Think of “continuum” as something you start and you never stop improving upon. In my mind, Business Analytics Continuum is continuous investment of resources to take business analytics capabilities to next level. So what are these levels?
Here are the visual representation of the concept:
I love mental model and frameworks. I have shared some frameworks on this blog already like 3 W’s (What, Why, what’s next) and 3 P’s (Platform, People, Process) focused on helping analytics leader figure what their analytics roadmap should be. I was reading ‘competing on analytics’ book and came across the 5 stages of Analytical competition which seemed like another framework worth sharing.
The two end of the spectrum are org is flying blind to org is competing through analytics. Stages are:
You can read about each one of these here: Five Stages of Analytic Competition and you can read a synopsis of the book here.
There are two broad range of algorithms that can help you detect fraud: 1) classification (supervised) 2) clustering (unsupervised)
Now it’s a fair assumption that fraud is pretty rare and it’s an outlier in your data. In other words, it’s a anomaly and the process of identifying them is called Anomaly Detection.
So under classification, there are algorithms out there specialize in “anomaly” detection like one-class SVM and PCA based anomaly detection. Try them out on your dataset and see if it’s able to capture “anomalies” in your dataset. While you are at it, don’t discount traditional classification algorithms either, they may be useful as well. You will have to train these algorithms and that’s why they are called “supervised”.
There an alternate approach. Which is to use unsupervised algorithms called “clustering” techniques. You could try something as simple as K-means or something more sophisticated. I haven’t used clustering much for solving fraud problems and have usually deferred to anomaly detection algorithms for this. But I am throwing this out there for making sure you know all the options! I can see these algorithms being applied to exploratory analysis where you are just exploring your data to find outliers to study them.
Hope that helps!
I watched this video over the weekend and wanted to share this very well done presentation by a Venture Capital (VC) firm with you — that’s why I love following VC’s (especially one’s who invest in Data/Analytics theme) since they tend to share some amazing insights on where the industry is going.
Abstract:
“One person, in a literal garage, building a self-driving car.” That happened in 2015. Now to put that fact in context, compare this to 2004, when DARPA sponsored the very first driverless car Grand Challenge. Of the 20 entries they received then, the winning entry went 7.2 miles; in 2007, in the Urban Challenge, the winning entries went 60 miles under city-like constraints.
Things are clearly progressing rapidly when it comes to machine intelligence. But how did we get here, after not one but multiple “A.I. winters”? What’s the breakthrough? And why is Silicon Valley buzzing about artificial intelligence again?
From types of machine intelligence to a tour of algorithms, a16z Deal and Research team head Frank Chen walks us through the basics (and beyond) of AI and deep learning in this slide presentation.
I classify analytics into four kinds:
With that overview, let’s look at Descriptive analysis a little bit more. This is usually the first step for any organization to start getting value out of all their data. They should be able to answer questions like: What were my sales last quarter? How about same quarter last year? Then compare them to see if they made progress. They can also report on sales (Actual vs goals) for last n months and see if they are trending in the right direction. Things like this! Once you have a good process and platform to get this right then the organization is ready to advance to next step which is diagnostic and this is where you start analyzing the key drivers and underlying reasons to figure “why” it’s happening. But you need start at Descriptive!
Hope that helps!
Originally published on Quora. Link Here
“Machine Learning” is a subset of “Data Analysis” — it’s just one of the activities that you could apply to solve a data analysis problem, you just need to find a problem that can use machine learning wizardry! What kind of activities?, you say — well, to answer that we will need to step back and categorize what problems could be solved by Data Analysis. There are broadly three kinds of problems:
To conclude: Depending on the analytics maturity of your organization and the business problem at hand, you might have to use Machine learning to solve a data analyis problem…And it never hurts to pick up Machine learning basics along with other data analysis skills that you might have.
Hope that helps.
A machin
e learning algorithm in Azure ML has few parameter settings that you can set — in this post, we will talk about 1) why you should NOT stick with default settings 2) How can “Tune model hyperparameters” module help you do so?
So first up, why you should not be using the default setting? The parameter settings that are applied to a model impacts the accuracy (or call it predictive power) of the algorithm…sometimes it may be significant and sometimes not but either case, you won’t know until you try changing the default values. In other words, by tuning the hyperparameters you could significantly boost your model’s performance!
Now, how do you go about setting the parameters such that it gives optimal performance? Let’s say that there are 3 parameters then that is 27 different combinations! How do you know which one is the best? You could dig a little deeper into the mechanism of how algorithms works and narrow down your list but that would still take some time. So, there should be a better way, right? Luckily there is: This where “Tune Model Hyperparameters” comes in! You can use it with any algorithm in Azure ML. This module helps you tune the hyper-parameters. There’s some things that you still need do like decide: Do you want the module to just try random n combinations? OR Do you want the module to try all combinations (fyi: this is compute-intensive operation!)? … AND you will have to decide what are you are optimizing for? Depending on the algorithm it would let you pick the evaluation metric that you want to optimize.
Now, there are some good articles already written that walks you through how to get about doing this so I am going to share these links with you:
I want to conclude by sharing few tips:
Notes from the field:
Most organizations could benefit from Analytics but before you set the Analytics road-map for your organization, it’s important to figure out your current stage and then build the road-map to achieve your vision. So how do we figure out the analytics maturity of an organization? Let me share a framework to think about this:
I have blogged about “Business Analytics Continuum” before — it’s a great framework to think about Analytics maturity in an organization — BUT the issue is that it’s harder for business people to remember the stages: Descriptive -> Diagnostics -> Predictive -> Prescriptive — And so there’s a simpler (but equally effective) framework that I have been using over past few months (What -> Why -> What’s next aka “3W” framework). And recently at a Microsoft Analytics conference, I saw this framework with an extra “W” which makes total sense that I liked a lot! So i thought I will share that with you all. So here you go — 4W framework:
Stage 1: What Happened?
Stage 2: Why did it happen?
Stage 3: What will happen?
Stage 4: What should I do?
I hope the framework as you think about your organization’s analytics vision/road-map and stages that you need to go through to help your org succeed with data!
Recommendations:
Building data driven companies — 3 P’s framework.
I wrote a post on BigDataCloud.com community comparing cloud based Machine Learning platforms: Amazon ML vs Microsoft Azure ML vs Databricks Cloud
My goal for the post were to: 1) share a framework to compare cloud-based Machine Learning platforms 2) Apply the framework to three platforms to see how they stack up.
Here’s the framework:
Please read the rest of the post on BigDataCloud.com. Azure ML vs Amazon ML vs Databricks Cloud.
If you’re getting started with Data Science & Machine Learning then I think this would be a great resource for you. This “cheat sheet” helps you select the “algorithm” to test depending on the problem you are trying to solve and the data-set that you have.
Download link: http://aka.ms/MLCheatSheet (Courtesy: Azure Machine Learning)
Also, even though the cheat sheet was created to help you with “Azure Machine learning” product, it’s still valid if you use other machine learning tools.
