CN105302859B - A kind of intelligent interactive system Internet-based - Google Patents

A kind of intelligent interactive system Internet-based Download PDF

Info

Publication number
CN105302859B
CN105302859B CN201510603622.8A CN201510603622A CN105302859B CN 105302859 B CN105302859 B CN 105302859B CN 201510603622 A CN201510603622 A CN 201510603622A CN 105302859 B CN105302859 B CN 105302859B
Authority
CN
China
Prior art keywords
word
user
information
phrase
carry out
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201510603622.8A
Other languages
Chinese (zh)
Other versions
CN105302859A (en
Inventor
李波
曾永梅
姚贡之
朱频频
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Zhizhen Intelligent Network Technology Co Ltd
Original Assignee
Shanghai Zhizhen Intelligent Network Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Zhizhen Intelligent Network Technology Co Ltd filed Critical Shanghai Zhizhen Intelligent Network Technology Co Ltd
Priority to CN201510603622.8A priority Critical patent/CN105302859B/en
Publication of CN105302859A publication Critical patent/CN105302859A/en
Application granted granted Critical
Publication of CN105302859B publication Critical patent/CN105302859B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36Creation of semantic tools, e.g. ontology or thesauri
    • G06F16/367Ontology
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33Querying
    • G06F16/332Query formulation
    • G06F16/3329Natural language query formulation or dialogue systems
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/30Semantic analysis

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Animal Behavior & Ethology (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Computer Interaction (AREA)
  • Machine Translation (AREA)

Abstract

A kind of intelligent interactive system Internet-based, the system are handled user information using following steps:A, the information issued to user segments;B, whether word, word and phrase after segmenting described in step A, which belong to entity, identifies;C, word, word and phrase after segmenting described in step A carry out semantic tagger analysis;D, word, word and phrase after segmenting described in step A carry out text error correction;E, syntactic analysis is carried out to the information that user issues;F, word, word and phrase after segmenting described in the information and step A of user's sending carry out weight correction;G, context processing is carried out to the information that user issues;H, according to the step B-G's as a result, to user issue information carry out similarity calculation, obtain threshold value;I, knowledge base is preset according to threshold value result queries, returns result to user.

