CN115659929B - Annotating interaction method and system based on multiple documents - Google Patents
Annotating interaction method and system based on multiple documents Download PDFInfo
- Publication number
- CN115659929B CN115659929B CN202211301753.7A CN202211301753A CN115659929B CN 115659929 B CN115659929 B CN 115659929B CN 202211301753 A CN202211301753 A CN 202211301753A CN 115659929 B CN115659929 B CN 115659929B
- Authority
- CN
- China
- Prior art keywords
- annotation
- information
- user
- port
- document
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000003993 interaction Effects 0.000 title claims abstract description 20
- 238000004458 analytical method Methods 0.000 claims abstract description 49
- 238000012216 screening Methods 0.000 claims abstract description 12
- 238000006243 chemical reaction Methods 0.000 claims description 23
- 238000010801 machine learning Methods 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 11
- 238000002372 labelling Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 7
- 230000002452 interceptive effect Effects 0.000 claims 1
- 230000010365 information processing Effects 0.000 abstract description 3
- 238000012552 review Methods 0.000 description 13
- 238000003672 processing method Methods 0.000 description 4
- 238000012795 verification Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000012553 document review Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 2
- 238000012098 association analyses Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000002224 dissection Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Document Processing Apparatus (AREA)
Abstract
The application provides an annotation interaction method and system based on multiple documents, which relate to the technical field of computer information processing, and are used for acquiring the annotation information of the multiple documents, extracting the source of the annotation user and generating a label for annotation, screening the annotation information of the multiple documents based on port user information, acquiring the annotation information to be consulted, inputting an annotation record set of the port user into an annotation analysis model, acquiring the annotation characteristics of the port user, further identifying and converting the annotation information to be consulted, and displaying the annotation information through the display port information.
Description
Technical Field
The application relates to the technical field of computer information processing, in particular to a multi-document-based annotation interaction method and system.
Background
At present, when document reading can be carried out at any time through a terminal display device, analysis on notes appearing in the document can only be carried out through manual comparison and analysis at present, the document reading process is more complicated due to subjectivity of personal notes habit and semantic understanding, the reading efficiency is low, meanwhile, the situation of judging errors can possibly appear, the matching degree of a document notes mode and a consulting user can be improved through conversion of the notes mode and the notes content of the document notes, however, when the notes information is processed, the final processing result can not reach the expected standard due to the limitation of the prior art, and a certain liftable space exists in the prior art.
In the prior art, when document annotation is referred, the document review efficiency is low and the analysis deviation risk exists due to the fact that the processing method of annotation information is not intelligent enough, the diversity of annotation modes and the subjective assumption of information of the referring personnel.
Disclosure of Invention
The application provides a multi-document-based annotation interaction method and system, which are used for solving the technical problems that the processing method of annotation information in the prior art is not intelligent enough, the diversity of annotation modes and the subjective speculation of information of consultants lead to low document consultation efficiency and the risk of analysis deviation.
In view of the above problems, the present application provides a multi-document-based annotation interaction method and system.
In a first aspect, the present application provides a multi-document-based annotation interaction method, where the method includes: obtaining multi-document information; identifying multi-document annotation information based on the multi-document information, wherein the multi-document annotation information comprises annotation positions and annotation document reference contents; extracting annotation user sources according to the multi-document annotation information, determining annotation user information, generating a label based on the annotation user information, and labeling the multi-document annotation information; display port information is obtained, and port user information is determined based on the display port information; screening the annotation user information of the multi-document annotation information according to the port user information to obtain annotation information to be referred, wherein the annotation information to be referred comprises annotation reply information of a port user and annotation information of a non-port user; the method comprises the steps of obtaining an annotation record set of a port user, inputting the annotation record set of the port user into an annotation analysis model, and obtaining annotation characteristics of the port user; and carrying out recognition conversion on the annotation information to be referred based on the annotation characteristics of the port user, and displaying the annotation information through the display port information.
