Task 9: Network Assignment

Task completed using Voyager’s Golden Record Curation Quiz Data

The Process

This task was challenging for me. Having no prior knowledge in statistical data and network visualization, I wasn’t sure how to use or read the results given in Palladio. The Systems Innovations videos (2015) were helpful in giving me a bare bones introductory level understanding of nodes, edges, the network connectivity and visualization of data. Stages in my learning process included a basic understanding of:

• Nodes, edges, connectivity, undirected graphs, etc.
• Palladio
• downloading .json data
• network visualization
• making new attempts at regenerating graphs in Palladio and changing the results by concentrating on weighting nodes, changing facets, etc.
• reflecting on my new basic understanding of how Palladio works and the information it can provide

What I was able to understand was that the assignment’s data is demonstrated as vertices (nodes) and edges (relationship of 2 nodes) resulting in an undirected graph. The more connected a node, the more connectivity it has on the graph. Through trial and error with Palladio, I began to change some of the parametres of the data (target, source, etc.). Although I don’t understand all the functions, I do understand the significance of data to stakeholders in obtaining information that can be used for gaining information for a set purpose, or for decision-making if that’s the end goal.

The Data

The first geodesic graph that appeared was difficult for me to grasp. It appeared “jumbled” and “static” (screenshot #1).

I searched for my name to see if I could make any sense from it — I couldn’t. Once I played with the facets, I determined that I was in modularity class network #1 with Deborah and Chris. I cannot clearly understand why we were grouped since I don’t see more commonalities in our song choices over that of any other participants.

From the data, I think I’ve determined some basic structures. Each participant is a “source” and each song is a “target”. By choosing just the target the data shares which/how many participants chose specific songs. “Size nodes” provides a dot visualization that indicates how often a track was selected based on the size of the dot (screenshot #4). Data indicates that Track 20 had the highest selections, while Track 19 had the fewest selections with only 1 participant (Lubna Yasin) in Group 3. In #1 screen shot, it’s clear that Track 19 is only connected to Lubna.

In terms of the null choices (missing or unknown), I would think that without further data this cannot be interpreted with the current data. It would require additional information from and about the participants. There’s no sufficient information on why I chose certain songs over others.

Reflection

I would suggest that there’s not enough data to determine why any given participant chose a specific curation of songs. This would require additional criteria such as the use of a questionnaire to compile data on reasons for song choices. Since all participants for Task 8 were given an established list of songs, the data results are limited in that the data cannot make assumptions about participants’ behaviour or interests. The information collected is not driven by our internet search habits or social media “likes”, making it difficult to make any inferences, which may result in misinterpretations. Code.org (2017) video points out that, “Our interests, behaviours, activities change the relationship within the network constantly.” This is not evident in the information collected in this task.

Implications

The exercise of taking the Golden Record soundtrack provides an example, albeit benign, of how data can be collected. In terms of political implications, one might contemplate the purpose of the data collection. If this assignment were a real-life situation, we might consider who developed the question, and why. How might the collected data be used in a social or global context?  In a real-life situation, the results could affect policy changes. Could the results affect the participants, the artists/songs, or the countries represented in some fashion? How might the data be manipulated? Based on the reason for collection, were the right questions being asked? Is the data being read and understood properly by those who are using it? Is there an ulterior motive for the collection?

Further Reflection

My husband and I run DJ’ing and karaoke services. We’ve been hosting weddings, family reunions, school graduations, company events, and fun dances for two decades in our community. For DJ’ing, we do our research and try to put together playlists that are representative of the attending demographic. This includes an initial interview with the organizer to discuss specific songs, styles, and genres to be included. Palladio could be a helpful tool for compiling a geodesic graph. Through observation, for example, by inputting specific nodes into the system (i.e. demographic of event crowd, song titles, artists, which songs had the most/least crowd approval), Palladio could target songs to specific demographic. The data could assist in creating improved playlists. Over time songs could be removed or added to develop a new graph so that the playlists remains current.

References

Code.org. (June 13, 2017). The Internet: How search works. Links to an external site. [Video]. YouTube.

Maslin, M. (November 22, 2019). The beautifully macabre cartoons of Gahan Wilson. The New Yorker. Retrieved from https://www.newyorker.com/culture/postscript/the-beautifully-macabre-cartoons-of-gahan-wilson

Music from Earth. (n.d.). NASA. Retrieved from https://voyager.jpl.nasa.gov/golden-record/whats-on-the-record/music/

Palladio. (n.d.). Retrieved from https://hdlab.stanford.edu/palladio-app/#/upload

Systems Innovation. (April 18, 2015). Graph theory overview. Links to an external site. [Video]. YouTube.

Systems Innovation. (April 19, 2015). Network connections. Links to an external site. [Video]. YouTube.