EPISODIC AND SEMANTIC MEMORY BASED REMEMBRANCE AGENT MODELING METHOD AND SYSTEM FOR VIRTUAL COMPANIONS

• US 20150278688A1
• Filed: 03/25/2015
• Published: 10/01/2015
• Est. Priority Date: 03/25/2014
• Status: Active Grant

• Current Assignee: Nanyang Technology University
• Original Assignee: Nanyang Technology University
• Inventors: MIAO, Chunyan, SHEN, Zhiqi, HAN, Xiaogang
• Granted Patent: US 9,928,463 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 1/163 Wearable computers, e.g. on a belt
• CPC Class Codes: G06F 16/345 Summarisation for human users
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

A remembrance agent is proposed, to be hosted on a wearable computer. The remembrance agent employs a first database (WK) of knowledge about the world in the form of concept maps. Events (“episodes”) experienced by a user are each classified as relating to one or more of the concepts in WK. The episodes are used to produce a second database (episodic memory, EM), and the classification is also used to update a third database (semantic memory, SM) organised using the concepts. The semantic memory thus summarizes the user'"'"'s interaction with the concepts during the episodes. A current situation of the user is classified according to the concepts, and the classification is used, with the EM and SM, to provide to the user information relevant to the current situation.

Claims

1. A computer system, comprising:

a computer processor;

at least one sensor for determining at least one characteristic of the user;

at least one information transmission device for transmitting information to the user; and

a data storage device, the data storage device storing;

(i) an episodic memory database of documents, each document representing a respective previous episode experienced by the user;

(ii) a semantic memory database indicative of the incidence in the previous episodes of each of a plurality of predefined concepts C; and

(iii) program instructions operative by the processor to cause the processor to;

(a) use data obtained using the at least one sensor to classify a current situation according to the plurality of predefined concepts C, to derive a current concept set which is one or more of said predefined concepts C, the current concept set being those of the predefined concepts C which are relevant to the current situation; and

(b) use the current concept set, the episodic memory and the semantic memory to obtain additional information relevant to the current situation, and transmit at least some of the additional information to the user using the at least one information transmission device.

2. A system according to claim 1 further comprising at least one data input device for receiving input from the user, the program instructions being operative based on the input to allow the user to select, from among the additional information, a subset to be transmitted to the user using the at least one information transmission device.

3. A system according to claim 1 which is a wearable computer system, adapted to be worn by the user and carried by the user as the user moves, the at least one sensor including a sensor to determine the user'"'"'s location.

4. A method of obtaining information relevant to a current situation, the method being performed by a computer system comprising:

a computer processor;

at least one sensor for determining at least one characteristic of the user;

at least one data input device for receiving input from the user;

at least one information transmission device for transmitting information to the user;

a data storage device, the data storage device storing;

(i) an episodic memory database of documents, each document representing a respective previous episode experienced by the user;

(ii) a semantic memory database indicative of the incidence in the previous episodes of each of a plurality of predefined concepts C;

the method comprising;

(a) using data obtained using the at least one sensor to classify the current situation according to the plurality of predefined concepts C, to derive a current concept set which is one or more of said predefined concepts C, the current concept set being those of the predefined concepts C which are relevant to the current situation; and

(b) using the current concept set, the episodic memory and the semantic memory to obtain additional data relevant to the current situation.

5. A method according to claim 4 further comprising a preliminary step of generating the semantic memory by:

classifying the documents in the episodic memory according to the predefined concepts C to determine a sub-set of the predefined concepts relevant to each of the documents, and

for each document, updating the semantic memory according to the corresponding derived sub-set of the predefined concepts.

6. A method according to claim 4 which is performed when the user is in a retail environment, the additional data comprising at least data describing products which are for sale in the retail environment.