/Exploratory_Data_Analysis_autoscout_MachineLearning

Primary LanguageJupyter Notebook

CLRSWY

Exploratory Data Analysis (EDA) Project

EDA for Car Price Prediction Model

Descriptions :

  • A .json file containing a dataset consisting of 15919 rows and 54 columns is provided.
  • This dataset, scraped from the on-line car trading company in 2019, contains many features of 9 different car models.
  • The features (variables) of this dataset are too messy and distored.

What is expected of you?

  • Read the .json file and assign the dataset into a DataFrame using pandas.
  • Implement all aspects of the EDA process to the dataset.
    • Fix corrupted data formats,
    • Handle with outliers and missing values,
      • Domain (automobiles) knowledge is important.
      • Always use the internet to do the research that you need.
      • Think carefully to decide whether a data is outliers or not. Examples :
        • There is no conventional car model with an average fuel consumption of 1 - 1.5 liters per 100 km.
        • Or you need to know that it cannot be a 300 euro car.
        • Or if there is only one car with 3 doors out of the 15919 cars, this is what you should pay attention to and examine.
    • Drop the columns / rows you determined unnecessary as a result of your analysis,
    • Use visualization tools while doing all these processes.
  • As a result, get the dataset ready to provide an appropriate input to the ML models.
  • Save cleaned dataset into a .csv file.

Need to Study :

  • .str.method,
  • .contains(),
  • .extract(),
  • .to_datetime(),
  • .get_dummies(),
  • .add_prefix(),
  • .sample(),
  • regex,
  • .to_numeric(),
  • .isin(),
  • .corr().