In a second aspect, the present application provides a multi-document based annotation interaction system, the system comprising: the information acquisition module is used for acquiring multi-document information; the information identification module is used for identifying multi-document annotation information based on the multi-document information, wherein the multi-document annotation information comprises annotation positions and annotation document reference contents; the information labeling module is used for extracting annotation user sources according to the multi-document annotation information, determining annotation user information and labeling the multi-document annotation information based on the annotation user information generating labels; the information determining module is used for obtaining the display port information and determining port user information based on the display port information; the information screening module is used for screening the annotation user information of the multi-document annotation information according to the port user information to obtain annotation information to be referred, wherein the annotation information to be referred comprises annotation reply information of a port user and non-port user annotation information; the characteristic acquisition module is used for acquiring an annotation record set of the port user, inputting the annotation record set of the port user into an annotation analysis model and acquiring the annotation characteristics of the port user; and the information conversion module is used for carrying out identification conversion on the annotation information to be referred based on the annotation characteristics of the port user and displaying the annotation information through the display port information.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
the multi-document-based annotation interaction method provided by the embodiment of the application comprises the steps of obtaining multi-document information and identifying annotation information, wherein the annotation information comprises annotation positions and annotation document reference contents, extracting annotation user sources to determine annotation user information, generating labels to annotate the multi-document annotation information, obtaining display port information and determining port user information, screening the multi-document annotation information to obtain annotation information to be referred, and the annotation information comprises annotation reply information of a port user and non-port user annotation information; inputting an annotation record set of a port user into an annotation analysis model to obtain annotation characteristics of the port user, further identifying and converting the annotation information to be consulted, and displaying the annotation information through the display port information, so that the problems that in the prior art, a processing method of the annotation information is not intelligent enough, the diversity of annotation modes and the subjective speculation of information of consultants are solved, the document consulting efficiency is low, the technical problem of analysis deviation risk exists, the intelligent conversion of document annotation is carried out based on the annotation habit of the consultants, the agreements of the document annotating and the consultants are improved, and the efficient and accurate consulting of the document is carried out.
Drawings
FIG. 1 is a schematic flow chart of a multi-document-based annotation interaction method;
FIG. 2 is a schematic diagram of a process for identifying and converting annotation information to be referred in an annotation interaction method based on multiple documents;
FIG. 3 is a schematic diagram of a process for marking annotation reply information in an annotation interaction method based on multiple documents;
FIG. 4 is a schematic diagram of a multi-document-based annotation interaction system.
Reference numerals illustrate: the system comprises an information acquisition module 11, an information identification module 12, an information labeling module 13, an information determination module 14, an information screening module 15, a characteristic acquisition module 16 and an information conversion module 17.
Detailed Description
The application provides a multi-document-based annotation interaction method and system, which are used for acquiring multi-document annotation information, extracting annotation user sources, generating labels for annotation marking, screening the multi-document annotation information based on port user information, acquiring annotation information to be consulted, inputting an annotation record set of the port user into an annotation analysis model, acquiring annotation characteristics of the port user, further identifying and converting the annotation information to be consulted, and displaying the annotation information through the display port information, so that the technical problems that the processing method of the annotation information in the prior art is not intelligent enough, the diversity of annotation modes and the subjective speculation of information of consulters are low in document consulting efficiency and the risk of analysis deviation exist are solved.
Example 1
As shown in FIG. 1, the application provides a multi-document-based annotation interaction method, which comprises the following steps:
step S100: obtaining multi-document information;
in particular, at present, document reading can be performed at any time through a terminal display device, but due to subjectivity of personal annotation habit, the annotation in the document can only be judged through manual comparison and analysis in the reading process, so that the reading efficiency is low, meanwhile, the situation of misjudgment can possibly occur.
Step S200: identifying multi-document annotation information based on the multi-document information, wherein the multi-document annotation information comprises annotation positions and annotation document reference contents;
step S300: extracting annotation user sources according to the multi-document annotation information, determining annotation user information, generating a label based on the annotation user information, and labeling the multi-document annotation information;
further, information identification is performed on the acquired multi-document information, and the annotation position corresponding to each piece of annotation information and the annotation document reference content are determined, wherein the annotation position is correspondingly associated with the annotation document reference content, optionally, the identified annotation information may have a plurality of different annotation users, and due to different personal annotation habits, the identification features such as the corresponding annotation information format and the like have differences, the annotation users and the corresponding document annotation information are correspondingly identified, and the annotation information identification result is acquired so as to facilitate the subsequent identification and distinction, and targeted processing is performed, so that the identified annotation information is integrated to generate the multi-document annotation information.
Further processing is carried out based on the multi-document annotation information, information extraction is carried out on the multi-document annotation information based on the annotation information identification result, one or more annotation users in the multi-document annotation information are determined, optionally, the annotation sources of the annotation users are extracted, information such as user names, annotation characteristics and personal styles of the annotation users are determined to serve as the annotation user information, simplification and normalization are carried out on the annotation user information, labels capable of reflecting subjective characteristics of the annotation users are determined, and the multi-document annotation information is marked based on the generated labels, so that information identification conversion can be directly carried out according to the marked labels, and information processing efficiency is improved.
Step S400: display port information is obtained, and port user information is determined based on the display port information;
step S500: screening the annotation user information of the multi-document annotation information according to the port user information to obtain annotation information to be referred, wherein the annotation information to be referred comprises annotation reply information of a port user and annotation information of a non-port user;
the terminal display device can be electronic devices such as a computer and a mobile phone, information extraction is further carried out on a display port of the terminal display device, such as login user, device display characteristics, display format and the like, the information is used as the display port information, user information of the login user including user identity information, document consulting content information and the like is further extracted based on the display port information, the user information is used as the port user information, the port user is a user currently operating, further, the port user information is used as a reference, the multi-document annotating information is screened according to the document consulting content information, annotating reply information of the port user and non-port user annotating information which are covered by document content of the port user are determined, the non-port user annotating information possibly comprises annotating information of a plurality of different users, identification consulting judgment can be directly carried out based on labels of the labels, the information is integrated to generate the to-be-consulted information, and the to-be-consulted annotated information is converted into real basic annotating information.
