AC - Method for displaying insights from input data on mobile applications
Publication Date: 2013-Jun-24
The IP.com Prior Art Database
Disclosed is a system that enables a mobile device user to obtain in-context insights about a large data set while interacting with a table of data. The user can employ the standard mobile gestures of pinch/swipe to navigate the table, and then tap to gain insight. The system then generates insight on the table, and overlays the insight most relevant to the user.
Page 01 of 5
Viewing a large data table within a document is extremely difficult from a mobile device
with a small screen size. For example, reading a stockholder's report for a Fortune 500 company on a smart phone is nearly impossible. The best available approach is to pinch, pan, or zoom across the table and search for the interesting figures. This is quite cumbersome. Additionally, the raw data is not as interesting or valuable as the key insights that can be extracted from the data.
The invention is a system that enables a mobile device user to obtain in-context insights about a large data set while interacting with a table of data. The user can employ the standard mobile gestures of pinch/swipe to navigate the table, and then tap to gain insight. The system then generates insight on the table, and overlays the insight most relevant to the user. The user defines the meaning of "most relevant" in several ways: closest to touch point, type of insight (e.g., max, total, variance, etc.), etc. per application type.
In the preferred embodiment, upon encountering a data table, the invention parses the input data in its original formatting and builds a standard two-dimensional model of the data. The data is then searched for "interesting" characteristics. The system presents the user with a simple thumbnail of the data. The user taps on the thumbnail to see the automatically extracted insights; double-tap to return to the original large table version of the data.
The steps for implementing the invention in a preferred embodiment follow:
1. User loads a document or web page with a large tabular data set on a mobile device
2. System generates a thumbnail of that data using standard image processing
3. The system analyzes the original table data, to gather insight about the data and the most relevant data
A. Insight is pulled out of the data, and stored in a readable short statement.
B. The insight information is weighted based on most likely relevance the user. Examples of "interesting" points include:
i. a "total"/"summary" row
ii.data that is marked up (boldfaced data is interesting)
iii. a "total" row that is automatically calculated
C. Any other method of finding insights (ex: analyze totals/variance/etc.). The insights are tagged with an insight type and the location in the original table.
4. User configures the preferred types of insights; this is a saved configuration setting
A. General insights (e.g., show the insights that are geographically closest to the screen interaction location)
B. Domain insights (e.g., sports fans are interested in totals/averages, financial advisors are l...