CN112232085A - Cross-DIKW modal text ambiguity processing method oriented to essential computing and reasoning - Google Patents
Cross-DIKW modal text ambiguity processing method oriented to essential computing and reasoning Download PDFInfo
- Publication number
- CN112232085A CN112232085A CN202011103480.6A CN202011103480A CN112232085A CN 112232085 A CN112232085 A CN 112232085A CN 202011103480 A CN202011103480 A CN 202011103480A CN 112232085 A CN112232085 A CN 112232085A
- Authority
- CN
- China
- Prior art keywords
- text
- target
- resources
- resource
- information
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application discloses a method for processing text ambiguity of a DIKW-crossing mode facing to essential computing and reasoning, which comprises the following steps: acquiring a target text, and determining a target data resource and a target information resource in the target text; inquiring related resources of the target text according to the target data resources and/or the target information resources, and determining the text meaning of the target text according to the related resources; if the number of the text meanings of the target text is larger than 1, acquiring supplementary resources of the target text, and generating a conditional restriction text of the target text according to the supplementary resources; and taking the text meaning of the text conforming to the condition limitation as the actual text meaning of the target text, and modifying the target text according to the actual text meaning. The method and the device can accurately identify and eliminate the ambiguity existing in the text. The application also discloses a system for processing the text ambiguity of the cross DIKW mode facing to essential computing and reasoning, electronic equipment and a storage medium, and the system has the beneficial effects.
Description
Technical Field
The application relates to the technical field of software engineering, in particular to a method and a system for processing ambiguity of a DIKW-crossing modal text oriented to essential computing and reasoning, electronic equipment and a storage medium.
Background
The advent of the big data age has made the size of data ever more enormous. The data can be associated and analyzed to obtain a lot of information, even important contents such as privacy, confidentiality and the like, the data and information resources can be summarized and logically inferred to become knowledge, the knowledge resources can act on the data resources and the information resources in turn, more new data resources and information resources which have values to specific targets can be calculated and inferred, and even some specific targets can be subjected to predictive analysis.
Ambiguity refers to the understanding of text content with different purposes, that is, information resources with different purposes can be obtained by carrying out various deductions on type resources in the content. The reasons for the ambiguity are two: one is that the content is lost, and a part of data resources or information resources are lacked, so that the understanding range of the content is wide, and different purposes of understanding can be generated during derivation; another is that there is redundancy in the content, which may lead to different purpose understandings when derived in combination with different types of resources. In the related technology, the processing of text ambiguity is mainly realized through a machine learning model, but the recognition accuracy of the machine learning model excessively depends on the richness of training samples, and the text ambiguity cannot be effectively processed.
Therefore, how to accurately identify and resolve ambiguities present in text is a technical problem that those skilled in the art need to solve at present.
Disclosure of Invention
The application aims to provide a method and a system for processing ambiguity of a cross-DIKW modal text facing essential computing and reasoning, an electronic device and a storage medium, which can accurately identify and eliminate ambiguity existing in the text.
In order to solve the technical problem, the application provides an essential computing and reasoning oriented method for processing text ambiguity in a cross-DIKW mode, which comprises the following steps:
acquiring a target text, and determining a target data resource and a target information resource in the target text;
inquiring related resources of the target text according to the target data resources and/or the target information resources, and determining the text meaning of the target text according to the related resources;
if the number of the text meanings of the target text is larger than 1, acquiring supplementary resources of the target text, and generating a conditional restriction text of the target text according to the supplementary resources;
and taking the text meaning conforming to the conditional restriction text as the actual text meaning of the target text, and modifying the target text according to the actual text meaning.
Optionally, determining the target data resource and the target information resource in the target text includes:
determining a resource type of the target text; the resource types comprise data resources, information resources and knowledge resources, the data resources are resources in a data map, the information resources are resources in an information map, and the knowledge resources are resources in a knowledge map;
performing cross-modal transformation on the target text to obtain the target data resource and the target information resource; and converting the cross-modal transformation into the transformation operation between any two resources of the data resource, the information resource, the knowledge resource and the data information mixed resource.
Optionally, performing cross-modal transformation on the target text to obtain the target data resource and the target information resource, including:
judging whether the target text is a data resource; if yes, setting the target text as the target data resource; if not, performing cross-modal transformation on the target text to obtain the target data resource;
judging whether the target text is an information resource; if yes, setting the target text as the target information resource; if not, performing cross-modal transformation on the target text to obtain the target information resource.
Optionally, querying related resources of the target text according to the target data resource and/or the target information resource, includes:
acquiring a related text of the target text;
and inquiring related resources of the target text from the associated text according to the target data resources and/or the target information resources.
Optionally, obtaining the supplementary resource of the target text includes:
taking the data resource with the association degree with the target data resource larger than a preset value in the data map as a supplementary resource of the target text;
and/or using the information resource with the degree of association with the target information resource in the information map larger than the preset value as the related resource of the target text.
Optionally, after determining that the number of text meanings of the target text is greater than 1, the method further includes:
judging that the target text is a text of missing data resources or information resources;
or, judging that the target text is a text with data resource redundancy or information resource redundancy.
Optionally, determining the text meaning of the target text according to the related resources, including
And respectively combining the related resources with each target data resource and each target information resource to deduce the text meaning of the target text.
The application also provides a system for processing the text ambiguity of the DIKW-crossing mode facing to essential computing and reasoning, which comprises:
the text analysis module is used for acquiring a target text and determining a target data resource and a target information resource in the target text;
the meaning determining module is used for inquiring related resources of the target text according to the target data resources and/or the target information resources and determining the text meaning of the target text according to the related resources;
the resource supplement module is used for acquiring supplement resources of the target text and generating a conditional restriction text of the target text according to the supplement resources if the number of the text meanings of the target text is greater than 1;
and the text modification module is used for taking the text meaning conforming to the conditional restriction text as the actual text meaning of the target text and modifying the target text according to the actual text meaning.
The application also provides a storage medium, on which a computer program is stored, which when executed, realizes the steps executed by the method for processing text ambiguity across DIKW modalities facing essential computation and inference.
The application also provides electronic equipment which comprises a memory and a processor, wherein the memory is stored with a computer program, and the processor calls the computer program in the memory to realize the steps executed by the method for processing the text ambiguity across the DIKW modality facing the essential computing and reasoning.
The application provides a method for processing text ambiguity of a DIKW-crossing mode facing to essential computing and reasoning, which comprises the following steps: acquiring a target text, and determining a target data resource and a target information resource in the target text; inquiring related resources of the target text according to the target data resources and/or the target information resources, and determining the text meaning of the target text according to the related resources; if the number of the text meanings of the target text is larger than 1, acquiring supplementary resources of the target text, and generating a conditional restriction text of the target text according to the supplementary resources; and taking the text meaning conforming to the conditional restriction text as the actual text meaning of the target text, and modifying the target text according to the actual text meaning.
