CN116108169B - Hot wire work order intelligent dispatching method based on knowledge graph - Google Patents
Hot wire work order intelligent dispatching method based on knowledge graph Download PDFInfo
- Publication number
- CN116108169B CN116108169B CN202211590719.6A CN202211590719A CN116108169B CN 116108169 B CN116108169 B CN 116108169B CN 202211590719 A CN202211590719 A CN 202211590719A CN 116108169 B CN116108169 B CN 116108169B
- Authority
- CN
- China
- Prior art keywords
- event
- text
- hotline
- word
- entity
- 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 35
- 238000005516 engineering process Methods 0.000 claims abstract description 6
- 239000013598 vector Substances 0.000 claims description 46
- 238000012545 processing Methods 0.000 claims description 41
- 230000011218 segmentation Effects 0.000 claims description 21
- 238000000605 extraction Methods 0.000 claims description 12
- 238000001514 detection method Methods 0.000 claims description 7
- 230000004927 fusion Effects 0.000 claims description 7
- 239000013604 expression vector Substances 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 230000002950 deficient Effects 0.000 abstract description 2
- 238000003058 natural language processing Methods 0.000 abstract description 2
- 238000010187 selection method Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 7
- 238000013527 convolutional neural network Methods 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000000306 recurrent effect Effects 0.000 description 3
- 238000013459 approach Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
- G06F16/353—Clustering; Classification into predefined classes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention discloses a hotline work order intelligent dispatching method based on a knowledge graph, and belongs to the field of work order dispatching. The invention adopts natural language processing technology to extract the event description of the work order text, and obtains the trigger word and event role content of the hotline event, thereby obtaining the key information of the event text, the key information of the hotline event is used as the basic information for constructing the knowledge graph, a plurality of questions are designed according to the information of the hotline event such as event-event type, event-trigger word, event-event role and the like obtained by extracting the hotline event, and the candidate dispatching departments of the event are obtained by combining the information of the key information entity in the knowledge graph and using the answer selection method. Aiming at the situation that the manual dispatch accuracy of the thermal wire work orders in the prior art is deficient, the invention is beneficial to improving the dispatch accuracy of the thermal wire work orders and better realizing 'instant complaints'.
Description
Technical Field
The invention relates to the technical field of work order dispatching, in particular to a hotline work order intelligent dispatching method based on a knowledge graph.
Background
The city long heat line work order (hereinafter referred to as "work order") is text information recorded by an operator according to the incoming call of the city, and its elements are time, place, person and event. The work order text information presents layer-by-layer diffusion characteristics of buildings, cells, communities and streets on a spatial sequence, and the continuous expansion characteristics of various aspects of urban life are generally covered on a theme sequence. Operators need to find responsibility departments to be allocated for worksheets, and the number of selectable responsibility departments is often tens, so that the link of confirming the responsibility departments is time-consuming and labor-consuming. In addition, the accuracy of manual work order assignment depends on the accurate judgment of the classification of citizen appeal by operators and the deep knowledge of the functions of each responsible department. In urban hot-line operation practice, false dispatch frequently occurs, resulting in unnecessary secondary dispatch. Along with the increasing variety and quantity of city events accepted by the city long hotline, the intelligent dispatching of worksheets is imperative.
In order to realize automatic allocation of the city long hot line work orders, a machine learning-based method is generally adopted: firstly, extracting features from the worksheet text description, then connecting a plurality of features in series to form a high-dimension feature vector, and finally completing classification by using a classifier. This approach requires a lot of feature engineering, and the selection and analysis of features is complex, resulting in significant effort to design features that may not be relevant to the task specified. Still further, work order text feature extraction and classification tasks may be accomplished automatically using convolutional neural networks (CNN, convolutional Neural Networks), recurrent neural networks (RNN, recurrent Neural Network), long Short Term Memory recurrent neural networks (LSTM). However, worksheets belonging to the same general class but different subclasses have many similarities in text information descriptions. For example, the description of "road occupation parking" and "road occupation repair" belonging to the class of "street order" is very similar, but it belongs to different responsible departments (public security office and city administration), and it is difficult to find such a small difference by using a single neural network method, and it is difficult to judge the responsible department to which it belongs.