Step S600: the method comprises the steps of obtaining an annotation record set of a port user, inputting the annotation record set of the port user into an annotation analysis model, and obtaining annotation characteristics of the port user;
step S700: and carrying out recognition conversion on the annotation information to be referred based on the annotation characteristics of the port user, and displaying the annotation information through the display port information.
The method comprises the steps of constructing an annotation analysis model, namely, a virtual model for carrying out annotation information feature recognition analysis, carrying out history annotation record retrieval on the port user based on reference content and annotation content, obtaining an annotation record set of the port user, dividing the annotation record set of the port user into a training set and a verification set as sample data, carrying out model training and verification on the constructed annotation analysis model until feature recognition analysis accuracy of the annotation analysis model reaches a preset standard, stopping carrying out model training, taking the trained model as the finally determined annotation analysis model, and outputting the annotation features of the port user.
Further, the annotation characteristics of the port user are determined through model analysis, the annotation characteristics can reflect the annotation habit of the port user laterally, the annotation characteristics are divided to determine various annotation types, such as annotation formats and annotation sequences of serious difficulties, keyword annotation modes, meanings of different identifications and the like, the annotation information of the pieces of annotation information to be referred is classified based on the various annotation types of the port user, information conversion is further performed based on the corresponding annotation characteristics, the annotation information to be referred is converted into the annotation modes conforming to the annotation habit of the port user, smoothness and reference efficiency in document reference by the port user are guaranteed, the annotation information to be referred after information conversion is further displayed based on the display port information, and the display port information is related information of terminal display equipment for document reference and comprises a display port for information display.
Further, the method includes inputting the annotation record set of the port user into an annotation analysis model to obtain the annotation characteristics of the port user, and the step S600 of the present application further includes:
step S610: according to the annotation record set of the port user, taking the reference content and the annotation content as input parameters, taking the annotation characteristics of the user as output results, establishing a machine learning model, and constructing a training data set by utilizing the annotation record set to train and learn the machine learning model;
step S620: and carrying out feature recognition analysis on the reference content and the annotation content, determining annotation type features, carrying out annotation format and sentence structure sequence recognition processing based on the annotation type features, and determining the port user annotation features.
Specifically, the port user is a user who operates currently, the reference content in the extracted annotation record is associated with the annotation content, the annotation record sets of the port user are generated by classifying and integrating the same type of annotation record, the machine learning model is further constructed based on a machine learning algorithm, the machine learning model is an auxiliary tool for extracting the characteristics of input content, and optionally, the machine learning model can be a multi-level network layer and comprises an information identification layer, a characteristic comparison analysis layer and an input/output layer, the input layer and the output layer are necessarily structures of the model, no special significance exists, the annotation record sets are used as sample data, the training set and the verification set are further input into the machine learning model, and the model training verification is performed until the simulation accuracy of the model reaches a preset accuracy degree, for example, 95%, so that the annotation analysis model is determined.
Further, the reference content and the annotation content are input into the information identification layer of the annotation analysis model based on the input layer, annotation state information is determined and then transmitted into the feature comparison analysis layer, the annotation type features, such as characters, graphs, charts, transverse lines, numbers and the like, are determined, the expression ideas corresponding to different annotation types are different, such as key annotation, keyword annotation and the like, the annotation format is regulated based on the annotation type features, the custom annotation content habit is determined, the arrangement sequence of the characters, graphs and icons, the sentence structure sequence and the like are exemplified, such as the sequence structure of the keyword, verb and graph as the individual expression, the reference content is directly added with the annotation feature words based on the expression structure, the annotation feature words are determined as the annotation features of the port user, the annotation standardized format of the port user is obtained, and meanwhile, the accuracy and objectivity of the annotation features which can be effectively output are ensured by carrying out feature dissection based on the annotation analysis model.
Further, as shown in fig. 2, before the identifying and converting the annotation information to be referred to based on the port user annotation feature, step S700 of the present application further includes:
step S710-1: obtaining a viewing requirement of a port user;
step S720-1: identifying and searching the annotation information to be searched according to the viewing requirement of the port user, and determining the annotation information to be searched;
step S730-1: determining annotation feature types according to the port user annotation features, and identifying and matching the annotation information required to be checked based on the annotation feature types to obtain matched annotation information to be converted;
step S740-1: and carrying out recognition conversion on the matched annotation information to be converted based on the port user annotation characteristics.
