Analytics Maturity Models

In the world of big data analytics, there are several emerging standards for measuring Analytics Capability Maturity within organizations.  One of these has been presented in the TIBCO Analytics Maturity Journey – their six steps toward analytics maturity are:  Measure, Diagnose, Predict and Optimize, Operationalize, Automate, and Transform.  Another example is presented through the SAS Analytics Assessment, which evaluates business analytics readiness and capabilities in several areas.  The B-eye Network Analytics Maturity Model mimics software engineering’s CMM (Capability Maturity Model) – their 6 levels of maturity are:  Level 0 = Incomplete; Level 1 = Performed; Level 2 = Managed; Level 3 = Defined, Level 4 = Quantitatively Managed; and Level 5 = Optimizing.

The most “mature” standard in the field is probably the IDC Big Data and Analytics (BDA) MaturityScape Framework.  This BDA framework (measured across the five core dimensions of intent, data, technology, process, and people) consists of five stages of maturity, which essentially parallel the others mentioned above:  Ad hoc, Opportunistic, Repeatable, Managed, and Optimized.

All of these are excellent models for analytics maturity.  But, if you find these different models to be too theoretical or opaque or unattainable, then I suggest a more practical model for your business analytics progression from ground zero all of the way up to cognitive analytics:  from Descriptive and Diagnostic, to Predictive, to Prescriptive, and finally to Cognitive.

A specific example from the field of Marketing is SYNTASA‘s PMI (Personalization Maturity Index). Personalization Capability Maturity parallels the Analytics Capability Maturity frameworks within the specific context of data-driven customer-centric one-to-one marketing and segmentation of one. Read more about this in the article The Battle for Customer Personalization – Divisive Clustering is Good For Youand in much more detail within SYNTASA’s PMI white paper linked above.

(continue reading here:  https://www.mapr.com/blog/raising-standard-big-data-analytics-profession)

Follow Kirk Borne on Twitter @KirkDBorne

 

A Day in the Life of Confounding Factors and Explanatory Variables

Would we trust an insurance provider who sets motorbike insurance rates based on the sales of sour cream? Or would we schedule our space launches according to the number of doctoral degrees awarded in Sociology?

Probably all of us would agree that this kind of decision-making is unjustified. A specific decision like this appears to be only superficially supported by the evidence of correlations between those various factors, but is there more to the story? Does it go any deeper? What if there exists a hidden causal factor that induces the apparently spurious correlation?

For example, suppose the increase in space launches and the increase in doctoral degrees in Sociology were both related to an increase in government investments in research studies on the sociological impacts of establishing a permanent human colony on the Moon. This case reveals a hidden causal connection in an otherwise strange correlation. The explanatory variable (which is a hidden confounding factor) is the research investment, and the response variables are the space launches and doctoral degrees.

What about other cases? What about the evidence that sour cream sales correlate with motorbike accidents? In such cases, shouldn’t we all be pleased to see organizations making evidence-based data-driven objective decisions (especially in this brave new world of exploding data volumes and ubiquitous analytics)? No, I don’t think so!!

So, what kind of world is this?

Welcome to the world of explanatory variables and confounding factors!

Statistical literacy is needed now more than ever (to paraphrase H. G. Wells). This includes awareness of and adherence to common principles of statistical reasoning. For example…

(continue reading here http://www.statisticsviews.com/details/feature/7914611/A-Day-in-the-Life-of-Explanatory-Variables-and-Confounding-Factors.html)

Follow Kirk Borne on Twitter @KirkDBorne

Drilling Through Data Silos with Apache Drill

Enterprise data collections are typically stored in silos belonging to different business divisions. Sometimes these silos belong to different projects within the same division. These silos may be further segmented by services/products and functions. Silos (which stifle data-sharing and innovation) are often identified as a primary impediment (both practically and culturally) to business progress and thus they may be the cause of numerous difficulties. For example, streamlining important business processes are rendered more challenging, ranging from compliance to data discovery. But, breaking down the silos may not be so easy to accomplish. In fact, it is often infeasible due to ownership issues, governance practices, and regulatory concerns.

