Description

Explanation

Exploratory Data Analysis refers to the critical process of performing initial investigations on data so as to discover patterns,to spot anomalies,to test hypothesis and to check assumptions with the help of summary statistics and graphical representations.

before entering the next process. please download the required dataset first if you want to try it yourself.

Kaggle links (https://www.kaggle.com/datasets/deepcontractor/top-video-games-19952021-metacritic)

maybe we should know about the dataset we are using. This dataset contains a list of video games dating from 1995 to 2021, it also provides things such as release dates, user review ratings, and critical review ratings. This dataset was taken by scraping from the website https://www.metacritic.com.

to continue into the EDA process, the first thing we do is make preparations

mount google collab with our own google drive.

import important libraries that will be used for the EDA process

extract and upload the dataset that we previously downloaded to the google drive that we connected to the google collab.

then try to read the dataset with the command as shown below.

at this stage I want to check the dataset information used. seen here there are 18800 data and has 6 columns. The data type for each column is also shown

then in this command we will look for empty data in each column. can be seen in the picture in the summary column there are 114 blank data.

because there is empty data, we will fill in the value of the data with N/A. after that check again the empty data to prove

then we change the data type of the user review which previously made our object a float. after that we also add a new special column, namely the year obtained from the release date value

in the current experiment we will be looking at 5 research questions.

Q1: Meta score each year?

Q2: Top games by meta score

Q3: meta score of each platform

Q4: Top PC game by meta score

Q5: Platform shares

The following are the results of experiments conducted to answer the questions above

Q1:

Q2: 

Q3: 

Q4: 

Q5: 

From the results of the previous results we can conclude that :

• It can be seen that meta scores and reviews start to fall over time. This could be due to the number of games released, or simply more people being able to leave their reviews.
• The Legend of Zelda: Ocarina of Time which was released on the Nintendo 64 platform has the highest meta score at 99
• the PC platform has the most games compared to other platforms. but in PC platform meta score only got 71.79 in average.