WO2021196520A1 - 一种面向税务领域知识图谱的构建方法及系统 - Google Patents
一种面向税务领域知识图谱的构建方法及系统 Download PDFInfo
- Publication number
- WO2021196520A1 WO2021196520A1 PCT/CN2020/113308 CN2020113308W WO2021196520A1 WO 2021196520 A1 WO2021196520 A1 WO 2021196520A1 CN 2020113308 W CN2020113308 W CN 2020113308W WO 2021196520 A1 WO2021196520 A1 WO 2021196520A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- tax
- knowledge
- data
- information
- graph
- Prior art date
Links
- 238000010276 construction Methods 0.000 title claims abstract description 22
- 238000000605 extraction Methods 0.000 claims abstract description 43
- 238000010586 diagram Methods 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims abstract description 17
- 230000004927 fusion Effects 0.000 claims abstract description 9
- 238000000275 quality assurance Methods 0.000 claims abstract description 7
- 238000000034 method Methods 0.000 claims description 57
- 239000013598 vector Substances 0.000 claims description 17
- 239000000284 extract Substances 0.000 claims description 13
- 238000005516 engineering process Methods 0.000 claims description 10
- 230000010354 integration Effects 0.000 claims description 8
- 238000013179 statistical model Methods 0.000 claims description 7
- 238000013135 deep learning Methods 0.000 claims description 5
- 238000004519 manufacturing process Methods 0.000 claims description 5
- 238000013075 data extraction Methods 0.000 claims description 2
- 230000006698 induction Effects 0.000 claims description 2
- 238000007689 inspection Methods 0.000 claims description 2
- 238000002372 labelling Methods 0.000 claims description 2
- 238000010801 machine learning Methods 0.000 claims description 2
- 238000003860 storage Methods 0.000 claims description 2
- 239000003814 drug Substances 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000012550 audit Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/10—Tax strategies
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Definitions
- the invention belongs to the field of taxation technology, and in particular relates to a method and system for constructing a knowledge graph in the taxation field.
- the existing tax service system requires more and more customized services and management of taxpayers, and human resources are becoming more and more tense, making it difficult to reduce the burden.
- the existing tax information system collects a large amount of taxpayer data, and analyzes the data according to the existing model to obtain information results, but the information is poor in interpretability and is not friendly to grassroots tax personnel. From the perspective of taxpayers, on the one hand, taxation policies are highly professional. Ordinary taxpayers cannot accurately understand the content of the policy, and can only understand the specific content of the policy through expert interpretation.
- the knowledge graph technology proposed by Google in May 2012 can express Internet information in a form closer to the human cognitive world, providing a better ability to organize, manage, and understand the massive amount of information on the Internet.
- Knowledge graph technology extracts entities and their attribute information and relationships between entities from web pages, and interprets massive amounts of data and knowledge from a semantic level.
- the existing knowledge graphs are general domain knowledge graphs such as Baidu Zhixin, Sogou Zhilifang, etc.
- Baidu Zhixin Baidu Zhixin
- Sogou Zhilifang etc.
- Related research For this reason, referring to the construction methods of knowledge graphs in other fields has certain reference significance for solving the problem of intelligent tax construction.
- Literature 1 Construction and Application of Traditional Chinese Medicine Knowledge Graph[J].Journal of Medical Informatics,2016,37(04):8-13;
- Literature 2 A method for constructing conceptual knowledge graphs in the field of water affairs based on DBpedia (201910161944.X).
- Literature 1 provides a method for constructing a knowledge graph using structured information of traditional Chinese medicine in the field of traditional Chinese medicine. This method is mainly based on the existing medical data set combined with the structured information of traditional Chinese medicine in the relational database and other data source information. Medical knowledge model, complete the knowledge map construction.
- Literature 2 provides a method of artificially constructing a conceptual dictionary and a general data set to construct a domain knowledge graph.
- the method of using the above documents mainly has the following problems: the data sources used in document 1 are mostly structured and mature medical data sets, which cannot be processed with unlabeled unstructured data; document 2 combines general data sets Constructing a domain knowledge graph cannot meet the requirements of the professional domain for the depth of knowledge.
- the present invention provides a method and system for constructing a knowledge graph for the tax field.
- the construction method adopts a combination of top-down and bottom-up.
- Top-down is to focus on the knowledge ontology structure through the expert experience of the tax expert system, and the conceptual model is designed through the definition of the pattern diagram through the ontology editor;
- bottom-up is to use big data technology to combine small knowledge and big data into big knowledge , Focus on examples of knowledge content and define data graphs.
- the present invention adopts the following technical solutions:
- a method and system for constructing a knowledge graph in the tax field mainly includes the following modules: pattern editing module, data processing module, information extraction module, fusion disambiguation module, quality assurance module, and knowledge service module.
- the process of the method is as follows: First, use the ontology sentence to construct the tax knowledge graph model based on the tax knowledge of the tax expert system in the pattern editing module; then perform data processing in the data processing module according to the designed tax knowledge graph model, including tax data sources The selection and acquisition of data, the cleaning of data, etc.; then the data obtained in the data processing module is used as input in the information extraction module, and the processed data is extracted according to different types according to the pattern diagram; then the extracted information needs to be in The fusion disambiguation module adopts the tax knowledge fusion method.