Big Data silos create additional complications including data duplication (and associated increased costs), complicated data replication solutions, high data latency, and data quality concerns, not to mention being an enabler of the real problematic situation where your data repositories could hold different versions of the truth. The silos also put a limit on business intelligence (discovery and actionable insights). As big data best practices rise above the hype and noise, we now know that actionable value is more easily extracted when multiple data sets can be integrated and viewed holistically.

Data analysts naturally want to break down silos to combine data from multiple data sources. Unfortunately, this can create its own bottleneck: a complex integration labyrinth—which is costly to maintain, rarely performs well, and can’t be guaranteed to provide consistent results.

In response, many companies have deployed Apache Hadoop to address the problem of segregated data. Hadoop enables multiple types of data to be directly processed in place, and it fully supports data integration from multiple sources across different data storage technologies.

Organizations that use Hadoop are finding additional benefits with Apache Drill, which is the open source version of Google’s Dremel system…

(continue reading here https://www.mapr.com/blog/drive-innovation-breaking-down-data-silos-apache-drill)

Follow Kirk Borne on Twitter @KirkDBorne

4 Reasons why an Accurate Analytics Model may not be Good Enough

Here are four reasons why the result of your analytics modeling might be correct (according to some accuracy metric), but it might not be the right answer:

  1. Your model may be underfit.

  2. Your model may be overfit.

  3. Your model may be biased.

  4. Your model may be suffering from the false positive paradox.

In data science, we are trained to keep searching even after we find a model that appears to be accurate. Data Scientists should continue searching for a better solution, for at least the four reasons listed above. Please note that I am not advocating “paralysis of analysis”, where never-ending searches for new and better solutions are just an excuse (or a behavior pattern) that prevents one from making a final decision. Good leaders know when an answer is “good enough”. We discussed this in a previous article: “Machine Unlearning – The Value of Imperfect Models”…

(For more discussion of the four cases listed above, continue reading herehttps://www.mapr.com/blog/4-reasons-look-further-accurate-answer-your-analytics-question)

Follow Kirk Borne on Twitter @KirkDBorne

The Definitive Q&A for Aspiring Data Scientists

I was recently asked five questions by Alex Woodie of Datanami for the article, “So You Want To Be A Data Scientist” that he was preparing. He used a few snippets from my full set of answers. The longer version of my answers provided additional advice. For aspiring data scientists of all ages, I provide here the full, unabridged version of my answers, which may help you even more to achieve your goal. (Note: I paraphrase Alex’s original questions in quotes below.)

1. “What is the number one piece of advice you give to aspiring data scientists?”

My number one piece of advice always is to follow your passions first. Know what you are good at and what you care about, and pursue that. So, you might be good at math, or programming, or data manipulation, or problem solving, or communications (data journalism), or whatever. You can do that flavor of data science within the context of any domain: scientific research, government, media communications, marketing, business, healthcare, finance, cybersecurity, law enforcement, manufacturing, transportation, or whatever. As a successful data scientist, your day can begin and end with you counting your blessings that you are living your dream by solving real-world problems with data. I saw a quote recently that summarizes this: “If you think your scarce data science skills could be better used elsewhere, be bold and make the move.” (Reference).

2. “What are the most important skills for an aspiring data scientist to acquire?”

There are many skills under the umbrella of data science, and we should not expect any one single person to be a master of them all. The best solution to the data science talent shortage is a team of data scientists. So I suggest…

(continue reading herehttps://www.mapr.com/blog/definitive-qa-aspiring-data-scientists)

Follow Kirk Borne on Twitter @KirkDBorne

Clear and Obvious Analytics for Clear and Present Dangers

Not every industry has found their clear and obvious applications of big data analytics. But the clear and present dangers of risk and fraud in financial transactions demand fast predictive modeling. Precisely because we live in the ubiquitous digital era, where most business (and non-business) transactions are rarely (if ever) in analog form and those transactions no longer move at the pace of humans (but at the speed of light), consequently the volume of digital signals as well as lurking dangers is enormous.

Digital signals (from sensors everywhere in our operational business systems) carry transactional information (what happened to what?), as well as metadata (descriptors) and analytics information (data-encoded knowledge and insights).  These analytics can be behavioral (providing insights into the interests, intentions, and preferences of the actors in a given transaction) as well as functional (providing insights into the actions or events associated with the transaction).

