Join us to see how we implemented boosting personalized search results and re-engineered the legacy solution.
We’ve achieved 40%–60% less effort by our users to find the content they’re looking for among 40 million documents within 100–200 milliseconds, including search, popularity, and personalization times. The average number of letters used in searches decreased from 9 to 4.
In this live-coding session, we’ll go over:
- Elasticsearch: basics, analyzers, char filters, token filters
- Ranking-based boosting
- Personalized (behavior-based) boosting
- Kafka: real-time user profile generation
- Spring Boot: putting them all together