Establishing and querying methods of knowledge library engine based on emergency management

• US 9,679,045 B2
• Filed: 12/27/2012
• Issued: 06/13/2017
• Est. Priority Date: 12/21/2012
• Status: Expired due to Fees

• Current Assignee: Cloud Computing Center Chinese Academy of Sciences
• Original Assignee: Cloud Computing Center Chinese Academy of Sciences
• Inventors: He, Zhongtang, Li, Xinan, Zhao, Fengwei, Li, Zhiyong
• Primary Examiner(s): KUDDUS, DANIEL A
• Application Number:
  US14/398,468
• Publication Number:
  US 20150095331A1
• Time in Patent Office: 1,629 Days
• Field of Search: 709225, 709231
• CPC Class Codes: G06F 16/33 Querying
• CPC Class Codes: G06F 16/335 Filtering based on additional data, e.g. user or group profiles filtering in web context G06F16/9535, G06F16/9536
• CPC Class Codes: G06F 16/353 into predefined classes
• CPC Class Codes: G06F 40/157 using dictionaries or tables
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/10 Office automation; Time management

Abstract

An establishing method of a knowledge library engine based on emergency management includes steps of extracting relevant data from unstructured text information, filtering out unwanted data and improving a structure of the relevant data, so as to ensure homogeneity of the relevant data; structurally analyzing a text content of obtained information, extracting a feature thereof and storing the feature in the feature library as a key word of a knowledge; and processing the obtained information with the automatic word segmentation module, then classifying a result according to an information classification standard, and establishing an association of the knowledge and information classification. A querying method of the knowledge library engine based on emergency management includes steps of: understanding semanteme, understanding knowledge, and querying knowledge.

Claims

1. A querying method of a knowledge library engine, comprising steps of:

understanding semanteme, understanding knowledge, and querying knowledge;

wherein the step of understanding semanteme specifically comprises steps of;

extracting core words which form a sentence with an intelligent segmentation technique, for being used by a semantic analyzing module;

during word segmentation, firstly dividing the whole sentence into clauses according to punctuation;

then dividing the clauses into words according to character lengths; and

comparing a traversed word with a word in a dictionary library, wherein when the traversed word appears in a dictionary, extracting the traversed word;

when the traversed word does not appear in a dictionary, filtering off the word as redundancy information;

wherein the step of understanding knowledge specifically comprises a step of;

automatically associating a synonym of the word extracted during the step of understanding semanteme with a synonym dictionary, then querying a key word in a feature library with the synonym, for providing query results with high accuracy, high relativity and high comprehensiveness to a user question; and

wherein the step of querying knowledge specifically comprises steps of;

querying the key word and querying a type;

wherein the step of querying the key word specifically comprises a step of;

querying the feature library according to the synonym dictionary, then finding out related knowledge;

the step of querying a type specifically comprises a step of;

determining a type of the knowledge, wherein detailed knowledge information found out by understanding the knowledge is classified mainly according to structural association of the knowledge library, for clearly displaying information to a user;

wherein the knowledge library engine is established by;

dividing an application frame into an application layer, an application supporting layer, a data layer, and a basic supporting layer;

extracting relevant data from unstructured text information, separating a natural language with a binary tree, and treating a synonym with the synonym dictionary;

filtering off unwanted data and improving a structure of the relevant data, for ensuring homogeneity of the relevant data;

structurally analyzing a text content of obtained information, extracting a feature of the text content and storing the feature in the feature library as a key word of a knowledge;

processing the obtained information with the automatic word segmentation module, then classifying according to an information classification standard, and establishing an association of the knowledge and information classification; and

matching a mode during a user information query with the feature library and the type library;

analyzing an information query log, then finding out a user behavior habit, and feeding back to the information input module;

wherein the application layer comprises an information input module and an information query module;

wherein the application supporting layer comprises an information extraction module, an automatic word segmentation module, a classification module, a semantic understanding module, a knowledge understanding module, and a knowledge query module;

wherein the data layer comprises a knowledge library, a feature library, and a type library;

the data layer further comprises a synonym dictionary and a type word dictionary; and

wherein the basic supporting layer comprises a distributed database, a storage device, a server, an application workstation, a network, an operation system, and a middleware;

wherein the middleware is placed on a heterogeneous hardware platform and an operation system platform for shielding difference of the heterogeneous hardware platform and the operation system platform, and providing services comprising load balancing, transaction management and reliable messaging for an up-layer application.

2. The querying method, as recited in claim 1, further comprising a step of:

establishing a user information feedback mechanism in a query system, for finding out user information, wherein the user evaluates and scores according to the query results obtained, information judgment depends on a scoring result.