SIMS 247: Information Visualization

Group Project: Bibliographic Information Visualization and Analysis (BIVA)

Team: Chitra Madhwacharyula, Colleen Whitney, and Lulu Guo
[Project Goals][Related Work][Data] Visual Mappings [Rationale][Evaluation][Future Work][Appendices]

Visual Mappings and Interaction

The BIVA Project uses the following visualization techniques to visualize and analyze bibliographic item circulation patterns.

Bar graphs

Bar graphs are used to show the subject wise circulation patterns of items across the years or within the months of a specific year. The length of the bar graph indicates the number of times the items in the subject group were circulated. A sample bar graph generated by our system is as shown in Figure 1.


Figure 1: Visualization of Circulation patterns by subject and time

Figure 1 shows the item circulation patters for the subject category “Languages and Literature across the years 1999 to 2005. The Y-axis of the graph denotes the number of charges or circulations while the X-axis denotes the time period (years or months within an year). The bar graphs provide an easy way to compare the circulation patterns between various subject areas as well as across time within a specific subject area.

Scatter Plots

Our system supports the process of drilling down into the selected information to retrieve greater details or add additional data of interest. For example a user can select a particular time period and subject of interest from the bar graphs and click to access a scatter plot that adds a new dimension: holdings.


Figure 2: Comparative View of Item-wise holding and circulation data

Figure 2 shows a scatter plot that compares the circulation and holdings data in the subject area “Languages and Literature” for the year 2000. This plot is created when a user clicks on the year 2000 bar in the graph shown in Figure 1. The scatter plot would give a comparative view of the holdings (how many libraries hold the item, on a log scale) and circulation (how many times has the item been circulated) in the selected time period for all the items in a specific subject category.

The X-axis of the scatter plot denotes the “number of holdings” of the item and the Y-axis indicates the number of circulations of the item. So this visualization provides a very simple and intuitive way to see, at a glance, the characteristics of the relationship between holdings and circulations for specific items and helps in answering interesting questions like:

  • Are there any items that have circulated very little at UCLA but are widely held?
  • Which books are the most popular within a subject area?
  • Are there any outliers?

Item Detail

The BIVA system also offers an item detail view that allows the user to select a specific item in the scatter plot and view details its details like item title and specific circulation transactions. Figure 3 shows an example of the information displayed in the item detail view when a specific item is selected from the scatter plot shown in Figure 2.


Figure 3: Item Detail View

Filters

The BIVA System provides two types of filters that allow the users to selectively view the bibliographic information. Both the filters are applicable for the highest view of the data i.e. the bar graphs that show the subject-wise item circulation patterns over time.

Subject Filter

Allows the user to select the subject categories that they want to view. Subject Filter is provided through check boxes that the users select or deselect based on the subjects they want to view.

Time Filter

Allows the user to select the time period for which they want to view the circulation patterns within a subject. We support two kinds of time-based views:

  • View of circulation patters across the years (from 1999-2005)
  • View of circulation patters within a year (by month)

Arrangement of Visual Components

The BIVA interface consists of 3 different panels that are used to display the three levels of information that our system supports. Figure 4 is a snapshot of how the different visual components are arranged in our system.


Figure 4: Arrangement of Visual Components

The left panel is used to display the item circulation patterns by subject and time. One bar graph is created for each subject and the default view shows the circulation patters across the years from 1999 to 2005. The graphs are placed adjacent to each other making it easy to compare them. The Subject and Time filters are also present in this panel and can be used by the user to change the default view.

The right top panel is used to display the next level of information, i.e. the scatter plot view that compares the holdings and circulation data. This plot is dynamically generated whenever the user selects a specific region of a bar graph in the left panel.

The right bottom panel displays item specific information depending on which individual item is selected by the user from the scatter plot. We have tried to eliminate the need for scrolling as much as possible by making optimal use of the screen space. We have also experimented with the layout and sizes of the different graphs to ensure that the information display and flow is simple and intuitive to the end user.