Functionalities in PeopleMap
Map View
The central visualization of PeopleMap, the Map View, hosts a set of different functionalities that allow you to explore the researcher datasets and better understand the diversity of topics among its individuals. Here is an overview of its different functionalities:
Show Distributions: In order to help you better understand how each cluster of researchers is formed, this toggle allows you to see the Gaussian distributions calculated by the Gaussian mixture model algorithm, allowing you to understand what the model found to be the appropriate split for the researcher embedding. Each distribution is colored differently according to the dots within it. Additionally, each distribution visualizes the space covered by three standard deviations of the distribution along each of its axes.
Clusters: To assist you in the exploration of the researcher embedding clusters, this slider allows you to manipulate the number of Gaussian distributions generated by the Gaussian mixture model algorithm. The slider itself does not change the embeddings of the researchers. But by increasing the number of clusters, the Gaussian distributions generated becomes increasingly tight and coherent but they also can create arbitrary splits for the researcher embeddings. Likewise, by decreasing the number of clusters, the Gaussian distributions become more expansive but also decrease in tightness and coherence of the clusters. Therefore, it is wise to balance both when finding an ideal cluster count.
Show All Names: To help you find specific researchers and recognize individuals in different clusters, this toggle displays the names of researchers alongside their respective dot within the Map View. It can be used to find researchers without hovering over each dot individually, but it can lead to cluttered names if there are a great amount of dots in the visualization.
Keywords Emphasis: This drop-down allows you to adjust the emphasis that is placed on a researcher’s Google Keywords, compared to their titles and abstracts, when generating their TFIDF embedding. By increasing the emphasis, more multiples of a researcher’s keyword are concatenated into their original combined document that is used to generate their TFIDF embedding. By decreasing the value, less multiples of a researcher’s keywords are concatenated into their original combined document; it can be set to zero. The purpose of this drop-down is to increase or decrease the weight placed on a researcher’s self-identified topics of study when calculating their position in the visualization, allowing you to better understand the characteristics of each researcher’s fields of interest.
Publication Set: This drop-down allows you to select which publications you would like to use for the Map View: the top fifty most cited publications of each researcher or the top fifty most recent publications of each researcher. The first choice will use a researcher’s most recognized publications to characterize their research, while the second choice will use a researcher’s latest work in their characterization. This allows you to explore the researcher dataset from two different angles.
Researcher View
In addition to the Map View functionalities, the Researcher View component (which is the panel to the right of the Map View visualization) displays the following information about a researcher when their corresponding dot is hovered over:
Name
Affiliation
Google Scholar Profile Keywords
Keywords
Citation Count
Google Scholar Profile Link
Google Scholar Profile Photo
Research Query
In addition to the tools above, the Research Query component, which is found in the top right of the PeopleMap platform, allows you to see what researchers are aligned with each of the Google Scholar keywords collected from the entire dataset of researchers as well as locate individual researchers in the Map View.
When a specified Google Scholar keyword is inputted from the total list:
The Map View will update and display the top five researchers most aligned with the topic in various shades of purple.
The shades of purple decrease in darkness linearly as the ranking declines from the most aligned researcher to the fifth most aligned researcher with the specified research topic.
The rest of the dots are simply assigned the lightest shade of purple to indicate that they are not part of the ranking.
When a specified researcher is inputted from the total dataset:
The researcher's dot will be pinned in the Map View.
Their dot will also be enlarged and outlined to clearly indicate their position.
Their profile information will be loaded into the Researcher View.
Last updated
