Unleashing Python's Power Inside Oracle: A New Era of Machine Learning with OML4Py

Oracle Machine Learning for Python (OML4Py) represents a significant advancement in data science and machine learning by seamlessly integrating Python's extensive machine learning libraries with Oracle Database's robust infrastructure. This paper explores the architecture, key features, an...

Full description

Saved in:
Bibliographic Details
Main Authors: Krishnappa, Manjunatha Sughaturu, Harve, Bindu Mohan, Jayaram, Vivekananda, Pandy, Gokul, Ingole, Balaji Shesharao, Ramineni, Vishnu, Joseph, Shenson, Bangad, Nikhil
Format: Conference Proceeding
Language:English
Subjects:
Big Data Analytics
Data science
Database Integration
Machine learning
Machine learning algorithms
Multicore processing
OML4Py
Oracle Machine Learning
Predictive analytics
Python
Risk management
Scalability
Security
Telecommunications
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Oracle Machine Learning for Python (OML4Py) represents a significant advancement in data science and machine learning by seamlessly integrating Python's extensive machine learning libraries with Oracle Database's robust infrastructure. This paper explores the architecture, key features, and practical applications of OML4Py, highlighting its ability to execute Python scripts and machine learning algorithms directly within the database environment. By reducing data movement and leveraging the database's processing power, OML4Py enhances efficiency and scalability in data analysis workflows. We provide a detailed guide on implementing OML4Py, from installation and configuration to model training and deployment. A case study on predicting customer churn in the telecom industry demonstrates the practical benefits and performance improvements achieved with OML4Py. This integration not only streamlines the machine learning process but also ensures secure, high-performance data processing, making it a valuable tool for data scientists and developers working with large-scale data.
ISSN:2771-3075
DOI:10.1109/MCSoC64144.2024.00068