Through retrieval, the patent of publication No. CN112541351A discloses a method and a system for dispatching government hot-line work orders in the field of construction, document entry matrixes are constructed by segmenting the texts of the government hot-line work orders in the field of construction, and the responsibility units of the work orders are predicted by inputting a pre-trained support vector machine model. The method is only applicable to related worksheets of building departments, but not applicable to other departments.
Disclosure of Invention
1. Technical problem to be solved by the invention
Aiming at the situation that the manual dispatching accuracy of the hot wire work orders in the prior art is deficient, the invention aims to provide the hot wire work order intelligent dispatching method based on the knowledge graph, which is beneficial to improving the dispatching accuracy of the hot wire work orders and better realizing 'follow-up and instant-on'.
2. Technical proposal
In order to achieve the above purpose, the technical scheme provided by the invention is as follows:
according to the hotline work order intelligent dispatching method based on the knowledge graph, the natural language processing technology is adopted to extract the events of the work order text, trigger words and event role contents of hotline events are obtained, so that key information of the event text is obtained, and the key information of the hotline events is used as basic information for constructing the knowledge graph. The event extraction result mainly comprises information such as an event-event type, an event-trigger word, an event-event role and the like, event triples are constructed according to the information, and then the triples are fused and associated by using knowledge graph related technologies such as entity fusion, relationship discovery, relationship reasoning and the like, so that a knowledge graph of event key information is constructed. According to the information such as 'event-event type', 'event-trigger word', 'event-event role', etc. extracted from the hotline event, designing a plurality of questions according to the key information of the event, combining the information of the key information entity in the knowledge graph, and using the answer selection method to obtain the candidate dispatch departments of the event.
The key of department assignment is analysis of the content of the work order text itself, and mining the relation between the events in the work order text and the government departments, the heat-line work order intelligent assignment method based on the event extraction and the knowledge graph provided by the invention comprises three main parts: s100, extracting events based on historical heat line work order texts; s200, constructing a knowledge graph based on event key information; s300, predicting the dispatching department based on the knowledge graph questions and answers.
The heat line work order text event extraction technology in S100 comprises two parts of event detection and event role identification, wherein the event detection process is as follows:
s111, sentence segmentation is carried out on the event text, and the jieba word segmentation is used for word segmentation processing to obtain documetteokens and sentencetoken;
s112, encoding the documetteokens by using a word embedding method, respectively extracting local features and global features of the hotline event text by using a CNN network and an RNN network, and obtaining a representation vector V of the hotline event text by combining an Attention mechanism 1 ；
S113, carrying out word embedding processing on sendencetokens, generating AMR graphs for each sentence by using an AMR algorithm, and connecting a plurality of AMR graphs by combining keywords, AMR root nodes and the like to obtain an event text integerEncoding AMR graph by GCN network to obtain representation vector V of hot line event text 2 ；
S114, processing each sentence of the hotline event text by using the BERT pre-training language model to generate word vectors of all words in the sentences, then using SIF weights to form the word vectors into a plurality of sentence vectors, and combining the sentence vectors into a representation vector V of the hotline event text by using a weighted average method according to the content hierarchical distribution of the text (such as the central content of the text is mainly concentrated in the first sentence or the last sentence, the sentence length and the like) 3 ；
S115, fusing three types of expression vectors V by adopting a gating mechanism 1 ，V 2 ，V 3 Obtaining a corresponding fusion vector V fuse And finally, carrying out event multi-label classification and event trigger word recognition on the fusion vector by using a fully connected network to obtain a corresponding multi-label event type and trigger word set.
The event role recognition process is as follows:
s121, performing jieba word segmentation on the event text to obtain a word segmentation result, and processing the event text by using a word embedding mode to obtain a corresponding coding result;
s122, using a BiLSTM+CRF model to carry out named entity recognition on the coded result, and recognizing all entity vectors in the hotline event text;
s123, for different event types, firstly, according to an embellishing table, finding an embellishing vector of a corresponding trigger word, and using a BiLSTM network to encode to obtain an event trigger word representation vector V t ；
S124, trigger word representing vector V corresponding to splicing event type t And entity vectors, calculating event roles of different event types.