After the target text is obtained, the target data resource and the target information resource contained in the target text are determined, then the related resource of the target text is inquired according to the target data resource and the target information resource, and the text meaning of the target text is determined according to the related resource. The method and the device for correcting the target text have the advantages that the conditional restriction text of the target text is generated according to the supplementary resources of the target text, the text meaning conforming to the conditional restriction text is used as the actual text meaning of the target text, the target text is further modified according to the actual text meaning, and ambiguity in the target text is eliminated. As can be seen, the method and the device can accurately identify and eliminate the ambiguity existing in the text. The application also provides a system for processing the text ambiguity of the DIKW-crossing mode oriented to essential computing and reasoning, a storage medium and electronic equipment, which have the beneficial effects and are not repeated herein.
Drawings
In order to more clearly illustrate the embodiments of the present application, the drawings needed for the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings can be obtained by those skilled in the art without inventive effort.
Fig. 1 is a flowchart of an essential computing and reasoning oriented method for processing text ambiguity across DIKW modalities according to an embodiment of the present application;
fig. 2 is a schematic structural diagram of an essential computing and reasoning oriented cross-DIKW modal text ambiguity processing system according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The following embodiments of the present application may be implemented with a multimodal content disambiguation system based on data graphs, information graphs, and knowledge graphs. The resource elements can be in three forms of data resources, information resources and knowledge resources, the atlas refers to a result of integrating the resource elements, and the atlas of the resource elements comprises a data atlas, an information atlas and a knowledge atlas. The DIKW refers to Data, Information, knowledge, and Wisdom, and the DIKW model is a model that can be used to help understand the relationship between Data, Information, knowledge, and Wisdom.
The data map is a collection of data resources with various data structures including arrays, linked lists, stacks, queues, trees, graphs and the like. Data maps are the basic individual items of numeric or other types of information obtained by observation. The information graph is information conveyed through the context after the data resources and the data resources are combined, and the information is suitable for analysis and interpretation after the concept mapping and the related relation combination. The essence of a knowledge graph is a semantic network, comprising a set of statistical rules summarized by information resources. The knowledge graph contains rich semantic relations, the edge density and the node density of the knowledge graph can be improved through information reasoning and entity linking on the knowledge graph, and the non-structural characteristics of the knowledge graph enable the knowledge graph to be seamlessly linked. For those skilled in the art, the concepts of Data atlas, Information atlas, Knowledge atlas, Data resource, Information resource, Knowledge resource, etc. are all clear and clear, and refer to the introduction in the documents "input-driven transaction efficiency optimization method integrating storage and computation", modeling Data, Information and Knowledge for Security Protection of Hybrid IoT and Edge Resources ", etc.
The embodiment provided by the application can be applied to the field of remote sensing, namely, the text ambiguity processing is realized based on the data map, the information map and the knowledge map related to the field of remote sensing.
Referring to fig. 1, fig. 1 is a flowchart of an essential computing and reasoning-oriented cross-DIKW modal text ambiguity processing method according to an embodiment of the present application.
The specific steps may include:
s101: acquiring a target text, and determining a target data resource and a target information resource in the target text;
after the target text is obtained, the embodiment may perform a data resource extraction operation on the target text to obtain the target data resource, and may also perform an information resource extraction operation on the target text to obtain the target information resource. As a possible implementation manner, the present embodiment may determine the target data resource and the target information resource that conform to the template by using a template that includes data resources and information resources. The embodiment can perform text analysis on the target text, and determine the target data resource and the target information resource according to the text analysis result. In this embodiment, a resource set including sample data resources and sample information resources may also be text-matched with the target text to obtain target data resources and target information resources.
The process of extracting the target data resource and the target information resource from the target text in this embodiment is illustrated as follows:
the target text "night in summer, user a stays in the study. "the target data resource may include(site: study room),(time: night),(season: summer), the information resources may include: i is0(in the study in summer at night);
s102: inquiring related resources of the target text according to the target data resources and/or the target information resources, and determining the text meaning of the target text according to the related resources;
the embodiment can acquire the associated text of the target text, determine the text meaning of the target text based on the data resource and the information resource in the associated text, and make any two text meanings of the target text different from each other, that is, the ambiguity of the target text. The embodiment can use the context of the target text as the associated text, and can also use other texts which are in contact with the target text as the associated text. After obtaining the associated text of the target text, the present embodiment may query, according to the target data resource and/or the target information resource, the relevant resource of the target text from the associated text.
After obtaining the relevant resources of the target text, this embodiment may combine the relevant resources with each target data resource and each target information resource respectively to derive the text meaning of the target text. For example, if the target text is "night, xiao ming in study", and the associated resources determined in the associated text are "spring festival" and "exam", the following two text meanings "night, xiao ming in study, and" night, xiao ming in study "will be obtained. As a possible implementation manner, the present embodiment may display the relevant resource and the target text to the human-computer interaction interface, so that the user can determine the text meaning of the target text.
S103: if the number of the text meanings of the target text is larger than 1, acquiring supplementary resources of the target text, and generating a conditional restriction text of the target text according to the supplementary resources;
if the number of the text meanings of the target text is 1, the target text is free of ambiguity; and if the number of the text meanings of the target text is more than 1, indicating that ambiguity exists in the target text. The ambiguous target text mentioned in this embodiment may be text of missing data resources or information resources, and the lack of content in the text results in reduction of the limitation of the content understanding range. Under the condition of wider understanding range, different information resources with different purposes can be deduced by combining different knowledge resources. By increasing the limit to the content understanding, the content understanding range can be narrowed, so that only one of the deduced information resources with different purposes is reserved, and the ambiguity is eliminated. Modeling is carried out on the content based on the data map, the information map and the knowledge map, and the condition that the content is lost can be divided into two types of information resource loss and data resource loss on the map. The ambiguous target text mentioned in this embodiment may also be text with redundant data resources or redundant information resources. The existence of redundant data or information resources in the text has many different objectives to be understood on one and the same problem. Content is modeled based on the data graph, the information graph, and the knowledge graph. The situation that the content has redundancy can be divided into two types of information resource redundancy and data resource redundancy on the map correspondingly.
The embodiment can query the supplementary resource of the target text from the data map according to the target data resource; the embodiment can also query the supplementary resource of the target text from the information map according to the target information resource.