Description

A kind of intelligent interactive system Internet-based
Technical field
The present invention relates to a kind of intelligent interactive method, more particularly to a kind of intelligent answer side Internet-based Method.
Background technique
In traditional intelligent interaction, the general of intelligent interaction copes with complicated dialogue, accuracy using template way It is lower, or analyzed after carrying out various participles to information, but general word segmentation result type is more, accuracy is lower.
Summary of the invention
The invention discloses a kind of intelligent interactive systems Internet-based, include the following steps:
A, the information issued to user segments;
B, whether word, word and phrase after segmenting described in step A, which belong to entity, identifies;
C, word, word and phrase after segmenting described in step A carry out semantic tagger analysis;
D, word, word and phrase after segmenting described in step A carry out text error correction;
E, syntactic analysis is carried out to the information that user issues;
F, word, word and phrase after segmenting described in the information and step A of user's sending carry out weight correction;
G, context processing is carried out to the information that user issues;
H, according to the step B-G's as a result, to user issue information carry out similarity calculation, obtain threshold value;
I, knowledge base is preset according to threshold value result queries, returns result to user.
Semantic tagger analysis in the step C includes field, different degree, similar word, synonym, cyberspeak.
Text error correction in the step D includes that the service class word and phrase in field carry out phonetic error correction;
Syntactic analysis in the step F is using rule and mask method.
Detailed description of the invention
Fig. 1-ontology and the example of instantiation, succession
Fig. 2-part of speech management
Fig. 3-synonym, weight correction
Specific embodiment
The invention discloses a kind of intelligent interactive systems Internet-based, include the following steps:
A, the information issued to user segments;
Participle is the common technology means of Computational Linguistics or artificial intelligence field, general using " maximum matching participle Method " or " most probable number method participle ",
B, whether word, word and phrase after segmenting described in step A, which belong to entity, identifies;
It is the instantiation of ontology for entity,
So-called ontology is clear to one kind of concept and is described in detail, is a kind of description method to real world.Or Person says that ontology is actually the Formal Representation to certain set concept and its mutual relationship among specific area.General packet Contain:
--- specific instances of ontology (object Object)
--- the attribute of ontology
--- affiliated Ontological classifications.
After instances of ontology, so that it may the attribute of ontology is inherited, for semantic tagger ready for analysis thereafter;
Specifically, such as attached drawing 1, there are many basic businesses for banking, all basic businesses are exactly a kind of Body for a certain specific ontology, for example handles rule, and marketing activity is exactly the succession of a kind of pair of basic business, is owned Attribute can inherit.
C, word, word and phrase after segmenting described in step A carry out semantic tagger analysis;
Semantic tagger is analyzed, including two parts of part-of-speech tagging and word sense tagging:
For part-of-speech tagging:The magnetic mark side of the general mistake driving using Hidden Markov Model or based on conversion Method;
For word sense tagging:Generally use word sense disambiguation method based on mutual information or row's discrimination method based on dictionary;
For the step, when user inputs a problem in robot front end, this problem can carry out word segmentation processing first, Then it is matched according to the result of participle, therefore the construction superiority and inferiority of part of speech, it is closely coupled with the degree of intelligence of robot.To word The additions and deletions of class and modification all can be【Part of speech management】Middle realization.
Such as Fig. 2,【Part of speech management】There is " public part of speech " under label, " proprietary part of speech ", wherein being ontology under " public part of speech " The corresponding part of speech of generic attribute is the customized peculiar part of speech of project under " proprietary part of speech ".
D, word, word and phrase after segmenting described in step A carry out text error correction;
E, syntactic analysis is carried out to the information that user issues;
F, word, word and phrase after segmenting described in the information and step A of user's sending carry out weight correction;
Such as Fig. 3, selection needs the classification right click being linked into, selects in a menu【Newly-built subclassification】, filled out in pop-up box Enter typonym and saves completion.
In this system, " * " " # " marked beside item name is respectively intended to distinguish the different degree and similarity of part of speech, " * " Represent important, weight is higher;" # " represents dissmilarity, and similarity is very low;"@" represents the word under the classification with phonetic error correction function Energy.Subclassification inherits " * " " # " "@" setting of parent classification automatically.
This system can also adjust weight according to user data log.Such as:" no " it is inessential to be based on statistics for word, but passes through Statistical analysis is crossed, " no " word occurs and sentence tail ratio is higher, and meaning is entirely different, so when " no " word appears in tail, Such as " I can open CRBT not " adjustment " no " word weight.
G, context processing is carried out to the information that user issues;
H, according to the step B-G's as a result, to user issue information carry out similarity calculation, obtain threshold value;
In addition, this system can also realize " hybrid operation of semantic formula and common question sentence ",
Such as:One standard ask for:" cosmetics mark exaggerates effect, false expression, how to investigate and prosecute?"
The semantic formula that the corresponding standard is asked can be analyzed to:[cosmetics | makeup brand] [falseness] [mark] [punishment] [method?]
The a certain extension that the corresponding standard is asked ask for:" the false information of cosmetics mark mark, it is industrial and commercial for this behavior Office takes any method to punish "
Assuming that including above-mentioned knowledge in knowledge base, system can be mixed processing to the information that user provides.Judge The problem of user, is such as close to standard and asks and can directly answer;It is such as decomposed into semantic formula, then is answered according to semantic formula; Cannot such as semantic formula be resolved into and be close to extension and ask, then ask answer according to extension;And non-individual is using above-mentioned any one Mode, to obtain max-thresholds.The answer of i.e. most identical user demand.
I, knowledge base is preset according to threshold value result queries, returns result to user.
Semantic tagger analysis in the step C includes field, different degree, similar word, synonym, cyberspeak.
Specifically, after by carrying out semantic tagger analysis according to above-mentioned aspect, the semanteme of the word divided is accurate, ambiguity It substantially eliminates.
Text error correction in the step D includes that the service class word and phrase in field carry out phonetic error correction;
Syntactic analysis in the step E is using rule and mask method.

Claims (4)

1. a kind of intelligent interactive system Internet-based, the system is handled user information using following steps:
A, the information issued to user segments;
B, whether word, word and phrase after segmenting described in step A, which belong to entity, identifies;
C, word, word and phrase after segmenting described in step A carry out semantic tagger analysis;
D, word, word and phrase after segmenting described in step A carry out text error correction;
E, syntactic analysis is carried out to the information that user issues;
F, according to user data log, word, word and phrase after segmenting described in the information and step A of user's sending carry out weight Correction;
G, context processing is carried out to the information that user issues;
H, according to the step B-G's as a result, to user issue information carry out similarity calculation, obtain threshold value;In the phase Mixed processing is carried out to the information that user issues like in degree calculating, including:The problem of judging user is such as close to that standard asks can be straight It takes back and answers;It is such as decomposed into semantic formula, then is answered according to semantic formula;Semantic formula cannot such as be resolved into and close to It is asked in extension, then asks answer according to extension;
I, knowledge base is preset according to threshold value result queries, returns result to user.
2. a kind of intelligent interactive system Internet-based according to claim 1, it is characterised in that:
Semantic tagger analysis in the step C includes field, different degree, similar word, synonym, cyberspeak.
3. a kind of intelligent interactive system Internet-based according to claim 1, it is characterised in that:
Text error correction in the step D includes that the service class word and phrase in field carry out phonetic error correction.
4. a kind of intelligent interactive system Internet-based according to claim 1, it is characterised in that:
Syntactic analysis in the step E is using rule and mask method.
CN201510603622.8A 2015-09-21 2015-09-21 A kind of intelligent interactive system Internet-based Active CN105302859B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510603622.8A CN105302859B (en) 2015-09-21 2015-09-21 A kind of intelligent interactive system Internet-based