Behavioral analytics is developing into a major component of digital marketing, as firms seek to sell, cross-sell, and up-sell their products to the right customer at the right time.  Behavioral analytics is also critical in risk mitigation of all sorts: financial, cybersecurity, health (individual and population), supply chain, machine performance, and so on.

Here are 10 examples of where fast predictive analytics can play a vital role in most industries (with a focus on financial):

  1. Predict credit risk and fraud in real-time!
  2. Use Social Media for deeper understanding (likes and dislikes) of your customers.
  3. Personalize customer interactions in real-time, across multiple channels.
  4. Stop improper insurance payments before claims are paid!
  5. Spot insurance rate evasion tactics during the quote process – before you issue a policy!
  6. Predict High Health Risk versus Low Health Risk to better manage healthcare decision-making.
  7. Generate better predictive models of health, car, and home insurance eligibility fraud, underwriting fraud, and improper payments.
  8. Spot adversarial and anomalous behavior in cyber networks – stop the data breach or illegal funds transfer before it happens!
  9. Eliminate your Supply Chain hiccups – move the right products to the right locations in the right quantities – and at the right time!
  10. Make better business decisions regarding merchandising, demand forecasting, and pricing – don’t leave money on the table, or products in your warehouse.

Let us look a little more closely at the financial services industry…

One of the common conditions in traditional financial services (including home, health, and auto insurance) has been the “pay and chase” — i.e., you make the payment to the claimant, and then (after making the payment) you find out that the claim is fraudulent, thus beginning the chase to get your money back.

The new world of predictive modeling and advanced analytics allows for a new mantra in the financial and insurance industries: “Do Not Pay!” — i.e., you do not pay the claim until you have analyzed its likelihood for claim fraud, extraordinary financial risk, or payment anomalies (e.g., duplicate payments).

Predictive analytics modeling delivers a better financial risk posture for your organization than the “pay and chase”. With access to greater and more diverse data sources, it is now possible to develop better models of your customers’ credit risk regardless of the industry. This is certainly true in the financial services industry where there is so much data available: credit scores, credit history, court records, tax records, health records, insurance claims, and more. There is no excuse for not examining as much “public data” as you can in conjunction with other data sources that are available to you internally within your organization. Moderate outlays of your organization’s funds that are incurred in acquiring access to diverse external data sources should be offset by the savings accrued by “not sending your funds out the door” erroneously (either to intentional fraudsters or in unintentional duplicate claims).

An analytics-driven predictive model can predict fraud more efficiently (with fast automatic statistical software packages) and more effectively (with higher precision and higher recall: fewer false positives and false negatives) than traditional business processes. A good predictive analytics model should: (a) detect claims that “smell funny”, (b) prevent the “pay and chase” mode of operations, and (c) stop claims fraud abruptly by empowering a “do not pay” mode of operations. Predictive analytics modeling should aim to satisfy the following business requirements:

  • Detect and prevent both opportunistic and professional fraud throughout the claims process.
  • Detect underwriting fraud, to prevent premium leakage at the point of sale and renewal.
  • Spot rate evasion tactics during the quote process – before you issue a policy.

Many more examples of use cases in the financial services industry (and elsewhere) where fast predictive analytics is important (and what you can do about it) have been expertly enumerated by the fast statistical modeling folks from Soft10, Inc. Check out their fast analytics products (including the Instant Online Overbilling Claims Detector) at http://soft10ware.com/.

Follow Kirk Borne on Twitter @KirkDBorne

A Growth Hacker’s Journey Through the Recent History of Data Science

In 1998, I was attending a conference when an astronomer that I knew from across the country sought me out and asked if his group could send the data from their large astronomy experiment to NASA’s ADF (Astrophysics Data Facility, where I was working at the time). Their data set was two Terabytes in total. That seemed big (like the birth of “Astronomy Big Data”) to me, especially for 1998, but I didn’t know how big until I went back to work a few days later. When I mentioned this opportunity to the NASA facility senior managers, they looked at me like I was unaware of something really obvious and important. They were right! They “reminded” me that the data facility was the home for 15,000 NASA space science mission data sets, and the aggregate sum total data volume for all of those data sets combined(!) was less than one Terabyte! They couldn’t possibly accept one single experiment’s data that single-handedly eclipsed the total volume of all of the other 15,000 experiments’ data sets combined.

