System and method of determining topics of a communication
Systems and methods include using a clustering engine to determine topics in a communication. An example system includes a user interface module that receives user input relating to a criteria to define a set of communications, wherein the criteria is at least one of a category, a score, a sentiment, an agent, an agent grouping, a speaker, a location, an event attribute, a call center, a time of communication, or a date of communication, an acoustic analysis module that analyzes the set of communications to determine one or more acoustic characteristics of one or more communications in the set of communications, and a clustering engine that analyzes words and phrases in the set of communications and the one or more acoustic characteristics, and to determine a topic of the set of communications based on at least one commonality in words, phrases, or the one or more acoustic characteristics.
1. A system, comprising:
a user interface module that receives a user input providing a criteria defining a set of communications, wherein the criteria is selected by the user via the user input such that the criteria is user-defined and includes a category, the category includes both a language category and an acoustic category, the language category is defined by at least one specific language characteristic, the acoustic category is defined by at least one specific acoustic characteristic, and the user interface module is configured to receive a selection of both the language category and the acoustic category from the user input;
a language module that analyzes the set of communications, which are defined by the criteria selected by the user, to determine one or more language characteristics of one or more communications in the set of communications;
an acoustic analysis module that analyzes the set of communications, which are defined by the criteria selected by the user, to determine one or more acoustic characteristics of the one or more communications in the set of communications; and
a determination module that categorizes the one or more communications in the set of communications by determining to associate at least one of the language category or the acoustic category with the one or more communications in the set of communications using the determined one or more language characteristics and the determined one or more acoustic characteristics, wherein the determined one or more language characteristics includes a language pattern comprising a proximity of a topic, and the determined one or more acoustic characteristics comprises a proximity of an event;
a clustering engine that analyzes words and phrases in the set of communications and the one or more acoustic characteristics, and determines a topic of the set of communications, which are defined by the criteria selected by the user, based on at least one commonality in words, phrases, the categorization of the one or more communications in the set of communications, or the one or more acoustic characteristics in the set of communications.
2. The system of claim 1 , wherein the one or more acoustic characteristics are at least one of a silence block, a stress of words, an aggregated stress of a plurality of words, an emphasis, an agitation, a tempo, a change in tempo, a gain in volume or energy of words, a tone, an overtalk, a time lag between words or symbols, a time dependency between key words or phrases, a proximity to an event, an inter-word timing, an inter-word sequencing, an inter-word distance, an inflexion of words, or a temporal pattern.
3. The system of claim 1 , wherein the one or more language characteristics is at least one of a phoneme, a word, a phrase, a placeholder word used in redaction, a language pattern, or a specific language characteristic comprising a presence or absence of specific language.
4. The system of claim 3 , wherein the one or more language characteristics is used to determine the category.
5. The system of claim 3 , wherein the one or more language characteristics is a language pattern comprising an order of occurrence of words.
6. The system of claim 3 , wherein the one or more language characteristics is a language pattern comprising an order of occurrence of words within a time frame.
7. The system of claim 3 , wherein the one or more language characteristics is a language pattern comprising a Boolean relationship of words.
8. The system of claim 7 , wherein the Boolean relationship is AND, OR, or NOT.
9. The system of claim 3 , wherein the one or more language characteristics is a language pattern comprising words at a specific location in the set of communications.
10. The system of claim 3 , wherein the one or more language characteristics is a language pattern comprising a proximity of a topic.
11. The system of claim 1 , wherein the category is at least one of a behavior, a reason, a procedure, a competitor, a dissatisfaction, an empathy, a repeat contact, a transferred call, a politeness, or a unique label assigned by the user or the system.
12. The system of claim 1 , wherein the set of communications are voice communications.
13. The system of claim 12 , wherein each voice communication is selected from a group consisting of VOIP, TDM, and SIP.
14. The system of claim 12 , wherein the system is adapted to convert the voice communications to text using a computer-based speech recognition module.
15. A method, comprising:
receiving, from a user interface, a user input that provides a criteria to define a set of communications, wherein the criteria is selected by the user via the user input such that the criteria is user-defined and includes a category, the category includes both a language category and an acoustic category, the language category is defined by at least one specific language characteristic, the acoustic category is defined by at least one specific acoustic characteristic, and the user interface is configured to receive a selection of both the language category and the acoustic category from the user input;
analyzing the set of communications, which are defined by the criteria selected by the user, to determine one or more language characteristics of one or more communications in the set of communications;
analyzing the set of communications, which are defined by the criteria selected by the user, to determine one or more acoustic characteristics of the one or more communications in the set of communications;
categorizing at least one of the set of communications by determining to associate at least one of the language category or the acoustic category with the one or more communications in the set of communications using the determined one or more language characteristics and the determined one or more acoustic characteristics, wherein the determined one or more language characteristics includes a language pattern comprising a proximity of a topic, and the determined one or more acoustic characteristics comprises a proximity of an event;
analyzing words and phrases in the set of communications and the one or more acoustic characteristics; and
determining a topic of the set of communications, which are defined by the criteria selected by the user, based on at least one commonality in words, phrases, the categorization of the one or more communications in the set of communications, or the one or more acoustic characteristics in the set of communications.
16. The method of claim 15 , wherein the one or more acoustic characteristics are at least one of a silence block, a stress of words, an aggregated stress of a plurality of words, an emphasis, an agitation, a tempo, a change in tempo, a gain in volume or energy of words, a tone, an overtalk, a time lag between words or symbols, a time dependency between key words or phrases, a proximity to an event, an inter-word timing, an inter-word sequencing, an inter-word distance, an inflexion of words, or a temporal pattern.
17. The method of claim 15 , wherein the one or more language characteristics is at least one of a phoneme, a word, a phrase, a placeholder word used in redaction, a language pattern, or a specific language characteristic comprising a presence or absence of specific language.
18. The method of claim 17 , wherein the one or more language characteristics is used to determine the category.
19. The method of claim 17 , wherein the one or more language characteristics is a language pattern comprising at least one of an order of occurrence of words, an order of occurrence of words within a time frame, a Boolean relationship of words, words at a specific location in the set of communications, or a proximity of a topic.
20. The method of claim 15 , wherein the category is at least one of a behavior, a reason, a procedure, a competitor, a dissatisfaction, an empathy, a repeat contact, a transferred call, a politeness, or a unique label assigned by the user or a processor executing the method.
21. The system of claim 1 , wherein the criteria includes the category, and the set of communications share the category such that the clustering engine determines the topic of the set of communications having a shared category.
22. The system of claim 1 , wherein the criteria includes the category, and the category is a new category identified by the system from at least of a repeating word, a repeating phrase, or a repeating acoustic characteristic.
23. The system of claim 1 , wherein the topic of the set of communications includes a plurality of topics, and the system displays the plurality of topics according to a degree of commonality.
24. The system of claim 23 , wherein the degree of commonality is indicated by a respective font size of each of the plurality of topics.
25. The system of claim 1 , wherein the criteria additionally includes at least one of a score, a sentiment, an agent, an agent grouping, a speaker, a location, an event attribute, a call center, a time of communication, or a date of communication.