DATA DRIVEN COMPONENT REPUTATION

• US 20090281819A1
• Filed: 05/12/2008
• Published: 11/12/2009
• Est. Priority Date: 05/12/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Garg, Neeraj
• Granted Patent: US 9,070,086 B2
• Time in Patent Office: Days
• US Class Current: 705/309
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06Q 50/182 Alternative dispute resolution

Abstract

Various technologies and techniques are disclosed for observing the real-world performance of components and using the collected data to improve other user experiences. Configuration data, usage data, and user experience data are collected for components on observed solution instances. The configuration data, usage data, and user experience data are analyzed to generate component diagnostic data that can be used to make recommendations to one or more advised solution instances as to how to correct a problem being experienced on the one or more advised solution instances. In one implementation, the component diagnostic data generated includes a reputation rating for each of the components.

Claims

1. A method for generating component diagnostic data for making recommendations to improve component performance on advised solution instances based on real world usage data of the components on observed solution instances comprising the steps of:

receiving configuration data that was collected for components on a plurality of observed solution instances;

receiving usage data that was collected for components on the plurality of observed solution instances;

receiving user experience data that was collected for components on the plurality of observed solution instances; and

generating component diagnostic data that can be used to make recommendations to one or more advised solution instances as to how to correct a problem being experienced on the one or more advised solution instances, the component diagnostic data being generated upon analyzing the configuration data, usage data, and user experience data.

2. The method of claim 1, wherein the component diagnostic data that is generated from the generating step includes a reputation rating for each of the components.

3. The method of claim 1, wherein the component diagnostic data that is generated from the generating step includes compatibility information for the components with respect to other components.

4. The method of claim 1, wherein the component diagnostic data that is generated from the generating step includes hardware requirements of the components based upon user experience data.

5. The method of claim 1, wherein the configuration data that is collected includes details regarding software and hardware installed on observed solution instances.

6. The method of claim 1, wherein the usage data that is collected includes details on what purposes the observed solution instances are used for most frequently.

7. The method of claim 1, wherein the user experience data is collected automatically based upon component performance details.

8. The method of claim 1, wherein a number of times a particular component crashed is included as part of the user experience data that is collected.

9. The method of claim 1, wherein at least a portion of the user experience data is collected as opinion data that is received from one or more users of the observed solution instances.

10. The method of claim 1, wherein the component diagnostic data is made available to third parties through a searchable component diagnostic database.

11. The method of claim 1, wherein the configuration data, usage data, and user experience data is collected by an observation agent installed on each of the observed solution instances and then transmitted to a separate solution instance for the analyzing step.

12. A method for making specific recommendations to a user for how to solve a problem occurring on an advised solution instance based upon data collected from observed solution instances comprising the steps of:

receiving a request from an advised solution instance for a recommendation on solving a problem being experienced by a user of the advised solution instance; and

when there are one or more specific components identified on the advised solution instance that are related with a bad customer experience from component diagnostic data gathered from observed solution instances, then providing one or more suggestions to the user for how to solve the problem with component changes.

13. The method of claim 12, further comprising the steps of:

when there are one or more specific usage patterns identified on the advised solution instance that are related with the bad customer experience from the component diagnostic data gathered from the observed solution instances, then providing one or more suggestions to the user for how to solve the problem with usage pattern changes.

14. The method of claim 13, wherein the one or more suggestions for solving the problem with usage pattern changes includes a suggested maximum number of component windows to have open at a concurrent point in time.

15. The method of claim 12, wherein the one or more suggestions for solving the problem with component changes includes one or more hardware replacement suggestions.

16. The method of claim 12, wherein the one or more suggestions for solving the problem with component changes includes one or more software replacement suggestions.

17. The method of claim 12, wherein the component diagnostic data is generated from configuration data, usage data, and user experience data collected from the observed solution instances.

18. A computer-readable medium having computer-executable instructions for causing a computer to perform steps comprising:

identifying a sub-set of solution instances of a plurality of observed solution instances that have a similar usage and configuration;

comparing solution instances in the sub-set that are associated with a bad customer experience with solution instances in the sub-set that are associated with a good customer experience to identify which components are different; and

creating a reputation rating for each of the components as a result of analysis performed during the comparing step.

19. The computer-readable medium of claim 18, further having computer-executable instructions for causing a computer to perform steps comprising:

repeating the identifying, comparing, and creating steps for one or more sub-sets of solution instances of the plurality of observed solution instances that have a similar usage and configuration.

20. The computer-readable medium of claim 18, wherein the reputation rating provides an indication of how frequently a respective component crashes based upon data previously collected from the plurality of observed solution instances.