In S200, based on the dispatching result of the historical hot line work order, information such as an event processing department and an event result can be obtained, information such as a trigger word, an event type, an event role and the like of a hot line event can be obtained according to the result of the event extraction algorithm, and a plurality of triples can be obtained by taking entities such as an event number, a department name, an event type name and the like as nodes:
"event-processing department-department name", "event-processing mode", "event-event type name", "event-trigger word name", "event-event role-role content";
the triples can obtain a relation diagram of a certain historical event, and the relation diagrams of the events are connected to obtain a knowledge graph of the hot-line work order event, wherein the same event can be associated with a plurality of types, a plurality of trigger words, a plurality of processing departments and the like, and the relation diagrams of the events are connected through information such as the event types, the trigger words, the processing departments and the like; the specific construction process comprises the following steps:
s201, processing the event role content, the processing result and other entities by adopting an entity alignment technology, discovering potential alignment entities, and renaming the potential alignment entities by using the same naming format;
s202, fusing the same event roles and event processing modes for the results after entity alignment, and associating different event relationship graphs through the event roles and the event processing modes.
According to the process, a knowledge graph based on the event key information can be constructed.
The dispatch department prediction process based on knowledge graph questions and answers in S300 is as follows:
s301, performing word segmentation processing on the designed problem, and performing feature extraction on a word segmentation result by adopting a BiLSTM network to obtain a corresponding problem expression vector;
s302, performing graph embedding processing on a multilayer neighborhood relation containing key entities (such as XX types and XX trigger words) in the problems in the knowledge graph by using a TransR to obtain corresponding embedded vectors;
s303, regarding the multi-layer neighborhood relation entity of the input entity as a candidate answer, predicting the probability scores of the questions and the relation entity, and selecting the entity with the highest probability score as the answer (dispatch department) of the questions;
s304, obtaining a plurality of candidate departments according to the designed general questions, setting a specific threshold value, and selecting an entity with probability score exceeding the given threshold value as an answer candidate set (an optional dispatch department and an associated department).
3. Advantageous effects
Compared with the prior art, the technical scheme provided by the invention has the following beneficial effects:
the knowledge-graph-based heat line work order intelligent dispatching method can improve the accuracy of manual dispatching of the city long heat lines and can better realize 'order-taking and-handling'.
Drawings
FIG. 1 is a flow chart of event detection according to the present invention;
FIG. 2 is a schematic flow chart of event character recognition in the present invention;
FIG. 3 is a schematic diagram of a dispatch department prediction model based on knowledge graph questions and answers in the invention.
Detailed Description
For a further understanding of the present invention, the present invention will be described in detail with reference to the drawings.
In the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
The invention is further described below with reference to examples.
Example 1
The method for intelligently assigning the heat line work orders based on event extraction and knowledge graph in the embodiment comprises the following steps:
s100, extracting events based on historical heat line work order texts;
s200, constructing a knowledge graph based on event key information;
s300, predicting the dispatching department based on the knowledge graph questions and answers.
Specifically, the hot work order text event extraction technique in S100 includes two parts, event detection and event role identification:
the event detection process is as follows:
s111, sentence segmentation is carried out on the event text, and the jieba word segmentation is used for word segmentation processing to obtain documetteokens and sentencetoken;
s112, encoding the documetteokens by using a word embedding method, respectively extracting local features and global features of the hotline event text by using a CNN network and an RNN network, and obtaining a representation vector V of the hotline event text by combining an Attention mechanism 1 ；
S113, carrying out word embedding processing on sendencetokens, generating AMR graphs for each sentence by using an AMR algorithm, connecting a plurality of AMR graphs by combining a keyword, AMR root nodes and other modes to obtain an event text integral AMR graph, and encoding the AMR graph by using a GCN network to obtain a representation vector V of a hotline event text 2 ；
S114, processing each sentence of the hotline event text by using the BERT pre-training language model to generate word vectors of all words in the sentences, then using SIF weights to form the word vectors into a plurality of sentence vectors, and combining the sentence vectors into a representation vector V of the hotline event text by using a weighted average method according to the content hierarchical distribution of the text (such as the central content of the text is mainly concentrated in the first sentence or the last sentence, the sentence length and the like) 3 ；
S115, fusing three types of expression vectors V by adopting a gating mechanism 1 ，V 2 ，V 3 Obtaining a corresponding fusion vector V fuse And finally, carrying out event multi-label classification and event trigger word recognition on the fusion vector by using a fully connected network to obtain a corresponding multi-label event type and trigger word set.
