Erik on Software

Rave Run: Arkansas River Trail and Clinton Presidential Park (Little Rock)

The fifth post in the series I started a couple years ago, unabashedly based on a regular spread in Runner's World Magazine called Rave Runs, in order to share interesting routes with other runners. Why a post on running within a blog that concentates on business and technology? One habit that I picked up on the road as a consultant over the years is always inquiring about local running paths while visiting unfamiliar locales for the first time. As discussed in other posts, I find running a pleasurable way to explore new places, and running has been a gift to me from an early age by helping me focus and providing stress relief.

A few weeks ago, I visited Little Rock, Arkansas. This visit was not my first. Perchance, my first visit to the area was due to a Fayetteville based consulting project, which resulted in visits to several other Arkansas locations for on-site testing of the software the team had designed and developed. The weather was incredibly hot and humid during this time period, and subsequent visits have revived those memories, but a rather large storm passed through the area the first day of this recent trip, leaving in its wake some great running weather for this time of year.

The Arkansas River Trail System (second map below) is an expansive 88-mile loop in Central Arkansas, although the portion on which I recently ran is the 15.6-mile smaller loop (first map below) linking Big Dam Bridge in the west and Clinton Presidental Park Bridge in the east.

My daily runs started at the Residence Inn downtown location, immediately in front of which runs the River Rail Streetcar along its 3.4-mile route that links Little Rock and North Little Rock (although as the hotel concierge noted, it does not run during inclement weather, as was the case during the first day of my stay, and so the photo below was taken toward the tail end of my trip).     

If you know me or have been following this series of posts, you already know that I am an early morning runner. The photos that I have included here were taken later in the day and not during my runs, due to lack of sunlight and the need to minimize disruption. While there are lamp posts in this area of downtown Little Rock, be aware that it can get quite dark. While I prefer to run solo, since it provides a time for meditation, you might want to consider having a fellow runner join you if running at night. While retracing a portion of my runs to take these photos, it is always amazing to me how much different a new locale looks with a bit of sunlight. Running under the first overpass in an easterly direction from downtown, for example, was a bit foreboding, and most of the artwork was not visible.

Unfortunately, most of the paths in this portion of the Arkansas River Trail System are paved. While the paths can be used for biking, running, walking, and a number of other pedestrian forms of travel (some Segways made an apparence, for example), most of the trails that I traversed are arguably not for biking, due to speed and visibility, although designers probably did not have runners in mind. Runners should mimimize running on concrete and asphalt. Just east of the first overpass, a fork in the path leads to either the William E. "Bill" Clark Presidential Park Wetlands, to the left, or to the Clinton Presidential Center by continuing straight ahead.

Although most of the paths in this portion of the Akransas River Trail System are paved, much of what makes this a rave run are the varied running paths. The first photo below shows the four types of running paths, which include the aforementioned concrete, stairways, wooden boardwalks, and pedestrian-only bridges. Several stairways dot the landcape, and the ones following the pattern depicted in the photo below are in my opinion well suited for running. Note that while the size of the wetlands might be 13 acres, from a running perspective the boardwalk is a relatively short distance.
 

 

A path to the Clinton Presidential Park Bridge immediately follows the wetlands. This pedestrian bridge is 1600 feet long and 17.5 feet wide. With expansive views of the Arkansas River and downtown Little Rock, traversing this bridge is a must, especially for runners new to the area. Little Rock is the only city in the United States with four pedestrian bridges that stretch over a navigable body of water. As with the aforementioned underpass, be aware that visits outside of daylight hours might be a bit foreboding, especially because the northern entrance to the bridge connects to a densely populated residential trailer park.

According to the sign at the entrance to the bridge, the Choctaw, Oklahoma and Gulf Railroad (CO&G) originally built the railroad bridge in 1899. After a hostile takeover by the Chicago, Rock Island and Pacific Railroad a few years later, the bridge was renamed the Rock Island Bridge. Later purchased by the Arkansas Gazette in 1967 after passenger service ended, five of the six bridges in Little Rock were slated for improvement due to the implementation of the McClellan-Kerr Arkansas River Navigation System, which created a safe and reliable navigation channel for barge traffic.

