Situation-aware recommendation using limited cluster sizes

• US 8,019,782 B2
• Filed: 04/08/2008
• Issued: 09/13/2011
• Est. Priority Date: 10/22/2007
• Status: Active Grant

• Current Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Original Assignee: Samsung Electronics Co. Ltd. (Samsung Group)
• Inventors: Cheng, Doreen, Song, Yu, Kalasapur, Swaroop, Jeong, Sangoh
• Primary Examiner(s): Fleurantin; Jean B
• Application Number:
  US12/099,716
• Publication Number:
  US 20090105987A1
• Time in Patent Office: 1,253 Days
• Field of Search: 707600-831, 706 45- 47, 706/52, 706/53, 706/62
• US Class Current: 707/791
• CPC Class Codes: G06F 18/24137 Distances to cluster centroïds
• CPC Class Codes: H04L 67/306 User profiles
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34
• CPC Class Codes: H04M 1/72457 according to geographic location
• CPC Class Codes: H04M 1/72472 wherein the items are sorted according to specific criteria, e.g. frequency of use
• CPC Class Codes: H04W 4/02 Services making use of location information
• CPC Class Codes: H04W 4/029 Location-based management or tracking services

Abstract

In one embodiment, data relating to usage patterns of the user is stored, wherein the data includes information as to items which were used and the context in which they were used. The data is then clustered into clusters of data points, wherein the number of data points per cluster is limited based on a preset value. Then a centroid is determined for each of the clusters. Clusters similar to the current context of the user are then selected by comparing a data point representing the current context of the user to one or more of the centroids. Then, for each of the one or more items, a probability that the user wishes to use the corresponding item is computed, based on the selected similar clusters, wherein the probabilities are used to recommend one or more of the items.

Claims

1. A computer-implemented method for making recommendations to a user, the method comprising:

storing data relating to usage patterns of the user, wherein the data includes information as to items which were used and the context in which they were used;

clustering the data into clusters of data points, wherein the number of data points per cluster is artificially limited to a preset maximum number of data points per cluster;

determining a centroid for each of the clusters;

selecting clusters similar to a current context of the user by comparing a data point representing the current context of the user to one or more of the centroids; and

computing, for each of one or more items, a probability that the user wishes to use the corresponding item, based on the selected similar clusters, wherein the probabilities are used for recommending one or more of the items.

2. The computer-implemented method of claim 1, wherein the items comprise software applications, computing services, or data.

3. The computer-implemented method of claim 1, wherein the recommending includes recommending the x items having the highest probability that the user wishes to use, wherein x is dynamically determined by using the average number of items that have been used in the selected clusters.

4. The computer-implemented method of claim 1, wherein the recommending includes recommending the x items having the highest probability that the user wishes to use, wherein x is a preset value.

5. The computer-implemented method of claim 1, wherein the computing and recommending are performed only for items not currently being used by the user.

6. The computer-implemented method of claim 1, wherein the clustering includes limiting the number of data points per cluster based on a preset average number of data points per cluster.

7. The computer-implemented method of claim 1, wherein the clustering includes limiting the number of data points per cluster such that the number of data points per cluster fall with a preset range.

8. The computer-implemented method of claim 1, wherein the selecting includes determining a Euclidean distance between the data point representing the current context and each of the one or more of the centroids and similar clusters are the ones with the least Euclidean distance from their centroids to the data point representing the current context.

9. The computer-implemented method of claim 1, wherein the context includes the user'"'"'s location.

10. The computer-implemented method of claim 1, wherein the user is operating a mobile device.

11. The computer-implemented method of claim 1, wherein the storing, clustering, and determining are performed in an offline mode.

12. An apparatus comprising:

an interface; and

one or more processors configured to;

store data relating to usage patterns of the user, wherein the data includes information as to items which were used and the context in which they were used;

cluster the data into clusters of data points, wherein the number of data points per cluster is artificially limited to a preset maximum number of data points per cluster;

determine a centroid for each of the clusters;

select clusters similar to a current context of the user by comparing a data point representing the current context of the user to one or more of the centroids; and

compute, for each of one or more items, a probability that the user wishes to use the corresponding item, based on the selected similar clusters, wherein the probabilities are used to recommend one or more of the items.

13. The apparatus of claim 12, wherein the apparatus is a mobile device.

14. The apparatus of claim 13, further including:

a load detection module operable to communicate with a network;

wherein the processor is further configured to perform the storing, clustering, and determining when the load detection module indicates the device is idle to perform the selecting, computing, and recommending when a request for recommendation is received.

15. A system for making recommendations to a user, the system comprising:

means for storing data relating to usage patterns of the user, wherein the data includes information as to items which were used and the context in which they were used;

means for clustering the data into clusters of data points, wherein the number of data points per cluster is artificially limited to a preset maximum number of data points per cluster;

means for determining a centroid for each of the clusters;

means for selecting clusters similar to a current context of the user by comparing a data point representing the current context of the user to one or more of the centroids;

means for computing, for each of one or more items, a probability that the user wishes to use the corresponding item, based on the selected similar clusters, wherein the probabilities are used to recommend one or more of the items; and

a processor coupled to the means for storing, means for clustering, means for determining, means for selecting, and means for computing.

16. A non-transitory program storage device readable by a machine tangibly embodying a program of instructions executable by the machine to perform a method for making recommendations to a user, the method comprising:

storing data relating to usage patterns of the user, wherein the data includes information as to items which were used and the context in which they were used;

clustering the data into clusters of data points, wherein the number of data points per cluster is artificially limited to a preset maximum number of data points per cluster;

determining a centroid for each of the clusters;

selecting clusters similar to a current context of the user by comparing a data point representing the current context of the user to one or more of the centroids; and

computing, for each of one or more items, a probability that the user wishes to use the corresponding item, based on the selected similar clusters.