The event role recognition process is as follows:
s121, performing jieba word segmentation on the event text to obtain a word segmentation result, and processing the event text by using a word embedding mode to obtain a corresponding coding result;
s122, using a BiLSTM+CRF model to carry out named entity recognition on the coded result, and recognizing all entity vectors in the hotline event text;
s123, for different event types, firstly, according to an embellishing table, finding an embellishing vector of a corresponding trigger word, and using a BiLSTM network to encode to obtain an event trigger word representation vector V t ；
S124, trigger word representing vector V corresponding to splicing event type t And entity vectors, calculating event roles of different event types.
In S200, based on the dispatching result of the historical hot line work order, information such as an event processing department and an event result can be obtained, information such as a trigger word, an event type, an event role and the like of a hot line event can be obtained according to the result of the event extraction algorithm, and a plurality of triples can be obtained by taking entities such as an event number, a department name, an event type name and the like as nodes:
"event-processing department-department name", "event-processing mode", "event-event type name", "event-trigger word name", "event-event role-role content";
the triples can obtain a relation diagram of a certain historical event, and the relation diagrams of the events are connected to obtain a knowledge graph of the hot-line work order event, wherein the same event can be associated with a plurality of types, a plurality of trigger words, a plurality of processing departments and the like, and the relation diagrams of the events are connected through information such as the event types, the trigger words, the processing departments and the like; the specific construction process comprises the following steps:
s201, processing the event role content, the processing result and other entities by adopting an entity alignment technology, discovering potential alignment entities, and renaming the potential alignment entities by using the same naming format;
s202, fusing the same event roles and event processing modes for the results after entity alignment, and associating different event relationship graphs through the event roles and the event processing modes.
According to the process, a knowledge graph based on the event key information can be constructed.
The dispatch department prediction process based on knowledge graph questions and answers in S300 is as follows:
s301, performing word segmentation processing on the designed problem, and performing feature extraction on a word segmentation result by adopting a BiLSTM network to obtain a corresponding problem expression vector;
s302, performing graph embedding processing on a multilayer neighborhood relation containing key entities (such as XX types and XX trigger words) in the problems in the knowledge graph by using a TransR to obtain corresponding embedded vectors;
s303, regarding the multi-layer neighborhood relation entity of the input entity as a candidate answer, predicting the probability scores of the questions and the relation entity, and selecting the entity with the highest probability score as the answer (dispatch department) of the questions;
s304, obtaining a plurality of candidate departments according to the designed general questions, setting a specific threshold value, and selecting an entity with probability score exceeding the given threshold value as an answer candidate set (an optional dispatch department and an associated department).
The invention and its embodiments have been described above by way of illustration and not limitation, but rather one of the embodiments of the invention is not limited thereto. Therefore, if one of ordinary skill in the art is informed by this disclosure, the structural mode and the embodiments similar to the technical scheme are not creatively designed without departing from the gist of the present invention.
Claims (1)
1. A hotline work order intelligent dispatching method based on a knowledge graph is characterized in that: comprising the following steps:
s100, extracting events based on historical heat line work order texts; including event detection and event role recognition, wherein the event detection process is as follows:
s111, sentence segmentation is carried out on the event text, and the jieba word segmentation is used for word segmentation processing to obtain documetteokens and sentencetoken;
