Thinking about what to build this weekend? Check out Highlights in the Objective API — enabling Highlights on your Index cranks up the fidelity of deep text searching on your search index, letting you do things like semantically search a collection of thousands of long text articles and return the exact paragraph that matches the human meaning of your query. All the power of AI-native search, with all of the messy chunking handled intelligently for you 🔎⚡
Objective, Inc.’s Post
More Relevant Posts
-
Upgrade your text search game with Hybrid Search! 📚🔍 Experience the best of both worlds: semantic search + term matching for spot-on results. Learn more about how to use hybrid Search in #MyScale:��https://lnkd.in/gQkAEHnb #LLM #VectorDatabase #vectorsearch
MyScale | Docs
docs.myscale.com
To view or add a comment, sign in
-
We built a Wikipedia search engine in 30 minutes with Objective that outperforms the Elastic-based solution used on wikipedia.org by 34%. Here’s how we did it.
How We Built Search Over All Of Wikipedia in 30 minutes with 34% Better Relevance
objective.inc
To view or add a comment, sign in
-
Search isn’t exclusively the domain of high tech. Search is an everyday activity, be it grocery shopping, shopping for content online or looking for a good book to read. If Stackoverflow sees the potential for an improved AI based search, no reason a site like Goodreads shouldn’t. Search as of today is a poorly understood problem. #aisearch #conversationalsearch #naturallanguagesearch
StackOverflow implemented its semantic search solution with Weaviate. How did they do it? They used a pre-trained BERT model from the SentenceTransformers library to generate the embeddings. Their reasons for using Weaviate: it's open source and you can host it on your own infrastructure so no third party sees your data plus they needed hybrid search - lexical and semantic on the same data. Read the full story on the StackOverflow blog https://lnkd.in/eGMtS9DE.
Ask like a human: Implementing semantic search on Stack Overflow
stackoverflow.blog
To view or add a comment, sign in
-
Simple solution 👍🏼. I used BERT sentence transformer 2 years ago in itsm solution proposal use case and it’s really useful where a semantic search along with keyword search is required.
StackOverflow implemented its semantic search solution with Weaviate. How did they do it? They used a pre-trained BERT model from the SentenceTransformers library to generate the embeddings. Their reasons for using Weaviate: it's open source and you can host it on your own infrastructure so no third party sees your data plus they needed hybrid search - lexical and semantic on the same data. Read the full story on the StackOverflow blog https://lnkd.in/eGMtS9DE.
Ask like a human: Implementing semantic search on Stack Overflow
stackoverflow.blog
To view or add a comment, sign in
-
StackOverflow implemented its semantic search solution with Weaviate. How did they do it? They used a pre-trained BERT model from the SentenceTransformers library to generate the embeddings. Their reasons for using Weaviate: it's open source and you can host it on your own infrastructure so no third party sees your data plus they needed hybrid search - lexical and semantic on the same data. Read the full story on the StackOverflow blog https://lnkd.in/eGMtS9DE.
Ask like a human: Implementing semantic search on Stack Overflow
stackoverflow.blog
To view or add a comment, sign in
-
Co-Founder, Chief AI & Analytics Advisor @ InstaDataHelp | Innovator and Patent-Holder in Gen AI and LLM | Data Science Thought Leader and Blogger | FRSS(UK) FSASS FROASD | 16+ Years of Excellence
The Future of Web Search: Exploring the Potential of the Semantic Web The Future of Web Search: Exploring the Potential of the Semantic Web Introduction: The internet has become an integral part of our lives, providing us with access to an enormous amount of information. Web search engines have played a crucial role in helping us navigate this vast sea of data. However, traditional search engines are […] https://lnkd.in/dZrtAbJq
The Future of Web Search: Exploring the Potential of the Semantic Web
https://instadatahelp.com
To view or add a comment, sign in
-
Is hybrid search enough? While encoder powered semantic search is great, it struggles to keep up with numeric requirements. For instance for a query like "shoes for less than $100", semantic search won't work. Similarly BM25 does a great job with keywords but lacks context. A solution is potentially hybrid search but we stack performance based on BM25 and Semantic search one after the one. To get the best of the two worlds, we developed a new approach, QAM: Query Attribute Modelling, a novel approach which begins when we are encoding the catalog data and involves query rewriting and attribute augmentation. This approach has proven to have substantial improvement over the current State of the Art retrievals. Learn more about our experiment: https://lnkd.in/d5Zc7Z5m
🛒Enhancing E-Commerce Search: Leveraging Keywords and Semantics for Better Results beyond Hybrid Search
blog.traversaal.ai
To view or add a comment, sign in
-
The Future of Web Search: Exploring the Potential of the Semantic Web The Future of Web Search: Exploring the Potential of the Semantic Web Introduction: The internet has become an integral part of our lives, providing us with access to an enormous amount of information. Web search engines have played a crucial role in helping us navigate this vast sea of data. However, traditional search engines are […] https://lnkd.in/d2v7ms6Z
The Future of Web Search: Exploring the Potential of the Semantic Web
https://instadatahelp.com
To view or add a comment, sign in
-
Checkout the live demo in the comments. The side by side demo really highlights how semantic search shines compared to keywords search.
We built a Wikipedia search engine in 30 minutes with Objective that outperforms the Elastic-based solution used on wikipedia.org by 34%. Here’s how we did it.
How We Built Search Over All Of Wikipedia in 30 minutes with 34% Better Relevance
objective.inc
To view or add a comment, sign in
-
Automate the growth of your sales funnel / WordPress & WooCommerce plugins for AI search and recommendations / @eostis @wpsolr
Don't use a semantic search yet? You'll be slaughtered by search engines !! StackOverflow had to change its strategy: rather than letting search engines slash its traffic by answering questions on spot (#ChatGPT), its new semantic internal search will do the same on their site ! Congratulations to Weaviate: Open source and hybrid were the key choice factors. But I wonder how they manage to integrate user signals to the ranking ? Want to do the same with WooCommerce and Weaviate: wpsolr.com #chagpt #gpt4 #weaviate #wpsolr #semanticsearch #stackoverflow
StackOverflow implemented its semantic search solution with Weaviate. How did they do it? They used a pre-trained BERT model from the SentenceTransformers library to generate the embeddings. Their reasons for using Weaviate: it's open source and you can host it on your own infrastructure so no third party sees your data plus they needed hybrid search - lexical and semantic on the same data. Read the full story on the StackOverflow blog https://lnkd.in/eGMtS9DE.
Ask like a human: Implementing semantic search on Stack Overflow
stackoverflow.blog
To view or add a comment, sign in