ASP.NET core Elasticsearch web based search engine
- BE SURE TO START THIS FIRST!
- navigate to folder with docker-compose.yml in it (should be the main folder)
docker-compose up
- give it time to run takes 2-5 minutes to start for me on an older windows machine
- For simplicity open the .sln file in Visual Studio
- run without debugging
- will take about a minute to start as it has to load data into Elasticsearch
- C-Max
- Tesla
- seatle
- WA
These should get you enough info and you can have fun after that, data is all electric car data so try searching anything related.
Data from Data.gov
- The website should load and index some randomly generated or selected test data.
- Your test data can be anything you want but should be at least several thousand records.
- The app should allow the user to enter keywords and find your indexed content.
- Your app should also feature a responsive design to ensure it renders well for a variety of screen sizes and resolutions.
- Stemming should be used – for example, when searching for “engineer”, the search should also return results for “engineering”, “engineers”, and “engineered”.
- Spell checking should present corrected search terms for user misspellings.
- Related search terms could be displayed when a user has searched, suggesting alternate keywords related to what they’ve already searched for.
- Auto-complete should suggest search terms as the user types in the search box.
- Attach meta-data to your indexed records and provide faceting options for the user to filter by.
- Data isn't very unique there is alot of repitition. Better data could be found and models adjusted I guess?
- could be bugs with pagination, it was a last minute addition but from my testing it's working
- slow start up, not worth fixing as test data loads in.
- need to dockerize ASP.NET core app Not sure if that would make running it any easier due to HTTPS certs.
- elasticsearch docker container is quite memory intensive. (around 7GB)
- no persistant elastic search data.
- data is rebuilt every time program is run
- only 2100 entries, can add more by reformatting data with the python file but start up times increase quite a bit