Scribd
Upload
539 views16 pages

FIFA 18 - Data Analysis: - Harsh Takrani - Pranay Lulla

The document outlines a data analysis project on FIFA 18 player data from sofifa.com and Twitter sentiment data. The goals are to cluster top players, identify best squads based on formations, find underrated young players, and do sentiment analysis and machine learning predictions. Data was preprocessed from sofifa.com and Twitter. Basic visualizations and analyses were done, including finding top 20 players, potential vs overall, and histograms. Young underrated players were identified. Best squads were analyzed based on formations. Machine learning methods like linear regression and SVR were used for value predictions. Sentiment analysis using TextBlob classified tweets as positive, negative or neutral.

Uploaded by

Sovan Dash
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
539 views16 pages

FIFA 18 - Data Analysis: - Harsh Takrani - Pranay Lulla

The document outlines a data analysis project on FIFA 18 player data from sofifa.com and Twitter sentiment data. The goals are to cluster top players, identify best squads based on formations, find underrated young players, and do sentiment analysis and machine learning predictions. Data was preprocessed from sofifa.com and Twitter. Basic visualizations and analyses were done, including finding top 20 players, potential vs overall, and histograms. Young underrated players were identified. Best squads were analyzed based on formations. Machine learning methods like linear regression and SVR were used for value predictions. Sentiment analysis using TextBlob classified tweets as positive, negative or neutral.

Uploaded by

Sovan Dash
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 16

FIFA 18 – Data

Analysis
-Harsh Takrani
-Pranay Lulla
Project Goals
• Collecting, preprocessing and loading complete dataset of players
• Clustering top players based on Overall Rating
• To identify best possible squad based on two formations
• Clustering young underrated players with high potential
• Sentiment analysis using Twitter to understand feedback of FIFA
18
• Applying Machine Learning Algorithm for prediction purposes
• Creating basic visualizations like Histograms and Line charts
Data Sources

www.sofifa.com – To
www.twitter.com – To
get complete dataset of
extract tweets to
FIFA 18, flexible data
analyze the sentiments
source as per
of customers using the
parameters required for
product
analysis
Collected data from sofifa.com and
analyzed the important parameters and
removed the redundant ones
Changing data types as per requirement

Data
Preprocessing Cleaned the Value and Wage columns
using functions

Eliminating special characters in Skills and


Finishing columns
Basic Analysis:
Finding Top 20
Players
Basic Visualization:
Potential vs
Overall
• Visualizing the peak of player using
overall
1. Grouping the players by
age
2. Getting a visualization by
age of player and overall
Basic
Visualization:
Histogram
Finding Young
Underrated
Players
• Here we create a growth
column that helps us
understanding how much a
player’s rating can be
improved
• This helps us finding young
underrated players which can
be used by players to build a
team in FIFA ULTIMATE TEAM
(FUT) for cheap prices
Best Squad Analysis:
Formation Dependent
• Here, we have created three
functions based on three
formations: 4-3-3, 4-2-3-1 and 3-5-2
• The function identifies one position
from preferred positions
• Then it returns output which
contains a data frame of 11 players
with position, name and rating
Divided the data into test and train
datasets using library model selection

Train data was used to train the


model
Machine
Learning Test data was used to compare the
values of prediction

The model is used to predict Value of


player using Overall rating parameter
Linear Regression
• Used linear regression to predict the values
• Mean Squared error:29.13
ML
Techniques Support vector Regression
• Used another modelling technique: SVR using
radial basis function for non-linear problems
• Mean Squared Error: 13.47
Twitter Analysis

• Extracted the data of FIFA from twitter

• Converted the data in to text

• Created a data frame with attributes like


Number of likes, Retweets count
• Used Textblob package to calculate the polarity
of tweets

Sentiment • The polarity was used to analyze the feedbacks


Analysis on the tweets

• The tweets were identified as positive, negative


and neutral
• Based on number of likes and retweets we can
Sentiment estimate the following of the game
• Sentiments are used to identify the polarity of
Analysis the game = .i.e. whether the tweets are
Demo positive, negative or neutral
Any
Questions?
Thank You

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy