1

ITI220 [Section 09] [Fall 2015] [10/14/15] – [Nikita, Abraham], [ICP3–Annotated Bibliography]

Introduction My client, Shibu Daniel, currently works as Vice President of User Experience (UX) in

Bank of New York Mellon (BNY Mellon). As an investment company with global reach, BNY

Mellon has a significant amount of data about client activities. Understanding, organizing, and

using data to make educated decisions is essential to the company’s success. As a UX designer

and a member of the “Client Technology Solutions” sector of BNY Mellon, my client’s

information need is most closely related to Big Data. For preliminary understanding, he wanted

me to research the following topics within the scope of Big Data: what is Big Data and why it is

important, how can companies use visualizations to organize data, how have companies done so

(real-world examples or case studies), and what frameworks or software are available for data

processing in this way. The information gathered will be useful in helping my client understand

how to represent large data sets more effectively, how to make use of BNY Mellon’s current data

analytics software, and how to organize data when designing technology solutions. In addition,

insights can be gained on consumer behavior, business models that work well with data analytics

implementation, and in risk mitigation.

For relevant but industry-specific articles, I used the Rutgers University Libraries

Articles+, a resource offered by the University that searches multiple databases at once, and

specifically, the ScienceDirect and Business Source Premier databases. To find articles

specifically in the field of business and IT, rather than health and sciences, I used the

“Publication” and “Database” section to limit the articles to the aforementioned disciplines.

Considering the ever changing nature of the IT industry, I used the “Publication Date” tab to

only bring up articles or resources within the last ten years. In doing so, all of my articles seemed

2

to still be relevant to the industry. I used Google Scholar briefly, but many of the articles only

included the publication information, not the actual text. In addition, several of the resources

were outdated. Google’s search engine yielded several extremely helpful resources, even if they

were not scholarly, using keywords with quotations such as “big data” with other keywords such

as: business AND management. Many of the articles I found online were from reputable online

magazines within the IT field. With a combination of the resources mentioned above, I was

successfully able to find articles and sources that pertained to my client’s information need.

Annotated Bibliography

Childs, H. (2013). VisIT: An end-user tool for visualizing and analyzing very large data. eScholarship University of Alabama. Retrieved from http://escholarship.org/uc/item/69r5m58v#page-7.

This particular article discusses the tool, VisIT that was originally developed in response to Big Data processing. It is an open source tool that focuses on three main functions: enabling data understanding, providing scalable support for large sets of data, and creating a robust product that is simple for the end user. In addition, the paper outlines the software’s design, architecture, and user interface concepts. In addition to the software’s initial successes, the author also outlines companies who have made use of VisIT. The article was found on eScholarship, an open-access repository of scholarly sources under California Digital Library. Even though it is not in a traditional database, the material itself is scholarly and the original study was published in a scholarly journal. Childs’ and his colleague’s work was also supported by the U.S. Department of Energy; the document was prepared as an account of the work sponsored by the United States Government. In addition to the aforementioned points, the comprehensive list of references provided at the end emphasizes the author’s credibility. While the source does not cover a wide variety of topics as some of the other articles do, this particular case study fulfills my client’s need for real world examples of companies who have implemented visualization software, or have made data processing software. Depending on the long-term goals for my client’s team, perhaps the insights gained from this case study could provide recommendations for BNY Mellon’s data processing tools. Choudhury, S. (2014). The future of information dashboards. UX Magazine. Retrieved from

https://uxmag.com/articles/the-future-of-information-dashboards The entire article focuses on information dashboards as a means of visualizing large data sets. There are five primary categories that she divides the article into: (1) dashboards offered by

3

independent software vendors as part of their data analytics tools, (2) the dashboard for the user on the go (mobile smartphones and tablets), (3) dashboards with high location intelligence quotients, (4) dashboards for real time data and (5) dashboards future predictive data. Shilpa Choudhury has published a number of articles in other well-known blogs such as Wired, ReadWrite, and Visual.ly. In addition, UX Magazine is a well-known online magazine that focuses on experience design. Practitioners and industry leaders who are well versed in the UX field write the majority of their articles. At one point Choudhury mentioned nine analytic predictions made by IIA (The International Institute for Analytics) for 2014. While the information has not been updated, many of the elements from the previous predictions were repeated or relevant for the following year. This article is relevant for my client because it provides examples of different kinds of dashboards (a specific visualization), along with real-world examples of companies that have implemented them. By analyzing this article, my client will be able to assess what his user needs are, what his team is trying to accomplish, how to execute it, and additional resources. Franks, B. (2014). Making big data actionable: How data visualization and other tools change the

game [Webinar]. Harvard Business Review. Retrieved from https://hbr.org/2014/05/making-big-data-actionable-how-data-visualization-and-other-tools-change-the-game