- the knowledge sources in the tax knowledge graph are different, and there are problems such as knowledge duplication and relationship redundancy, which require pattern matching and entity alignment to complete the knowledge fusion and store it in the knowledge base.
- a quality assurance module is added for knowledge feedback, and the tax expert system is used to resolve the knowledge conflicts in the construction of an intelligent tax model.
- the application of the system focuses on the follow-up knowledge service module to provide tax knowledge intelligent recommendation and tax question answering services.
- the present invention adopts the following technical solutions to achieve:
- Step 1 Construct a tax knowledge graph pattern diagram
- the tax expert system determines the overall concept of the tax knowledge graph based on the input tax knowledge, and formally defines the tax knowledge graph as a graph G, G ⁇ G s , G d , R>, which consists of the tax model graph G s , tax
- the data graph G d and the relationship between the two are composed of R
- the tax model graph is a diagram describing various abstract concepts in the tax field and their mutual relations.
- the formal definition is G s ⁇ N s , E s >, where N s is the set of nodes in the graph representing the abstract tax concept in the tax map, and E s is the set of attribute edges representing the semantic relationship between the concepts, and then the determined tax abstract concept is filled into the tax model map according to the hierarchical relationship;
- the tax data contained in the tax knowledge map comes from invoice information, basic taxpayer information, Chinese tax websites, national tax law textbooks, national tax term collections, and tax preferential cases; these tax data include structured data, semi-structured data and unstructured data Data is collected, sorted, stored and cleaned according to data types;
- Information extraction takes processed data as input, and the goal of extraction is to extract triples of form E, or entity, relationship, and attribute/entity; extraction methods formulated according to different data sources include: rule-based methods and Based on statistical models and deep learning methods;
- the construction of the tax knowledge graph is a process of continuous iteration and update. Due to different data sources and different knowledge bases, the data in the tax knowledge graph is diversified and heterogeneous; from step 2, the invoice information and basic taxpayer information are obtained Use the above steps as data to construct the enterprise production and operation knowledge graph and the industrial chain graph as a subgraph of the tax knowledge graph;
- Step 5 Tax knowledge feedback
- step 2 includes the following steps:
- Step 201 Store structured data, namely invoices and basic taxpayer information in a relational database
- the taxpayer’s basic information table depicts the detailed status of the company’s production and operation, and the invoice reflects the facts of the transaction relationship between taxpayers, that is, the flow of the industry chain; first, the taxpayer’s basic information is stored in the Mysql database, and then the taxpayer’s basic information
- the id field corresponds to the foreign key of the invoice and is stored in the Mysql database according to the specifications;
- Step 202 Crawl and collect relevant website URLs about Chinese taxation
- Step 203 collect and sort out national tax law textbooks, national tax term collections, and tax preferential cases
- step 3 The specific process of information extraction in step 3 is as follows:
- Step 301 Use a rule-based method to extract information on Chinese tax webpages and national tax law textbooks
- Step 302 extract using methods based on statistical models and deep learning
- Entity extraction is performed first, using tax preference cases and unstructured data in other taxation as input. If there is labeled data, conditional random field models, hidden Markov models, and maximum entropy model statistical models can be used to extract information. For labeling data, you can use two-way LSTM-CRF and two-way LSTM-CNNs-CRF to directly take word vectors as input, and output new vectors of words in an end-to-end manner, and then output word recognition results through the CRF layer; then use regular expression extraction Hierarchical structure or extraction of relationships through unlabeled remote supervision, and finally attribute extraction. For tax entities such as taxpayers, the extracted content includes business scope, credit rating, and risk score.
- the present invention has the following beneficial effects:
- the present invention can well solve the problem of difficulty in processing knowledge graph data in the tax field.
- Tax data mainly comes from Chinese tax policy-related websites, national tax law textbooks, national taxation bureau terminology collections, and specific cases of collecting and sorting out tax incentives.
- the characteristics of the data are the coexistence of structured data, semi-structured data and unlabeled unstructured data. Therefore, there are two solutions to data in a method for constructing a knowledge map for the tax field proposed by the present invention: one is a rule-based tax triplet extraction method; the other is a tax triplet based on statistical models and deep learning. Group extraction method. It has achieved the effect of accurately processing data and improving the quality of map construction.
- the present invention can be applied to a variety of complex tax scenarios and has strong adaptability.
- tax business scenarios are complex, and a single tax knowledge map cannot cover all application scenarios. Therefore, in the process of constructing an intelligent tax model based on the knowledge map, a tax sub-graph method is proposed to automatically construct different sub-graphs for different business scenarios. Map, and multiple sub-maps can be automatically updated and merged into a large tax knowledge map. It has achieved the effect of providing personalized services based on different tax scenarios.
- Fig. 1 is a flowchart of a method for constructing a knowledge map for the tax field according to the present invention
- Figure 2 is the definition diagram of the tax knowledge map
- Figure 3 is a flow chart of tax data preprocessing
- Figure 4 is a flowchart of tax data information extraction.
- Fig. 5 is a structural block diagram of a system for constructing a knowledge map for the tax field designed by the present invention.
- Fig. 1 shows a flowchart of a method for constructing a tax domain knowledge graph provided by an embodiment of the present invention. As shown in Fig. 1, in this embodiment, the present invention provides a tax domain knowledge
- the method of constructing the atlas includes the following steps:
- Step 1 Formulation of tax knowledge map model diagram
- the expert system formulates the definition rules of the pattern diagram as the input of the pattern editing module. Select the representative keywords in the tax field and the semantic relationship between them. The most important relationship is the parent-child relationship, which is reflected in the inheritance relationship.
- the schema diagram follows the RDF framework standard, and the representation of the relationship is rdfs: subclassof represents the inheritance relationship between the two. The side pointed to by the arrow is the inherited object
- the abstract concepts of taxation are as follows: transaction, transaction is the collective name of all entities in the field; the subcategories of transaction include taxation and media, taxation is the construction object of this patent, and the media is the interaction of entities
- the object of transmission; the tax field can be roughly divided into two categories: taxation and tax law; tax law includes tax-related policies and regulations, etc.; media includes people, organizations, and objects; taxpayers include natural persons, taxpayers, legal persons, Unincorporated persons; unincorporated persons include partnerships and sole proprietorships.
- the tax field also has the tax calculation relationship (tax) and the type relationship (type), expressed in the form of rdfs:tax and rdf:type.
- the tax data graph describes the specific facts in the knowledge graph.
- the nodes represent instance nodes and attribute values, and the edges represent the relationship between instance nodes and attribute values.
- the formal expression is G d ⁇ N d , E d >.
- N d is the set of nodes, and E d is the set of edges.
- the relationship between the schema diagram and the data diagram is represented by rdf:type, which represents the relationship between the instance in the data diagram and the concept to which it belongs.
- the fact is that the enterprise A is the entity of the taxpayer. , Enterprise A meets the requirement of the vacancy cancellation rule with a credit rating of A, which is represented in the data graph as shown in Figure 2.
- the data target is the tax data of Zhejiang Province
- the tax knowledge graph pattern obtained by the pattern editing module is used as the input of the data processing module to determine the type of data to be prepared. The specific steps are shown in Figure 3:
- the taxpayer basic information table contains the following fields ⁇ NSRDZDAH, NSRSBH, SHXYDM, NSRMC, NSRBM, HY_ID, HY_DM, HYMC, MXHY_ID, MXHY_DM ⁇ , which means ⁇ taxpayer electronic file number, taxpayer identification number, social credit code, taxpayer name, taxpayer code, industry serial number, industry code, industry name, detailed industry, detailed industry code ⁇ .
- the taxpayer’s electronic file number field in the taxpayer’s basic information table is used as the foreign key of the invoice information table.
- the invoice information table contains the following fields ⁇ FP_ID, FPHM, GFNSRDZDAH, XFNSRDZDAH, JE, SE ⁇ , which represent the meaning of ⁇ invoice number , Invoice goods, electronic identification number of the taxpayer of the purchaser, electronic identification number of the taxpayer of the seller, amount, tax amount ⁇ .
- Use python crawler to crawl the URL of Chinese tax website, save the URL in the form of a list as a txt file, and then use the PageRank algorithm to rank the importance and relevance of the list pages, and select the top ten websites as the semi-structured data data source.
- the data processed in step 2 is used as input, that is, the word vector result of the data processing module is used as the input of the information extraction module, and the tax information is extracted according to the data type according to the business scenario.
- Specific steps are as follows:
- Process structured data Zhejiang province invoice data and taxpayer basic information table extract transaction information to confirm that the company id and transaction relationship are transaction details, that is, the transaction product.
- the last step is to process unstructured data.
- tax preference cases and unstructured data such as text information generated in the process of extracting other types of data are included. Specific steps are as follows:
- the word vector of the text obtained in the previous step is used as input, and the system is shown as the tax entity and relation word vector obtained after information extraction.
- the semantic model uses singular value decomposition technology to decompose the vector space to obtain semantic features, and then takes the words near the name word as the feature vector and uses the vector cosine similarity comparison, that is, the method of combining clustering and semantic similarity Calculate the entity similarity to achieve the purpose of entity disambiguation.
- Step 5 Tax knowledge feedback
- the data of this basic model is passed as input to the quality assurance module, that is, the expert system.
- the expert system determines the problems in the model, marks the problem data, and gives solutions . Finally achieve the purpose of ensuring the quality of the model.
- the tax knowledge graph model output by the quality assurance module is input into the knowledge service module as the final model result, and the knowledge service module is used as the carrier to realize various services such as personalized tax knowledge recommendation and tax knowledge question and answer.
- a taxation domain-oriented knowledge graph construction and the system includes:
- Mode editing module used to obtain knowledge in the tax field to formulate tax mode diagrams.
- Data processing module used to process the required data extracted from the web page and the data in the database.
- Information extraction module used to extract tax entity, attribute and relationship information from the processed data.
- Fusion disambiguation module used to align the extracted entities with the original knowledge graph, eliminate redundancy, and update entities.
- Quality assurance module used to feed back knowledge, ensure knowledge quality, and resolve knowledge conflicts in construction.
- Knowledge service module used for knowledge prediction, tax inspection, tax question and answer services.
- the information extraction module includes a structured information extraction sub-module, a semi-structured information extraction sub-module, and an unstructured information extraction sub-module:
- the structured information extraction sub-module is used to directly map existing structured data or use R2RML to map RDF data.
- the semi-structured information extraction sub-module is used to extract existing webpage tax entity values and attribute value mapping relationships.
- the conversion of the unstructured information extraction sub-module into word vectors relies on a remote supervised learning algorithm for extracting key information from the text corpus.
- Knowledge service module According to the business scenarios of the tax knowledge graph, it is divided into the production and operation knowledge sub-graph, the industrial chain knowledge sub-graph, and the enterprise preferential sub-graph. According to the different sub-graphs, it provides tax audit, tax Q&A, and preferential policy intelligent recommendation services.
- the embodiment of the present invention provides a knowledge map system oriented to the tax field, which completes the task of building a tax knowledge map by collecting and processing invoice information tables, taxpayer information tables, and tax policy web pages, using information extraction, and knowledge fusion and disambiguation methods. Intelligent recommendation service for Q&A and preferential policies.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Development Economics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Economics (AREA)
- Marketing (AREA)
- Technology Law (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Educational Administration (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
Claims (5)
- 一种面向税务领域知识图谱的构建方法,其特征在于,包括以下步骤:步骤1,构建税务知识图谱模式图首先税务专家系统根据输入的税务知识确定税务知识图谱整体的概念,将税务知识图谱形式化定义为一张图G,G≤G s,G d,R>,其由税务模式图G s、税务数据图G d以及二者之间的关系R组成;税务模式图是描述税务领域中各种抽象概念及其相互之间关系的图,形式化定义为G s≤N s,E s>,其中N s为图中的结点集合代表税务图谱中税务抽象概念,E s为属性边集合代表概念之间的语义关系,之后将确定的税务抽象概念按照层次关系填入税务模式图中;步骤2,数据处理税务知识图谱中包含的税务数据来源于发票信息、纳税人基本信息、中国税务网站、国家税法教材、国家税务术语集以及税收优惠案例;这些税务数据包含结构化数据、半结构化数据和非结构化数据,按照数据类型收集、整理、存储和清洗数据;步骤3,税务数据信息抽取信息抽取将处理后的数据作为输入,抽取的目标是抽取出形为E,或实体、关系和属性/实体的三元组;根据数据源的不同制定的抽取方法有:有基于规则的方法和基于统计模型和深度学习的方法;步骤4,税务知识融合税务知识图谱的构建是一个不断迭代不断更新的过程,由于数据源不同、知识库不同导致税务知识图谱中的数据具有多样性和异构性;由步骤2,得到的发票信息和纳税人基本信息作为数据采用上述步骤构建出企业生产经营知识图谱和产业链图谱,作为税务知识图谱的子图;步骤5,税务知识反馈当构建税务知识图谱过程中出现数据冲突、知识质量难以确定以及知识无法抽取问题时税务知识反馈具体的方法是将问题收集起来分门别类转发到专家系统中,由税务专家系统给出解决方案从而保证知识库的质量。
- 根据权利要求1所述的一种面向税务领域知识图谱的构建方法,其特征在于,步骤2中的具体实现方法包括以下步骤:步骤201,将结构化数据即发票、纳税人基本信息存储到关系型数据库中纳税人基本信息表刻画出企业生产经营的详细状况,发票反映纳税人之间的交易关系事实即产业链流动情况;首先将纳税人基本信息存储到Mysql数据库中,之后将于纳税人基本信息中的id字段对应于发票的外键按照规格存储到Mysql数据库中;步骤202,爬取搜集关于中国税务的相关网站网址首先利用爬虫技术搜集所有有关于中国税务相关网站的网址,之后将这些网址按照信任度等级排序,最后去除信任度低于80%的网址并存储;步骤203,收集整理国家税法教材、国家税务术语集以及税收优惠案例首先将统一所有非结构化数据的格式将其转换成文本文件存储,然后根据制定的模式图使用人工标注,最后将国家税务术语集使用Bert工具将文字预训练产生词向量文件。
- 根据权利要求2所述的一种面向税务领域知识图谱的构建方法,其特征在于,步骤3中的信息抽取的具体流程如下:步骤301,使用基于规则的方法抽取中国税务网页信息、国家税法教材信息将数据处理中得到的信任度较高的中国税务网站的网址作为输入,使用有监 督的机器学习技术,学习每个网站中标注好的网页的数据抽取规则,即包装器归纳法,抽取出税务关键词、税务关系词和税务属性词,从而对相似结构的web页面直接抽取出所需的三元组信息;接着,学习税法教材的半结构化信息比如章节标题、段落标题和层级关系学习到抽取规则,之后抽取所需的税务概念三元组信息;步骤302,使用基于统计模型和深度学习的方法抽取首先进行实体抽取,将税收优惠案例以及其他税务中的非结构化数据作为输入,如果有标注的数据则能够使用条件随机场模型、隐马尔可夫模型和最大熵模型统计模型抽取信息,如果没有标注数据则可以使用双向LSTM-CRF与双向LSTM-CNNs-CRF直接将词向量作为输入,通过端到端的方式输出词的新的向量再经过CRF层输出词的识别结果;接着使用正则表达式抽取分层结构或是通过无标签远程监督的方法抽取关系,最后属性抽取对于税务实体比如纳税人,抽取的内容有经营范围、信用等级和风险分值。
- 根据权利要求3所述的一种面向税务领域知识图谱的构建方法,其特征在于,子图与税务知识融合的具体方法如下:首先对相似字符串计算编辑距离计算属性相似度,然后根据属性相似度采用回归或者聚类的方法计算实体相似度,最终达到税务知识融合的目的。
- 一种面向税务领域知识图谱构建系统,该系统包括:模式编辑模块:用于获取税务领域知识制定税务模式图。数据处理模块:用于处理网页提取出的所需数据和数据库中的数据。信息抽取模块:用于将所述处理后的数据提取税务实体、属性和关系信息。融合消歧模块:用于将抽取实体与原有知识图谱对齐、消除冗余、更新实体。 质量保证模块:用于反馈知识、保证知识质量,解决构建中的知识冲突。知识服务模块:用于知识预测、提供税务稽查、税收问答服务。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010238326.3 | 2020-03-30 | ||
CN202010238326.3A CN111428053B (zh) | 2020-03-30 | 2020-03-30 | 一种面向税务领域知识图谱的构建方法 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2021196520A1 true WO2021196520A1 (zh) | 2021-10-07 |
Family
ID=71549940
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2020/113308 WO2021196520A1 (zh) | 2020-03-30 | 2020-09-03 | 一种面向税务领域知识图谱的构建方法及系统 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN111428053B (zh) |
WO (1) | WO2021196520A1 (zh) |
Cited By (55)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113269331A (zh) * | 2021-04-25 | 2021-08-17 | 云南电网有限责任公司信息中心 | 一种变压器检修作业的知识库构建方法 |
CN113902533A (zh) * | 2021-10-11 | 2022-01-07 | 税安科技(杭州)有限公司 | 一种适用于财税领域指标自定义和自动化运行的应用 |
CN113918677A (zh) * | 2021-10-18 | 2022-01-11 | 智联(无锡)信息技术有限公司 | 基于知识图谱自动化链路分层的数据处理方法、装置及计算机可读介质 |
CN114055451A (zh) * | 2021-11-24 | 2022-02-18 | 深圳大学 | 基于知识图谱的机器人操作技能表达方法 |
CN114186759A (zh) * | 2022-02-16 | 2022-03-15 | 杭州杰牌传动科技有限公司 | 基于减速机知识图谱的物料调度控制方法及其系统 |
CN114201619A (zh) * | 2022-02-18 | 2022-03-18 | 中国电子技术标准化研究院 | 多层结构标准知识图谱构建、标准检索方法及装置 |
CN114417020A (zh) * | 2022-03-29 | 2022-04-29 | 浙江省标准化研究院(金砖国家标准化(浙江)研究中心 浙江省物品编码中心) | 一种产业链图谱构建系统及方法 |
CN114547324A (zh) * | 2021-12-31 | 2022-05-27 | 华中农业大学 | 食品安全国家标准领域本体的构建方法、系统和介质 |
CN114580418A (zh) * | 2022-03-04 | 2022-06-03 | 河北师范大学 | 一种警察体能训练知识图谱系统 |
CN114647743A (zh) * | 2022-05-20 | 2022-06-21 | 国网浙江省电力有限公司 | 电力营销全业务门禁规则图谱生成及处理方法、装置 |
CN114676840A (zh) * | 2022-01-27 | 2022-06-28 | 广州数说故事信息科技有限公司 | 一种基于海量数据的因果事理组提取方法、系统及计算机可读存储介质 |
CN114780083A (zh) * | 2022-06-17 | 2022-07-22 | 之江实验室 | 一种知识图谱系统的可视化构建方法及装置 |
CN114896426A (zh) * | 2022-07-14 | 2022-08-12 | 中国人民解放军国防科技大学 | 一种电子目标认知图谱的构建方法 |
CN114912637A (zh) * | 2022-05-21 | 2022-08-16 | 重庆大学 | 人机物知识图谱制造产线运维决策方法及系统、存储介质 |
CN114936295A (zh) * | 2022-07-21 | 2022-08-23 | 北斗天下卫星导航有限公司 | 一种基于领域知识图谱的通用热点分析方法和系统 |
CN114969548A (zh) * | 2022-06-24 | 2022-08-30 | 天津联创科技发展有限公司 | 一种产业图谱数据智能获取方法及系统 |
CN115079588A (zh) * | 2022-08-05 | 2022-09-20 | 福州大学 | 基于时空数据驱动的智能家居服务管控方法 |
CN115114458A (zh) * | 2022-08-29 | 2022-09-27 | 北京北投智慧城市科技有限公司 | 一种基于智慧建筑知识图谱的设备控制方法及系统 |
CN115203436A (zh) * | 2022-07-15 | 2022-10-18 | 国网江苏省电力有限公司信息通信分公司 | 一种基于有向图数据融合的电力知识图谱构建方法和装置 |
CN115510245A (zh) * | 2022-10-14 | 2022-12-23 | 北京理工大学 | 一种面向非结构化数据的领域知识抽取方法 |
CN115545799A (zh) * | 2022-11-04 | 2022-12-30 | 北京赛西科技发展有限责任公司 | 信息技术服务质量评估方法、装置、设备及介质 |
CN115759256A (zh) * | 2022-11-24 | 2023-03-07 | 中安华邦(北京)安全生产技术研究院股份有限公司 | 一种安全生产数字化知识库构建方法、系统、介质及设备 |
CN116010587A (zh) * | 2023-03-23 | 2023-04-25 | 中国人民解放军63921部队 | 航天测发保障性条件知识推送方法、装置、介质及设备 |
CN116010583A (zh) * | 2023-03-17 | 2023-04-25 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | 一种级联耦合的知识增强对话生成方法 |
CN116136861A (zh) * | 2023-04-18 | 2023-05-19 | 中国电子科技集团公司第十研究所 | 基于知识图谱的分布式多源异构数据管理系统及方法 |
WO2023088249A1 (zh) * | 2021-11-18 | 2023-05-25 | 华为技术有限公司 | 一种数据处理的合规性检测方法、装置和相关设备 |
CN116611903A (zh) * | 2023-07-18 | 2023-08-18 | 太仓市律点信息技术有限公司 | 基于数字金融服务的大数据处理方法、服务器及存储介质 |
CN116701665A (zh) * | 2023-08-08 | 2023-09-05 | 滨州医学院 | 基于深度学习的中医古籍知识图谱构建方法 |
CN116702899A (zh) * | 2023-08-07 | 2023-09-05 | 上海银行股份有限公司 | 一种适用于公私联动场景的实体融合方法 |
CN116701643A (zh) * | 2022-12-15 | 2023-09-05 | 重庆交通大学 | 基于社会传感数据的人群出行行为解析方法及存储介质 |
CN116719955A (zh) * | 2023-08-09 | 2023-09-08 | 北京国电通网络技术有限公司 | 标签标注信息生成方法、装置、电子设备和可读介质 |
CN116756396A (zh) * | 2023-06-29 | 2023-09-15 | 广东齐峰信息科技有限公司 | 基于知识图谱的数字档案管理系统及方法 |
CN116821376A (zh) * | 2023-08-30 | 2023-09-29 | 北京华琦远航国际咨询有限公司 | 煤矿安全生产领域的知识图谱构建方法及系统 |
CN116821712A (zh) * | 2023-08-25 | 2023-09-29 | 中电科大数据研究院有限公司 | 非结构化文本与知识图谱的语义匹配方法及装置 |
CN116842099A (zh) * | 2023-05-09 | 2023-10-03 | 中电科大数据研究院有限公司 | 一种多源异构数据处理方法和系统 |
CN116955589A (zh) * | 2023-09-19 | 2023-10-27 | 山东山大鸥玛软件股份有限公司 | 一种基于教材知识图谱的智能命题方法、系统、命题终端及存储介质 |
CN116992959A (zh) * | 2023-06-02 | 2023-11-03 | 广州数说故事信息科技有限公司 | 一种基于知识图谱的食品类产品创意概念生成验证方法 |
CN117112806A (zh) * | 2023-10-12 | 2023-11-24 | 北京大学深圳研究生院 | 一种基于知识图谱的信息结构化方法和装置 |
CN117131245A (zh) * | 2023-10-27 | 2023-11-28 | 江苏数兑科技有限公司 | 一种运用知识图谱技术实现目录资源推荐机制的方法 |
WO2023225960A1 (en) * | 2022-05-26 | 2023-11-30 | Siemens Aktiengesellschaft | Industrial data processing method and apparatus for edge device |
CN117150049A (zh) * | 2023-09-15 | 2023-12-01 | 上海歆广数据科技有限公司 | 一种个案图谱架构系统 |
CN117151122A (zh) * | 2023-10-30 | 2023-12-01 | 湖南三湘银行股份有限公司 | 一种基于自然语言处理的银行客服会话问答处理方法及系统 |
CN117236432A (zh) * | 2023-09-26 | 2023-12-15 | 中国科学院沈阳自动化研究所 | 一种面向多模态数据的制造工艺知识图谱构建方法及系统 |
CN117313855A (zh) * | 2023-11-28 | 2023-12-29 | 支付宝(杭州)信息技术有限公司 | 规则决策方法和装置 |
CN117312565A (zh) * | 2023-11-28 | 2023-12-29 | 山东科技大学 | 一种基于关系融合与表示学习的文献作者姓名消歧方法 |
CN117391313A (zh) * | 2023-12-12 | 2024-01-12 | 广东正迪科技股份有限公司 | 基于ai的智能决策方法、系统、设备以及介质 |
CN117473431A (zh) * | 2023-12-22 | 2024-01-30 | 青岛民航凯亚系统集成有限公司 | 一种基于知识图谱的机场数据分类分级方法及系统 |
CN117520796A (zh) * | 2023-11-23 | 2024-02-06 | 交通运输部规划研究院 | 一种基于知识图谱的公路路基水毁评估方法及系统 |
CN117633518A (zh) * | 2024-01-25 | 2024-03-01 | 北京大学 | 一种产业链构建方法及系统 |
CN117708680A (zh) * | 2024-02-06 | 2024-03-15 | 青岛海尔科技有限公司 | 一种用于提升分类模型准确度的方法及装置、存储介质、电子装置 |
CN117743558A (zh) * | 2024-02-20 | 2024-03-22 | 青岛海尔科技有限公司 | 基于大模型的知识加工、知识问答方法、装置及介质 |
CN117909492A (zh) * | 2024-03-19 | 2024-04-19 | 国网山东省电力公司信息通信公司 | 一种电网非结构化信息抽取方法、系统、设备及介质 |
CN117973520A (zh) * | 2024-03-29 | 2024-05-03 | 山东云力信息科技有限公司 | 基于大数据可视化的智慧社区知识图谱的构建方法 |
CN118075267A (zh) * | 2024-04-22 | 2024-05-24 | 青岛青软晶尊微电子科技有限公司 | 一种基于设计电路的数字后端服务平台构建方法 |
CN116204660B (zh) * | 2023-03-28 | 2024-06-11 | 北京航空航天大学 | 一种多源异构数据驱动的领域知识图谱构建方法 |
Families Citing this family (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111428053B (zh) * | 2020-03-30 | 2023-10-20 | 西安交通大学 | 一种面向税务领域知识图谱的构建方法 |
CN111813961B (zh) * | 2020-08-25 | 2020-12-18 | 腾讯科技(深圳)有限公司 | 基于人工智能的数据处理方法、装置及电子设备 |
CN112148890B (zh) * | 2020-09-23 | 2023-07-25 | 中国科学院自动化研究所 | 基于网络群体智能的教学知识点图谱系统 |
CN112148840A (zh) * | 2020-09-30 | 2020-12-29 | 神思电子技术股份有限公司 | 一种基于自然语言理解的税务知识中台及其构建方法 |
CN112231418B (zh) * | 2020-10-15 | 2024-06-28 | 南方电网人工智能科技有限公司 | 电力标准知识图谱构建方法、装置、计算机设备和介质 |
CN112434532B (zh) * | 2020-11-05 | 2024-05-28 | 西安交通大学 | 一种支持人机双向理解的电网环境模型及建模方法 |
CN112199515B (zh) * | 2020-11-17 | 2023-08-15 | 西安交通大学 | 一种多形态知识图谱驱动的知识服务创新方法 |
CN112434224A (zh) * | 2020-12-08 | 2021-03-02 | 神州数码信息系统有限公司 | 一种基于知识图谱的税收优惠政策推荐方法及其系统 |
CN112613611A (zh) * | 2020-12-29 | 2021-04-06 | 上海永骁智能技术有限公司 | 一种基于知识图谱的税务知识库系统 |
CN112612906B (zh) * | 2020-12-29 | 2022-01-11 | 广东电网有限责任公司中山供电局 | 一种基于最大熵模型的规则知识图谱构建方法及系统 |
CN112784064B (zh) * | 2021-01-28 | 2023-05-12 | 电子科技大学 | 一种面向社交网络的层次化超维知识图谱构建方法 |
CN112905746A (zh) * | 2021-03-08 | 2021-06-04 | 国能大渡河流域水电开发有限公司 | 一种基于知识图谱技术的制度档案知识挖掘处理方法 |
CN112966057B (zh) * | 2021-04-22 | 2022-08-12 | 上海深杳智能科技有限公司 | 知识图谱构建方法、系统、信息处理系统、终端及介质 |
CN113139022B (zh) * | 2021-04-29 | 2022-09-13 | 同济大学 | 一种基于混合规则的企业物流数据按需融合方法 |
CN113849659A (zh) * | 2021-08-18 | 2021-12-28 | 国网天津市电力公司 | 一种审计制度时序知识图谱的构建方法 |
CN114118483A (zh) * | 2021-12-08 | 2022-03-01 | 北京鼎兴达信息科技股份有限公司 | 一种铁路维修作业指导方法及系统 |
CN114626368B (zh) * | 2022-03-18 | 2023-06-09 | 中国电子科技集团公司第十研究所 | 一种垂直领域规则常识知识获取方法及系统 |
CN114969364B (zh) * | 2022-05-23 | 2024-05-31 | 冶金自动化研究设计院有限公司 | 企业生产物料跟踪系统知识图谱构建方法 |
CN115525776A (zh) * | 2022-10-31 | 2022-12-27 | 中国电信股份有限公司 | 事件抽取模型训练方法、事件抽取方法以及相关设备 |
CN115600601B (zh) * | 2022-11-08 | 2023-03-31 | 税友软件集团股份有限公司 | 一种税法知识库构建方法、装置、设备及介质 |
