Diffbot

Graph RAG or vector-based RAG? There's some discussions (debate) around which is better. (but we're not continuing this conversation...) Thanks to Tomaz Bratanic, one of the thought leaders and trailblazers in the graph world with cool applications, we jointly explore a project that realizes and visualizes the idea of"bringing unstructured and structured data" together. You'll see side-by-side comparisons of AI-generated responses between vector-only searches and vector+knowledge graph in the video, showing how the latter can outperform the former. Neo4j can store both graph data and text embeddings for vector-based searches, and Diffbot provides verified data from the web — the exact opposite of hallucinated LLM-based sources, so AI applications don't suffer from unreliable outputs. To see the cool network visualization in the app, jump to 3:51 and check out Tomaz and Anej's work. You can always go back and see how we use Diffbot's Article, Natural Language Processing, and Enhance APIs to enrich knowledge graphs with accurate and reliable information. p.s. Who's also going to AI Engineer World's Fair the next 2 days in SF? DM if you're around! #neo4j #diffbot #graphrag

Hi everyone. Have you ever had a pile of personal or work documents that you wanted to make sense of or find all relevant information from? In three well-documented projects, I have addressed this challenge. I implemented a Retrieval-Augmented Generation (RAG) system using a recent architecture, RAPTOR. However, this system could not answer all questions. To cover these limitations, I used a Knowledge Graph. Finally, I implemented everything locally to ensure the privacy of proprietary documents, which is of paramount importance. Visit my webpage to read more and watch the videos of the three projects: https://lnkd.in/gNnnH3Ne In the implementation of RAG, I did not sweep anything under the rug. All limitations are exposed, and I provide suggestions for improvements to set realistic expectations for the deployment of this AI system. Some technologies I used for implementation include: Neo4j Graph DBMS, Diffbot API, Cypher query language, LangChain, Mistral Language model, Nomic AI embeddings, Beautiful Soup for web scraping, Streamlit for UI, Vector Database, Credits: + Inventures 2024 website is the source of my training data. + YouTube thumbnails are made by Sombilon Studios + I am thankful for Javier A. Jaime Serrano, who answered my questions on Graph DBMS. + Soundtrack: Persian Version of Game Of Thrones Song Other references on my page ⚠️ Disclaimers: + This is an educational purpose porjece, in-line with Alberta Innovates privacy policies. Any further use of the sessions data should be checked with their latest privacy policy. + Language models may modify the text during summarization process. The only source of data for reference is their webpage at https://lnkd.in/gHk7k4r6

What's the best IDE to view huge Diffbot data dumps? Hint —

👋 Hi friends, we're going to be hanging out at the upcoming AI Engineer World's Fair in San Francisco the week of June 24-28. If you're in San Francisco, we'd love to meet you! Leave a comment to let us know. We may even have a t-shirt in your size! Here, I even modeled for you.

Am I procrastinating on a more important project? Yes. By shopping for overpriced shirts? Also yes.

Neo4j + Diffbot is an incredible combination. Here's how you can learn more about it! This month on the new episode of GraphStuff.fm, the official Neo4j Podcast, my colleagues Jennifer Reif and Jason Koo welcome Leann Chen to talk about knowledge graphs and Diffbot. Leann is a developer advocate at Diffbot and focused on using knowledge graphs to improve LLM based applications. This is a must-listen if you are interested in the intersection of knowledge graphs and LLMs. You can find the full episode here: https://buff.ly/4e7RwB6 Follow me Daniel Bukowski for daily content about the intersection of graphs, data science, and GenAI. #neo4j #diffbot #genai #llm #knowledgegraphs

Looking good!

📃 Fresh updates to the LangChain + Diffbot docs this week! Previously, you might've followed the GraphTransformer guide to turn Wikipedia pages into queryable Neo4j knowledge graphs. But did you know you can also use the Diffbot Document Loader to turn text from any website into a KG? Let us know if you'd like a video walkthrough 👇

It's here! The UI sandbox for Natural Language API is now generally available on the Diffbot Dashboard. This doesn't add any new NL API features. It's primary objective is to make it easy to test NL API without having to write an API call. - Click through disambiguated entities to access their entity profile in the Diffbot KG - Visualize and isolate extracted facts - Generate NL API requests in Python, Javascript, and cURL - View and copy the complete NL JSON response Up next, we're going to share some fresh open-source NL showcase projects like the sanctions tracker we built back in 2022. Stay tuned! 📣 🙌 Shout out to our fans — Leann Chen: Confidence scores are listed in the entity table! William Bui: IAB categories with confidence scores can be found in the JSON response. Tomaz Bratanic: Find the diffbotUris for each entity in the JSON or just click on them in the entity/facts table.

This has been on our roadmap for a minute, but I've finally got a scaffold in place for a Natural Language UI in the dashboard! Planned features: - Entity highlighting and linking - API request code generator for Python, Javascript & cURL - View complete NL API JSON response Anything else ya'll would like to see here? Maybe some #LangChain specific tools?