING REFLEX:The reflex of goalkeeper and players.

• LS:Long snapper or left striker.
• ST:Striker
• RS:Right striker
• LW:Left sided wingers.
• LF:Left forward
• CF:Center forward
• RF:Right forward
• RW:The RW is usually on the right end of the attacking trident, with the Striker and Left Winger, which mainly contributes to the team in terms of goals and assists.
• LAM:Left attacking midfield
• CAM:Center attacking midfield
• RAM:Right attacking midfield
• LM:Left midfield
• LCM:Left center midfield
• CM:Center Midfield
• RCM:Right center midfield
• RM:Right midfield
• LWB:Left Wing Back
• LDM:Left defensive midfield
• CDM:Center defensive midfield
• Right defensive midfield
• RWB:Right wing back
• LB:Left back
• LCB:Left center back
• CB:Center back
• RCB:Right center back

• Impute Missing value with median because feature is not normal distributed
• Handling categorical data with mannual encoding
• Handling outlier with Empherical rule and IQR.
• Scaling the data with the help of min-max scaler

• Drop unique and constant column
• Remove highly correalted feature, In this data lots of highly correlated feature is available.
• Their is no duplicate is present.
• At last apply PCA to reduce the dimension of feature.

• He we are use K-means algorithm to make cluster/group.
• With 3 cluster sillhoutte score is 0.61
• With 5 cluster sillhoutte score is 0.50