Well, this was embarrassing! What could we do? I was told that ADF could accept the data if I would write a research grant proposal and win some funds to pay for all of the new I.T. resources that would be required. “What kind of research proposal would pay for such a thing?” I asked myself. This led me to investigate a field of research that I had only briefly heard in conversation once or twice previously — Data Mining (= Machine Learning applied to large data sets). The more I read about this topic (now called Data Science), the more I became convinced that this is what I wanted to do for the rest of my research career. I was hooked. I was at the right place at the right time…

(continue reading herehttps://www.mapr.com/blog/growth-hackers-journey-right-place-right-time)

Follow Kirk Borne on Twitter @KirkDBorne

These are a few of my favorite things… in Big Data and Data Science: A to Z

A while back, we made a list from A to Z of a few of our favorite things in big data and data science. We have made a lot of progress toward covering several of these topics. Here’s a handy list of the write-ups that I have completed so far:

AAssociation rule mining:  described in the article “Association Rule Mining – Not Your Typical Data Science Algorithm.”

C – Characterization:  described in the article “The Big C of Big Data: Top 8 Reasons that Characterization is ‘ROIght’ for Your Data.”

H – Hadoop (of course!):  described in the article “H is for Hadoop, along with a Huge Heap of Helpful Big Data Capabilities.” To learn more, check out the Executive’s Guide to Big Data and Apache Hadoop, available as a free download from MapR.

K – K-anything in data mining:  described in the article “The K’s of Data Mining – Great Things Come in Pairs.”

L – Local linear embedding (LLE):  is described in detail in the blog post series “When Big Data Goes Local, Small Data Gets Big – Part 1” and “Part 2

N – Novelty detection (also known as “Surprise Discovery”):  described in the articles “Outlier Detection Gets a Makeover – Surprise Discovery in Scientific Big Data” and “N is for Novelty Detection…” To learn more, check out the book Practical Machine Learning: A New Look at Anomaly Detection, available as a free download from MapR.

P – Profiling (specifically, data profiling):  described in the article “Data Profiling – Four Steps to Knowing Your Big Data.”

Q – Quantified and Tracked:  described in the article “Big Data is Everything, Quantified and Tracked: What this Means for You.”

R – Recommender engines:  described in two articles: “Design Patterns for Recommendation Systems – Everyone Wants a Pony” and “Personalization – It’s Not Just for Hamburgers Anymore.” To learn more, check out the book Practical Machine Learning: Innovations in Recommendation, available as a free download from MapR.

S – SVM (Support Vector Machines):  described in the article “The Importance of Location in Real Estate, Weather, and Machine Learning.”

Z – Zero bias, Zero variance:  described in the article “Statistical Truisms in the Age of Big Data.”

The Goldilocks Principle in Predictive Modeling and Data Science

In the field of statistics, there has been a lot written about statistical fallacies, logical fallacies, and fallacious reasoning. The following big list of fallacies is one that I like to use in my own undergraduate data science courses, particularly in my Data Ethics class where I teach my students about “lying with statistics”:

http://en.wikipedia.org/wiki/List_of_fallacies

Many of these fallacies are relevant to data science modeling, including this one: Circular Reasoning, where the reasoner “begins with what he or she is trying to end up with; sometimes called assuming the conclusion.”

A broken clock is truly an example of circular reasoning (as the dial is circular, and the clock represents a particular measurement in a repeating circular perspective): “Even a broken clock is right twice a day.”

(source: http://tvtropes.org/pmwiki/pmwiki.php/Main/StoppedClock)

In the following article, I use the broken clock analogy for circular reasoning in describing the importance of verification and validation in predictive analytics models: Are your predictive models like broken clocks? Here’s how to fix them.”  The article also discusses the importance of training vs. test data sets, the bias-variance tradeoff in data science modeling, underfitting vs. overfitting, and the Goldilocks Principle applied to data science.

(continue reading here) 

Follow Kirk Borne on Twitter @KirkDBorne

Learning from Data Big and Small — What’s the Shape of Your Data?

(A version of this article was originally published on BigDataRepublic.com in July 2013 — that site no longer exists.)

Does discovery depend on the scale of your experiment? In some cases, no! Whether Christopher Columbus sailed with 3 ships or 3000, he still would have found the New World, probably in the same amount of time. In this case, the existence of the Americas is independent of the scale of the exploration resources. Conversely, there are many more cases where the potential for discovery does scale with the size of your resources. If those resources are Big Data, then prepare to say “hello, world” to many more new worlds (and new discoveries). The good news for small-to-mid scale projects is that, even without Big Data, you can still be a Columbus.