The data map includes a large number of data resources, and a certain degree of association exists between the data resources in the data map, and this embodiment may query, according to the degree of association of the data resources in the data map, a supplemental resource of the target text, for example, a data resource in the data map whose degree of association with the target data resource is greater than a preset value may be used as the supplemental resource of the target text. For example, the target data resource is "winter", and the data resource in the data map, the association degree of which with the target data resource is greater than the preset value, may include "warm keeping index", "snowfall probability", and the like. Correspondingly, the information map includes a large number of information resources, and a certain degree of association exists between the information resources in the information map, and this embodiment may query, according to the degree of association of the information resources in the information map, a supplemental resource of the target text, for example, an information resource in the information map whose degree of association with the target information resource is greater than the preset value may be used as the supplemental resource of the target text. For example, the target information resource is "einstein is in class", and the information resources in the information graph that are more associated with the target information resource than a preset value may include "einstein is physicist", "einstein is married at that time", and the like.
S104: and taking the text meaning conforming to the conditional restriction text as the actual text meaning of the target text, and modifying the target text according to the actual text meaning.
After obtaining the conditional text of the target text, the present embodiment may use the text meaning that conforms to the conditional text as the actual text meaning of the target text. Continuing with the example of the target information resource "einstein is in class", the target text will have the following two text meanings "1, the career of einstein is a student", "2, the career of einstein is a teacher"; if the conditional restriction text corresponding to the supplementary resource is that "einstein is a physicist and then einstein is married", the actual text meaning of the target text can be determined by using the conditional restriction text. The present embodiment may modify the target text based on the actual text meaning in order to disambiguate the target text. As a possible implementation manner, the present embodiment may display each text meaning of the conditional restriction text and the target text to the human-computer interaction interface, so that the user determines the text meaning that will conform to the conditional restriction text.
In this embodiment, after the target text is acquired, the target data resource and the target information resource included in the target text are determined, and then the relevant resource of the target text is queried according to the target data resource and the target information resource, and the text meaning of the target text is determined according to the relevant resource. In this embodiment, a conditional restriction text of a target text is generated according to a complementary resource of the target text, a text meaning conforming to the conditional restriction text is used as an actual text meaning of the target text, and the target text is modified according to the actual text meaning, so that ambiguity in the target text is eliminated. Therefore, the method and the device can accurately identify and eliminate the ambiguity existing in the text.
As a further introduction to the corresponding embodiment of fig. 1, the target data resource and the target information resource in the target text may be determined by: determining a resource type of the target text; performing cross-modal transformation on the target text to obtain the target data resource and the target information resource; the resource types comprise data resources, information resources and knowledge resources, the data resources are resources in a data map, the information resources are resources in an information map, and the knowledge resources are resources in a knowledge map; the cross-modal transformation is a transformation operation between any two resources of data resources, information resources, knowledge resources and data information mixed resources. The data information mixed resource is a resource in which a data resource and an information resource are mixed.
Specifically, in the process of performing cross-modal transformation, the following operations may be performed: judging whether the target text is a data resource; if yes, setting the target text as the target data resource; if not, performing cross-modal transformation on the target text to obtain the target data resource; judging whether the target text is an information resource; if yes, setting the target text as the target information resource; if not, performing cross-modal transformation on the target text to obtain the target information resource.
The flow described in the above embodiment is explained below by an embodiment in practical use.
Scene 1: and processing ambiguity caused by missing of the information resources.
Text content: at night in summer, user A stays in the study. "may correspond to the following data and information resources:
there is ambiguity in understanding the content because the content lacks the information resource of "what user a is going to do at the study". E.g. in connection with data resources"user stays in study" and knowledge resources K1: "study is the place to learn," it can be deduced that user a is at study for the purpose of learning. To combine data resources"evening" and knowledge resource K2: "a person would normally sleep at night," it can be deduced that user A may be aiming to sleep in the study. Both of these derivations are correct without other related resources, but result in different purpose information resources, leading to ambiguity.
The above is symbolically expressed as follows:
it is known that: k1=RIN(TACTIVITY(Study),TPLACE(Studyroom))
K2=RAT(TACTIVITY(Sleep),TTIME(Night))
The derivation can be made:
the embodiment can narrow the content understanding range by increasing related data resources or information resources, thereby eliminating ambiguity. The following discussion is directed to the case of adding data resources and adding information resources, respectively.
Mode a 1: increasing data resources
If the related data resource D is known1: the air-conditioning of the bedroom is bad; d2: the air conditioning of the study is good. Incorporating known data resources D03: summer and knowledge resources K3: "very hot summer" can deduce that the bedroom temperature is high and the study room temperature is low. The above-described data resources increase the constraints on the study environment, thereby narrowing the scope of content understanding to the temperature-related domain. Then combines with knowledge resources K4: the fact that people like to sleep in a cool place can be deduced that the study room is low in temperature and suitable for sleeping, and the previously deduced information resource that the purpose of the user A staying in the study room is sleeping is supported, so that ambiguity is eliminated.
The above is symbolically expressed as follows:
it is known that: d1=(TFACILITY(INS(AIR_CONDITIONBedroom))|TCONDITION(Broken))
D2=(TFACILITY(INS(AIR_CONDITIONStudyroom))|TCONDITION(Normal))
K3=RIS(TSEASON(Summer),TTEMPERATURE(High))
K4=RLIKE(TPERSON,RIN(TACTIVITY(Sleep),RIS(TPLACE,TTEMPERATURE(Low))))
The derivation can be made:
the algorithm implementation of the above mode a1 is as follows:
from known data resources D0Information resource I0Combine the related knowledge resources (i.e. the related resources mentioned above) to derive the information resources I of different purposesnew1And Inew2。
Retrieving relevant data resources D in a data maprelated(i.e., the supplemental resources mentioned above).
From DrelatedThe information resource I capable of reducing the understanding range is further deduced by combining the related information resource and the knowledge resourcenew3。
Judgment of Inew3And Inew1、Inew2The relation between, retention Inew3And the supported information resources delete other information resources.
And setting the residual unique information resources as final results to eliminate ambiguity.
Mode a 2: increasing information resources
If the related information resources I are known1: user a dislikes learning. Binding D02: "evening", and knowledge resource K5: "talent liking learning may learn at night," it can be deduced that user a is unlikely to learn at the study at this time. The information resource excludes the understanding range of the content from the information resource of ' the user A is in the study and aims at learning ', and the only information resource left is that the user A is in the study and aims at sleeping ' which is the final result, so that the ambiguity is eliminated.
The above is symbolically expressed as follows:
it is known that: i is1=!RLIKE(A,TACTIVITY(Study))
K5=RAT(RDO(TPERSON(RLIKE(TPERSON,TACTIVITY(Study))),TACTIVITY(Study)),TTIME(Night))
The derivation can be made:
the algorithm implementation of the above mode a2 is as follows:
from known data resources D0Information resource I0Combine the related knowledge resources (i.e. the related resources mentioned above) to derive the information resources I of different purposesnew1And Inew2。
Retrieving related information resources I in an information graphrelated(i.e., the supplemental resources mentioned above).
