Max Abouchar’s Post

Worth the investment to take one of Lori's AI Classes

Productivity. Performance. Trust. ✨ Find out more about how SAS delivers on these fundamental principles for AI success! ✨Here first about our data and AI roadmap! ✨Learn how we empower Python at enterprise scale! Join us a SAS Innovate on Tour - at The Oval 👇 https://lnkd.in/dGDAa_Qd

Check out the latest "Quarterly What’s New Product Update - June 2024 Episode" to explore the cutting-edge features and enhancements in the MicroStrategy ONE platform. Discover how these updates can drive your business intelligence and analytics to the next level. 📈 Highlights include advanced AI capabilities, enhancements to data visualization, and much more. Watch the full episode here: https://ow.ly/wPp150StMR5 #BusinessIntelligence #DataAnalytics #AI #MicroStrategy #TechInnovation #ProductUpdate

Everyone knows how to use RAG to "talk to your documents," but did you know you can also use it to predict the future? Yes, you really can! 🚀 Read on to discover the potential of RAG in making informed, data-driven forecasts that might just outsmart human predictions: 🤔 What’s the most unexpected use of RAG you’ve encountered? Drop your thoughts or questions in the comments! #RAG #VectorDB #GenAI #ArtificialIntelligence #PredictionMarkets

As actuaries, we thrive on data. Machine learning enhances our ability to forecast, measure risk, and make data-driven decisions more effectively than ever before. "To predict the future, we must unlock patterns hidden in the past, and machine learning is the key." Excited to continue this learning journey and apply these tools to drive greater insights! #ActuarialScience #MachineLearning #DataScience #Sklearn #RiskManagement #PredictiveModeling #DataDriven #Innovation #AI #TechForActuaries #ContinuousLearning

📈Transforming Financial Data into Predictive Power with Decision Trees🌳 I recently applied machine learning to predict stock movements using key financial metrics. Here's a breakdown of the work: 🔍 Analyzed Competitor Data: Incorporated data from balance sheets, P&L statements, and cash flow metrics to get a complete financial picture. 💻 Used yfinance for Stock Data: Pulled real-time stock data using yfinance and created target variables by calculating means. 🌳 Built a Decision Tree Model: Trained a decision tree model to predict stock performance based on quarterly financial reports and competitor analysis. 📈 Results: The model effectively predicted stock trends, providing valuable insights for financial decision-making. This project combines financial analysis with machine learning to offer a robust method for predicting stock price movements. #MachineLearning #DataScience #StockMarket #Finance #DecisionTree #FinancialModeling #AI #CompetitorAnalysis #Python #yfinance #FinTech

🚀 New article is out! 🚀 This weekend, I decided to dive into a challenge I received last week and developed an article to share a step-by-step guide on document segmentation using LLMs and graph theory. This is a simple approach but filled with interesting insights for those who want to explore the world of language models and graphs to organize and analyze large volumes of text. I invite you to check out the article! I did my best to be as clear as possible, presenting visual elements, code snippets, and straightforward explanations to make it easier to understand. 📖 See you on Medium! 📖 https://lnkd.in/eVvJGMTW

Monotonic constraints in machine learning models can significantly enhance their reliability and causal accuracy, yet they remain underused. - 🔍 Causal modeling vs. traditional ML: Causal models offer more dependable “what-if” predictions, crucial for making data-driven business decisions. - 🏡 Real estate example: Using a simple house price dataset, adding monotonic constraints transforms predictions, improving actionable insights for businesses. - 📈 One-line code improvement: Integrating monotonic constraints into models like LightGBM or CatBoost is simple but drastically improves the accuracy of causal questions. #MachineLearning #CausalInference #AIInnovation - 💡 Causal models focus on understanding how changes in inputs affect outcomes, essential for predictive analytics. - 📊 Monotonic constraints ensure that model outputs behave in a predictable way, making them more interpretable for business decisions. - ⚙️ This method is easily applicable to existing models, requiring minimal code adjustments. https://lnkd.in/gNhcRvUj

Curious how you can visualize millions of glucose measurements from hundreds of thousands of people? Check out this visualization I made in a few hours today thanks to DuckDB and Observable: https://jaan.li/glucose It's interactive, and shows the glucose levels from 300,000+ hospitalizations of 150,000+ people. Code is in the link. What other types of data would you want to be interactive like this to find patterns? Comment below or message me :) I'm continuing to build my portfolio like this to work at the intersection of large scale machine learning, large language models, and interactive visualization of complex data.

Dive into this simple approach of using LLMs ,spaCy and spacy-llm to extract entities and relationships from text data. You won't believe what little code you end up using because of the amazing abstractions spacy-llm provides! https://lnkd.in/dJEPhzcy Thanks to Sofie Van Landeghem and Raphael Mitsch for their insights and review of the blog. And thanks to MantisNLP for being awesome and pushing us to do this as part of company research time!