Our Instant Search solution provides an instant search experience including typo handling, filters, custom ranking, and many more features.
Instant Search is designed to perform searches on every keystroke. Each keystroke triggers a new query in the search engine. For example, if a user types "beatles", the search engine will perform seven queries: "b", "be", "bea", "beat", "beatl", "beatle", and (finally) "beatles".
Each search engine must have access to one or more sources of data. This data is divided into documents, such as pages or products.
In order to use Instant Search it is important to know where to find what in our platform. The rule of thumb is that the one-time setup is done in the management environment and your day-to-day business is done in the app. For Instant Search, this boils down to the following distribution:
- Management of resources
- Managing search engines
- Linking resources with search engines
- Searching your sources
- Setting up synonyms
- Setting of stop words
- Managing attributes
- Setting up ranking rules
- Setting up filters
- Lightning Fast (Responses < 50 milliseconds): Priority is given to fast responses for a smooth search experience.
- Search as you type: Results are updated with each keystroke. To make this possible, we use prefix search.
- Typo tolerance: Understands typos and spelling errors.
- Tokenization in English, Chinese, and all languages that use spaces as word separators.
- Whole document results: whole document is returned on search.
Custom search and indexing capabilities:
- Custom ranking: Adjust the relevance of the search engine and the ranking of the search results.
- Stop words: Ignore common non-relevant words such as of or the.
- Highlighting search terms: Highlight search results in documents
- Synonyms: Define synonyms for a better search experience.
- REST-full and GraphQL API
- Integrated web interface in our platform: allows you to test your search settings without implementing a front-end