From IrelatedThe information resource I capable of reducing the understanding range is further deduced by combining the related data resource and the knowledge resourcenew3。
Judgment of Inew3And Inew1、Inew2Relation between, delete Inew3The objectionable information resources reserve other information resources.
And setting the residual unique information resources as final results to eliminate ambiguity.
Scene 2: and processing ambiguity caused by missing of the information resources.
Text content: "user A's ancestor is larger than user B's ancestor. "may correspond to the following data and information resources:
since the content lacks a data resource related to "age of user a and age of user B", there is a different purpose of understanding the information resource "age size of user a and user B" based on this content. Although there is a knowledge resource K _ 1: "high ancestry is likely to be older" and an information resource "user A is likely to be older than user B" can be deduced. But there are many examples of high and low age generations, so the information resource "user a may be older than user B" still cannot be excluded.
The above is symbolically expressed as follows:
it is known that: k1=RPROBABLY_GREATER_THAN(TAGE(TPERSON(TSENIORITY(High)),TAGE(TPERSON(TSENIORITY(Low)) can be derived:
it is also possible to narrow the content understanding by adding related data or information resources, thereby disambiguating. The following discussion is directed to the case of adding data resources and adding information resources, respectively.
Mode B1: increasing data resources
If the related data resource D is known1: the user A has mature mind; d2: user B is mentally savory. Binding of D1、D2An information resource "user a is more mature than user B" can be deduced. The data resource increases the limit to the mental maturity relationship of the user A and the user B when judging the age of the user A and the user B, and further expands the understanding range of the contentAnd (5) shrinking. The previously derived information resource "user a is older than user B" is supported.
The above is symbolically expressed as follows:
it is known that: d1=(A|TMIND(Mature))
D2=(B|TMIND(Naieve))
The derivation can be made:
the algorithm implementation of the above mode B1 is as follows:
from known data resources D0Information resource I0Combines related knowledge resources to deduce information resources I of different purposesnew1And Inew2。
Retrieving relevant data resources D in a data maprelated。
From DrelatedThe information resource I capable of reducing the understanding range is further deduced by combining the related information resource and the knowledge resourcenew3。
Judgment of Inew3And Inew1、Inew2Relation between, delete Inew3The objectionable information resources reserve other information resources.
And setting the residual unique information resources as final results to eliminate ambiguity.
Mode B2: increasing information resources
If the related information resources I are known1: user B honors user a's attitude. Having knowledge resources K3: "the low-position person honors the high-position attitude". Binding of I1And K3The information resource "user a is more dominant than user B" can be deduced. The above data resources are increased in judging "age size of user a and user B"In this case, the restriction on the status relationship between the user a and the user B further narrows the understanding range of the content. The previously derived information resource "user a is older than user B" is supported.
The above is symbolically expressed as follows:
it is known that: i is1=RRESPECT(B,A)
K3=RRESPECT(TPERSON(TSTATUS(Low)),TPERSON(TSTATUS(High)))
The derivation can be made:
the algorithm implementation of the above mode B2 is as follows:
from known data resources D0Information resource I0Combine the related knowledge resources (i.e. the related resources mentioned above) to derive the information resources I of different purposesnew1And Inew2。
Retrieving related information resources I in an information graphrelated(i.e., the supplemental resources mentioned above).
From IrelatedThe information resource I capable of reducing the understanding range is further deduced by combining the related data resource and the knowledge resourcenew3。
Judgment of Inew3And Inew1、Inew2Relation between, delete Inew3The objectionable information resources reserve other information resources.
And setting the residual unique information resources as final results to eliminate ambiguity.
Scene 3: there is a process of ambiguity caused by redundancy for information resources.
Text content: "user A likes playing basketball, user A dislikes sports. "may correspond to the following information resources:
having knowledge resources K1: playing basketball belongs to sports; k2: the relationship "annoying" contradicts the relationship "like". From I02And K1If the user A disagrees the sports and the basketball playing belongs to one of the sports, a new information resource I can be deducednew1: user A hates playing basketball, by K2Knowing I01 and Inew1Are in contradiction. So for the question of "attitude of user A playing basketball", I01 and I02There is a different purpose to understand that there is redundancy in the information resources in the content.
The above symbolized representation is as follows:
it is known that: k1=RBELONGTO(TACTIVITY(PlayBasketball),TACTIVITY(Sports))
K2=ROPPOSE(TRELATION(Like),TRELATION(Hate))
The derivation can be made:
from the above derivation: redundant information resourcesAndare contradictory, so one must have an error. Disambiguation can be achieved by adding related data resources or information resources to help determine whether redundant information resources are correct or incorrect. The following discussion is directed to the case of adding data resources and adding information resources, respectively.
Mode C1: increasing data resources
If the spatial data resource D related to the user A is known1: a basketball court. Has related knowledge resources K3: the main purpose of the basketball court is to play basketball; k4: people who play basketball often like playing basketball. Binding of D1And K3User a is often present at a basketball court, so user a is often basketball. Recombination of K4The user A plays basketball frequently, and people who play basketball frequently probably like playing basketball, which indicates that the user A probably likes playing basketball, supports information resourcesOn information resourcesWith mutually supported data, but information resourcesWithout the data supported, it is prone to decideIs right and correctError and thus disambiguation.
The above symbolized representation is as follows:
known as D1=(A|TPLACE(INS(BasketballCourt))
K3=RIN(TACTIVITY(PlayBasketball),TPLACE(BasketballCourt))
K4=RLIKE(TPERSON(RDO(person,TACTIVITY(PlayBasketball))),TACTIVITY(PlayBasketball))
The derivation can be made:
the algorithm implementation of the above mode C1 is as follows:
retrieving relevant data resources D in a data maprelated。
From DrelatedThe information resource I which helps to judge the correctness is further deduced by combining the related information resource and the knowledge resourcenew。
Judgment of InewAndthe relation between, retention InewSupported results, and delete another result.
Will InewAnd setting the supported result as a final result to eliminate ambiguity.
Mode C2: increasing information resources
If the related information resources I are known1: user a is a member of a basketball team. Has related knowledge resources K5: members of a basketball team often play basketball. Binding of I1And K5User A is a member of the basketball team, so user A plays basketball often. User A plays basketball frequently and then combines K4The user A plays basketball frequently, and people who play basketball frequently probably like playing basketball, which indicates that the user A probably likes playing basketball, supports information resourcesOn information resourcesWith mutually supported data, but information resourcesWithout the data supported, it is prone to decideIs right and correctError and thus disambiguation.