In 1980, the Rock Island Bridge ceased operation and was acquired by the Missouri Pacific Railroad (MoPac). By the mid-1990s, the Union Pacific Railroad prepared plans to tear down the bridge due to the liability that it represented, but in 2001 the City of Little Rock assumed ownership and retains ownership today. In 2010, the City of Little Rock officially changed the name of the bridge to the Clinton Presidential Park Bridge as a tribute to the Clinton Foundation's efforts to renovate it into a ramped pedestrian pathway.

As the above maps show, the Clinton Presidential Center resides at the easternmost location of the Arkansas River Trail System, and will probably prove to be a good resting stop for many a runner, with water fountains and shade for daytime runners on paths largely without shelter. Although I did not actually enter the building, this is only the second presidential library I have visited in the United States, after having visited the Abraham Lincoln Presidential Library and Museum during opening weekend in April 2005.

---

Rave Run: Buffalo Bayou (Houston, Texas)

Rave Run: C & O Canal Towpath and Potomac River Waterfront (Georgetown)

Rave Run: Great Western Trail (Lombard, Illinois)

Rave Run: Ridges Road in Door County (Baileys Harbor, Wisconsin)

New Book Review: "Practical Machine Learning"

New book review for Practical Machine Learning: Innovations in Recommendation, by Ellen Friedman and Ted Dunning, O'Reilly, 2014, reposted here:

This work is a short, white paper sized text that walks the reader through practical aspects of machine learning recommendation, specifically for those who are new to this space. After discussing the build versus buy dilemma with respect to incorporation of a recommendation system into the enterprise, the authors stress simplicity for ease of adoption and maintenance, and state that smart simplification in the case of recommendation is the focus of this paper, focusing on user behavior, co-occurrence, and text retrieval. Following this introductory discussion, the authors delve into a high-level discussion of choosing and collecting data for input into a recommendation system, building the recommendation model using co-occurrence, how Apache Mahout builds a model, and use of Apache Mahout with Apache Lucene (composed of Solr and Lucene-Core).

For readers already somewhat familiar with this problem space, it is not until Chapter 6 ("Example: Music Recommender") that the authors walk through an example that is used for a machine learning course developed by MapR Technologies. While still at a high level, this example discusses data sources, recommendations at scale using an Apache Hadoop based cluster, use of search to make recommendations, dithering, anti-flood measures, and multimodal and cross recommendation. While some technologists will likely be inclined to initiate work in this area by hitting community websites for the open source projects discussed, this book provides an accessible introduction to machine learning recommendation, and references to commercial implementations such as LucidWorks Search and the MapR distribution of Hadoop can be largely ignored.

New Book Review: "Star Schema"

New book review for Star Schema: The Complete Reference, by Christopher Adamson, McGraw-Hill Osborne Media, 2010, reposted here:

For technologists looking for a thorough exploration of the traditional data warehouse star schema, nothing compares to this book. Other reviewers are correct that although works by Ralph Kimball are classics in this space, the content that Adamson provides here is presented in much greater detail with a far greater quantity of examples, and written very well. While I found myself referencing "The Data Warehouse Toolkit: The Definitive Guide to Dimensional Modeling" by the aforementioned author during a recent data warehouse project, because that resource is introductory many practitioners will likely end up referring to the design tips that The Kimball Group website offers as supplementary material on a more frequent basis, and many of these design tips are really only a point in the right direction.

So to get to the point, Adamson offers great advice and explicit step-by-step explanations for many star schema topics. The only area that the author does not delve into is material around specific database products, but readers should be able to research this information after following along with the presentation that the author offers with this work. After discussing the fundamentals of analytic databases and dimensional design, data warehouse architectures, and stars and cubes, the author offers detailed presentations on fact tables, dimensions, hierarchies, bridges, snowflakes, and slowly changing dimensions, followed by discussions on performance and additional topics of interest for developers of dimensional models, including an exploration of some common dimensional features that often strain business intelligence tools, techniques to mitigate any shortcomings, and short discussions on process, design, and documentation.

One great example of the thoroughness of this book is the presentation that the author offers in Chapter 10 ("Recursive Hierarchies and Bridges"). After discussing types of recursive hierarchies (balanced versus unbalanced and attribute-based versus instance-based), reporting challenges associated with recursive hierarchies, and the process and implications with flattening such hierarchies, Adamson delves into the hierarchy bridge, providing a detailed walkthrough of how to construct it and use it, how to avoid double-counting, how to hide the bridge from novice users, how to resolve the many-to-many relationship, and how to handle Type 2 changes to the hierarchy. The step-by-step explanation of this last item is especially well done, with great diagrams and a detailed discussion of the ripple effect of such changes and why you should not resist the ripple effect, despite the additional data involved. This explanation contrasts markedly with other material on this subject. While a single design tip called "Help for Hierarchies" from The Kimball Group, for example, discusses this subject, it is done so at a very high level.