Specifically, the multi-document annotation information is screened based on the annotation user information to obtain the annotation information to be checked, and further, the viewing requirements of the port user, namely the current operation user, are determined, for example, the annotation of the document with heavy difficulty, the information identification is further carried out on the annotation information to be checked, the annotation information meeting the viewing requirements of the port user is determined, and the annotation information is used as part of information for viewing the annotation information required to ensure viewing smoothness and the annotation information conversion necessity exists.
Further, based on the annotation characteristics of the port user, the annotation characteristic types are determined, for example, different annotation format sequences and the like are adopted according to different annotation meanings, wherein the annotation characteristic types possibly comprise a plurality of types, further, information matching is carried out on the annotation characteristic types and the required annotation checking information, corresponding identification is carried out on the required annotation checking information corresponding to the matching result and the annotation characteristic types so as to directly identify and distinguish, one annotation characteristic type can correspond to one or more pieces of required annotation checking information, the one or more pieces of annotation characteristic types are used as the annotation information to be converted, further, information identification conversion is carried out on the matched annotation information to be converted according to the corresponding annotation characteristic types of the port user, and the document annotation information is converted into annotation information conforming to the annotation habit of the port user on the basis that the content and the integrity of the annotation information are not influenced, so that the reading smoothness and the comfort of the document of the port user can be effectively improved, and the annotation content is prevented from being caused by subjective analysis error.
Further, step S700 of the present application further includes:
step S710-2: analyzing the quotation content of the annotation document according to the annotation information to be consulted, and extracting the annotation information set with the same root;
step S720-2: carrying out annotation user information analysis on the same-root annotation information set, and determining the user association degree of the annotation user information and the port user information;
step S730-2: generating different display features based on different user associations;
step S740-2: and carrying out content semantic analysis based on the same-root annotation information set, determining semantic similarity, and carrying out same-type marking on annotations with high similarity based on the semantic similarity.
Specifically, the annotation information of the plurality of pieces of annotation information in the annotation information to be referred is respectively subjected to annotation source searching, whether the annotation document reference content corresponding to the annotation information belongs to the same reference content is determined, namely, the same reference content or the annotation information of the plurality of pieces of annotation information corresponding to the keywords are generated by carrying out information corresponding integration and annotation, wherein the annotation information corresponding to the same reference content in the same-root annotation information set may be annotated by the same annotation user or a plurality of different annotation users.
Further, the same-root annotation information set is subjected to annotation user information analysis, user association analysis is performed on the annotation user information and the port user information, for example, the same-root annotation and annotation content attention degree exist, different display characteristics are set based on the difference of the user association degree, the display priority is decreased according to the association degree, the higher the association degree is, the more obvious the display characteristics are, further, the same-root annotation information set is subjected to semantic analysis on annotation information based on the reference content respectively, the semantic similarity of each group of same-root annotations is determined, the same-root annotations with higher similarity can be regarded as the same-type annotations, optionally, multiple annotation types are set based on multiple-level similarity so as to conduct selective review of annotation information, the repeated review rate of the information is reduced, and the document review efficiency can be effectively improved.
Further, as shown in fig. 3, step S700 of the present application further includes:
step S710-3: when the display port displays the annotation reply information of the port user, carrying out semantic recognition on the annotation reply information to obtain reply semantic information;
step S720-3: according to the reply semantic information, displaying the same display characteristics of the semantic similarity meeting the preset requirement;
step S730-3: the method comprises the steps of obtaining annotation user information of annotation reply information;
step S740-3: determining a reply user relationship according to the endorsement user information and the port user information;
step S750-3: and marking the annotation reply information of the port user based on the reply user relationship and the semantic similarity.
Specifically, when information conversion is performed on the to-be-referred annotation information, wherein part of annotation reply information exists, after conversion is completed, annotation information is displayed based on the display port, if the display port is displayed as the annotation reply information of the port user, the annotation reply information is extracted and subjected to semantic recognition, a specific reply content direction is determined, for example, annotation supplement, annotation judgment and the like are determined as the reply semantic information, semantic similarity judgment is further performed on the reply semantic information, a preset requirement, namely, a critical value for semantic similarity limitation is set, each piece of information in the reply semantic information is judged to be subjected to check analysis in pairs respectively, whether the semantic similarity meets the preset requirement is determined, when the preset requirement is met, the fact that two pieces of information subjected to comparison belong to the same-feature semantic information is indicated, the judgment result is further classified, and the same-feature semantic information is displayed.
Further, the annotation information corresponding to the annotation reply information is obtained, the annotation user information of the annotation information is determined, the annotation user information, the annotation reply information and the annotation information correspond to each other, the information reply relation analysis is carried out on the annotation user information and the port user information, the reply user relation is determined, the annotation user possibly annotates for the port user or other users, the annotation reply information is the port user annotation, the annotation reply information of the port user is further classified, integrated and identified according to the reply user relation and the semantic similarity, the same category is identified based on the same identification information, and the user can conveniently check pertinently.