The above symbolized representation is as follows:
it is known that: i is1=RIS_A_MEMBER_OF(A,TGROUP(INS(BasketballTeam))
K5=RDO(TPERSON(RIS_A_MEMBER_OF(person,TGROUP(BasketballTeam)),TACTIVITY(PlayBasketball))
The derivation can be made:
the algorithm implementation of the above mode C2 is as follows:
retrieval of related data resources I in information mapsrelated。
From IrelatedThe information resource I which helps to judge the correctness is further deduced by combining the related data resource and the knowledge resourcenew。
Judgment of InewAndthe relation between, retention InewSupported results, and delete another result.
Will InewAnd setting the supported result as a final result to eliminate ambiguity.
Scene 3: there is a handling of ambiguities caused by redundancy for data resources.
Simultaneous presence of data resources in contentToday the temperature is 30 degrees;today the temperature is 20 degrees. May correspond to the following data resources:
for the temperature of today "This problem, data resourcesAndthe expressed contents are contradictory, and redundant data resources are explainedAndone of which must have an error. Disambiguation can be achieved by adding related data resources or information resources to help determine whether redundant data resources are correct or incorrect. The following discussion is directed to the case of adding data resources and adding information resources, respectively.
Mode D1: increasing data resources
If data resource D is known1: summer season; d2: in the south of the sea. Having knowledge resources K1: the temperature in the summer of Hainan is higher. Joining data resources D1、D2And knowledge resource K1It can be deduced that today's air temperature should be high. Support data resourcesIn data resourcesWith mutually supported data, and data resourcesWithout the data supported, it is prone to decideIs right and correctError and thus disambiguation.
The above symbolized representation is as follows:
it is known that: d1=(TSEASON(Summer))
D2=(TPLACE(Hainan))
K1=RIS(RIN(TPLACE(Hainan),TSEASON(Summer)),TTEMPERATURE(High))
The derivation can be made:
the algorithm implementation of the above mode D1 is as follows:
retrieving relevant data resources D in a data maprelated。
From DrelatedThe data resource D which helps to judge the correctness is further deduced by combining the related information resource and the knowledge resourcenew。
Judgment of DnewAndthe relation between, retention DnewSupported results, and delete another result.
Will DnewAnd setting the supported result as a final result to eliminate ambiguity.
Mode D2: increasing information resources
If the information resource I is known1: data resourcesFrom the meteorological office; information resource I2: data resourcesOriginating from the network. Having knowledge resources K2: data originating from professional organizations is more reliable than data originating from networks. In connection with information resources I1,I2And knowledge resources K2From which data resources can be derivedComparing data resourcesTo be more reliable. From which it can be determinedIs right and correctError and thus disambiguation.
The above symbolized representation is as follows:
it is known that: i is1=RFROM(D01,TINSTITUTE(INS(MeteorologicalBureau))
I2=RFROM(D02,TINTERNET(INS(Website))
K2=RRELIABLE_THAN(TDATA(RFROM(data,TINSTITUTE)),TDATA(RFROM(data,TINTERNET)))
The derivation can be made:
the algorithm implementation of the above mode D2 is as follows:
retrieving related information resources I in an information graphrelated。
From IrelatedCombining related data resources, further deducing information resources I for helping to judge correctnessnew。
Judgment of InewAndthe relation between, retention InewSupported results, and delete another result.
Will InewAnd setting the supported result as a final result to eliminate ambiguity.
Whether the detection of the ambiguity phenomenon or the addition of the related type resource for disambiguation needs to complete the cross-modal transformation from the original type resource to the new type resource. The type resource as the conversion object can be mainly divided into two types, namely, a data resource and an information resource, and the following discussion is made on two cases, namely, the conversion object is a data resource and the conversion object is an information resource.
Modality conversion case 1:
if the conversion object is a data resource: "occupation of user A". Symbolized as follows:
D0=(A|TOCCUPATION(INS(Student))
there are three ways in which D can be derived0The method comprises the following steps: the derivation is performed by combining data resources with knowledge resources, the derivation is performed by combining information resources with knowledge resources, and the derivation is performed by combining data resources with information resources with knowledge resources. The following are directed to these threeThe derivation modes are discussed separately.
The derivation process of the data resource and the knowledge resource is as follows:
if there is a related data resource D1: user a is 10 years old this year. Having related knowledge resources K1: people under the age of 15 should go to school. Binding of D1And K1: user a is 10 years old this year and his age is less than 15 years old, so user a should go to school. The target data resource of "the occupation of the user a is a student" can be further deduced.
The above symbolized representation is as follows:
known as D1=(A|TAGE(10))
K1=RSHOULD(TPERSON(RLESS THAN(TAGE,15)),TACTIVITY(Education))
The derivation can be made:
I0→D0=(A|TOCCUPATION(INS(Student))
the derivation process of the information resource and the knowledge resource is as follows:
if there is related information resource I1: user A often goes to school; i is2: user a has no teacher qualifications. Having knowledge resources K2: students and teachers need to go to school frequently; k3: the teacher has teacher qualification certificate. Binding of I1And K2: user a often goes to school, so user a is a student or a teacher. Binding of I2And K3: user a does not have a teacher qualification, so user a is not a teacher. The user A is a student or a teacher, and the user A is not a teacher, so that the target number of the user A who careers for students can be further deducedAccording to the resource.
The above symbolized representation is as follows:
is known as I1=RGO_TO(A,TPLACE(INS(School)))
I2=!ROWN(A,TLICENCE(INS(TeacherCertification)))
K2=RGO_TO(TOCCUPATION(Student)AND TOCCUPATION(Teacher),TPLACE(School))
K3=ROWN(TOCCUPATION(Teacher),TLICENCE(INS(TeacherCertification))
The derivation can be made:
I0→D0=(A|TOCCUPATION(INS(Student))
the derivation process of the data resource mixed information resource combined with the knowledge resource is as follows:
if there is a related data resource D1: user A is 10 years old this year; related information resources I1: user a often goes to school. Having knowledge resources K2: students and teachers need to go to school frequently; k4: the teacher is typically over 20 years old. Binding of I1And K2: user a often goes to school, so user a is a student or a teacher. Binding of D1And K2: user a is 10 years old this year, and the teacher is typically older than 20 years old, so user a is not a teacher. User A is a student or a teacher, and user A is not a teacher, and can further deduce "The profession of user a is a student "this target data resource.
The above symbolized representation is as follows:
known as D1=(A|TAGE(10))
I1=RGO_TO(A,TPLACE(INS(School)))
K2=RGO_TO(TOCCUPATION(Student)AND TOCCUPATION(Teacher),TPLACE(School))
K4=RGREATER_THAN(TAGE(TOCCUPATION(Teacher)),20)
The derivation can be made:
I0→D0=(A|TOCCUPATION(INS(Student))
modality conversion case 2:
if the conversion object is an information resource: "user A likes playing football". Symbolized as follows:
$$I0=R_{LIKE}(A,\T_{ACTIVITY}(INS(Play Soccer))\\$$
there are three derivations I0The method comprises the following steps: the derivation is performed by combining data resources with knowledge resources, the derivation is performed by combining information resources with knowledge resources, and the derivation is performed by combining data resources with information resources with knowledge resources. The following discussion is separately directed to these three derivation modes.
