Update week10.md

week10.md

@@ -0,0 +1,6 @@
+## Source
+https://arxiv.org/pdf/2202.10520.pdf
+
+## Reflection
+
+The paper I am reviewing this week is "Making Data Tangible: A Cross-disciplinary Design Space for Data Physicalization" by S. Sandra Bae, Clement Zheng and more. The reason I am reviewing this paper is because of my general interest in the topic of data physicalization and other forms of unique data visualization beyond paper. While this paper did not create or test any data physicalizations itself it reviewed and "Coded" popular papers in the field. This coding process aimed to categorize each of the physicalizations by context, structure and interactions. Personally the two fields I am most interested in are context and interactions to see if utilizing the two in specific ways can help data visualizations more effectivly effect its audience. We see that data physicalizations created for a specific audience may have more influence in them, but most of the works were created for a general audience. Another finding was the material choices were often dependent on the type of data as some materials could speak more while other projects were more focused on shape/ interactability than material. Lastly sensorary interactions were the most significant in affecting the users understanability of the data. Overall maybe a little disapointed in paper as it was not very conclusive in many cases, but still find the field rather interesting. 

week11.md

@@ -0,0 +1,6 @@
+## Source
+https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6728899/
+
+## Reflection
+
+I wanted to start looking at differnt ways that are currently available to visualise dna alignment/ omics data which led me to this paper. This paper showcases many different tools and resources in R, Python and Java that could be used for ploting omics data. Main things I was looking for was ways to utilize omics data in chord diagrams. Because of the size of the data for genomics research many of the platforms are not web based. I would be interested in trying to find a way to make some of these plots with d3 as their components do not seem overely complex, but further research needs to be done. Either way this was a good insight into some of the available technologies regarding circular omics data portrayal.

week12.md

@@ -0,0 +1,4 @@
+## Source
+https://onlinelibrary.wiley.com/doi/full/10.1111/cgf.13610
+## Resource
+The article begins by trying to define what a multilayer network is to standardize some of the different terminoligies used in different fields. Utilimitly the definition is fluid depending on the field. The researchers tested different multilayer network visualizations and grouped them on categories like, Cross-layer entity connectivity, Cross-layer entity comparison, Layer manipulation, reconfiguration, Layer comparison based on numerical attributes, Layer comparison based on topological, connectivity. One dimensional views looked at chord diagrams and parallel node-link bands like those made in graph trail for a starting point. 2d, 2.5D and 3D graphs utilize colors, depth, or differnt link designs to plot mutliple layers next to each other, but seperated by their layers. Can be helpful with small visualizations, but become overwhelming when displayed in large. Another interesting plot was the matrix based plots, using hybrid matrix node link plots you can reap the benfits of quick readability from node links as well as pros from the matrix visualizations. They are still doing research into defining a best practices, but recomend user testing for any plots, they think field and uses have grown greatly in past decade and will become more accessable. 

week13.md

@@ -0,0 +1,5 @@
+## Source
+https://ieeexplore.ieee.org/abstract/document/8017587
+## Reflection
+
+While I initially started reading this paper to get a better of idea of how to use small multiple for genomics data sets I found the first half of the paper rather intersting as it was about the computer scientists working with the biologists to get an idea of what types of visualizations they wanted. In a few sessions of just talking to the scientists and them showing the developers what they were working with they had an idea of the type of application they wanted to create. One tool that they mentioned they used in the creation of their application was Node Trix, a hybrid representation of networks based on the node- link diagram where communities can be represented as matrices. Another cool tool that they mentioned getting inspiration from was MultiPiles which is a visualization to explore time-series of dense, weighted networks and is based on the physical analogy of piling adjacency matrices, each one representing a single temporal snapshot. Resulting product is a scatterplot of small multiples that use NodeTrix to link between each other and MultiPiles to summarise and group data at each point. The program seems pretty thorough in the sense that it has many options for filtering and different views for looking at data individually or grouped and seems to have made a significant impact in how these scientists do work.

week14.md

@@ -0,0 +1,4 @@
+## Source
+https://www.sciencedirect.com/science/article/pii/S2468502X21000048
+## Reflection
+On creating our initial network visualization for our final project I noticed the browser was struggling a little to render and handle the large network visualization in d3. After doing a little research I found this paper on NetV.js which is a library that was built for high-efficiency visualizations of large-scale graphs and networks. The paper begins by addressing the gap in the field for an easy to use large-scale network visalizer, as others have been made but are typically more reserved for lower level languages lik c# or C++ because of their performance speed. To achieve the rendering speeds and performance they are looking for the researchers utilized the users GPU to render the networks. NetV.js had a few other goals with regards to usability, they built and support a large amount of interactivity as well as lasso selection of nodes. Ultimitly this JS library looks pretty interesting and easy enough to use that I may utilize it for my final project.

week15.md

@@ -0,0 +1,4 @@
+## Source
+https://ieeexplore.ieee.org/document/6876017
+## Reflection
+Wanted to look at UpSet as an alternative to vendiagrams for visualizing how data is connected between sets. Design reminds me of some visualizations I have seen for time series data. Using a visualization like this may be helpful for showing set interactions between the different proteins and their relations to each group. Initial attempts to decode the relations are a little difficult, but with some training it is effective. It also takes up a lot less space than a vendiagram and shows it in a more succinct way.