s112, encoding the documetteokens by using a word embedding method, respectively extracting local features and global features of the hotline event text by using a CNN network and an RNN network, and then combining an Attention mechanism to obtain a representation vector V1 of the hotline event text;
s113, carrying out word embedding processing on sendencetokens, generating AMR graphs for each sentence by using an AMR algorithm, connecting a plurality of AMR graphs to obtain an integral AMR graph of an event text, and encoding the AMR graph by using a GCN network to obtain a representation vector V2 of the hotline event text;
s114, processing each sentence of the hotline event text by using the BERT pre-training language model, generating word vectors of all words in the sentences, then forming the word vectors into a plurality of sentence vectors by using SIF weights, and combining the sentence vectors into a representation vector V3 of the hotline event text by using a weighted average method according to the content hierarchical distribution of the text;
s115, fusing three types of expression vectors V1, V2 and V3 by adopting a gating mechanism to obtain a corresponding fusion vector Vuse, and finally carrying out event multi-label classification and event trigger word recognition on the fusion vector by using a fully-connected network to obtain a corresponding multi-label event type and trigger word set;
the event role recognition process is as follows:
s121, performing jieba word segmentation on the event text to obtain a word segmentation result, and processing the event text by using a word embedding mode to obtain a corresponding coding result;
s122, using a BiLSTM+CRF model to carry out named entity recognition on the coded result, and recognizing all entity vectors in the hotline event text;
s123, for different event types, firstly finding an embellishing vector of a corresponding trigger word according to the embellishing table, and encoding by using a BiLSTM network to obtain an event trigger word representation vector Vt;
s124, the trigger words corresponding to the spliced event types represent vectors Vt and entity vectors, and event roles of different event types are calculated;
s200, constructing a knowledge graph based on event key information; based on the dispatching result of the historical hot line work order, the related information of the hot line event can be obtained by combining an event extraction algorithm, and a plurality of event node triples can be obtained by taking a plurality of event information entities as nodes, wherein the event node triples comprise event-processing department-department names, event-processing modes, event-event types-event type names, event-trigger words-trigger word names and event-event role-role contents;
the knowledge graph construction process is as follows:
s201, processing event information entities by adopting an entity alignment technology, discovering potential alignment entities, and renaming the potential alignment entities by using the same naming format;
s202, fusing the same event roles and event processing modes for the results after entity alignment, and associating different event relationship graphs through the event roles and the event processing modes;
s300, predicting a dispatching department based on knowledge graph questions and answers;
the process is as follows:
s301, performing word segmentation processing on the designed problem, and performing feature extraction on a word segmentation result by adopting a BiLSTM network to obtain a corresponding problem expression vector;
s302, performing graph embedding processing on a multilayer neighborhood relation containing key entities in the problem in the knowledge graph by using a TransR to obtain a corresponding embedded vector;
s303, regarding the multi-layer neighborhood relation entity of the input entity as a candidate answer, predicting the probability scores of the questions and the relation entity, and selecting the entity with the highest probability score as the question answer;
s304, obtaining a plurality of candidate departments according to the designed general questions, setting a specific threshold value, and selecting an entity with probability score exceeding the given threshold value as an answer candidate set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211590719.6A CN116108169B (en) | 2022-12-12 | 2022-12-12 | Hot wire work order intelligent dispatching method based on knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211590719.6A CN116108169B (en) | 2022-12-12 | 2022-12-12 | Hot wire work order intelligent dispatching method based on knowledge graph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116108169A CN116108169A (en) | 2023-05-12 |
| CN116108169B true CN116108169B (en) | 2024-02-20 |
Family
ID=86258819
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211590719.6A Active CN116108169B (en) | 2022-12-12 | 2022-12-12 | Hot wire work order intelligent dispatching method based on knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116108169B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116795963A (en) * | 2023-05-25 | 2023-09-22 | 上海蜂威科技发展有限公司 | Hotline allocation method based on knowledge graph question and answer |
| CN117094688B (en) * | 2023-10-20 | 2023-12-19 | 国网信通亿力科技有限责任公司 | Digital control method and system for power supply station |
| CN117435697B (en) * | 2023-12-21 | 2024-03-22 | 中科雨辰科技有限公司 | Data processing system for acquiring core event |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110543562A (en) * | 2019-08-19 | 2019-12-06 | 武大吉奥信息技术有限公司 | Event map-based automatic urban management event distribution method and system |
| CN112632223A (en) * | 2020-12-29 | 2021-04-09 | 天津汇智星源信息技术有限公司 | Case and event knowledge graph construction method and related equipment |
| CN112699681A (en) * | 2020-12-17 | 2021-04-23 | 国网冀北电力有限公司信息通信分公司 | Power communication system defect fault order dispatching method and device based on knowledge graph |
| CN112836120A (en) * | 2021-01-27 | 2021-05-25 | 深圳大学 | Multi-mode knowledge graph-based movie recommendation method, system and terminal |
| CN113420117A (en) * | 2021-06-23 | 2021-09-21 | 北京交通大学 | Emergency classification method based on multivariate feature fusion |
| CN113590837A (en) * | 2021-07-29 | 2021-11-02 | 华中农业大学 | Deep learning-based food and health knowledge map construction method |
| CN113657091A (en) * | 2021-09-02 | 2021-11-16 | 长三角信息智能创新研究院 | Government affair hot line work order allocation method based on event extraction and authority list |
| CN113868427A (en) * | 2020-06-30 | 2021-12-31 | 中国移动通信集团天津有限公司 | Data processing method and device and electronic equipment |
| CN113919811A (en) * | 2021-10-15 | 2022-01-11 | 长三角信息智能创新研究院 | Hot line event distribution method based on strengthened correlation |
| CN114064918A (en) * | 2021-11-06 | 2022-02-18 | 中国电子科技集团公司第五十四研究所 | Multi-modal event knowledge graph construction method |
| CN114676233A (en) * | 2021-12-15 | 2022-06-28 | 清华大学 | Medical automatic question-answering method based on skeletal muscle knowledge graph |
| CN114936727A (en) * | 2022-01-18 | 2022-08-23 | 中国电信集团工会上海市委员会 | Work order distribution system, method and computer equipment |
| CN115034224A (en) * | 2022-01-26 | 2022-09-09 | 华东师范大学 | News event detection method and system integrating representation of multiple text semantic structure diagrams |
| CN115114409A (en) * | 2022-07-19 | 2022-09-27 | 中国民航大学 | Civil aviation unsafe event combined extraction method based on soft parameter sharing |
| CN115203532A (en) * | 2022-06-07 | 2022-10-18 | 深圳路演时刻网络数据有限公司 | Project recommendation method and device, electronic equipment and storage medium |
| CN115438995A (en) * | 2022-09-21 | 2022-12-06 | 青岛酷特智能股份有限公司 | Service processing method and equipment for garment customization enterprise based on knowledge graph |
| CN115470867A (en) * | 2022-10-17 | 2022-12-13 | 云南电网有限责任公司信息中心 | Agent matching method, device, equipment and storage medium based on knowledge graph |
-
2022
- 2022-12-12 CN CN202211590719.6A patent/CN116108169B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110543562A (en) * | 2019-08-19 | 2019-12-06 | 武大吉奥信息技术有限公司 | Event map-based automatic urban management event distribution method and system |
| CN113868427A (en) * | 2020-06-30 | 2021-12-31 | 中国移动通信集团天津有限公司 | Data processing method and device and electronic equipment |
| CN112699681A (en) * | 2020-12-17 | 2021-04-23 | 国网冀北电力有限公司信息通信分公司 | Power communication system defect fault order dispatching method and device based on knowledge graph |
| CN112632223A (en) * | 2020-12-29 | 2021-04-09 | 天津汇智星源信息技术有限公司 | Case and event knowledge graph construction method and related equipment |
| CN112836120A (en) * | 2021-01-27 | 2021-05-25 | 深圳大学 | Multi-mode knowledge graph-based movie recommendation method, system and terminal |
| CN113420117A (en) * | 2021-06-23 | 2021-09-21 | 北京交通大学 | Emergency classification method based on multivariate feature fusion |
| CN113590837A (en) * | 2021-07-29 | 2021-11-02 | 华中农业大学 | Deep learning-based food and health knowledge map construction method |
| CN113657091A (en) * | 2021-09-02 | 2021-11-16 | 长三角信息智能创新研究院 | Government affair hot line work order allocation method based on event extraction and authority list |
| CN113919811A (en) * | 2021-10-15 | 2022-01-11 | 长三角信息智能创新研究院 | Hot line event distribution method based on strengthened correlation |
| CN114064918A (en) * | 2021-11-06 | 2022-02-18 | 中国电子科技集团公司第五十四研究所 | Multi-modal event knowledge graph construction method |
| CN114676233A (en) * | 2021-12-15 | 2022-06-28 | 清华大学 | Medical automatic question-answering method based on skeletal muscle knowledge graph |