In this webinar, hosted by Harvard Business Review (HBR), Franks begins by defining big data and discussing how visualizations were used in the past to aid cognition. He transitions to modern day, outlining how data visualizations have changed according to the needs of the modern user as well as advances in technology. Today, customized visualizations are tied to a specific analysis – in that way, modern visualization tools must be interactive, interconnected, collaborative, flexible and enabling. Franks’ greater point is essentially that businesses need to use data to support decision making, rather than just reporting what occurred – this calls for a change in methodology. Perhaps it is the nature of a webinar session, but there were no direct links or references provided at the end of the presentation. Nonetheless, Franks does an excellent job maintaining currency by discussing contemporary software vendors, companies who have implemented visualizations, and other relevant anecdotes. He also works for Teradata, a company that provides services in data warehousing, big data and analytics, and marketing applications. His experience within the field of big data can clearly be seen. In addition, the magazine that hosted the webinar, HBR, is a reputable organization under Harvard University that provides a bridge between academia and various enterprises. As such, the articles are reliable but also relevant to the industry. I would recommend this webinar primarily because it emphasizes a change in the current business model (the need to make decisions, not just analyze them). Franks provides practical advice as an industry professional that would be relevant to my client’s need for integrating visualizations into big data processing.

Hoffer, D. (2014). What does Big Data look like? Visualization is key for humans. Wired

Magazine. Retrieved from http://www.wired.com/insights/2014/01/big-data-look-like-visualization-key-humans/

4

The basis of Hoffer’s entire article rests on the idea that big data needs to be more human. In developing his points, he establishes that information visualization is a means of wayfinding, not simply a means of organizing data. In doing so, he also reiterates the need for data visualizations to be comprehensive (displaying multiple levels of information according to user needs), to be scalable, and to be simple. Hoffer provides several examples of companies who are successfully humanizing big data and how they have implemented the considerations mentioned above. At the end of the article, his call to action seems to be that since big data itself is constantly evolving, the technology we use to organize that data must be robust enough to process it – just like every other piece of software the IT industry uses. The article was written in 2014 and is still very current considering the real world examples that Hoffer mentions (such as Google Maps). That being said, each of the sources that he mentioned had a link to the original article or a related source that provided further information. When the article was first written, the author was the head of User Experience at Declara, but is currently a Design Director at McKinsey & Company. His experience is well translated in the article itself. This would be a very relevant source for my client considering that Hoffer has a background in User Experience and the article seems to focus on data processing within that particular subfield of IT. The article as a whole is a good baseline for the technology necessary to implement data processing through information visualizations. Intel. (2013). Big data visualization: Turning big data into big insights [Web document]. Intel.

Retrieved from http://www.intel.com/content/dam/www/public/us/en/documents/white-papers/big-data-visualization-turning-big-data-into-big-insights.pdf

The paper starts off with an overview of the current IT landscape as well as the problems that are leading industry professionals to data discovery tools such as information visualization. Essentially this new form of business intelligence involves integrating data from a variety of user-based sources that are then displayed in an interactive and easily understood format. The author(s) describe(s) how challenges with big data have to do with its volume, variety and velocity (the three V's of big data) as well as the increasing availability of mobile devices. The paper then outlines key features of visualization-based data discovery, how it addresses the aforementioned challenges, how to protect data quality, and how to establish data governance policies. While no author is provided, the source document is from Intel – a well-known company that has always stood at the forefront of technological innovation. In addition to providing a comprehensive list of endnotes, the article mentions the Gartner BI (business intelligence) Summit often. Gartner is a leading information technology research and advisory company; the Summit meeting is an important event within the industry. The inclusion of such sources reiterates the document’s reliability and credibility. This is relevant to my client because it provides a list of necessary conditions from a business perspective, including security. Considering that my client works in the banking industry, security is an important part of data processing and risk mitigation.

5

Ishwarappa, & Anuradha, J. (2015). A brief introduction on big data: 5Vs characteristics and Hadoop technology. Procedia Computer Science, 48. Retrieved from ScienceDirect database, through Rutgers University Libraries Articles+.

This paper discusses the 5V’s of big data (volume, velocity, value, veracity, and variety) in detail. Moving forward, the paper also looks at challenges that companies face when handling big data, namely, challenges involves capturing, analyzing, storing, searching, sharing, visualization, and transferring data. The article primarily features Hadoop, an open source distributed data processing tool that is being used by many industry professionals. This article was found on the ScienceDirect database, making it a reliable and accurate source. Considering that Apache’s Hadoop is still a popular solution to companies for their information needs, the article is also timely and current. This would benefit my client by going into detail about Hadoop and its data processing frameworks. His own team can implement similar frameworks whether they use in-house software or a third party source. Keahey, A.T. (2013). Using visualization to understand big data. IBM Software Business

Analytics. Retrieved from https://tdwi.org/~/media/E3362B4A0E184F75AB29403 676C4C3CD.pdf.

