The Power of Systematized Knowledge: How I'm Applying Obsidian and Local Large Language Models to Drive Industry Insights

## N1_LLM output:        

The Power of Systematizing Data

As someone who works in the construction industry, I've always been fascinated by the potential of data to drive innovation and automation. With the increasing use of building information modeling (BIM) and other digital technologies, I believe that every aspect of our work can benefit from a more systematic approach to data collection, analysis, and application. By organizing and optimizing our processes through automation, governance, and analytics, we can unlock new efficiencies and improve the quality of our work. For me, this means finding ways to apply machine learning algorithms to our project management workflows, or using data visualization tools to better understand the relationships between different data components. I'm excited to explore these possibilities.

Visualizing Data as a Tree

For me, data is like a tree - we know its components, but only when we organize and visualize it in a meaningful way can we truly understand its underlying structure. This is where Obsidian comes in. By creating a customized knowledge graph, I can navigate and connect the various aspects of BIM, CGI, AEC, and emerging technologies.

Running Local Large Language Models

I've also been experimenting with running open-source local large language models offline on my laptop from 2021 using Open-WebUI/Alpaca. This allows me to access a range of models like Llama3.2, Gemma2, and Nemotron Mini on Fedora and Linux

Goals

With Obsidian and the power of local large language models, I aim to achieve the following goals:

1. Documenting Industry Insights: Logging my experiences, implementations, and industry knowledge to create a valuable resource for others in BIM/AEC.
2. Data Connection Analysis: Identifying relationships between different pieces of information to extract deeper insights from my data.
3. LLM-Gated Governance: Using LLMs to analyze and refine data and suggest solutions
4. Augmenting Data with Consistency and Scalability

I'll use local large language models to:

• Suggest missing connections and refine existing information
• Analyze data entered to identify patterns and trends
• Augment existing knowledge with consistency and scalability

This article was initially written by me and then rewritten with AI assistance.

#BIM #AEC #CGI #ArtificialIntelligence #MachineLearning #NaturalLanguageProcessing #DataVisualization #DataAnalytics #ProductivityHacks #FutureOfWork #InnovationInAction #Obsidian #LLMs