A search engine for getting price statistics of motorcycles sold on Blocket
This search engine takes a query and filtering arguments and performs a tf-idf based search through a dataset of motorcycle ads. The ads that match the query best have their price average, maximum and minimum taken and presented to the user.
HOW TO USE:
-
Install pillow and elasticsearch modules for python.
pip install pillow,pip install elasticsearch
-
Download and run Elasticsearch according to these instructions: https://www.elastic.co/downloads/elasticsearch
-
Start the program with :
python3 SearchGUI.py -
What you can filter on in the fields:
- Query: Specify the search in free text (e.g. brand, model, color, condition - whatever you want)
- City: Enter a location/city that you want to filter the search on. THIS FIELD IS NOT RECOMMENDED TO USE SINCE SOME CITIES ARE DIVIDED INTO SMALLER REGIONS AND THE MAIN CITY DOES NOT GIVE ANY HITS. SHOULD BE REPLACED BY DROPDOWN.
- Type: Select a specific motorcycle type or search amongst all types ("All")
- Maximum Model Year: Select how new motorcycles you want to include in the search result
- Minimum Model Year: Select how old motorcycles you want to include in the search result