Method for visualisation of status data in an electronic system

• US 7,788,280 B2
• Filed: 11/15/2007
• Issued: 08/31/2010
• Est. Priority Date: 11/15/2007
• Status: Expired due to Fees

• Current Assignee: LinkedIn Corporation (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Krishnamurthy, Bharat, Neogi, Anindya, Kothari, Ravi, Singh, Raghavendra
• Primary Examiner(s): Woo; Isaac M
• Application Number:
  US11/940,680
• Publication Number:
  US 20090132555A1
• Time in Patent Office: 1,020 Days
• Field of Search: 707600-831
• US Class Current: 707/791
• CPC Class Codes: G06F 11/324 Display of status information
• CPC Class Codes: G06F 16/904 Browsing; Visualisation therefor for navigating the web G06F16/954; browsing optimisation for the web G06F16/957

Abstract

A method is disclosed for facilitating visualisation of status data in an electronic system. The status data comprises metadata including a plurality of information context categories. The status data also comprises data nodes, each data node including specific instances of a respective context category. The method starts with selecting possible data combinations and permutations of the information context categories, each permutation defining a respective hierarchy. For each permutation, information content of data for each node of the respective hierarchy is estimated and an informative tree and total entropy for all informative nodes are found. The permutation and the informative tree that has the lowest total entropy are then presented to the user.

Claims

1. A computer-implemented method for facilitating visualisation of status data in an electronic system, the status data comprising metadata, the metadata including a plurality of information context categories, and data nodes, each data node including specific instances of an information context category, the method comprising:

selecting, by a computer, all possible data combinations and permutations of the information context categories, each data combination and permutation defining a respective hierarchy; and

for the each data combination and permutation performing by said computer;

estimating information content of data for each node of the respective hierarchy;

finding an informative tree corresponding to the respective hierarchy;

finding total entropy for all informative nodes in the informative tree; and

presenting a data combination and permutation corresponding to the informative tree with a lowest total entropy to the user.

2. The method of claim 1, wherein the estimating information content of data for each node of the respective hierarchy comprises:

using electronic system data history analysis to establish historical set point patterns in measurement time series; and

computing a deviation of set point patterns in a current time series from the historical set point patterns.

3. The method of claim 2, wherein the using electronic system data history analysis to establish historical set point patterns in measurement time series, includes:

observing behaviour of data over representative periods of activity/time; and

filtering spontaneous variations observed in the data.

4. The method of claim 3, wherein the filtering comprises:

using frequency domain and optimization techniques to find regularity;

using any detected regularity to form patterns; and

storing the patterns as one or more rules.

5. The method of claim 4, wherein the formed patterns are time-based.

6. The method of claim 2, wherein the computing the deviation of the set point patterns in current time series from the historical set point patterns, includes:

using stored patterns and current patterns to estimate a distance measure;

finding a node where the distance measure is the largest; and

computing running statistics to invoke incremental pattern adaptation.

7. A computer system for facilitating visualisation of status data in an electronic system, the status data comprising metadata, the metadata including a plurality of information context categories, and data nodes, each data node including specific instances of an information context category, the system comprising:

a memory for storing the status data; and

a processor configured to;

select all possible data combinations and permutations of the information context categories, each data combination and permutation defining a respective hierarchy; and

for the each data combination and permutation,

estimate information content of data for each node of the respective hierarchy;

find an informative tree corresponding to the respective hierarchy;

find total entropy for all informative nodes in the informative tree;

and

present a data combination and permutation informative corresponding to the informative tree with a lowest total entropy to the user.

8. The system of claim 7 wherein the estimate of the information content of data for each node of the respective hierarchy comprises:

using electronic system data history analysis to establish historical set point patterns in measurement time series; and

computing a deviation of set point patterns in a current time series from the historical set point patterns.

9. The system of claim 8, wherein the using electronic system data history analysis to establish historical set point patterns in measurement time series comprises:

observing behaviour of data over representative periods of activity/time; and

filtering spontaneous variations observed in the data.

10. The system of claim 9 wherein filtering comprises:

using frequency domain and optimization techniques to find regularity;

using any detected regularity to form patterns; and

storing the patterns as one or more rules.

11. The system of claim 10, wherein the formed patterns are time-based.

12. The system of claim 8, wherein the computing the deviation of the set point patterns in current time series from the historical set point patterns comprises:

using stored patterns and current patterns to estimate a distance measure;

finding a node where the distance measure is the largest; and

computing running statistics to invoke incremental pattern adaptation.

13. A computer program storage medium readable by computer, tangibly embodying a program of instructions executable by said computer for effecting a method for facilitating visualisation of status data in an electronic system, the status data comprising metadata, the metadata including a plurality of information context categories, and data nodes, each data node including specific instances of an information context category, the method comprising:

selecting all possible data combinations and permutations of the information context categories, each data combination and permutation defining a respective hierarchy; and

for the each data combination and permutation performing;

estimating information content of data for each node of the respective hierarchy;

finding an informative tree corresponding to the respective hierarchy;

finding total entropy for all informative nodes in the informative tree; and

presenting a data combination and permutation corresponding to the informative tree with a lowest total entropy to the user.