Fresh Links Sundae encapsulates information I have come across during the past week. Often they are from the people whose work I admire or resonate with me. I hope you will find these ideas thought-provoking at the minimum. Even better, I hope these ideas will, over time, help my fellow IT pros make better decisions, be awesome, and kick ass!
Predictive analytics can never offer any ironclad guarantees around prediction, so how do you evaluate a new tool or a new model? Theos Evgeniou offers some basic questions in evaluating new predictive models to help you get the most out of your predictive analytics. How to Tell If You Should Trust Your Statistical Models (Harvard Business Review)
Some organizations implement ITSM with a number of sophisticated processes that end up too bureaucratic to be effective. Stuart Rance suggests how you can simplify things but stay effective at the same time. Do you really need all those cumbersome processes? (Optimal Service Management)
Jason Brownlee believes that, in order to get good at applying machine learning algorithms, you need to build up an intuition of how an algorithm behaves on real data. He describes the process we should follow when studying machines learning algorithms. How to Build an Intuition for Machine Learning Algorithms (Machine Learning Mastery)
Some organizations’ metrics programs fail while others are successful. Phyllis Drucker outlines the steps for organizing your metric framework and how you can kick your metrics game up a notch. A five-step framework for business oriented metrics (The ITSM Review)
As a product manager for cloud services, Alex Bordei’s mission is to make sure his team gets the highest performance possible out of the technologies used in their services. He discusses how NoSQL databases can scale vertically and horizontally, and what you should consider when building a cluster. Scaling NoSQL databases: 5 tips for increasing performance (O’Reilly Radar)
When we train and deploy machine learning models for big data analytics, we run several risks when we strive for perfection and over-train the model. Kirk Borne advocates that we can reap great benefits from data analytics by having fast, simple, slightly imperfect machine learning. Machine Unlearning: The Value of Imperfect Models (MapR)
A number of IT organizations have transformed themselves from a technology-oriented to a services-oriented organization with the practice of IT service management. John Worthington discusses the various approaches and the continual cycle of service definition. What Does It Mean for IT to Be Customer-Focused? (VMware CloudOps)
Some people feel the traditional IT enterprise architecture (EA) can be too rigid or not flexible enough for today’s fast-changing environment. Charles Betz shares his ideas of what an Agile EA approach might look like. Agile and Enterprise Architecture (lean4it)
You must be logged in to post a comment.