Data Visualization

Final Revision

Link to Revision

Link to R Script

Link to Presentation

Revisions to the report are towards the end and in bold. Images of “Age of Acquisitions” have also been added.

There are conflicting opinions on what exactly is a successful company acquisition. Tomi Laamanen’s analysis of 111 small tech-based companies suggests that when a company acquires another, it is usually for a need for complementary resources. Barney and Turk, on the other hand, suggest that it is in when you combine unrelated resources into the operation of the company, it creates value in society. As for factors that identify acquirers, Blonigen and Taylor’s research shows substantial negative correlation between companies that ranked high in research and development with a high change to acquire. Bueller and Drori state that acquisitions can enhance the gradual development of a company’s capabilities and strategy. As for how acquisitions impact a company, Thatcher and Oliver’s research shows that increased productivity depends on the fixed size of the company and the size of their market.

Our visualization shows the progression of company acquisitions over time. This tells a story about the history of each company, the types of companies they are interested in, and paints an overall picture of where the tech industry is going. The visualization is relevant and unique because we haven’t seen a visualization that tells this story yet and we were curious to see what the data on the excel sheet would show.

The dataset is from Enigma.io and shows over 8,500 company acquisitions, dates, and the names of acquired companies from the Crunchable database. It is significant to us for a couple reasons. First, it gives us an idea of where the industries are going. Second, it gives us insight into the strategies of the individual companies. We had many issues we had to work through to get the data to work and tell a story. To make things more interesting, we went through and manually searched for the categories for the acquired companies because this would be the most interesting thing to visualize.

We used R to analyze the data and it quickly answered some of the questions we were asking ourselves. We found out which companies had the most acquisitions over time, including the months and years that the most acquisitions happened. More specifically, 2011 was the year with the most acquisitions (1487), with May being the month with the highest acquisition traffic (807). This helped us iterate through our ideas about the stories we could tell through the dataset.

We created a web visualization because we wanted people to interact with the different companies and get information like the name of the acquired company on mouseover. We brainstormed through some 4D implementations of the data and thought it would be interesting to have a tangible element in the future when we have more time to extend on the concept.

Finding the right form to tell this story was challenging and it required iterating through many concepts that didn’t work out. As our sketches show, we we started very broad and with time, trial and error, we narrowed it down. We initially wanted to show the relationship between companies and time through a circle with that had networked lines that connected based on what you clicked on.

The visualization came together when we begin limiting the information we used to tell the story. As much as we wanted to show the whole picture, we did not want to overwhelm the user. We have the ability to choose where to begin and end the story and made the decision to narrow the scope by only including companies with 10 or more acquisitions. We used D3 to implement this visualization, specifically a library called Gantt Chart. Once the user understands what colors correspond with what categories, they can play around with the different companies and get a sense for where they are headed.

The only related project we found online was a visualization called “The Age of Acquisitions” on TechCrunch. This visualization focused on the amount companies were sold for and used size to represent the amount. The visualization was also implemented through D3 and the categories they used made it easy for us to narrow down our own categories.

Screen-Shot-2015-12-16-at-6.42.26-PM Screen-Shot-2015-12-16-at-6.42.00-PM Screen-Shot-2015-12-16-at-6.42.09-PM

 

This visualization is for anyone interested in the field of technology, its past and where it’s going. It is also useful for startups looking at different acquisition options, and could be a good complement to the TechCrunch visualization so they can get a picture of what the companies have paid for other acquisitions.

In future iterations, we would like to incorporate information from the Crunchbase API, once granted access. A multi dimensional graph that gives users the ability to switch on and off what information to display would revel more information and allow the user to play with the data. It would also be interesting to see this information on a map to see the locations of the companies that get acquired.

In working with startups, it is useful to know the type of companies being acquired by bigger companies so they would get an idea of which companies would be a good fit. In my testing of the visualization after the presentation, I got feedback that it would be nice to also have additional information regarding sectors, i.e. biological hardware, health SaaS, etc. It will be tricky to find direct, accurate, and concise data on this but it will be worth the curation.

Additional feedback was very useful. It would be interesting to know what was happening in the social media world during the time of the acquisition and pull that information in through an API like Twitter. Additional suggestions such as stock market prices would provided an interesting intersection of finance and data visualization.

Finally, an update on the Crunchbase API, we would need to apply to a Commercial license or join their Venture Program. The issue though is that even with the API calls, D3 will not be able to visualize data unless it is on one input file. There will be curation either way.

Works Cited

B., Barney J., and T. A. Turk. “Superior Performance from Implementing Merger and Acquisition Strategies: A Resource-based Analysis.” The Management of Corporate Acquisitions: International Perspectives (1994): n. pag. Print.