The derivation process of the data resource and the knowledge resource is as follows:
if there is a spatial data resource D related to the user A1: a football field. Has related knowledge resources K1: the football field is mainly used for playing football; k2: people who play football frequently like playing football. Binding of D1And K1User a is often present at a football pitch, so user a plays football often. Recombination of K2The user A plays football frequently, and people who play football frequently probably like playing football, so that the target information resource of 'the user A likes playing football' can be further deduced.
The above symbolized representation is as follows:
known as D1=(A|TPLACE(INS(SoccerCourt))
K1=RIN(TACTIVITY(PlaySoccer),TPLACE(SoccerCourt))
K2=RLIKE(TPERSON(RDO(person,TACTIVITY(PlaySoccer))),TACTIVITY(PlaySoccer))
The derivation can be made:
the derivation process of the information resource and the knowledge resource is as follows:
if there is information resource I1: user a is a member of a football team. Has related knowledge resources K2: people who often play football like playing football; k3: members of a football team often play football. Binding of I1And K3User a is a member of the football team, so user a plays football frequently. Recombination of K2The user A plays football frequently, and people who play football frequently probably like playing football, so that the target information resource of 'the user A likes playing football' can be further deduced.
The above symbolized representation is as follows:
it is known that: i is1=RIS_A_MEMBER_OF(A,TGROUP(INS(SoccerTeam))
K2=RLIKE(TPERSON(RDO(person,TACTIVITY(PlaySoccer))),T(PlaySoccer))
K3=RDO(TPERSON(RIS_A_MEMBER_OF(person,TGROUP(SoccerTeam)),TACTIVITY(PlaySoccer)) can be derived:
the derivation process of the data resource mixed information resource combined with the knowledge resource is as follows:
if there is a reading data resource D related to the user A2: football news; and information resource I2: user a likes sports. Having knowledge resources K4: people who often watch football news are interested in football sporting events; k5: sports include playing football, basketball, and the like. Binding of D2And K4User a reads football news often, so user a is interested in football events. Since the interest of the user a in the soccer may only stay on watching the soccer game, the information "the user a is interested in the soccer game" cannot directly deduce that the user a likes playing the soccer. Binding of I2And K5User a likes sports, which includes kicking football. This information is also not sufficient to directly deduce that user a likes playing football, because user a may be more interested in playing sports such as basketball. However, since the fact that the user A is interested in the football game is deduced in the past, the target information resource that the user A likes playing football can be deduced by combining the information that the user A likes playing sports.
The above symbolized representation is as follows:
known as D2=(A|TNEWS(Soccer))
I2=RLIKE(A|TACTIVITY(INS(SportsActivity)))
K4=RINTERESTED_IN(TPERSON(RREAD(person,TNEWS(Soccer))),TSPORTS(Soccer))
K5=RINCLUDE(TACTIVITY(SportsActivity),TACTIVITY(PlaySoccer,PlayBasketball,...))
The derivation can be made:
referring to fig. 2, fig. 2 is a schematic structural diagram of an essential computing and reasoning oriented cross-DIKW modal text ambiguity processing system according to an embodiment of the present application;
the system may include:
the text analysis module 100 is configured to obtain a target text and determine a target data resource and a target information resource in the target text;
a meaning determining module 200, configured to query, according to the target data resource and/or the target information resource, a relevant resource of the target text, and determine a text meaning of the target text according to the relevant resource;
a resource supplement module 300, configured to obtain a supplement resource of the target text and generate a conditional restriction text of the target text according to the supplement resource if the number of text meanings of the target text is greater than 1;
and the text modification module 400 is used for taking the text meaning conforming to the conditional restriction text as the actual text meaning of the target text and modifying the target text according to the actual text meaning.
In this embodiment, after the target text is acquired, the target data resource and the target information resource included in the target text are determined, and then the relevant resource of the target text is queried according to the target data resource and the target information resource, and the text meaning of the target text is determined according to the relevant resource. In this embodiment, a conditional restriction text of a target text is generated according to a complementary resource of the target text, a text meaning conforming to the conditional restriction text is used as an actual text meaning of the target text, and the target text is modified according to the actual text meaning, so that ambiguity in the target text is eliminated. Therefore, the method and the device can accurately identify and eliminate the ambiguity existing in the text.
Further, the text analysis module 100 includes:
the type determining unit is used for determining the resource type of the target text; the resource types comprise data resources, information resources and knowledge resources, the data resources are resources in a data map, the information resources are resources in an information map, and the knowledge resources are resources in a knowledge map;
the modal conversion unit is used for performing cross-modal conversion on the target text to obtain the target data resource and the target information resource; and converting the cross-modal transformation into the transformation operation between any two resources of the data resource, the information resource, the knowledge resource and the data information mixed resource.
Further, the modal transformation unit is used for judging whether the target text is a data resource; if yes, setting the target text as the target data resource; if not, performing cross-modal transformation on the target text to obtain the target data resource; the system is also used for judging whether the target text is an information resource; if yes, setting the target text as the target information resource; if not, performing cross-modal transformation on the target text to obtain the target information resource.
Further, the meaning determining module 200 is configured to obtain a text associated with the target text; and the system is also used for inquiring related resources of the target text from the associated text according to the target data resources and/or the target information resources.
Further, the resource supplementing module 300 is configured to use the data resource in the data map, of which the association degree with the target data resource is greater than a preset value, as a supplementing resource of the target text; and/or, the information resources in the information map, the association degree of which with the target information resources is greater than the preset value, are used as the related resources of the target text.
Further, the method also comprises the following steps:
the text type judging module is used for judging that the target text is a text lacking data resources or information resources after judging that the number of the text meanings of the target text is greater than 1; or, judging that the target text is a text with data resource redundancy or information resource redundancy.
Further, the meaning determining module 200 is configured to combine the related resource with each of the target data resources and each of the target information resources to derive a text meaning of the target text.
Since the embodiment of the system part corresponds to the embodiment of the method part, the embodiment of the system part is described with reference to the embodiment of the method part, and is not repeated here.
The present application also provides a storage medium having a computer program stored thereon, which when executed, may implement the steps provided by the above-described embodiments. The storage medium may include: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The application further provides an electronic device, which may include a memory and a processor, where the memory stores a computer program, and the processor may implement the steps provided by the foregoing embodiments when calling the computer program in the memory. Of course, the electronic device may also include various network interfaces, power supplies, and the like.