Further, after displaying the same display characteristics with the semantic similarity meeting the preset requirement according to the reply semantic information, step S720-3 of the present application further includes:
step S721-3: classifying the annotation reply information according to the semantic similarity;
step S722-3: determining classified display characteristics based on classification results of the annotation reply information;
step S723-3: and highlighting the semantic similarity lower than a preset threshold value.
Specifically, by performing semantic recognition on the annotation reply information, determining the reply semantic information, further performing semantic similarity analysis on the reply semantic information, by setting a similarity classification section to determine a plurality of similarity sections, for example, classifying the annotation reply information based on semantic similarity, obtaining a classification result of the annotation classification information, determining classification display characteristics corresponding to each classification level in the classification result, for example, uniformly displaying section information with higher overlapping degree, namely highest similarity, based on the same display characteristics, for general displayable similarity, further setting the preset threshold, namely, defining a base line critical value of the semantic similarity, highlighting the corresponding annotation reply information when the semantic similarity in the classification result is lower than the preset threshold, and performing classification display on the annotation reply information based on the semantic similarity, thereby extracting the targeted information according to actual reference requirements and improving the information recognition efficiency.
Further, step S700 of the present application further includes:
step S710-4: obtaining the checking requirement of the annotating user;
step S720-4: determining checking annotation user information based on the annotation user checking requirement, and performing traversal comparison on the multi-document annotation information based on the checking annotation user information to obtain a checking user annotation information set;
step S730-4: carrying out annotation reference content and annotation information analysis on the user annotation information set to obtain user attention content characteristics and annotation information characteristics;
step S740-4: and generating the demand display information according to the user annotation information set, the user attention content characteristics and the annotation information characteristics, and displaying the demand display information through the display port information.
Specifically, before document review, corresponding document information can be directly determined based on the requirement of review purpose to avoid invalid work, user review requirement corresponding to document annotation information is determined, namely, the purpose of review of document annotation information is achieved, the document annotation information is used as the annotation user review requirement, further, user information for review of the annotation is determined based on the annotation user review requirement, the user information for review is used as the annotation review user information, further, multi-document annotation information is subjected to traversal comparison based on the annotation review user information, the annotation information of the document by each piece of annotation review user information is determined, relevant annotation information and user information are correspondingly integrated, a plurality of groups of annotation information corresponding to different users in the user review information set are generated, further, joint analysis is conducted on the annotation reference content and the annotation information feature, comprehensive consideration is conducted on the frequency, the annotation direction and the like of reference content, further, the user attention content feature and the annotation information feature are obtained, the user information set, the point of interest of the user is further, the point of interest of the user is required to be reviewed, the point of view is displayed, and the point of view information is displayed, and the requirement of the user information is required to be further, and the point view information is displayed, and the point of view is displayed, and the point view information is required to be viewed, and the point view is displayed, and the point view information is further is displayed.
Example two
Based on the same inventive concept as the annotation interaction method based on multiple documents in the foregoing embodiment, as shown in fig. 4, the present application provides an annotation interaction system based on multiple documents, where the system includes:
an information acquisition module 11, wherein the information acquisition module 11 is used for acquiring multi-document information;
an information identifying module 12, wherein the information identifying module 12 is configured to identify multi-document annotation information based on the multi-document information, and the multi-document annotation information includes an annotation position and annotation document reference content;
the information labeling module 13 is used for extracting annotation user sources according to the multi-document annotation information, determining annotation user information, and labeling the multi-document annotation information based on the annotation user information generating labels;
the information determining module 14, the information determining module 14 is configured to obtain display port information, and determine port user information based on the display port information;
the information screening module 15 is configured to screen the multi-document annotation information for annotation user information according to the port user information, so as to obtain annotation information to be referred, where the annotation information to be referred includes annotation reply information of a port user and annotation information of a non-port user;
the feature acquisition module 16 is configured to obtain an annotation record set of a port user, input the annotation record set of the port user into an annotation analysis model, and obtain annotation features of the port user;
the information conversion module 17 is configured to identify and convert the annotation information to be referred to based on the annotation characteristics of the port user, and display the annotation information through the display port information.
Further, the system further comprises:
the request acquisition module is used for acquiring the viewing request of the port user;
the annotation information determining module is used for identifying and searching the annotation information to be referred according to the viewing requirement of the port user and determining the annotation information to be required to be viewed;
the information matching module is used for determining annotation feature types according to the port user annotation features, and identifying and matching the annotation information required to be checked based on the annotation feature types to obtain matching annotation information to be converted;
and the information identification and conversion module is used for carrying out identification and conversion on the matched annotation information to be converted based on the annotation characteristics of the port user.
Further, the system further comprises:
the reference analysis module is used for analyzing the reference content of the annotation document according to the annotation information to be consulted and extracting the annotation information set with the same root;
the association degree determining module is used for analyzing the endorsement user information of the same root endorsement information set and determining the user association degree of the endorsement user information and the port user information;
the display characteristic generation module is used for generating different display characteristics based on different user association degrees;
and the annotation marking module is used for carrying out content semantic analysis based on the same-root annotation information set, determining semantic similarity and carrying out the same-type marking on the annotations with high similarity based on the semantic similarity.
Further, the system further comprises:
the semantic recognition module is used for carrying out semantic recognition on the annotation reply information when the display port displays the annotation reply information of the port user, so as to obtain reply semantic information;
the feature display module is used for displaying the same display features of which the semantic similarity meets the preset requirement according to the reply semantic information;
the user information acquisition module is used for acquiring annotation user information of the annotation reply information;
the relationship determining module is used for determining a reply user relationship according to the annotating user information and the port user information;
and the reply information marking module is used for marking the annotation reply information of the port user based on the reply user relationship and the semantic similarity.
Further, the system further comprises:
the information classification module is used for classifying the annotation reply information according to the semantic similarity;
the display characteristic determining module is used for annotating the classification result of the reply information to determine classified display characteristics;
and the threshold judging module is used for highlighting the semantic similarity lower than a preset threshold.
Further, the system further comprises:
the requirement acquisition module is used for acquiring the checking requirement of the annotating user;
the information comparison module is used for determining the information of the user for checking the endorsements based on the checking requirement of the endorsements of the user, and carrying out traversal comparison on the information of the multi-document endorsements based on the information of the user for checking the endorsements to obtain an endorsement information set of the user for checking the endorsements;
the characteristic information acquisition module is used for analyzing annotation reference content and annotation information of the annotation information set of the user to obtain the attention content characteristics and annotation information characteristics of the user;
and the information display module is used for generating the demand display information according to the user annotation information set, the user attention content characteristics and the annotation information characteristics and displaying the demand display information through the display port information.
Further, the system further comprises:
the model building training module is used for building a machine learning model by taking the reference content and the annotation content as input parameters and the annotation characteristics of the user as output results according to the annotation record set of the port user, and building a training data set by utilizing the annotation record set to train and learn the machine learning model;
and the annotation feature determining module is used for carrying out feature recognition analysis on the reference content and the annotation content, determining annotation type features, carrying out annotation format and sentence structure sequence recognition processing based on the annotation type features, and determining the port user annotation features.
Through the foregoing detailed description of a multi-document-based annotation interaction method, those skilled in the art can clearly understand that a multi-document-based annotation interaction method and system in this embodiment, for the device disclosed in the embodiment, the description is relatively simple because it corresponds to the method disclosed in the embodiment, and relevant places refer to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (6)
1. A multi-document-based annotation interaction method, the method comprising:
obtaining multi-document information;
identifying multi-document annotation information based on the multi-document information, wherein the multi-document annotation information comprises annotation positions and annotation document reference contents;
extracting annotation user sources according to the multi-document annotation information, determining annotation user information, generating a label based on the annotation user information, and labeling the multi-document annotation information;
display port information is obtained, and port user information is determined based on the display port information;
screening the annotation user information of the multi-document annotation information according to the port user information to obtain annotation information to be referred, wherein the annotation information to be referred comprises annotation reply information of a port user and annotation information of a non-port user;
the method comprises the steps of obtaining an annotation record set of a port user, inputting the annotation record set of the port user into an annotation analysis model, and obtaining annotation characteristics of the port user;
identifying and converting the annotation information to be referred based on the annotation characteristics of the port user, and displaying the annotation information through the display port information;
inputting the annotation record set of the port user into an annotation analysis model to obtain port user annotation characteristics, wherein the method comprises the following steps:
according to the annotation record set of the port user, taking the reference content and the annotation content as input parameters, taking the annotation characteristics of the user as output results, establishing a machine learning model, and constructing a training data set by utilizing the annotation record set to train and learn the machine learning model;
performing feature recognition analysis on the reference content and the annotation content, determining annotation type features, performing annotation format and sentence structure sequence recognition processing based on the annotation type features, and determining the port user annotation features;
before the annotation information to be referred is identified and converted based on the port user annotation characteristics, the method comprises the following steps:
obtaining a viewing requirement of a port user;
identifying and searching the annotation information to be searched according to the viewing requirement of the port user, and determining the annotation information to be searched;
determining annotation feature types according to the port user annotation features, and identifying and matching the annotation information required to be checked based on the annotation feature types to obtain matched annotation information to be converted;
and carrying out recognition conversion on the matched annotation information to be converted based on the port user annotation characteristics.
2. The method of claim 1, wherein the method further comprises:
analyzing the quotation content of the annotation document according to the annotation information to be consulted, and extracting the annotation information set with the same root;
carrying out annotation user information analysis on the same-root annotation information set, and determining the user association degree of the annotation user information and the port user information;
generating different display features based on different user associations;
and carrying out content semantic analysis based on the same-root annotation information set, determining semantic similarity, and carrying out same-type marking on annotations with high similarity based on the semantic similarity.
