Predicting Call Quality

• US 20150142702A1
• Filed: 03/06/2014
• Published: 05/21/2015
• Est. Priority Date: 11/15/2013
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Nilsson, Mattias, Saabas, Ando, Vafin, Renat, Vaalgamaa, Markus, Dumitras, Adriana, Tamme, Teele, Veski, Andre
• Granted Patent: US 9,558,451 B2
• Time in Patent Office: Days
• US Class Current: 706/11
• CPC Class Codes: G06F 18/23 Clustering techniques
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/048 Fuzzy inferencing
• CPC Class Codes: H04L 41/00 Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
• CPC Class Codes: H04L 41/5025 by proactively reacting to service quality change, e.g. by reconfiguration after service quality degradation or upgrade
• CPC Class Codes: H04L 41/5067 Customer-centric QoS measurements
• CPC Class Codes: H04L 41/509 wherein the managed service relates to media content delivery, e.g. audio, video or TV
• CPC Class Codes: H04L 65/00 Network arrangements, protocols or services for supporting real-time applications in data packet communication real-time or near real-time messaging, e.g. instant messaging [IM] H04L51/04; selective video distribution H04N21/00
• CPC Class Codes: H04L 65/1069 Session establishment or de-establishment
• CPC Class Codes: H04L 65/61 for supporting one-way streaming services, e.g. Internet radio
• CPC Class Codes: H04L 65/70 Media network packetisation
• CPC Class Codes: H04L 65/80 Responding to QoS
• CPC Class Codes: H04M 3/2227 Quality of service monitoring

Abstract

The disclosure relates to a communication client application for running on a user terminal to conduct calls over a network. The client is configured to access a model which models quality of user experience for calls based on a set of technical parameters of each call. The model itself is based on user feedback indicating subjective quality of multiple past calls as experienced by multiple users, modeled with respect to the technical parameters collected from each of the past calls. The model generates a predicted call quality score predicting the quality of user experience for the call given its technical parameters. Based on this process, one or more of the technical parameters of the call can be adapted to try to increase the quality experienced by the user.

Claims

1. A communication client application embodied on a computer-readable medium and configured so as when executed on a user terminal to perform operations comprising:

conducting at least a first call over a network; and

accessing a model which models quality of user experience for calls based on a set of technical parameters of each call, the model being based on user feedback indicating subjective quality of multiple past calls as experienced by multiple users, modeled with respect to the technical parameters collected from each of the past calls;

wherein the model generates a predicted call quality score predicting a quality of user experience for the first call, based on the model applied to the technical parameters of the first call; and

the communication client application is configured to continue the first call with one or more of the technical parameters of the first call being adapted based on the generation of the predicted call quality score for the first call.

2. The communication client application of claim 1, configured so as when executed on the user terminal to perform further operations of:

receiving user feedback for the first call from a user of the user terminal, and submitting the user feedback for the first call to the model where the user feedback and technical parameters of the first call are incorporated to adapt the model; and

conducting a second call over the network;

wherein the model generates a predicted call quality score predicting the quality of user experience for the second call, based on the adapted model applied to the technical parameters of the second call; and

the communication client application is configured to continue the second call with one or more of the technical parameters of the second call being adapted based on the generation of the predicted call quality score for the second call.

3. The communication client of claim 1, wherein the model is implemented on a server and the communication client is configured to access the model over the network.

4. The communication client of claim 1, wherein the model is implemented on the user terminal.

5. The communication client of claim 1, wherein the set of technical parameters of each call comprises any of:

one or more parameters relating to an environment from which the call is being captured,

one or more parameters relating to capture of call media of the call,

one or more parameters relating to media content of the call,

one or more parameters relating to encoding and/or decoding of the call,

one or more parameters relating to quality of service for the call over the network,

one or more parameters relating to play-out of call media of the call, and/or

one or more parameters relating to processing resources of a transmitting and/or receiving terminal of the call.

