If you ever had a hang, performance issue, memory leak, crash, or cryptic exception in a .NET application that you couldn't fix, then this session is for you.
Join Tess as she walks through a number of these issues and talks about how to debug them in this demo-intensive session.
Everyone working in a team, knows this situation: you feel, there is something standing in the room, which stops the team from making a reasonable decision. But none wants to talk about it openly. This is beating around the bush is dangerous for the success of the project.
This talk explains the differences between Elephants and Mooses in the Room and how you can recognize them. You will learn about the importance to name those giants to get them out of the room and don´t let them be a brain blocker to your success anymore. And it is a plea for a culture of transparency and openness to avoid these awkward, tough and sometimes really expensive situations.
Every day we are noticing that applications are becoming more intelligent. They can predict your online shopping preferences, movies you want to watch or interesting articles. Nowadays it’s hard to imagine a successful business that is not making a profit from some data forecasts.
Usually, any predictive analytics requires in-depth knowledge on machine learning. Companies need to think of hiring skilled staff able to build and manage complex models. This is the place when Azure Machine Learning Studio comes in. It offers low-cost, easy to use and managed environment for developers of all skills levels. During this talk Barbara will provide information about the following topics:
- Creating an Azure ML experiment
- Using variety of data manipulation techniques
- Customising ML process by using R modules
- Publishing and consuming an endpoint for predictive analysis
- Retraining the ML model
After this talk, attendees will get to know basics of Azure ML Studio. They will be able to use a variety of data sources, create experiments and use the predictions in their systems. The talk will provide them with the information on how to enrich their system by reasoning from data using proven and highly scalable ML technologies in an easy and low-cost way.
AML (Anti-Money Laundering) solutions typically tend to be rule engine driven and involve significant manual follow-up activities. Using a Machine Learning approach, AML solutions can be enhanced to reduce false positives, as well as to better prioritize the items flagged for manual follow-up.
This one-hour session will be structured as below:
First, we will briefly discuss the AML domain and the typical AML detection workflow.
Secondly, we will have an in-depth look into how Machine Learning algorithms can help with enhancing the AML solutions through better detection and better prioritization of detected fraud activity items.
Thirdly, we will look at how this can be implemented with Azure Machine Learning to achieve qualitative as well as quantitative enhancement objectives.
Finally, we will briefly look at the applicability of the Machine Learning approach to other areas within Financial Services domain like Insurance Claims Fraud Detection, etc.
Writing libraries and toolchains for your developer peers can be a challenge. Easy to use, flexible, testable, and extensible are all qualities we strive for when writing software that others will use. Hardware related tools specifically can be especially complex, where extra empathy and thoughtful process is often needed.
This session will be a walk-through of several case studies sourced from personal open source work, written for both the NodeJS Hardware Working Group and the vibrant worldwide Nodebots community at large. Lessons learned will be aplenty, providing some fresh perspectives on how deeply your code can improve the first impressions of those in the communities you wish to cultivate.