3. The method of claim 2, wherein the method further comprises:
when the display port displays the annotation reply information of the port user, carrying out semantic recognition on the annotation reply information to obtain reply semantic information;
according to the reply semantic information, displaying the same display characteristics of the semantic similarity meeting the preset requirement;
the method comprises the steps of obtaining annotation user information of annotation reply information;
determining a reply user relationship according to the annotation user information of the annotation reply information and the port user information;
and marking the annotation reply information of the port user based on the reply user relationship and the semantic similarity.
4. The method of claim 3, wherein after displaying the co-display features with the semantic similarity satisfying the preset requirement according to the reply semantic information, the method comprises:
classifying the annotation reply information according to the semantic similarity;
determining classified display characteristics based on classification results of the annotation reply information;
and highlighting the semantic similarity lower than a preset threshold value.
5. The method of claim 1, wherein the method further comprises:
obtaining the checking requirement of the annotating user;
determining checking annotation user information based on the annotation user checking requirement, and performing traversal comparison on the multi-document annotation information based on the checking annotation user information to obtain a checking user annotation information set;
carrying out annotation reference content and annotation information analysis on the user annotation information set to obtain user attention content characteristics and annotation information characteristics;
and generating the demand display information according to the user annotation information set, the user attention content characteristics and the annotation information characteristics, and displaying the demand display information through the display port information.
6. A multi-document based annotation interactive system, the system comprising:
the information acquisition module is used for acquiring multi-document information;
the information identification module is used for identifying multi-document annotation information based on the multi-document information, wherein the multi-document annotation information comprises annotation positions and annotation document reference contents;
the information labeling module is used for extracting annotation user sources according to the multi-document annotation information, determining annotation user information and labeling the multi-document annotation information based on the annotation user information generating labels;
the information determining module is used for obtaining the display port information and determining port user information based on the display port information;
the information screening module is used for screening the annotation user information of the multi-document annotation information according to the port user information to obtain annotation information to be referred, wherein the annotation information to be referred comprises annotation reply information of a port user and non-port user annotation information;
the characteristic acquisition module is used for acquiring an annotation record set of the port user, inputting the annotation record set of the port user into an annotation analysis model and acquiring the annotation characteristics of the port user;
the information conversion module is used for carrying out identification conversion on the annotation information to be referred based on the annotation characteristics of the port user and displaying the annotation information through the display port information;
the model building training module is used for building a machine learning model by taking the reference content and the annotation content as input parameters and the annotation characteristics of the user as output results according to the annotation record set of the port user, and building a training data set by utilizing the annotation record set to train and learn the machine learning model;
the annotation feature determining module is used for carrying out feature recognition analysis on the reference content and the annotation content, determining annotation type features, carrying out annotation format and sentence structure sequence recognition processing based on the annotation type features, and determining the port user annotation features;
the request acquisition module is used for acquiring the viewing request of the port user;
the annotation information determining module is used for identifying and searching the annotation information to be referred according to the viewing requirement of the port user and determining the annotation information to be required to be viewed;
the information matching module is used for determining annotation feature types according to the port user annotation features, and identifying and matching the annotation information required to be checked based on the annotation feature types to obtain matching annotation information to be converted;
and the information identification and conversion module is used for carrying out identification and conversion on the matched annotation information to be converted based on the annotation characteristics of the port user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211301753.7A CN115659929B (en) | 2022-10-24 | 2022-10-24 | Annotating interaction method and system based on multiple documents |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211301753.7A CN115659929B (en) | 2022-10-24 | 2022-10-24 | Annotating interaction method and system based on multiple documents |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115659929A CN115659929A (en) | 2023-01-31 |
| CN115659929B true CN115659929B (en) | 2023-09-15 |
Family
ID=84991666
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211301753.7A Active CN115659929B (en) | 2022-10-24 | 2022-10-24 | Annotating interaction method and system based on multiple documents |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115659929B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102819531A (en) * | 2011-06-10 | 2012-12-12 | 北大方正集团有限公司 | Cloud reading service system, cloud reading service method and device |