Blonigen, Bruce A., and Christopher T. Taylor. “R&D Intensity and Acquisitions in High-Technology Industries: Evidence from the US Electronic and Electrical Equipment Industries.”The Journal of Industrial Economics 48.1 (2000): 47-70. Web.

Brueller, Nir N., Abraham Carmeli, and Israel Drori. “How Do Different Types of Mergers and Acquisitions Facilitate Strategic Agility?” California Management Review 56.3 (2014): 39-57. Web.

“Gantt-Chart.” Gantt-Chart. N.p., n.d. Web. 09 Dec. 2015. <http://dk8996.github.io/Gantt-Chart/>.

Laamanen, Tomi. “Option Nature of Company Acquisitions Motivated by Competence Acquisition.” Small Business Economics 12.2 (1999): 149-68. JSTOR. Web. 09 Dec. 2015.

Thatcher, Matt E., and Jim R. Oliver. “The Impact of Technology Investments on a Firm’s Production Efficiency, Product Quality, and Productivity.” Journal of Management Information Systems 18.2 (2001): 17-45. Print.

“Visualizing 15 Years Of Acquisitions By Apple, Google, Yahoo, Amazon, And Facebook.”TechCrunch. N.p., n.d. Web. 09 Dec. 2015. <http://techcrunch.com/2014/02/25/the-age-of-acquisitions/>.

Final-Acquisition Viz

There are conflicting opinions on what exactly is a successful company acquisition. Tomi Laamanen’s analysis of 111 small tech-based companies suggests that when a company acquires another, it is usually for a need for complementary resources. Barney and Turk, on the other hand, suggest that it is in when you combine unrelated resources into the operation of the company, it creates value in society. As for factors that identify acquirers, Blonigen and Taylor’s research shows substantial negative correlation between companies that ranked high in research and development with a high change to acquire. Bueller and Drori state that acquisitions can enhance the gradual development of a company’s capabilities and strategy. As for how acquisitions impact a company, Thatcher and Oliver’s research shows that increased productivity depends on the fixed size of the company and the size of their market.

Our visualization shows the progression of company acquisitions over time. This tells a story about the history of each company, the types of companies they are interested in, and paints an overall picture of where the tech industry is going. The visualization is relevant and unique because we haven’t seen a visualization that tells this story yet and we were curious to see what the data on the excel sheet would show.

The dataset is from Enigma.io and shows over 8,500 company acquisitions, dates, and the names of acquired companies from the Crunchable database. It is significant to us for a couple reasons. First, it gives us an idea of where the industries are going. Second, it gives us insight into the strategies of the individual companies. We had many issues we had to work through to get the data to work and tell a story. To make things more interesting, we went through and manually searched for the categories for the acquired companies because this would be the most interesting thing to visualize.

We used R to analyze the data and it quickly answered some of the questions we were asking ourselves. We found out which companies had the most acquisitions over time, including the months and years that the most acquisitions happened. More specifically, 2011 was the year with the most acquisitions (1487), with May being the month with the highest acquisition traffic (807). This helped us iterate through our ideas about the stories we could tell through the dataset.

We created a web visualization because we wanted people to interact with the different companies and get information like the name of the acquired company on mouseover. We brainstormed through some 4D implementations of the data and thought it would be interesting to have a tangible element in the future when we have more time to extend on the concept.

Finding the right form to tell this story was challenging and it required iterating through many concepts that didn’t work out. As our sketches show, we we started very broad and with time, trial and error, we narrowed it down. We initially wanted to show the relationship between companies and time through a circle with that had networked lines that connected based on what you clicked on.

The visualization came together when we begin limiting the information we used to tell the story. As much as we wanted to show the whole picture, we did not want to overwhelm the user. We have the ability to choose where to begin and end the story and made the decision to narrow the scope by only including companies with 10 or more acquisitions. We used D3 to implement this visualization, specifically a library called Gantt Chart. Once the user understands what colors correspond with what categories, they can play around with the different companies and get a sense for where they are headed.

The only related project we found online was a visualization called “The Age of Acquisitions” on TechCrunch. This visualization focused on the amount companies were sold for and used size to represent the amount. The visualization was also implemented through D3 and the categories they used made it easy for us to narrow down our own categories.

This visualization is for anyone interested in the field of technology, its past and where it’s going. It is also useful for startups looking at different acquisition options, and could be a good complement to the TechCrunch visualization so they can get a picture of what the companies have paid for other acquisitions.

In future iterations, we would like to incorporate information from the Crunchbase API, once granted access. A multi dimensional graph that gives users the ability to switch on and off what information to display would revel more information and allow the user to play with the data. It would also be interesting to see this information on a map to see the locations of the companies that get acquired.

Works Cited

B., Barney J., and T. A. Turk. “Superior Performance from Implementing Merger and Acquisition Strategies: A Resource-based Analysis.” The Management of Corporate Acquisitions: International Perspectives (1994): n. pag. Print.

