Unleashing the Power of Machine Learning in Google Cloud

Artificial Intelligence is a red-hot topic in today's tech landscape, but its practical application often remains a mystery. Let Martin Weber be your guide and take you on a journey that will teach you how to turn your own data into a strategic asset in our increasingly data-driven world. In this session, he will break down the basics of machine learning and will help you gain hands-on experience of Big Query and its machine learning capabilities.

Transcript

2. Ihr lokaler Cloud-native Experte
3. 3 ABOUT ME Engineer at Posedio since 2022 Developer for around 5 years Doing my masters at TU Wien I like lifting (somewhat) heavy stuff
4. GCP AND MACHINE LEARNING USING ML MODELS IN BIGQUERY MARTIN WEBER 13 June 2024
5. AGENDA 1. Introduction to ML 2. Dataset and Model 3. Using BigQuery 4.Inspecting the Model 5. Wrapping it all up
6. INTRODUCTION TO ML 1
7. 7 WHAT IS ML
8. 8 TYPES OF ML Supervised learning
9. 9 TYPES OF ML Supervised learning
10. 10 TYPES OF ML • Unsupervised Learning
11. 11 TYPES OF ML • Unsupervised Learning
12. 12 TYPES OF ML • Reinforcement learning
13. 13 BIGQUERY OVERVIEW What do you want to do? Predict between categories Time Series Forecasting Logistic Regression Boosted Trees Classifier ARIMA-PLUS Matrix Factorization Generate Recommendations Legend Start Tasks Models Examples Autoencoder Reduce Data Dimenstionality PCA DNN Classifier Wide & Deep Classifier AutoML Table Classifer Group data into clusters K-Means Predict values Linear Regression Find Anomalies Identify spam emails Identify tumor types Product recommendation Create personalized content Perform customer segmentation Predict sales figures Predict stock prices Predict housing prices based on historical data Detect fraud Predict credit risk Analysis of written text Analysis of DNA data Boosted Trees Regressor DNN Regressor AutoML Table Regressor Wide & Deep Regressor • Reinforcement learning
14. DATASET AND MODEL 2
15. 15 RECOMMENDER SYSTEMS
16. 16 MATRIX FACTORIZATION N k N M M
17. 17 EXPLORING THE DATASET Movies • id, title, genre User Ratings • user_id, item_id, rating, timestamp
18. 18 DISTRIBUTION OF MOVIE RATINGS
19. 19 NUMBER OF RATINGS PER MOVIE
20. 20 MOVIES BY GENRE
21. 21 MEAN RATING BY GENRE
22. USING BIG QUERY 3
23. 23 TRAIN MODEL
24. INSPECTING THE MODEL 4
25. WRAPPING IT UP 5
26. 26 HOW MUCH DID IT COST
27. 27 KEY TAKEAWAYS Easy to get into Fast results Cheap Training Potential to optimize further
28. THANK YOU! POSEDIO GMBH +43 660 9154419 contact@posedio.com Weyringergasse 1-3, 1040 Wien, Millenium Park 4, 6980 Lustenau www.posedio.com