Interactive acquisition of remote services

Patent metadata

Field	Value
Publication number	US-9122745-B2
Application number	US-201313890969-A
Country	US
Kind code	B2
Filing date	May 9, 2013
Priority date	May 9, 2013
Publication date	Sep 1, 2015
Grant date	Sep 1, 2015

Abstract

A natural language specification of at least one high level information technology services requirement is obtained from a user, via a conversational interface; the same is parsed into first pre-defined semi-structured data, using a conversation parser. Based on the first pre-defined semi-structured data, a subset of candidate information technology services is identified, with a dialog engine, from a plurality of candidate information technology services provided by a plurality of vendors, the dialog engine is used to formulate a response including second pre-defined semi-structured data. The response is reverse-parsed into a natural language response, using the conversation parser. The natural language response includes a question for the user to assist in further refining the subset of candidate information technology services; the natural language response is presented to the user via the conversational interface.

First claim

What is claimed is:

1. A method comprising:
obtaining from a user, via a conversational interface, a natural language specification of at least one high level information technology services requirement;
parsing said natural language specification of said at least one high level information technology services requirement into first pre-defined semi-structured data, using a conversation parser;
based on said first pre-defined semi-structured data:
identifying, with a dialog engine, a subset of candidate information technology services from a plurality of candidate information technology services provided by a plurality of vendors; and
formulating, with said dialog engine, a response comprising second pre-defined semi-structured data;

reverse parsing said response comprising said second pre-defined semi-structured data into a natural language response, using said conversation parser, said natural language response comprising a question for said user to assist in further refining said subset of candidate information technology services; and
causing said natural language response to be presented to said user via said conversational interface.

2. The method of claim 1 , further comprising:
obtaining from said user, via said conversational interface, a natural language input responsive to said natural language response;
parsing said natural language input into third pre-defined semi-structured data, using said conversation parser;
based on said third pre-defined semi-structured data:
filtering said subset of candidate information technology services from said plurality of candidate information technology services provided by a plurality of vendors; and
configuring at least one candidate information technology service of said further filtered subset of candidate information technology services to meet said specification of said user.

3. The method of claim 2 , wherein said filtering and said configuring are carried out simultaneously.

4. The method of claim 2 , wherein said filtering comprises iteration-min filtering.

5. The method of claim 4 , wherein said identifying comprises semantically querying a service knowledge base with a semantic query engine.

6. The method of claim 5 , further comprising building said service knowledge base by extracting features from service descriptions in accordance with a schema of an ontology.

7. The method of claim 6 , further comprising obtaining said service descriptions via a crawler.

8. The method of claim 6 , further comprising obtaining said service descriptions via a service provider registration interface.

9. The method of claim 1 , wherein said parsing comprises:
extracting at least one of a verb and a noun from at least one sentence of said natural language specification;
removing stop words from said extracted at least one of a verb and a noun;
dividing words remaining after said removing into include and exclude sets;
propagating said include and exclude sets; and
finding at least one of a corresponding category and a corresponding capability by tag annotating said include and exclude sets.

10. The method of claim 1 , further comprising providing a system, wherein the system comprises distinct software modules, each of the distinct software modules being embodied on a computer-readable storage medium, and wherein the distinct software modules comprise a conversational interface module, a conversation parser module, and a dialog engine module;
wherein:
said obtaining and said causing are carried out by said conversational interface module executing on at least one hardware processor, said conversational interface module executing on said at least one hardware processor comprising said conversational interface;
said parsing and said reverse parsing are carried out by said conversation parser module executing on said at least one hardware processor, said conversation parser module executing on said at least one hardware processor comprising said conversation parser; and
said identifying and formulating are carried out by said dialog engine module executing on said at least one hardware processor, said dialog engine module executing on said at least one hardware processor comprising said dialog engine.

11. A computer program product comprising a non-transitory computer readable storage medium having computer readable program code embodied therewith, said computer readable program code comprising:
computer readable program code configured to obtain from a user a natural language specification of at least one high level information technology services requirement;
computer readable program code configured to parse said natural language specification of said at least one high level information technology services requirement into first pre-defined semi-structured data;
computer readable program code configured to, based on said first pre-defined semi-structured data:
identify a subset of candidate information technology services from a plurality of candidate information technology services provided by a plurality of vendors; and
formulate a response comprising second pre-defined semi-structured data;

computer readable program code configured to reverse parse said response comprising said second pre-defined semi-structured data into a natural language response, said natural language response comprising a question for said user to assist in further refining said subset of candidate information technology services; and
computer readable program code configured to cause said natural language response to be presented to said user.

12. A system comprising:
a memory;
at least one processor, coupled to said memory; and
a non-transitory computer-readable storage medium, a plurality of distinct software modules being embodied on said non-transitory computer-readable medium, said distinct software modules being loadable into said memory for execution by said processor, said distinct software modules comprising a conversational interface module, a conversation parser module, and a dialog engine module;
wherein said at least one processor is operative to:
create a conversational interface by executing said conversational interface module, said conversational interface obtaining from a user a natural language specification of at least one high level information technology services requirement;
create a conversation parser by executing said conversation parser module, said conversation parser parsing said natural language specification of said at least one high level information technology services requirement into first pre-defined semi-structured data;
create a dialog engine by executing said dialog engine module, said dialog engine, based on said first pre-defined semi-structured data:
identifying a subset of candidate information technology services from a plurality of candidate information technology services provided by a plurality of vendors; and
formulating a response comprising second pre-defined semi-structured data;

wherein:
said conversation parser reverse parses said response comprising said second pre-defined semi-structured data into a natural language response, said natural language response comprising a question for said user to assist in further refining said subset of candidate information technology services; and
said conversational interface causes said natural language response to be presented to said user.

13. The system of claim 12 , wherein:
said conversational interface obtains from said user a natural language input responsive to said natural language response;
said conversation parser parses said natural language input into third pre-defined semi-st

Assignees

• IBM

Inventors

• Akolkar Rahul P
• Chefalas Thomas E
• Jiang Yexi
• Laredo Jim A
• Perng Chang-Shing
• Sailer Anca
• Schaffa Frank A
• Segal Alla
• Silva-Lepe Ignacio
• Tao Tao
• Zhou Yang

Classifications

• G06F16/3329 Primary

  Natural language query formulation · CPC title

• G06F17/30654 Primary

  Physics · mapped topic

• G06F16/243 Primary

  Natural language query formulation · CPC title