Conversation agent

Abstract

Various systems and methods for a conversation agent are described herein. A system for a conversation agent includes a context module to determine a context of a conversation involving a first participant and a second participant; a conversation history module to access a conversation history of the first participant and a conversation history of the second participant; a conversation topic module to identify a conversation topic using the context of the conversation, the conversation history of the first participant, and the conversation history of the second participant; and a presentation module to present the conversation topic to the first participant.

What is claimed is: 
     
         1 . A system for a conversation agent, the system comprising:
 a context module to determine a context of a conversation involving a first participant and a second participant;   a conversation history module to access a conversation history of the first participant and a conversation history of the second participant;   a conversation topic module to identify a conversation topic using the context of the conversation, the conversation history of the first participant, and the conversation history of the second participant; and   a presentation module to present the conversation topic to the first participant.   
     
     
         2 . The system of  claim 1 , wherein to determine the context of the conversation, the context module is to:
 receive sensor information describing a mental, emotional, or physical state of the first participant; and   determine an interest level of the first participant regarding the conversation.   
     
     
         3 . The system of  claim 1 , wherein to determine the context of the conversation, the context module is to:
 receive sensor information describing a mental, emotional, or physical state of the second participant; and   determine an interest level of the second participant regarding the conversation.   
     
     
         4 . The system of  claim 2 , wherein the sensor information is obtained from a brain computer interface sensor. 
     
     
         5 . The system of  claim 4 , wherein the brain computer interface detects alertness of the first participant, the alertness used to derive the interest level of the first participant. 
     
     
         6 . The system of  claim 2 , wherein the sensor information is obtained from an image recognition sensor. 
     
     
         7 . The system of  claim 6 , wherein the image recognition sensor detects gestures of the second participant, the gestures used to derive the interest level of the second participant. 
     
     
         8 . The system of  claim 6 , wherein the image recognition sensor tracks eye movement of the second participant, the eye movement used to derive the interest level of the second participant. 
     
     
         9 . The system of  claim 6 , wherein the image recognition sensor tracks facial expressions of the second participant, the facial expressions used to derive the interest level of the second participant. 
     
     
         10 . The system of  claim 2 , wherein the sensor information is obtained from an audio sensor. 
     
     
         11 . The system of  claim 10 , wherein the audio sensor tracks voice characteristics of the second participant, the voice characteristics used to derive the interest level of the second participant. 
     
     
         12 . The system of  claim 2 , wherein the sensor information is obtained from a physiological sensor. 
     
     
         13 . The system of  claim 12 , wherein the physiological sensor tracks heart rate of the first participant, the heart rate used to derive the interest level of the first participant. 
     
     
         14 . The system of  claim 13 , further comprising:
 an alert module to alert the first participant when the interest level of the first participant exceeds a threshold.   
     
     
         15 . The system of  claim 14 , wherein the alert comprises an audio cue. 
     
     
         16 . The system of  claim 15 , wherein the audio cue comprises a voice message. 
     
     
         17 . The system of  claim 16 , wherein the voice message is presented to the first participant and is in a native language native of the second participant. 
     
     
         18 . The system of  claim 14 , wherein the alert comprises a haptic cue. 
     
     
         19 . The system of  claim 1 , wherein the conversation history of the first participant comprises a summary of topics of a previous conversation with the second participant. 
     
     
         20 . The system of  claim 1 , wherein the conversation history of the first participant comprises follow up notes provided by the first participant regarding a previous conversation with the second participant. 
     
     
         21 . The system of  claim 1 , wherein to identify the conversation topic, the conversation topic module is to:
 model a conversation shape of the first participant based on the conversation history of the first participant;   model a conversation shape of the second participant based on the conversation history of the second participant; and   identify an intersection of the conversation shapes of the first and second participants, the intersection corresponding to the conversation topic.   
     
     
         22 . The system of  claim 21 , wherein the conversation topic module is to:
 identify a plurality of potential conversation topics from the intersection of the conversation shapes of the first and second participants; and   use a heuristic to identify the conversation topic from the plurality of potential conversation topics.   
     
     
         23 . A method of implementing a conversation agent on a compute device, the method comprising:
 determining a context of a conversation involving a first participant and a second participant;   accessing a conversation history of the first participant and a conversation history of the second participant;   identifying a conversation topic using the context of the conversation, the conversation history of the first participant, and the conversation history of the second participant; and   presenting the conversation topic to the first participant.   
     
     
         24 . The method of  claim 23 , wherein identifying the conversation topic comprises:
 modeling a conversation shape of the first participant based on the conversation history of the first participant;   modeling a conversation shape of the second participant based on the conversation history of the second participant; and   identifying an intersection of the conversation shapes of the first and second participants, the intersection corresponding to the conversation topic.   
     
     
         25 . The method of  claim 24 , further comprising:
 identifying a plurality of potential conversation topics from the intersection of the conversation shapes of the first and second participants; and   using a heuristic to identify the conversation topic from the plurality of potential conversation topics.   
     
     
         26 . At least one machine-readable medium including instructions, which when executed by a machine, cause the machine to:
 determine a context of a conversation involving a first participant and a second participant;   access a conversation history of the first participant and a conversation history of the second participant;   identify a conversation topic using the context of the conversation, the conversation history of the first participant, and the conversation history of the second participant; and   present the conversation topic to the first participant.   
     
     
         27 . The at least one machine-readable medium of  claim 26 , further comprising instructions to access a personal history of the first participant, and wherein the instructions to identify the conversation topic include instructions to use the personal history of the first participant. 
     
     
         28 . The at least one machine-readable medium of  claim 27 , wherein identifying the conversation topic comprises:
 identifying a situation from the personal history of the first participant that is analogous to the context of the conversation; and   identifying the conversation topic based on the situation.   
     
     
         29 . The at least one machine-readable medium of  claim 26 , further comprising instructions to access a personal history of the second participant, and wherein the instructions to identify the conversation topic comprise instructions to use the personal history of the second participant. 
     
     
         30 . The at least one machine-readable medium of  claim 29 , wherein the instructions to accessing the personal history of the second participant comprise instructions to access a social network profile of the second participant.