Analysis of skoda fabia sales offers, based on scraped data from the auction site
Tools used in this project:
sklearn : 0.23.2 pandas : 1.1.3 matplotlib : 3.3.2 selenium : 3.141.0 seaborn : 0.11.0
Using very simply web scraper, I downloaded about 900 Skoda Fabia offers. Made analyse to find voivodeship with lowest mean price, using pandas in jupyter notebook. Also applayed a regression model to predict price of Skoda Fabia on parameters like: mileage, year of production, location/voivodeship, fuel type.
Close future plans:
- implement adaline perceptron
Update v2:
- prepared data to use with linear regression model ( standarization and label encoding)
- added scikit-learn lib to predict car price but model score is very bad
Update v3:
- after apply IQR method to delete outliers, model accuracy is 78%
- added a plot with milage and price coefficient and with regression line
- example of prediction price of two cars with different parameters