Blonigen, Bruce A., and Christopher T. Taylor. “R&D Intensity and Acquisitions in High-Technology Industries: Evidence from the US Electronic and Electrical Equipment Industries.” The Journal of Industrial Economics 48.1 (2000): 47-70. Web.

Brueller, Nir N., Abraham Carmeli, and Israel Drori. “How Do Different Types of Mergers and Acquisitions Facilitate Strategic Agility?” California Management Review 56.3 (2014): 39-57. Web.

“Gantt-Chart.” Gantt-Chart. N.p., n.d. Web. 09 Dec. 2015. <http://dk8996.github.io/Gantt-Chart/>.

Laamanen, Tomi. “Option Nature of Company Acquisitions Motivated by Competence Acquisition.” Small Business Economics 12.2 (1999): 149-68. JSTOR. Web. 09 Dec. 2015.

Thatcher, Matt E., and Jim R. Oliver. “The Impact of Technology Investments on a Firm’s Production Efficiency, Product Quality, and Productivity.” Journal of Management Information Systems 18.2 (2001): 17-45. Print.

“Visualizing 15 Years Of Acquisitions By Apple, Google, Yahoo, Amazon, And Facebook.” TechCrunch. N.p., n.d. Web. 09 Dec. 2015. <http://techcrunch.com/2014/02/25/the-age-of-acquisitions/>.

 

 

Link to the project

Play with R

Kevin gave us a brief review of R, a professional statistics data analyse tool. I have known R for a long time from a statistical Phd friends, but I never though of touching it. After last week’s class, my impression of this tool is it is a powerful data analyse tool. The advantage of R is packages are wrapped up for us to use with learning correlated algorithms. It is pretty handy for people like us who don’t have statistics background but need to analyse data. The disadvantage is the visualization it does is not interactive. If we want to make interesting graph, R may be not a good option.

My statistical friend sent me a tutorial and I got my hand on it for a little bit. The important thing I learnt is the way R processing data is by matrix. In other words, it process data into matrix instead of objects or arrays like JavaScript or D3. After transferring original data into matrix, all analysis will be conducted on the matrix. Overall, R is more math oriented than programming oriented. I used basic analysing function to parse the NYC graduation data. The analysing process is fast and easy to read. Maybe we can use R to clear the data, then use D3 to do visualization.

Screen Shot 2015-11-19 at 4.52.44 PM

Geographic Visualization

In this week, I want to try geographic visualization since geographic and cartographic data visualizations are popular. I found a dataset of 4-year college graduation rate in U.S. I decided to play with them. What I wanted to do was to correlate information with the map. When doing mouseover on different states, the corresponding info will show up. The color of each state is based on its graduation rate.

I tried to use two different libraries and I chose uStates.js at the end because it is simpler. However, another problem is the library cannot process objects. So I modified it a little bit. But I haven’t figure out how to show the graduation rate when mouseover.

Timeline of Women in Aerospace and Manufacturing

Or you can see the timeline here!!!

This week’s activity was to create our first timeline project with timeline.js library. Patrick, Helen, Neill and I was a group. We decided to do something related to feminism and engineering. I found a timeline picture of women in aerospace and manufacturing field, then, we decided to use this topic.

timeline-2014

After wikipedia those amazing women, I got familiar to this topic. The history of women in aerospace field is much shorter than I thought. The first milestone happened in 1906. During the following half century, this topic was still not bloomed nor popular. Only a few of women participated in this field and most of them were pilot. The situation was so bad that the first woman in commercial air force was forced to step down from the cockpit by the all-male pilots union. The good thing was their families were supportive. After they got the pilot’s license, many of their parents bought them airplanes. Over more than a century, there are less than 30 milestones of this topic. We do need more women involve in aerospace industry.

Data Visualization Midterm Project

“Graphic designers support comprehension by simplifying visualizations to the extent possible. Simplification strategies include discouraging three-dimensional displays, removing extraneous gridlines and decimals, and avoiding color gradiation, of all which Tufte refers to as “visual noise” and “non-data ink”, terms he coined to refer to anything that does not directly aid the understanding of the data in the display.” (Azzan & Evergreen, 2013)

Radiators in New York City housing buildings have a problem of overheating some apartments and leaving other apartments cold during the winter time. This creates an imbalance of temperature that leaves tenants uncomfortable. Our visualization is for building owners retrofitted radiator installations in their apartment buildings. This tool communicates temperature imbalances in the entire building, on each apartment floor, and in individual units in a way that makes achieving tenant comfort easy for the building owner.

Research shows that a healthy temperature not only reduces costs, but increases health and reduces hospital visits (Mccormack, Mckeever & Syme, 2008). The American Heritage Dictionary of the English Language defines room temperature between 68-72°F, and West Midlands Public Health Observator defines comfortable wintertime temperature for a living room as 70°F. (Watkins, 2011) We used these as reference points but allowed for slight variability, keeping in mind that tenants and building owners will have control over their temperature preferences upon the tool’s deployment.

