SYSTEM AND METHOD FOR DISCOVERING AND EXPLORING CONCEPTS

• US 20150032452A1
• Filed: 07/26/2013
• Published: 01/29/2015
• Est. Priority Date: 07/26/2013
• Status: Active Grant

• Current Assignee: Genesys Cloud Services Incorporated
• Original Assignee: Genesys Telecommunications Laboratories Incorporated (Genesys Cloud Services Incorporated)
• Inventors: Lev-Tov, Amir, Faizakof, Avraham, Ollinger, David, Konig, Yochai
• Granted Patent: US 9,971,764 B2
• Time in Patent Office: Days
• US Class Current: 704/245
• CPC Class Codes: G06F 16/355 Class or cluster creation or modification
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G10L 15/1822 Parsing for meaning understanding

Abstract

A method for identifying concepts in a plurality of interactions includes: filtering, on a processor, the interactions based on intervals; creating, on the processor, a plurality of sentences from the filtered interactions; computing, on the processor, a saliency of each the sentences; pruning away, on the processor, sentences with low saliency for generating a set of informative sentences; clustering, on the processor, the sentences of the set of informative sentences for generating a plurality of sentence clusters, each of the clusters corresponding to a concept of the concepts; computing, on the processor, a saliency of each of the clusters; and naming, on the processor, each of the clusters.

Claims

1. A method for identifying concepts in a plurality of interactions, the method comprising:

filtering, on a processor, the interactions based on intervals;

creating, on the processor, a plurality of sentences from the filtered interactions;

computing, on the processor, a saliency of each the sentences;

pruning away, on the processor, sentences with low saliency for generating a set of informative sentences;

clustering, on the processor, the sentences of the set of informative sentences for generating a plurality of sentence clusters, each of the clusters corresponding to a concept of the concepts;

computing, on the processor, a saliency of each of the clusters; and

naming, on the processor, each of the clusters.

2. The method of claim 1, wherein the interactions comprise an output of a voice recognition system.

3. The method of claim 2, further comprising filtering the output of the voice recognition system based on word confidence.

4. The method of claim 2, wherein the voice recognition system is a large-vocabulary continuous speech recognition system.

5. The method of claim 1, wherein the intervals are time intervals.

6. The method of claim 1, wherein the clustering the sentences comprises:

selecting a plurality of template sentences from the set of informative sentences, each of the template sentences corresponding to one of the clusters;

computing a similarity of each sentence in the set of informative sentences to the template sentences;

assigning each of the sentences of the set of informative sentences to a cluster of the clusters in accordance with the computed similarities with the template sentences; and

removing clusters having fewer than a threshold number of assigned sentences.

7. The method of claim 6, wherein the clustering the sentences further comprises, iteratively, selecting additional template sentences from sentences that were not assigned to a cluster and repeating the selecting the plurality of template sentences, computing the similarity of each sentence to the template sentences;

the assigning the each of the sentences to the clusters; and

removing clusters having fewer than the threshold number of assigned sentences until all sentences are assigned or until an iteration limit is reached.

8. The method of claim 1, wherein the naming each of the clusters comprises, for each cluster:

computing a term frequency-inverse document frequency for each sentence in the cluster and

naming the cluster with a sentence of the cluster having the highest term frequency-inverse document frequency.

9. A method for exploring concepts automatically identified in a plurality of interactions, the method comprising:

receiving, on a processor, a query comprising a concept;

retrieving, by the processor, a cluster containing the concept; and

displaying the retrieved cluster,

wherein the concepts are automatically identified by;

filtering, on the processor, the interactions based on intervals;

creating, on the processor, a plurality of sentences from the filtered interactions;

computing, on the processor, a saliency of each the sentences;

pruning away, on the processor, sentences with low saliency for generating a set of informative sentences; and

clustering, on the processor, the sentences of the set of informative sentences for generating a plurality of sentence clusters, each of the clusters corresponding to a concept of the concepts.

10. A computer system comprising:

a processor; and

a memory, wherein the memory has stored thereon instructions that, when executed by the processor, causes the processor to identify concepts in a plurality of interactions by;

filtering the interactions based on intervals;

creating a plurality of sentences from the filtered interactions;

computing a saliency of each the sentences;

pruning away sentences with low saliency for generating a set of informative sentences;

clustering the sentences of the set of informative sentences for generating a plurality of sentence clusters, each of the clusters corresponding to a concept of the concepts;

computing a saliency of each of the clusters; and

naming each of the clusters.

11. The system of claim 10, wherein the interactions comprise an output of a voice recognition system.

12. The system of claim 11, wherein the instructions further comprise instructions for filtering the output of the voice recognition system based on word confidence.

13. The system of claim 11, wherein the voice recognition system is a large-vocabulary continuous speech recognition system.

14. The system of claim 10, wherein the intervals are time intervals.

15. The system of claim 10, wherein the clustering the sentences comprises:

selecting a plurality of template sentences from the set of informative sentences, each of the template sentences corresponding to one of the clusters;

computing a similarity of each sentence in the set of informative sentences to the template sentences;

assigning each of the sentences of the set of informative sentences to a cluster of the clusters in accordance with the computed similarities with the template sentences; and

removing clusters having fewer than a threshold number of assigned sentences.

16. The system of claim 15, wherein the clustering the sentences further comprises, iteratively, selecting additional template sentences from sentences that were not assigned to a cluster and repeating the selecting the plurality of template sentences, computing the similarity of each sentence to the template sentences;

the assigning the each of the sentences to the clusters; and

removing clusters having fewer than the threshold number of assigned sentences until all sentences are assigned or until an iteration limit is reached.

17. The system of claim 10, wherein the naming each of the clusters comprises, for each cluster:

computing a term frequency-inverse document frequency for each sentence in the cluster and

naming the cluster with a sentence of the cluster having the highest term frequency-inverse document frequency.

18. A computer system comprising:

a processor; and

a memory, wherein the memory has stored thereon instructions that, when executed by the processor, causes the processor to respond to requests for exploration of concepts by;

receiving a query comprising a concept;

retrieving a cluster containing the concept; and

displaying the retrieved cluster,

wherein the concepts are automatically identified by;

filtering the interactions based on intervals;

creating a plurality of sentences from the filtered interactions;

computing a saliency of each the sentences;

pruning away sentences with low saliency for generating a set of informative sentences; and

clustering the sentences of the set of informative sentences for generating a plurality of sentence clusters, each of the clusters corresponding to a concept of the concepts.