As a data architect working through this book for the first time during a project, most of my time was spent in Part 3 ("Dimensional Design"), which breaks down into Chapter 6 ("More on Dimensional Tables"), Chapter 7 ("Hierarchies and Snowflakes"), Chapter 8 ("More Slow Change Techniques"), Chapter 9 ("Multi-Valued Dimensions and Bridges"), and Chapter 10 ("Recursive Hierarchies and Bridges"), although I did spend significant time with Chapter 4 ("A Fact Table for Each Process"), Chapter 11 ("Transactions, Snapshots, and Accumulating Snapshots"), and Chapter 12 ("Factless Fact Tables"). While I will likely continue to reference a number of other resources in this space, this is the one book currently in the marketplace that has staying power for the long-term. Highly recommended.

Pro Bono: IEEE Chicago Section Statistician - Part 5

My last post in this series on my pro bono efforts outside of the workplace as Statistician for the IEEE Chicago Section mentioned that I walked the Membership Development Chair through the visualizations of my latest monthly presentation to the team, partially in preparation for his request of me to present to the executive committee. Unfortunately, the executive committee meeting that included a number of presentations by other individuals was already behind its already tight schedule at the outset, and so my presentation time ended up being reduced to 15 minutes.

While my presentation was originally intended to cover some of the visualizations and findings of the past year, delivery probably would have benefited from a scaled down scope that concentrated on data of recent months, including the month of March 2014 which was quite coincidentally the best month across the 8 years covered by the database in terms of total active membership net difference (membership additions less membership deletions), as well as April 2014 which was the first month during this same time period where total active membership "closed the gap" in YOY total active membership.

In the above plot, the line representing calendar year 2014 is on the bottom. Readers of this blog might recall that total active membership has been declining each year, but this trend has just been broken since total active membership is now above the same points of the previous year two months in a row. (Note that my plan is to overhaul these visualizations to account for color blindness. To be brief, two individuals - one of the executives discussed here, and one of my work colleagues in the workplace - recently mentioned that they are color blind. The first individual could not tell that the year 2014 was encoded with the color yellow, and the second could not distinguish between the blue and green colors I had chosen for Facts and Dimensions in the data warehouse model that I had built for the team.)

One positive result of my presentation to the executives is that the new Section Chair/President had a number of follow-up questions to ask later that same week which I was able to address. Essentially, what these questions encompassed was the ability to forecast/predict what the upcoming year is going to look like in terms of membership numbers. Questions such as this came as no surprise to someone in the trenches with the data. However, in response I emphasized that what I have been putting together for the team is descriptive analytics, not predictive analytics. No visualizations that have been provided to date should be seen as predictive analytics.

As explained to the new Section Chair/President, I could easily create what some might view as predictive graphs, such as trend lines etc, but I would be doing the Section a disservice by doing so, since these would not be accurate. In thinking about what helps predict future months, it really has a lot to do with individual members. It also has a lot to do with the economy and general industry trends, but these are fairly abstract concepts - even the government cannot predict unemployment rates and is always adjusting past months. In thinking about individual members - their demographics, education, fields of work, technical interests, etc - these are the core factors upon which models can be built to predict the types of individuals we are attracting and the types of individuals who choose to leave. Unfortunately, we do not have access to member credit ratings, but if we did, these might additionally help predict who will fail to keep up with their membership dues and go into arrears each year.

Assembling data sets created from the database in order to start capturing this information might be a start, but it would be a slow process because I would only be able to assemble data sets of the current month for which data is available. Once the data is updated for the next month, much of the information of the prior month is no longer made available in the database. The aggregate totals are still there, but not the data from which these aggregates were created. As discussed with the team, at minimum I would need the ability to query the database to determine whether an individual was active for a particular month in the past, but the historical member grades that the database tracks for each individual alongside the historical statuses of members (when corporate IEEE decides to provide this information) would only be a start. As discussed with corporate IEEE last year, essentially all of the data needs to be effective dated.