Learning from small data has justifiably acquired a faithful following of advocates (see this and this).  Let us illustrate this with a common example: Time Series Analysis.

In a simple single-parameter data stream, you can extract characterizations from the time series: (a) the change since the last value (y2-y1); (b) a running mean (e.g., the average of the last 3 data points = [y1+y2+y3]/2); (c) the slope of the trend line (= velocity = dy/dt = [y2-y1]/[t2-t1]); (d) the rate of change of the trend line slope (= acceleration = the 2nd derivative of the data d2y/dt = {[y3-y2]/[t3-t2] – [y2-y1]/[t2-t1]} / [t3-t1] ); (e) the rate of change of acceleration (= jerk); and so on.

Stock market day traders watch 2nd derivatives more closely than the other time series characterizations, since that parameter can signal an inflection point in the data series. Inflection points (a change in the sign of the 2nd derivative) can thus be used as a predictor of an impending turn-around point (maximum = time to sell; or minimum = time to buy) in the time series.

These simple statistical metrics are therefore valuable and informative in some circumstances.  Somewhat more interesting characterizations include the shape of the variation: e.g., U, V, or W. These symbolic representations of temporal behaviors can be quite powerful for sequence mining, pattern discovery, transition detection, and trend analysis in time series data, as well as for the all-important dimensionality reduction and indexing of massive complex data streams.

If the time series stream of data is dense (in time), then you can do a spectral (frequency) analysis to measure the strength of patterns in the time series on all scales (high-frequency to low-frequency) — this is called Fourier Analysis. This analysis gives you a large number of characterization metrics (e.g., the frequency components and their amplitudes) for dense time series.  You can monitor these metrics and alert the end-user only when the power spectrum of the different frequency components changes significantly, even if the change is in only one component (e.g., its phase or amplitude) or if a new component appears (e.g., an hourly fluctuation in data that previously only showed daily fluctuation).

Finally, imagine massive parallel streams of data: Big Time Series Data. Now the fun begins! Such parallel streams may be Twitter timelines for hundreds of millions of users, or streaming data from hundreds (or thousands) of sensors in an airplane or manufacturing plant, or streaming transaction data from millions of retail shoppers or for a large financial firm. Monitoring massively parallel data streams in this way may be a perfect job for a distributed computing environment: Map-Reduce and Hadoop.

At each step (or within each incremental time range) of such massive data streams, you can create a data distribution histogram of the data values Y (or a histogram of trend line slopes dY, or of 2nd derivatives d2Y) across the full ensemble of parallel data streams. You can then estimate a variety of statistical metrics for the separate data distributions (i.e., one set of metrics each for Y, dY, d2Y, and others) as a function of time: mean, median, mode, variance, skew, kurtosis, presence of a long tail, mixture models, and more.  (Of course, if the data are textual, as in Twitter comments, then some form of numerical coding of the text will yield a goldmine of value – that’s a story for another article.)

Exploiting a variety of statistical metrics (data stream characterizations) such as these is where the exploration and discovery potential expands significantly. Similar to the small-data cases described earlier, the values of these characteristic statistical metrics on massive data streams become a model for the state of the system that you are monitoring. The model itself can be monitored and flagged for significant changes in these characteristic statistical features or for the appearance of new features in the data streams. As long as the massive parallel data streams continue to behave in predictable consistent patterns (which is called a “stationary state”), then there is no need to alert the end-user. However, when the stationarity of the data stream model changes (perhaps triggered by a change in any one of the state parameters that exceeds a pre-specified threshold), then a signal is raised and the end-user verifies whether a truly new behavior or event has been discovered. Land ahoy! All hands on deck!

The point of these examples is to demonstrate that discovery and learning from small data is still useful and valuable. As the data set becomes increasingly larger, it is then possible (and likely) that more intricate, subtle, and descriptive features within the data will be revealed. The discovery potential of bigger data thereby increases (perhaps exponentially). Additionally, the nature and diversity of the discoveries become richer, and maybe so will you!

Follow Kirk Borne on Twitter @KirkDBorne