I would like to give a warm welcome again to the readers of The Information Age. Yesterday I decided not to post anything; not just because it was a religious holiday here where I am at the moment living, in my country of origin Portugal – and I normally do not bother with holidays, religious or otherwise, if I have something really important to do -, but also to give a little break as a pause to re-energize, and to help put ideas in order to find the best context possible for a post or share anything of real value.
With that said I decided to come back today. It will not be a very long post, like say reviewing a research paper, but it will be a re-share from a well-known other weblog/portal about Data Science/Big Data called Data Science Central. This portal is, as of now, well-known to me maybe for the good part of the last five years or so. I also have been posting some content there once in a while, and so it is no stranger to me and I am pretty sure and confident about its quality and editorial credibility.
The post is from Data Science Central’s chief editor and founder Vincent Granville, someone I also know well from social media connections. The topic he’s chosen is an important one: a course on implementing AI in the enterprise targeted for AI Engineers/Data Scientists. Artificial Intelligence (AI), as readers of this Blog and others quite already know, is deeply intertwined with Data Science’s theoretical concepts but also with Data Science practical implementation and data engineering issues.
Striking the very first image from the post: a cloud with the names of the major skills that any experienced AI/data scientist practitioner know to be required mastering for the success of any implementation (and then even some luck is also a requirement…😏):
The course described in the post will start this next January 2017, with editions scheduled for London and Berlin. It is created in partnership with H2O.ai, which is a leading open source machine learning platform that designs and implements smarter applications and data products. The course provides a certification of completion for successful participants.
I outline the further details below as an appetizer to you, but also as way to encourage further check the related content in Vincent’s post and in Data Science Central portal in general:
The course covers
- Design of Enterprise AI
- Technology foundations of Enterprise AI systems
- Specific AI use cases
- Development of AI services
- Deployment and Business models
The course targets developers and Architects who want to transition their career to Enterprise AI. The course correlates the new AI ideas with familiar concepts like ERP, Data warehousing etc and helps to make the transition easier. The course is based on a logical concept called an ‘Enterprise AI layer’. This AI layer is focused on solving domain specific problems for an Enterprise. We could see such a layer as an extension to the Data Warehouse or the ERP system (an Intelligent Data Warehouse/ Cognitive ERP system). Thus, the approach provides tangible and practical benefits for the Enterprise with a clear business model.
The implementation / development for the course is done using the H2O APIs for R, Python & Spark.
The course covers the following Enterprise AI Use Cases
- Fraud detection
- Anomaly detection
- Churn, classification
- Customer analytics
- Natural Language Processing, Bots and Virtual Assistants
The course comprises three parts
Section One: Implementing Enterprise AI
- Machine learning
- neural net and deep net
- Reinforcement learning
- Implementation in H2O of the use cases above
Section Two: Deploying Enterprise AI
Here, we cover the actual deployment issues for Enterprise AI including
- Acquiring Data and Training the Algorithm
- Processing and hardware considerations
- Business Models – High Performance Computing – Scaling and AI system
- Costing an AI system
- Creating a competitive advantage from AI
- Industry Barriers for AI
Section Three: Projects
Enterprise AI project created for AI use cases in teams.
I hope this share to be useful to as much readership as possible. And specially to anyone that usually checks and reads Data Science Central on a regular basis.
featured image: H2O.ai youtube channel