The embodiments are described in a progressive manner in the specification, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description. It should be noted that, for those skilled in the art, it is possible to make several improvements and modifications to the present application without departing from the principle of the present application, and such improvements and modifications also fall within the scope of the claims of the present application.
It is further noted that, in the present specification, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Claims (10)
1. An essential computing and reasoning oriented cross DIKW modal text ambiguity processing method is characterized by comprising the following steps:
acquiring a target text, and determining a target data resource and a target information resource in the target text;
inquiring related resources of the target text according to the target data resources and/or the target information resources, and determining the text meaning of the target text according to the related resources;
if the number of the text meanings of the target text is larger than 1, acquiring supplementary resources of the target text, and generating a conditional restriction text of the target text according to the supplementary resources;
and taking the text meaning conforming to the conditional restriction text as the actual text meaning of the target text, and modifying the target text according to the actual text meaning.
2. The essential computing and reasoning oriented cross-DIKW modality text ambiguity processing method of claim 1, wherein determining the target data resources and target information resources in the target text comprises:
determining a resource type of the target text; the resource types comprise data resources, information resources and knowledge resources, the data resources are resources in a data map, the information resources are resources in an information map, and the knowledge resources are resources in a knowledge map;
performing cross-modal transformation on the target text to obtain the target data resource and the target information resource; and converting the cross-modal transformation into the transformation operation between any two resources of the data resource, the information resource, the knowledge resource and the data information mixed resource.
3. The essential computing and reasoning oriented cross-DIKW modality text ambiguity processing method of claim 2, wherein the cross-modality conversion is performed on the target text to obtain the target data resource and the target information resource, and comprises:
judging whether the target text is a data resource; if yes, setting the target text as the target data resource; if not, performing cross-modal transformation on the target text to obtain the target data resource;
judging whether the target text is an information resource; if yes, setting the target text as the target information resource; if not, performing cross-modal transformation on the target text to obtain the target information resource.
4. The essential computing and reasoning oriented cross-DIKW modality text ambiguity processing method of claim 1, wherein querying relevant resources of the target text according to the target data resources and/or the target information resources comprises:
acquiring a related text of the target text;
and inquiring related resources of the target text from the associated text according to the target data resources and/or the target information resources.
5. The essential computing and reasoning oriented cross-DIKW modality text ambiguity processing method of claim 1, wherein obtaining the supplementary resources of the target text comprises:
taking the data resource with the association degree with the target data resource larger than a preset value in the data map as a supplementary resource of the target text;
and/or using the information resource with the degree of association with the target information resource in the information map larger than the preset value as the related resource of the target text.
6. The essential computing and reasoning oriented cross-DIKW modality text ambiguity handling method of claim 1, further comprising, after determining that the number of text meanings of the target text is greater than 1:
judging that the target text is a text of missing data resources or information resources;
or, judging that the target text is a text with data resource redundancy or information resource redundancy.
7. The essential computing and reasoning oriented cross-DIKW modality text ambiguity processing method of any of claims 1 to 6, wherein determining the text meaning of the target text according to the relevant resources comprises:
and respectively combining the related resources with each target data resource and each target information resource to deduce the text meaning of the target text.
8. An essential computing and reasoning oriented cross-DIKW modal text ambiguity processing system, comprising:
the text analysis module is used for acquiring a target text and determining a target data resource and a target information resource in the target text;
the meaning determining module is used for inquiring related resources of the target text according to the target data resources and/or the target information resources and determining the text meaning of the target text according to the related resources;
the resource supplement module is used for acquiring supplement resources of the target text and generating a conditional restriction text of the target text according to the supplement resources if the number of the text meanings of the target text is greater than 1;
and the text modification module is used for taking the text meaning conforming to the conditional restriction text as the actual text meaning of the target text and modifying the target text according to the actual text meaning.
9. An electronic device comprising a memory having a computer program stored therein and a processor, the processor implementing the steps of the essential computing and reasoning oriented cross-DIKW modality text ambiguity processing method according to any one of claims 1 to 7 when calling the computer program in the memory.
10. A storage medium having stored thereon computer-executable instructions which, when loaded and executed by a processor, carry out the steps of the essential computing and reasoning oriented cross-DIKW modal text disambiguation method of any of the preceding claims 1 to 7.
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011103480.6A CN112232085B (en) | 2020-10-15 | 2020-10-15 | Cross-DIKW modal text ambiguity processing method oriented to essential computing and reasoning |
CA3136527A CA3136527C (en) | 2020-10-15 | 2021-09-14 | Cross-dikw-mode ambiguity processing method oriented to essential computing and reasoning |
PCT/CN2021/118178 WO2022078145A1 (en) | 2020-10-15 | 2021-09-14 | Cross-dikw modal text ambiguity processing method for essential calculation and reasoning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011103480.6A CN112232085B (en) | 2020-10-15 | 2020-10-15 | Cross-DIKW modal text ambiguity processing method oriented to essential computing and reasoning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112232085A true CN112232085A (en) | 2021-01-15 |
CN112232085B CN112232085B (en) | 2021-10-08 |
Family
ID=74117326
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011103480.6A Active CN112232085B (en) | 2020-10-15 | 2020-10-15 | Cross-DIKW modal text ambiguity processing method oriented to essential computing and reasoning |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN112232085B (en) |
WO (1) | WO2022078145A1 (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113538179A (en) * | 2021-06-11 | 2021-10-22 | 海南大学 | DIKW-based intelligent patent application method and system |