6. The communication client of claim 1, wherein at least some of said calls are video calls, and at least one of the set of technical parameters of each video call relates to video of the call.

7. The communication client of claim 6, wherein at least one of the technical parameters of each video call relates to audio of the call.

8. The communication client of claim 1, wherein the model models said quality of user experience in further dependence on one or more user-related parameters being characteristic of a user of said user terminal.

9. The communication client of claim 8, wherein the one or more user-related parameters comprise any of:

a country or geographic region selected by the user,

a country or geographic region the user terminal is detected to be located in,

a language selected by the user,

a language of a user interface of the client application,

a language of a user interface of the user terminal,

age of the user,

a brand of the client application,

a brand of the user terminal,

an indication of advertisements in the client,

an indication of advertisements on the user terminal, length of call,

length of a contact list of the user,

a length of time the user has participated in a communication system over which the calls are made,

a regularity with which the user has participated in the communication system,

an amount of prior feedback given by the user, and/or

a relationship between the user and one or more other participants of the respective call.

10. The communication client of claim 1, wherein the model provides additionally functionality modeling call quality based on one or more a-priori parameters of the network and/or user terminal, and the client is configured to access the additional functionality of the model in order to begin the first call with one or more of the technical parameters of the first call being initially set based on said additional functionality.

11. The communication client of claim 1, wherein the user feedback upon which the model is based comprises explicit user feedback scores rating the subjective quality of the multiple past calls as experienced by the multiple users.

12. The communication client of claim 11, wherein each of the user feedback scores is a score out of five.

13. The communication client of claim 1, wherein the user feedback upon which the model is based comprises implicit feedback derived from at least one metric correlated with the quality of user experience.

14. The communication client of claim 13, wherein the metric correlated with user experience is call length.

15. The communication client of claim 1, wherein the model is based on clustering the user feedback scores into multi-dimensional cells, each cell being associated with a respective range of each of the set of technical parameters.

16. The communication client of claim 1, wherein the model is based on the technical parameters and one or more model parameters of the model itself, the model being based on a training of the model parameters by a training algorithm.

17. A user terminal comprising and being arranged to execute the communication client application of claim 1.

18. A communication system comprising the user terminal of claim 17 and the model.

19. A server comprising:

a model configured to model quality of user experience for calls based on a set of technical parameters of each call, the model being based on user feedback indicating subjective quality of multiple past calls as experienced by multiple users, modeled with respect to the technical parameters collected from each of the past calls;

a network interface arranged to receive further user feedback indicating the subjective quality of a plurality of further calls from a plurality of users, and to receive the technical parameters of the further calls;

wherein the model is configured to dynamically incorporate the user feedback and the technical parameters of the further calls and thereby dynamically adapt the model as the further user feedback is received.

20. A method of modeling quality of user experience for calls conducted over a network, wherein each call is associated with a set of technical parameters;

the method comprising;

obtaining a multiple user feedback scores, each rating subjective quality of a respective one of multiple calls experienced by multiple users;

modeling the user feedback scores in relation to the technical parameters of each of the respective calls, thereby generating a model for predicting quality of user experience;

obtaining the technical parameters of one or more first subsequent calls;

applying the model to the technical parameters of the first subsequent calls to determine a predicted call quality score predicting the quality of user experience for the each of the first subsequent calls;

adapting the first subsequent calls based on the predicted call quality scores of the first subsequent calls;

receiving, over the network, user feedback scores for the first subsequent calls from the respective client applications;

adapting said model by incorporating the user feedback scores and technical parameters of the first subsequent calls, thereby producing an adapted model;

obtaining the technical parameters of one or more second subsequent calls;

applying the adapted model to the technical parameters of the second subsequent calls to determine a predicted call quality score predicting the quality of user experience for each of the second subsequent calls; and

adapting the second subsequent calls based on the call quality scores of the second subsequent calls.