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510603622.8A CN105302859B (en) 2015-09-21 2015-09-21 A kind of intelligent interactive system Internet-based

Publications (2)

Publication Number Publication Date
CN105302859A CN105302859A (en) 2016-02-03
CN105302859B true CN105302859B (en) 2018-11-30

Family

ID=55200131

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510603622.8A Active CN105302859B (en) 2015-09-21 2015-09-21 A kind of intelligent interactive system Internet-based

Country Status (1)

Country Link
CN (1) CN105302859B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106485328B (en) * 2016-10-31 2020-06-19 上海智臻智能网络科技股份有限公司 Information processing system and method
CN108073587B (en) * 2016-11-09 2022-05-27 阿里巴巴集团控股有限公司 Automatic question answering method and device and electronic equipment
CN106599163B (en) * 2016-12-08 2019-11-22 上海云信留客信息科技有限公司 A kind of data digging method and device for big data
CN110175230A (en) * 2019-05-29 2019-08-27 广州伟宏智能科技有限公司 Intelligent robot interactive system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101174259A (en) * 2007-09-17 2008-05-07 张琰亮 Intelligent interactive request-answering system
CN101373532A (en) * 2008-07-10 2009-02-25 昆明理工大学 FAQ Chinese request-answering system implementing method in tourism field
CN101510221A (en) * 2009-02-17 2009-08-19 北京大学 Enquiry statement analytical method and system for information retrieval
CN104657346A (en) * 2015-01-15 2015-05-27 深圳市前海安测信息技术有限公司 Question matching system and question matching system in intelligent interaction system

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10515154B2 (en) * 2014-03-12 2019-12-24 Sap Se Systems and methods for natural language processing using machine-oriented inference rules

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101174259A (en) * 2007-09-17 2008-05-07 张琰亮 Intelligent interactive request-answering system
CN101373532A (en) * 2008-07-10 2009-02-25 昆明理工大学 FAQ Chinese request-answering system implementing method in tourism field
CN101510221A (en) * 2009-02-17 2009-08-19 北京大学 Enquiry statement analytical method and system for information retrieval
CN104657346A (en) * 2015-01-15 2015-05-27 深圳市前海安测信息技术有限公司 Question matching system and question matching system in intelligent interaction system

Also Published As

Publication number Publication date
CN105302859A (en) 2016-02-03

Similar Documents

Publication Publication Date Title
CN105117388B (en) A kind of intelligent robot interactive system
CN109493166B (en) Construction method for task type dialogue system aiming at e-commerce shopping guide scene
CN107122346B (en) The error correction method and device of a kind of read statement
Sidorov et al. Empirical study of machine learning based approach for opinion mining in tweets
CN105843897B (en) A kind of intelligent Answer System towards vertical field
US8548805B2 (en) System and method of semi-supervised learning for spoken language understanding using semantic role labeling
CN106339602B (en) A kind of intelligent consulting robot
WO2020233386A1 (en) Intelligent question-answering method and device employing aiml, computer apparatus, and storage medium
CN105302859B (en) A kind of intelligent interactive system Internet-based
CN111062220B (en) End-to-end intention recognition system and method based on memory forgetting device
CN102929860B (en) Chinese clause emotion polarity distinguishing method based on context
CN107391575A (en) A kind of implicit features recognition methods of word-based vector model
CN105868179A (en) Intelligent asking-answering method and device
US20230394247A1 (en) Human-machine collaborative conversation interaction system and method
CN106326307A (en) Language interaction method
Chao et al. Emerging Technologies of Natural Language‐Enabled Chatbots: A Review and Trend Forecast Using Intelligent Ontology Extraction and Patent Analytics
Zad et al. Systematic evaluation of a framework for unsupervised emotion recognition for narrative text
Umber et al. NL-based automated software requirements elicitation and specification
CN105183716B (en) A kind of intelligent interactive method based on abstract semantics
Veena et al. A graph-based relation extraction method for question answering system
CN110309513B (en) Text dependency analysis method and device
KR101851786B1 (en) Apparatus and method for generating undefined label for labeling training set of chatbot
CN105260178A (en) Intelligent cloud service application development method and system
Barbieri et al. Towards a natural language conversational interface for process mining
CN114579709A (en) Intelligent question-answering intention identification method based on knowledge graph

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant