Text to Knowledge Graph Made Easy with Graph Maker: What to Watch in AI

Text to Knowledge Graph Made Easy with Graph Maker: What to Watch in AI

In the world of artificial intelligence (AI), there are two exciting developments that have caught the attention of many researchers and developers. One is the ability to create knowledge graphs from text with the help of open-source libraries like Graph Maker. The other is the focus on serving the enterprise market with AI solutions.

Graph Maker is a Python library that utilizes open-source language models (LLMs) such as Llama3, Mistral, Mixtral, or Gemma to extract knowledge graphs from a given corpus of text. To create a knowledge graph, two pieces of information are needed: a knowledge base (a collection of text, code, articles, etc.) and an ontology (categories of entities and their relationships). Graph Maker simplifies this process and provides a user-friendly interface for building knowledge graphs.

On the other hand, the enterprise market is now a prime target for AI companies. While generative AI models have gained popularity among consumers, several companies are directly focusing on enterprises. These companies, including Glean, Lamini, Dust, and Lance, are building AI products that incorporate internal data and adhere to corporate guidelines.

Glean, for example, has developed a platform that indexes, embeds, and keeps companies' internal data (such as Notion, Slack, Drive, GitHub) updated in real-time. This allows LLM-backed products to access and utilize this data effectively. By leveraging proprietary data across multiple modalities, enterprises can create AI models that lead to differentiated services, insights, and increased operational efficiencies.

However, it's crucial for enterprises to focus on using their own data rather than relying solely on pre-trained language models. Labelbox addresses this challenge by simplifying the process of feeding datasets into AI models. This allows companies to leverage their proprietary data effectively and create AI applications that go beyond chatbots and augment existing applications.

Moreover, AI has the potential to revolutionize the user experience by fundamentally changing how we interact with products. Lamini, an LLM engine, makes it easy for developers to train, fine-tune, deploy, and improve their language models with human feedback. This enables the development of AI applications that reinvent product experiences and leverage the technology to enhance creative tools.

While these advancements in AI present exciting opportunities, there are also challenges that need to be addressed. One of the key obstacles preventing enterprises from shipping AI applications to production is the lack of appropriate governance controls. Questions like data permissions, source data ownership, and model inference location need to be answered. Glean, with its ability to plug into an enterprise's internal environment, provides a solution to these governance challenges and enables enterprises to confidently leverage their internal data for AI model training and inference.

In conclusion, the combination of Graph Maker and the focus on serving the enterprise market represents significant advancements in the field of AI. By simplifying the process of creating knowledge graphs from text and incorporating internal data, these developments enable enterprises to harness the power of AI and create differentiated services. To fully leverage these advancements, here are three actionable pieces of advice:

1. Invest in building and maintaining a robust knowledge base: A well-curated and up-to-date knowledge base is essential for creating accurate and meaningful knowledge graphs. Regularly update your corpus of text or data to ensure the knowledge graph remains relevant.

2. Embrace the power of proprietary data: While pre-trained language models are valuable, don't overlook the importance of your own data. Leverage your proprietary data across multiple modalities to create AI models that provide unique insights and services.

3. Prioritize governance and data control: As AI applications become more prevalent, it's crucial to establish appropriate governance controls. Ensure that you have clarity on data permissions, source data ownership, and model inference location to maintain data control and comply with regulations.

By following these pieces of advice, you can make the most of the advancements in AI and create impactful applications that drive innovation and efficiency in your enterprise.