Semantic Search: Building search beyond keyword matching
Users expect search that matches on "things not strings." We can transform your search from basic keyword matching into one able to manage concepts. We wrote Relevant Search the guide to smarter search. We've evaluated numerous semantic search methods. From machine learning to integrating and developing taxonomies, let our team improve your search.
How we can help
- Use machine learning and NLP methods such as topic-modeling to enhance search
- Develop and integrate taxonomies and controlled vocabularies into search
- Measure the impact of semantic search against your business metrics
- Integrate content expertise into a semantic search solution
- Develop models using latent dirichlet allocation, latent semantic analysis, or word2vec
- Integrate with Lucene, Solr, or Elasticsearch search solutions
Here's what our clients have to say
We leverage OSC’s deep expertise to solve relevance challenges unique to our industry. Partnering with OSC ensures that our scientific, technological, and medical publishing expertise and taxonomies are extended into the best-in-breed search technology that powers the targeted retrieval of our clients’ nuanced scholarly content.
I’ve rarely experienced such positive, results-oriented, cheerful customer service.
Our semantic search products & publications:
Quepid — a relevancy tuning platform. Monitor and improve your results.
Select semantic search blog articles:
A few years ago John Berryman and I experimented with integrating Latent Semantic Analysis (LSA) with Solr to build a...
In Chapter 4 of Relevant Search, we talk a LOT about Elasticsearch analyzers. Without analyzers, your search engine would be...
Built upon Lucene, Solr provides fast, highly scalable, and easily maintainable full-text search capabilities. However, under the hood, Solr is...