The dataset contains 8,354 data points that include a full month of temperature records. We used this dataset as an experiment to represent the data in a meaningful way for the building owner. In its current state, the data provides an instant glimpse into apartments with temperature issues. It adds an interactive way of getting the building temperature average, the average for each floor, each unit, and a performance recap of the month. In its future iterations, it will tap into live data to give building owners an up to date, actionable recap. The visualization answers (1) are temperatures in tenants’ apartments comfortable? (2) are there actionable items to increase tenant comfort? and (3) how is the longer term performance of the retrofitted radiators? The visualization is currently online for the purposes of this class.

We brainstormed the function of this visualization and the more we knew what we didn’t need the easier it was to rule out options of visual representation. We initially thought we could use leaflet to show where the apartments are located, but then realized that the target user will only have 1-5 apartment buildings and will be satisfied with a simple identifier instead of a map. We also thought through creating a custom SVG solution that had a floor plan type of visualization, but each apartment’s layout is inconsistent and may cause more difficulties in the long run.

CBA9799E-E86E-442D-9D2A-9FE92876062F-1-2048x1536-orientedB93E0CAD-33EB-4E92-9826-C8A6D8F0C67D-1-2048x1536-orientedIMG_6386
IMG_6793FullSizeRenderYelp_sunburst

The user journey of the initial prototypes is as follows: (1) a building owner visits the online dashboard that indicates whether something needs their immediate attention, (2) they navigate to their building(s) to see a detailed breakdown of the floors and units, (3) they jump around between features like performance, settings and contacts (tenants and utility services) in future iterations. Though we feel this visualization works for now, these mockups are what we want in the long run. The learning curve required to design an interactive SVG framework from scratch did not fit our timeframe.

A project we used as an inspiration is Sage.is, a site that is rethinking the way data on nutritional labels is designed. This site takes information and displays it in a simple and aesthetic way. In this train of thought, we knew that a lot could be communicated through simply color with our dataset. We explored different prototypes for our concept, and we came across the DashBoard visualization. We decided to use the bars as floors, using height to communicate temperature. We used the pie chart to show units that may need attention. We then parsed and pushed the data into relevant arrays to pass through our functions. Jumping into the DashBoard visualization pushed us to familiarize ourselves with D3’s functionality.

Our project’s intention is to simplify the way building data is presented by going beyond the traditional line and bar charts to deliver a simple message. We did not want to display any extraneous data that would be irrelevant to the user. We used D3 and javascript and published to the web.

The most important next step is to figure out if D3 is the right tool to keep moving with this project. If it is, we need to dive deeper into D3. The process had us asking questions as to what would be the best tool to create the most efficient user experience. Do we want to keep coding and designing on D3 or create a native mobile app? What is more effective in the long run? Do programming languages like Swift support D3? What kind of data visualization apps are available for mobile development?

Though the product has a target audience set, Sandra defined future tenants and building buyers as other target audiences. If a building has a history of comfortable temperature, the future tenant does not have to worry about being too hot or too cold when they live there because the track record can be shown to them. Future building owners can also see that the building will provide reliable and comfortable heating. We started thinking about whether or not these audiences would change the content of the visualization.

Assam and Evergreen (2013) discuss dashboard creation as a process that involves designing the interface (our sketches), building the framework (the current code), populating the dashboard (sample data), then repeating the process through evaluating and refining. With this project, we have created a base to repeat the process for future iterations.

Works Cited

The American Heritage Dictionary of the English Language. Boston: Houghton Mifflin, 1992. Print.

Assam, Tarek, and Stephanie D. H. Evergreen. Data Visualization, Part 2. N.p.: n.p., n.d. Print.

Lloyd, E. L., C. Mccormack, M. Mckeever, and M. Syme. “The Effect of Improving the Thermal Quality of Cold Housing on Blood Pressure and General Health: A Research Note.” Journal of Epidemiology & Community Health 62.9 (2008): 793-97. Web.

NPashaP. “DashBoard.” DashBoard. N.p., n.d. Web. 15 Oct. 2015. <http://bl.ocks.org/NPashaP/96447623ef4d342ee09b>.

“Smarter Food Labeling Is Here.” Sage. N.p., n.d. Web. 15 Oct. 2015. <http://sage.is/>.

Watkins, David E. Heating Services in Buildings: Design, Installation, Commissioning & Maintenance. Chichester, West Sussex, UK: Wiley-Blackwell, 2011. Print.

“WMPHO – Home Page.” WMPHO – Home Page. N.p., n.d. Web. 15 Oct. 2015. <http://www.wmpho.org.uk/>.

 

Link to the project

Link to data: current temp, monthly data

Slides