CN113810480A (en) * | 2021-09-03 | 2021-12-17 | 海南大学 | Emotion communication method based on DIKW content object |
CN114039865A (en) * | 2021-08-30 | 2022-02-11 | 海南大学 | Intention calculation oriented cross-DIKW modal transmission and optimization system |
WO2022078145A1 (en) * | 2020-10-15 | 2022-04-21 | 海南大学 | Cross-dikw modal text ambiguity processing method for essential calculation and reasoning |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080208864A1 (en) * | 2007-02-26 | 2008-08-28 | Microsoft Corporation | Automatic disambiguation based on a reference resource |
US20140316768A1 (en) * | 2012-12-14 | 2014-10-23 | Pramod Khandekar | Systems and methods for natural language processing |
CN106997399A (en) * | 2017-05-24 | 2017-08-01 | 海南大学 | A kind of classification question answering system design method that framework is associated based on data collection of illustrative plates, Information Atlas, knowledge mapping and wisdom collection of illustrative plates |
CN110633366A (en) * | 2019-07-31 | 2019-12-31 | 国家计算机网络与信息安全管理中心 | Short text classification method, device and storage medium |
CN110704641A (en) * | 2019-10-11 | 2020-01-17 | 零犀(北京)科技有限公司 | Ten-thousand-level intention classification method and device, storage medium and electronic equipment |
US20200073996A1 (en) * | 2018-08-28 | 2020-03-05 | Stitched.IO Limited | Methods and Systems for Domain-Specific Disambiguation of Acronyms or Homonyms |
CN110969022A (en) * | 2018-09-29 | 2020-04-07 | 北京国双科技有限公司 | Semantic determination method and related equipment |
CN111368548A (en) * | 2018-12-07 | 2020-07-03 | 北京京东尚科信息技术有限公司 | Semantic recognition method and device, electronic equipment and computer-readable storage medium |
CN111538844A (en) * | 2020-03-20 | 2020-08-14 | 华为技术有限公司 | Target field knowledge base generation and problem solution method and device |
CN111723188A (en) * | 2020-06-23 | 2020-09-29 | 宁波富万信息科技有限公司 | Sentence display method and electronic equipment based on artificial intelligence for question-answering system |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105630763B (en) * | 2014-10-31 | 2019-08-02 | 国际商业机器公司 | For referring to the method and system of the disambiguation in detection |
CN108280061B (en) * | 2018-01-17 | 2021-10-26 | 北京百度网讯科技有限公司 | Text processing method and device based on ambiguous entity words |
CN111651570A (en) * | 2020-05-13 | 2020-09-11 | 深圳追一科技有限公司 | Text sentence processing method and device, electronic equipment and storage medium |
CN112232085B (en) * | 2020-10-15 | 2021-10-08 | 海南大学 | Cross-DIKW modal text ambiguity processing method oriented to essential computing and reasoning |
-
2020
- 2020-10-15 CN CN202011103480.6A patent/CN112232085B/en active Active
-
2021
- 2021-09-14 WO PCT/CN2021/118178 patent/WO2022078145A1/en active Application Filing
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080208864A1 (en) * | 2007-02-26 | 2008-08-28 | Microsoft Corporation | Automatic disambiguation based on a reference resource |
US20140316768A1 (en) * | 2012-12-14 | 2014-10-23 | Pramod Khandekar | Systems and methods for natural language processing |
CN106997399A (en) * | 2017-05-24 | 2017-08-01 | 海南大学 | A kind of classification question answering system design method that framework is associated based on data collection of illustrative plates, Information Atlas, knowledge mapping and wisdom collection of illustrative plates |
US20200073996A1 (en) * | 2018-08-28 | 2020-03-05 | Stitched.IO Limited | Methods and Systems for Domain-Specific Disambiguation of Acronyms or Homonyms |
CN110969022A (en) * | 2018-09-29 | 2020-04-07 | 北京国双科技有限公司 | Semantic determination method and related equipment |
CN111368548A (en) * | 2018-12-07 | 2020-07-03 | 北京京东尚科信息技术有限公司 | Semantic recognition method and device, electronic equipment and computer-readable storage medium |
CN110633366A (en) * | 2019-07-31 | 2019-12-31 | 国家计算机网络与信息安全管理中心 | Short text classification method, device and storage medium |
CN110704641A (en) * | 2019-10-11 | 2020-01-17 | 零犀(北京)科技有限公司 | Ten-thousand-level intention classification method and device, storage medium and electronic equipment |
CN111538844A (en) * | 2020-03-20 | 2020-08-14 | 华为技术有限公司 | Target field knowledge base generation and problem solution method and device |
CN111723188A (en) * | 2020-06-23 | 2020-09-29 | 宁波富万信息科技有限公司 | Sentence display method and electronic equipment based on artificial intelligence for question-answering system |
Non-Patent Citations (2)
Title |
---|
YUCONG DUAN: "Existence Computation Revelation on Entity vs Relationship for Relationship Defined Everything of Semantics", 《2019 20TH IEEE/ACIS INTERNATIONAL CONFERENCE ON SOFTWARE ENGINEERING, ARTIFICIAL INTELLIGENCE, NETWORKING AND PARALLEL/DISTRIBUTED COMPUTING (SNPD)》 * |
邵礼旭 等: "数据、信息和知识三层图谱架构的推荐服务设计", 《计算机科学与探索》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2022078145A1 (en) * | 2020-10-15 | 2022-04-21 | 海南大学 | Cross-dikw modal text ambiguity processing method for essential calculation and reasoning |
CN113538179A (en) * | 2021-06-11 | 2021-10-22 | 海南大学 | DIKW-based intelligent patent application method and system |
CN114039865A (en) * | 2021-08-30 | 2022-02-11 | 海南大学 | Intention calculation oriented cross-DIKW modal transmission and optimization system |
CN113810480A (en) * | 2021-09-03 | 2021-12-17 | 海南大学 | Emotion communication method based on DIKW content object |
Also Published As
Publication number | Publication date |
---|---|
CN112232085B (en) | 2021-10-08 |
WO2022078145A1 (en) | 2022-04-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112232085B (en) | Cross-DIKW modal text ambiguity processing method oriented to essential computing and reasoning | |
CN103258000B (en) | Method and device for clustering high-frequency keywords in webpages | |
CN112819023B (en) | Sample set acquisition method, device, computer equipment and storage medium | |
Venkataramani et al. | Discovery of technical expertise from open source code repositories | |
Stamatatos et al. | Plagiarism and authorship analysis: introduction to the special issue | |
CN108304381B (en) | Entity edge establishing method, device and equipment based on artificial intelligence and storage medium | |
CN112559895A (en) | Data processing method and device, electronic equipment and storage medium | |
CN111369294B (en) | Software cost estimation method and device | |
KR20140015653A (en) | Contents recommendation system and contents recommendation method | |
CN110516164A (en) | A kind of information recommendation method, device, equipment and storage medium | |
CN111858962B (en) | Data processing method, device and computer readable storage medium | |
Faisal et al. | A novel framework for social web forums’ thread ranking based on semantics and post quality features | |
CN117077679A (en) | Named entity recognition method and device | |
CN103377381A (en) | Method and device for identifying content attribute of image | |
CN117272999A (en) | Model training method and device based on class incremental learning, equipment and storage medium | |
Hapazari | Challenges faced by unemployed people in Lesotho: A case of Manonyane rural community | |
CN112445895B (en) | Method and system for identifying user search scene | |
CN114328895A (en) | News abstract generation method and device and computer equipment | |
CN115878882A (en) | Hierarchical representation learning of user interests | |
Tamilin et al. | Context-driven semantic enrichment of italian news archive | |
CA3136527C (en) | Cross-dikw-mode ambiguity processing method oriented to essential computing and reasoning | |
CN112015982A (en) | Customized news generation method, device, equipment and storage medium | |
US20190304040A1 (en) | System and Method for Vetting Potential Jurors | |
CN116127086B (en) | Geographical science data demand analysis method and device based on scientific and technological literature resources | |
CN116306622B (en) | AIGC comment system for improving public opinion atmosphere |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |