{"id":227,"date":"2020-07-11T12:27:41","date_gmt":"2020-07-11T19:27:41","guid":{"rendered":"https:\/\/blogs.ubc.ca\/vaughan\/?p=227"},"modified":"2020-07-11T12:38:25","modified_gmt":"2020-07-11T19:38:25","slug":"task-9-network-assignment","status":"publish","type":"post","link":"https:\/\/blogs.ubc.ca\/vaughan\/2020\/07\/11\/task-9-network-assignment\/","title":{"rendered":"Task 9: Network Assignment"},"content":{"rendered":"

Making Connections<\/h2>\n
\n

When I examined the data in Paladio, it was clear that there were a few winning<\/em> musical pieces selected by most of us (Jaat Kahan Ho, Johhny B. Goode, and Morning Star Devil Bird, for example) and some not-so-popular selections as well (Kinds of Flowers, Panpipes (Solomon Islands), and String Quartet No. 13).<\/p>\n

I chose songs based on the presence of human voices (humanity).\u00a0 My thinking was, if we wanted to share humanity with other life forms, perhaps we should share the diversity of our voices and the unique sounds from across the planet.\u00a0 But, what do my song selections have in common (if anything) with others?\u00a0\u00a0What was it about the three most popular choices that drew most of us to these songs?<\/p>\n<\/div>\n

If you examine the image below, you can visually identify the connection between the three most popular songs and those of us who chose them.\u00a0 In particular, if you look at the cluster of names representing those of us who selected all three of the most popular musical pieces, the analysis suggests we might have something in common-either in our musical preferences or our selection criteria.\u00a0 However, specifically what<\/em> we have in common, cannot be ascertained through this exercise alone.\u00a0 What can this image tell us about those of us who selected all three pieces versus those who selected two or only one?\u00a0 For example, what is similar between Abe Kang and I that compelled us both to select\u00a0Jaat Kahan Ho, but what differed between our criteria that compelled me<\/i> to select Johnny B. Goode and Morning Star Devil Bird but not Abe?\u00a0 This analysis alone cannot reveal our selection criteria nor reveal the story behind why we’ve made these particular choices, but it can\u00a0<\/em>reveal the presence of a connection between the two of us.<\/div>\n
\n
\"Image<\/a>
Distribution of students who selected the three most popular songs: Jaat Kahan Ho, Johhny B. Goode, and Morning Star Devil Bird,<\/figcaption><\/figure>\n<\/div>\n
\n

Why Can’t We Determine the Reasons Behind Our Choices?<\/h2>\n
The network graphs we can produce using the .json file only tell a part of the story.\u00a0 As we’ve learned throughout this course, without having a method for telling the\u00a0complete\u00a0<\/em>story, we are left with a series of connections (or pieces of a story) without context.\u00a0 I think about the assumptions I made when creating my emoji story: that others understood the structure of a reality-TV-type elimination show (that each week after being presented with a new challenge, one couple is inevitably sent home); without having the necessary familiarity with (or schema for) reality shows, readers of my emoji story would be unable to guess the name of the piece I’d chosen .\u00a0 Like my emoji story, the graphs we produced in Paladio, for me, were missing key context and information that would have helped me make better connections between the data so I could ‘read’ the entire story behind our collective selections.<\/div>\n

Connecting the Unconnected<\/h2>\n
I spent a bit more time exploring some of the least popular selections.\u00a0 If you examine the two images below, you’ll note that only two people selected two of the least popular pieces:\u00a0 Kinds of Flowers was selected by Sukhjeevan and I; Panpipes (Solomon Islands), was selected by Sukhjeevan and Kevin.\u00a0 Again, what was it about these pieces that prevented others from selecting them?\u00a0 Why did the three of us choose them?\u00a0 Thinking a bit deeper, Sukhjeevan’s name was associated with the two least popular choices shared also by Kevin and I.\u00a0 Is it possible then, that Sukhjeevan, Kevin and I all share a commonality in our selection criteria or musical preference(s)?\u00a0 The network graphs do not provide any indication of why\u00a0<\/em>the three of us selected these not-so-popular tunes (nor why they weren’t popular in the first place) so we can’t know if there is\u00a0<\/em>a connection between us\/our selections (or not).<\/div>\n
\n
\"Image<\/a>
Only two students selected Kinds of Flowers<\/figcaption><\/figure>\n<\/div>\n
\n
\"Image<\/a>
Only two students selected Panpipes (Solomon Islands)<\/figcaption><\/figure>\n

Implications of Visualizations<\/h2>\n
\n

I read an interesting article this week about Big Data and the commercial use of our personal data (Bauer et al.’s 2017 article,\u00a0Ethical perspectives on recommending digital technology for patients with mental illness<\/em>).\u00a0 The article discusses the potential dangers of Google, Facebook, and others having access to our personal data because of the potential misrepresentation of our information and the scary implications associated with selling our misinterpreted data to third parties (such as insurance companies) (Bauer et al., 2017).<\/p>\n

Task 9 put into practice what Bauer et al.’s (2017) article suggested in theory:\u00a0 the danger of making connections and assumptions using incomplete or misinterpreted data.\u00a0 For example, Bauer et al. discuss third party acquisition of searches performed by Google users and what those searches may (incorrectly) imply.\u00a0 For example, if a Google user searches for the word depression\u00a0<\/em>repeatedly, one might assume they are struggling with mental health issues or that they’ve been recently diagnosed with depression.\u00a0 Without context, we do not know the whole\/complete story though: we can only guess why the Google user is performing a particular search, but we can’t know for sure.\u00a0 Someone might research\u00a0depression\u00a0<\/em>for any number of reasons!\u00a0 Could a family member or friend be suffering from depression?\u00a0 Could the user be working on a school project?\u00a0 Without providing the entire search context<\/em>, insurance companies who are privy to such private searches may make erroneous assumptions based on clients’ search history on Google (which could have massive implications on clients’ insurance costs and coverage).<\/p>\n

Thus, when grouping data and creating connections based on an incomplete picture lacking context, the resulting associations and story\u00a0<\/em>will be prone to error and depending on the potential use of information, could be quite damaging.<\/p>\n<\/div>\n

Next Steps<\/h2>\n
<\/div>\n
\n

I wanted to know more about data visualization and I was interested to see if I could better understand the connections between the data we analyzed this week.\u00a0 Through Paladio we were able to see the song selections everyone made, who else\u00a0selected those songs and which songs were the most and least popular.\u00a0 However, I was interested in determining whether I could tease out any connections between students based on our song choice.\u00a0 I am just beginning to learn about data analysis and visualization so I asked a friend to take our .json file and help me upload it to Gephi and see what connections we could make.\u00a0 From there we were able to ‘play’ with the data a bit and create a different network graph.<\/p>\n

The graph below represents the connections between students based on our song selections.\u00a0 The three distinct colours (orange, blue and green) indicate three separate groupings of students based on our song selections.\u00a0 In the center you’ll see a series of pinkish lines indicating where the orange and blue groups intersect (note that the green group is the outlier and only barely connects with the blue group).\u00a0 This suggests that members in these three groups have more commonalities than they do differences (their song selections are more similar to one another than they are different).<\/p>\n<\/div>\n

\n

This visualization also indicates that there is something common between the orange and blue groups and that Alexandra, Emma and Janice appear to be at the “center”.\u00a0 Since I can move the nodes around the page Alexandra, Emma and Janice are not the geographic<\/i> center, rather, the size<\/i> of their nodes suggests their song selections best represent or connect with the entire class’ overall song choices versus Kevin, for example who is the last node on the right in the green ‘group’ who least represents the class’ overall song selections.<\/p>\n<\/div>\n

\n

The thickness of the lines (edges) indicate the number of selected songs shared by two people.\u00a0 For example, Alexandra and Janice’s song selections must be quite similar and Tyler and\u00a0\u0160\u00e1rka must also have selected similar songs because the edges connecting them are thicker than the edges connecting other classmates to one another.\u00a0 Of note as well, the song selections belonging to Melody and I seem to cross over between the blue and orange groups so I imagine we must share some common songs (and song criteria) between both groups.<\/p>\n<\/div>\n

<\/div>\n
\n

To answer my earlier question about whether\u00a0Sukhjeevan, Kevin and I have anything in common, based on the graph below,\u00a0Sukhjeevan and I have one shared song in common and we are in two different groups (Sukhjeevan in orange, me in blue), however we are both<\/i>\u00a0connected to Alexandra so perhaps there is<\/i>\u00a0something about our song choices that, though not necessarily similar to one another, perhaps compliment each other?\u00a0 Kevin does not appear to share any grouping nor commonality with either\u00a0Sukhjeevan nor I (other than Panpipes).<\/p>\n

Thus, in taking the data one step further, I was able to visualize our class’ groupings a bit better and understand how we connect (or not!) through examining edge thickness, group colour and node size as well.<\/p>\n<\/div>\n

\n
\"An<\/a>
Three distinct groups emerged when we used Gephi to compare students’ song selections<\/figcaption><\/figure>\n

Reference<\/p>\n

Bauer, M., Glenn, T., Monteith, S., Bauer, R., Whybrow, P. C., & Geddes, J. (2017). Ethical perspectives on recommending digital technology for patients with mental illness.\u00a0International Journal of Bipolar Disorders, 5(1), 6. Retrieved from: https:\/\/doi-org.ezproxy.library.ubc.ca\/10.1186\/s40345-017-0073-9<\/a><\/div>\n<\/div>\n<\/div>\n<\/div>\n","protected":false},"excerpt":{"rendered":"

Making Connections When I examined the data in Paladio, it was clear that there were a few winning musical pieces selected by most of us (Jaat Kahan Ho, Johhny B. Goode, and Morning Star Devil Bird, for example) and some not-so-popular selections as well (Kinds of Flowers, Panpipes (Solomon Islands), and String Quartet No. 13). … <\/p>\n

Continue reading “Task 9: Network Assignment”<\/span><\/a><\/p>\n","protected":false},"author":60405,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[83,78,81,77,80,82,79,76],"class_list":["post-227","post","type-post","status-publish","format-standard","hentry","category-tasks","tag-edges","tag-gephi","tag-graph","tag-network-assignment","tag-network-graph","tag-nodes","tag-paladio","tag-task-9"],"_links":{"self":[{"href":"https:\/\/blogs.ubc.ca\/vaughan\/wp-json\/wp\/v2\/posts\/227","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/blogs.ubc.ca\/vaughan\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blogs.ubc.ca\/vaughan\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blogs.ubc.ca\/vaughan\/wp-json\/wp\/v2\/users\/60405"}],"replies":[{"embeddable":true,"href":"https:\/\/blogs.ubc.ca\/vaughan\/wp-json\/wp\/v2\/comments?post=227"}],"version-history":[{"count":20,"href":"https:\/\/blogs.ubc.ca\/vaughan\/wp-json\/wp\/v2\/posts\/227\/revisions"}],"predecessor-version":[{"id":251,"href":"https:\/\/blogs.ubc.ca\/vaughan\/wp-json\/wp\/v2\/posts\/227\/revisions\/251"}],"wp:attachment":[{"href":"https:\/\/blogs.ubc.ca\/vaughan\/wp-json\/wp\/v2\/media?parent=227"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blogs.ubc.ca\/vaughan\/wp-json\/wp\/v2\/categories?post=227"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blogs.ubc.ca\/vaughan\/wp-json\/wp\/v2\/tags?post=227"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}