To provide readers a better understanding of the current state of the IEEE membership database, below are some screenshots of the Oracle Business Intelligence Enterprise Edition (OBIEE) 11g user interface. The first screenshot shows a small subset of the data available about me that is specific to historical member grade effective dates. Starting as a Student Member during graduate school, I was later elevated to Associate Member, Member, and most recently, Senior Member. The memberbership status field apparently indicates that I was in Active status during all of these time periods, with no other statuses.

Now take a look at the next screen shot, which contains similar items, albeit for another member that I anonymized by not including any personally identifiable data. In quering data for this individual, I also ordered by historical member grade effective date. As you can see, this member has a long history with IEEE, and was elevated to Life Senior membership status over 25 years ago. While the membership status change date for my entry above is the same as the start date for my first historical member grade, however, notice that membership status change date for this other individual is just a few months ago. So why does the membership status field indicate Inactive for all time periods? Because the database does not track historical member status and only records current status.

This particular data is very unuseful, unless perhaps the user only cares about current state, or is willing to keep track of changes over time in a separately maintained data set outside of the IEEE membership database. But notice another apparent anomaly. The membership status change date is one day removed from the most recent historical member grade termination date. While this data might be accurate, it seems very odd that the two dates do not line up with one another. Essentially, what this example seems to imply is that a presumably active Life Senior member went into Inactive status with one day to spare. And this anomaly does not even address the fact that this time period is the second of two as Life Senior member. Why are there two time periods? The database simply does not divulge this information.

In thinking about the models that can be built with predictive analytics in mind, consider the data associated with personal updates that LinkedIn provides to its members. For readers not famliar with this feature of LinkedIn, personal updates are essentially a text box via which members can share messages with their connections. For this discussion, the content of these messages is not technically important, but keep in mind that these personal updates are often used to communicate career related events and share blog posts such as the one you are reading right now. For illustration, I captured some screenshots of another LinkedIn feature labeled "Who's Viewed Your Updates" that is directly linked to the personal updates feature.

The above four screenshots, clockwise starting with the upper left screenshot, are intended to show who viewed three personal updates that had been shared in early March 2014. The first screen shot is a summary of the latter three screenshots, which are broken down by each individual update. Fortunately, the numbers add up. There were 65 total views by my connections of these updates (31 + 15 + 19 = 65), 4 total likes (1 + 1 + 2 = 4), and 1 comment. While this is basic math, notice that the plots that LinkedIn provides do not match these numbers. At some point, LinkedIn must have determined that there is a saturation point with the number of data points being depicted. Although this is a general observation, because this only seems to apply to the total number of views, note the number of likes and comments, which are accurate.

So what does all of this have to do with predictive analytics? The key word in this LinkedIn feature is "who" - "who" has viewed particular personal updates. Where is the "who"? If I hover my cursor above one of these data points, does a depiction of the associated LinkedIn connection materialize nearby? No. All that can be reviewed by a member here are counts. A member cannot actually take a look at who viewed their updates. With the assumption that I have the same number of LinkedIn connections in the future when another personal update is posted, is there any way for me to determine how many will view, like, or comment? Probably not. As colleagues have heard me say on a number of occasions over the years, anything is possible - but not everything is probable. Now consider a recent personal update that is separated in time from the previous three updates by a few months.

What, no data points? We already established that there is no "who" per se in the equation. And we know from the first example from a few months ago that the total number of 47 data points would not be included if these were depicted. But why are there no data points? Well, it turns out that LinkedIn has decided that unless there are likes or comments, all of the data points would simply be views and so these would not provide any value. Perhaps. But one can argue that only displaying a subset of total data points alongside total likes and comments can be misleading, giving the impression that the ratio of likes and comments to total views is actually higher than it actually is in reality.

What all of these observations have in common is that in order to predict, one needs to be able to drill down to the individual data points to understand the characteristics of each so that the likelihood of appearance of other data points in the future exhibiting similar characteristics can be ascertained. In looking at what is currently made available to LinkedIn members, one cannot determine the audience of personal updates. And while granular information about IEEE members is technically being made available for the current month during which the database is being queried, this information is not being made available on a historical basis. Going back to my plot of total member count from May 2014 above, we do know that increasing senior member count has had a big impact in recent months, and the database does keep track of the types of member additions and deletions that take place over time, but we need to understand the characteristics of members behind these numbers. Looking at total counts does not really show the big picture.

---

Pro Bono: IEEE Chicago Section Statistician - Part 1, Part 2, Part 3, Part 4