This article discusses how businesses can make use of visualizations when making sense of the following five categories of user needs: (1) simple customer data (2) customer data involving time (3) customer sentiment, which is essentially how people feel about the company (4) measuring customer relationships and (5) customers at different levels. In each of these five categories, the author gives one or two examples of the most effective kind of visualization for that particular need. For example, when measuring customers at different levels, a hierarchal visualization conveys the information better than other forms. In addition to the simple data visualizations that the author advises his reader to use, Keahey also promotes IBM’s data analytical services. As far as accuracy and credibility, this particular article is one of several resources that IBM shows to potential clients who are interested in IBM’s data processing services. As such, I can see that the information is well researched and up-to-date to fit the current IT environment; it even has a list of references at the end of the article. In addition, the author is one of IBM’s Visualization Science and Systems experts, making him a reliable and knowledgeable source. I can see this being useful for my client in addressing his information need for what kinds of visualizations can be used to display data. Not only does Keahey showcase several visualization techniques, he also clearly addresses which situations each of them would be appropriate for. LaValle, S., Lesser, E., Shockley, R., Hopkins, M.S., & Krushwitz, N. (2011). Big Data,

analytics and the path from insights to value. MIT Sloan Management Review, 52(2). Retrieved from Business Source Premier database, through Rutgers University Libraries Articles+.

The purpose of this article is to answer the question, how are organizations using analytics to gain insight and make decisions from big data? Essentially, there were three main findings from

6

the report: that top organizations are more likely to apply analytics to activities, that managerial and cultural aspects are the biggest challenges to implementing analytics, and that being able to visualize data in different ways is increasing in overall value for the company. The article goes into detail about a new methodology for adopting analytics within an organization, along with several recommendations from a business perspective. Although the article was not found on a database, MIT Sloan Management Review is a reliable and dependable resource, known for their well-researched articles and relevant topics. In order to better understand the challenges that companies have faced and have overcome for analytics, MIT Sloan Management Review along with IBM Institute for Business Value conducted a survey with more than 3000 industry professionals (business managers, analysts, and executives). In addition both organizations interviewed a number of academic experts and subject matter experts from various related disciplines. In doing so, the article provides a holistic but specific viewpoint. All of the authors have a significant amount of experience in the Business Analytics realm of IT. It would be relevant to my client’s information need for business models that work with analytics implementation. The recommendations would provide practical business advice in how to overcome common problems with using visualizations in big data processing. Minelli, M., Dhiraj, A., & Chambers, M. (2013). Big data, big analytics: Emerging business

intelligence and analytic trends for today's businesses [ebook]. Wiley CIO Series. Retrieved from eBook Collection database, through Rutgers University Libraries Articles+.

This particular eBook provides a holistic approach to big data. The first couple of chapters consider what big data is and why it is relevant, along with some industry examples of it in action. Considering my client’s needs, the third chapter on “Big Data Technology” would be the most beneficial. This particular chapter goes into detail about Hadoop, a well-known data analytics company (including their business model, critical components, and overall goals). The chapter also considers the importance of data discovery when compared to more traditional data processing models. The book is written for business and IT professionals with the purpose of helping such individuals integrate big data analytics within their own organization. Combined, the authors have years of experience related to business analytics solutions, technology-based solutions, and related decision sciences. In addition to their expertise, the notes or references at the end of each chapter, and its addition in the EBSCOHost database all reiterate the eBook’s credibility. The article will be relevant in addressing the client’s need for different ways to visualize data, as well has the need to outline business processes (necessary parties, technologies, and management models) involved when implementing a data visualizing software. Not only are the topics expansive, they are explained in detail. Visualizations make big data meaningful. (2014). Communications of the ACM, 57(6), 19-21.

Retrieved from Business Source Premier database, through Rutgers University Libraries Articles+.

The author of this article provides a unique perspective by describing data visualization as an expressive medium, where ultimately the “artist” or data analyst can choose what pieces of

7

information to show or withhold from the viewer. Moving forward, the article as a whole discusses advances in technology that set the stage for more effective data visualization practices such as the availability of storage, the efficiency of cloud computing, and the rise of software tools such as Apache Hadoop – all of which, simplify data processing. Other important aspects that the article mentions include the importance of “humanizing” these visualizations, and looking for new and more dynamic ways to display data outside of simple charts and graphs.

While an author is not provided, the article was found on the Business Source Premier database and published in the academic journal Communications of the ACM, making it highly reputable. The article itself was written in 2014 and none of the technologies or techniques mentioned are outdated – which is to be expected considering the recency of big data in general. Even though it does not address the business aspect of big data processing, the article will be relevant in describing how data visualizations have changed to fit the current IT environment and what advances are now possible. The details mentioned in this article are not expansive in its coverage, but it will fulfill the client’s information need for general information about data visualizations, what can be accomplished with current technologies, as well as the future of big data.