| CN114936727A (en) * | 2022-01-18 | 2022-08-23 | 中国电信集团工会上海市委员会 | Work order distribution system, method and computer equipment |
| CN115034224A (en) * | 2022-01-26 | 2022-09-09 | 华东师范大学 | News event detection method and system integrating representation of multiple text semantic structure diagrams |
| CN115203532A (en) * | 2022-06-07 | 2022-10-18 | 深圳路演时刻网络数据有限公司 | Project recommendation method and device, electronic equipment and storage medium |
| CN115114409A (en) * | 2022-07-19 | 2022-09-27 | 中国民航大学 | Civil aviation unsafe event combined extraction method based on soft parameter sharing |
| CN115438995A (en) * | 2022-09-21 | 2022-12-06 | 青岛酷特智能股份有限公司 | Service processing method and equipment for garment customization enterprise based on knowledge graph |
| CN115470867A (en) * | 2022-10-17 | 2022-12-13 | 云南电网有限责任公司信息中心 | Agent matching method, device, equipment and storage medium based on knowledge graph |
Non-Patent Citations (7)
| Title |
|---|
| Gated Fully Fusion for Semantic Segmentation;Xiangtai Li 等;Proceedings of the AAAI conference on artificial intelligence;11418-11425 * |
| Li-Yuan Geng ; Wei Jin ; Han-Bing Qu.Emergency Events Classification Based on Minsheng Hotline Unbalanced Short-Text.《2017 IEEE International Conference on Big Knowledge(ICBK)》.2017,188-191. * |
| 利用门控机制融合依存与语义信息的事件检测方法;陈佳丽;洪宇;王捷;张婧丽;姚建民;;中文信息学报(第08期);51-60 * |
| 国内基于知识图谱的信息推荐研究进展;孙雨生;祝博;朱礼军;;情报理论与实践(第12期);163-169+149 * |
| 基于依存句法分析与分类器融合的触发词抽取方法;高源;席耀一;李弼程;;《计算机应用研究》;20160531;第 33 卷(第 5 期);1407-1410 * |
| 基于知识图谱的电信行业人才管理应用研究;许经伟;李雪迪;王文学;;《信息通信》;20200915(第 9 期);219-221 * |
| 融合RoBERTa和特征提取的政务热线工单分类;陈钢;《计算机与现代化》;20220630(第 6 期);21-26+31 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116108169A (en) | 2023-05-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116108169B (en) | Hot wire work order intelligent dispatching method based on knowledge graph | |
| CN111046132B (en) | Customer service question-answering processing method and system for searching multiple rounds of conversations | |
| CN111026842B (en) | Natural language processing method, natural language processing device and intelligent question-answering system | |
| CN112347222B (en) | Method and system for converting non-standard address into standard address based on knowledge base reasoning | |
| CN113392986B (en) | Highway bridge information extraction method based on big data and management maintenance system | |
| CN112581106B (en) | Government affair event automatic order dispatching method fusing grid semantics of handling organization | |
| CN105069047B (en) | A kind of search method and device of geography information | |
| CN109376227A (en) | A kind of prison term prediction technique based on multitask artificial neural network | |
| CN113312501A (en) | Construction method and device of safety knowledge self-service query system based on knowledge graph | |
| CN104464733A (en) | Multi-scene managing method and device of voice conversation | |
| Choi et al. | Conceptual cost-prediction model for public road planning via rough set theory and case-based reasoning | |
| CN109165275B (en) | Intelligent substation operation ticket information intelligent search matching method based on deep learning | |
| CN107145991B (en) | Time-varying random network dynamic path searching method considering road section correlation | |
| CN113159403B (en) | Intersection pedestrian track prediction method and device | |
| CN115017425B (en) | Location search method, location search device, electronic device, and storage medium | |
| CN111524353A (en) | Method for traffic text data for speed prediction and travel planning | |
| CN116484024A (en) | Multi-level knowledge base construction method based on knowledge graph | |
| CN115330268A (en) | Comprehensive emergency command method and system for dealing with mine disaster | |
| CN115100395A (en) | Urban block function classification method integrating POI pre-classification and graph neural network | |
| CN115953080B (en) | Engineer service level determination method, apparatus and storage medium | |
| CN115017144B (en) | Judicial document case element entity identification method based on graphic neural network | |
| CN113657091A (en) | Government affair hot line work order allocation method based on event extraction and authority list | |
| CN110096506B (en) | Tree cell structure description and storage method for multi-layer requirements | |
| Gajzler | Supporting the technical management of residential buildings in the process of their exploitation | |
| Bannur et al. | General transit feed specification assisted effective traffic congestion prediction using decision trees and recurrent neural networks |
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 |