Adnan Masood, a Machine Learning Phd's musings on Life, Technology, Research & Development
The idea of using of Bayesian Belief Networks in digital forensics to quantify the evidence has been around for a while now. To provide qualitative approaches to Bayesian evidential reasoning in the digital Meta-Forensics is however relatively new in the decision support systems research. For law enforcement, decision support and application of data mining techniques to “soft” forensic evidence is…
Last week I spoke to Southern California .NET Architecture Group regarding implementation of Service Oriented Design Patterns with Windows Communication Foundation. The talk was an architectural overview of so called SOA tenants and how to implement these best practices using WCF. Being technologists, we tend to focus more on underlying technologies and have tendency to avoid topics like…
I have recently came across Wainwright & Jordan's paper on exponential families, graphical models, and variational inference and found it to be quite comprehensive and unifying introduction of the topic. Probabilistic graphical models use a graph-based representation as the basis for compactly encoding a complex distribution over a high-dimensional space. If you are familiar with Koller and Friedman's work on…
I will be speaking at the next SoCal IASA chapter meeting will be Thursday May 17, 2012 at Rancho Santiago Community College District, 2323 N. Broadway, Santa Ana. Meeting starts at 7:00 pm iA, pizza and networking 6:30 pm. RSVP by emailing to mike.vincent@mvasoftware.com if you plan to attend. Implementing SOA Design Patterns with WCF…
In Bayesian Artificial Intelligence, authors Kevin B. Korb and Ann E. Nicholson points out the non-Bayesian nature of Belief networks. The researchers note Many AI researchers like to point out that Bayesian networks are not inherently Bayesian at all; some have even claimed that the label is a misnomer. At the 2002 Australasian Data Mining Workshop, for example, Geoff…
Like distraction-free-easy-on-eyes Dark IDE's, most developers prefer clutter free green on black background for their text editors as well. On Lyx, it's fairly easy to do with step by step instructions here. This color scheme is somewhat similar to Darkroom for Windows. All you'd need to do is to modify your lyx preferences file with the…
As discussed previous post on Customizing Conditional Probability using Code Generation with SamIam, I have touched upon importance of having programmatic and declarative control over the network. Working with SamIam (and with Infer.NET to some extent) gives a researcher provides this flexibility which is hard to find in proprietary tools. Here is a simple example of a…
While we are having fun visualizing with tag clouds, here is one on the following four key papers in the area of pattern mining with Bayesian networks as background knowledge and discovery of interesting patterns based on Bayesian network background knowledge. Fast discovery of interesting patterns based on Bayesian network background knowledge Fast discovery of…
The Discipline of Machine Learning by Tom Mitchell Introduction to Support Vector Machines - Dustin Boswell Fast Training of Support Vector Machines using Sequential Minimal Optimization Introduction to linear regression The elements of statistical learning (book) Survey of Clustering Algorithms Supervised Machine Learning: A Review of Classification Techniques Ensemble Methods in Machine Learning The Boosting Approach to…
Here is an interesting word cloud built using dissertation by Hei Chan of UCLA on Sensitivity analysis of probabilistic graphical models. Following is the summary of the thesis. Probabilistic belief systems are used in artificial intelligence to model uncertainty. A popular framework for realizing probabilistic belief systems is to use graphical models, such as Bayesian…