| CN103500158A (en) * | 2013-10-08 | 2014-01-08 | 北京百度网讯科技有限公司 | Method and device for annotating electronic document |
| JP2014139713A (en) * | 2013-01-21 | 2014-07-31 | Mitsubishi Electric Corp | Refereeing result table generating device and refereeing result table generating program |
| CN104391831A (en) * | 2014-11-12 | 2015-03-04 | 武汉传神信息技术有限公司 | Method and system for commenting file contents |
| JP2015138358A (en) * | 2014-01-22 | 2015-07-30 | 日本電気株式会社 | Information processing device, information processing method, and information processing program |
| CN110956023A (en) * | 2018-09-25 | 2020-04-03 | 珠海金山办公软件有限公司 | Comment display method and device |
| CN113392179A (en) * | 2020-12-21 | 2021-09-14 | 腾讯科技（深圳）有限公司 | Text labeling method and device, electronic equipment and storage medium |
| WO2021184574A1 (en) * | 2020-03-20 | 2021-09-23 | 平安国际智慧城市科技股份有限公司 | Document data importing method and apparatus, document data exporting method and apparatus, and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200160231A1 (en) * | 2018-11-19 | 2020-05-21 | International Business Machines Corporation | Method and System for Using a Multi-Factorial Analysis to Identify Optimal Annotators for Building a Supervised Machine Learning Model |
| US11243916B2 (en) * | 2019-02-27 | 2022-02-08 | Atlassian Pty Ltd. | Autonomous redundancy mitigation in knowledge-sharing features of a collaborative work tool |
-
2022
- 2022-10-24 CN CN202211301753.7A patent/CN115659929B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102819531A (en) * | 2011-06-10 | 2012-12-12 | 北大方正集团有限公司 | Cloud reading service system, cloud reading service method and device |
| JP2014139713A (en) * | 2013-01-21 | 2014-07-31 | Mitsubishi Electric Corp | Refereeing result table generating device and refereeing result table generating program |
| CN103500158A (en) * | 2013-10-08 | 2014-01-08 | 北京百度网讯科技有限公司 | Method and device for annotating electronic document |
| JP2015138358A (en) * | 2014-01-22 | 2015-07-30 | 日本電気株式会社 | Information processing device, information processing method, and information processing program |
| CN104391831A (en) * | 2014-11-12 | 2015-03-04 | 武汉传神信息技术有限公司 | Method and system for commenting file contents |
| CN110956023A (en) * | 2018-09-25 | 2020-04-03 | 珠海金山办公软件有限公司 | Comment display method and device |
| WO2021184574A1 (en) * | 2020-03-20 | 2021-09-23 | 平安国际智慧城市科技股份有限公司 | Document data importing method and apparatus, document data exporting method and apparatus, and storage medium |
| CN113392179A (en) * | 2020-12-21 | 2021-09-14 | 腾讯科技（深圳）有限公司 | Text labeling method and device, electronic equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| Markup: A Web-Based Annotation Tool Powered by Active Learning;Samuel Dobbie等;Frontiers;第3卷;第1-9页 * |
| The INCEpTION Platform: Machine-Assisted and Knowledge-Oriented Interactive Annotation;Jan-Christoph Klie等;Proceedings of the 27th International Conference on Computational Linguistics: System Demonstrations;第5–9页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115659929A (en) | 2023-01-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2018000269A1 (en) | Data annotation method and system based on data mining and crowdsourcing | |
| JP3856778B2 (en) | Document classification apparatus and document classification method for multiple languages | |
| CN1794234A (en) | Data semanticizer | |
| CN111930792A (en) | Data resource labeling method and device, storage medium and electronic equipment | |
| CN114880468A (en) | Building specification examination method and system based on BilSTM and knowledge graph | |
| CN113987125A (en) | Text structured information extraction method based on neural network and related equipment thereof | |
| CN113362072B (en) | Wind control data processing method and device, electronic equipment and storage medium | |
| CN116090560A (en) | Knowledge graph establishment method, device and system based on teaching materials | |
| CN113553419A (en) | Civil aviation knowledge map question-answering system | |
| CN115659929B (en) | Annotating interaction method and system based on multiple documents | |
| CN117592470A (en) | Low-cost gazette data extraction method driven by large language model | |
| CN113822040A (en) | Subjective question marking and scoring method and device, computer equipment and storage medium | |
| CN116644728B (en) | Contract generation method and system based on clause digitization | |
| CN113011154A (en) | Job duplicate checking method based on deep learning | |
| CN116578703A (en) | Intelligent identification system and method | |
| CN114118098A (en) | Contract review method, equipment and storage medium based on element extraction | |
| CN110119464B (en) | Intelligent recommendation method and device for numerical values in contract | |
| Blum et al. | A comparative wordlist for investigating distant relations among languages in Lowland South America | |
| CN118297360B (en) | Intelligent matching method for chest pain center construction | |
| CN118170933B (en) | Construction method and device of multi-mode corpus data oriented to scientific field | |
| CN113704405B (en) | Quality inspection scoring method, device, equipment and storage medium based on recorded content | |
| KR20120098081A (en) | System and method for managing mathematical contents | |
| CN116681057A (en) | Analysis method of mail structure, training method of recognition model and related equipment | |
| CN117632899A (en) | ICL special disease database construction method, device, equipment and storage medium | |
| CN118277532A (en) | Medical data processing suggestion generation method, device and equipment |
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 | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A method and system for interactive annotation based on multiple documents Granted publication date: 20230915 Pledgee: China Construction Bank Corporation Nanjing Jiangbei new area branch Pledgor: NANJING HENTOR INFORMATION TECHNOLOGY CO.,LTD. Registration number: Y2024980028531 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right |