CN116340547A (zh) * | 2023-05-24 | 2023-06-27 | 江苏微皓智能科技有限公司 | 一种对话式报表可视化生产方法及系统 |
CN116562265B (zh) * | 2023-07-04 | 2023-12-01 | 南京航空航天大学 | 一种信息智能解析方法、系统及存储介质 |
CN117688110B (zh) * | 2024-02-02 | 2024-04-26 | 山东再起数据科技有限公司 | 数据中台数据血缘图谱构建方法 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180366013A1 (en) * | 2014-08-28 | 2018-12-20 | Ideaphora India Private Limited | System and method for providing an interactive visual learning environment for creation, presentation, sharing, organizing and analysis of knowledge on subject matter |
CN109271530A (zh) * | 2018-10-17 | 2019-01-25 | 长沙瀚云信息科技有限公司 | 一种疾病知识图谱构建方法和平台系统、设备、存储介质 |
CN110347844A (zh) * | 2019-07-15 | 2019-10-18 | 中国人民解放军战略支援部队航天工程大学 | 一种太空目标知识图谱构建系统 |
CN110390023A (zh) * | 2019-07-02 | 2019-10-29 | 安徽继远软件有限公司 | 一种基于改进bert模型的知识图谱构建方法 |
CN111428053A (zh) * | 2020-03-30 | 2020-07-17 | 西安交通大学 | 一种面向税务领域知识图谱的构建方法 |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103218679B (zh) * | 2013-01-30 | 2017-06-20 | 广东源恒软件科技有限公司 | 一种企业税务知识管理平台 |
CN105335519B (zh) * | 2015-11-18 | 2021-08-17 | 百度在线网络技术(北京)有限公司 | 模型生成方法及装置、推荐方法及装置 |
CN106933814A (zh) * | 2015-12-28 | 2017-07-07 | 航天信息股份有限公司 | 税务数据异常分析方法及系统 |
CN107945024B (zh) * | 2017-12-12 | 2020-08-21 | 厦门市美亚柏科信息股份有限公司 | 识别互联网金融借贷企业经营异常的方法、终端设备及存储介质 |
CN109408643B (zh) * | 2018-09-03 | 2023-05-30 | 平安科技(深圳)有限公司 | 基金相似度计算方法、系统、计算机设备和存储介质 |
CN110197280B (zh) * | 2019-05-20 | 2021-08-06 | 中国银行股份有限公司 | 一种知识图谱构建方法、装置及系统 |
CN110825882B (zh) * | 2019-10-09 | 2022-03-01 | 西安交通大学 | 一种基于知识图谱的信息系统管理方法 |
CN110825890A (zh) * | 2020-01-13 | 2020-02-21 | 成都四方伟业软件股份有限公司 | 一种预训练模型知识图谱实体关系抽取方法及装置 |
-
2020
- 2020-03-30 CN CN202010238326.3A patent/CN111428053B/zh active Active
- 2020-09-03 WO PCT/CN2020/113308 patent/WO2021196520A1/zh active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180366013A1 (en) * | 2014-08-28 | 2018-12-20 | Ideaphora India Private Limited | System and method for providing an interactive visual learning environment for creation, presentation, sharing, organizing and analysis of knowledge on subject matter |
CN109271530A (zh) * | 2018-10-17 | 2019-01-25 | 长沙瀚云信息科技有限公司 | 一种疾病知识图谱构建方法和平台系统、设备、存储介质 |
CN110390023A (zh) * | 2019-07-02 | 2019-10-29 | 安徽继远软件有限公司 | 一种基于改进bert模型的知识图谱构建方法 |
CN110347844A (zh) * | 2019-07-15 | 2019-10-18 | 中国人民解放军战略支援部队航天工程大学 | 一种太空目标知识图谱构建系统 |
CN111428053A (zh) * | 2020-03-30 | 2020-07-17 | 西安交通大学 | 一种面向税务领域知识图谱的构建方法 |
Non-Patent Citations (1)
Title |
---|
SU DEGUO; WANG LIANG; WANG GUOXIAN: "Research of knowledge base of active statistical map cartography", FUZZY SYSTEMS AND KNOWLEDGE DISCOVERY (FSKD), 2010 SEVENTH INTERNATIONAL CONFERENCE ON, 10 August 2010 (2010-08-10), pages 2872 - 2874, XP031751676 * |
Cited By (88)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113269331A (zh) * | 2021-04-25 | 2021-08-17 | 云南电网有限责任公司信息中心 | 一种变压器检修作业的知识库构建方法 |
CN113902533A (zh) * | 2021-10-11 | 2022-01-07 | 税安科技(杭州)有限公司 | 一种适用于财税领域指标自定义和自动化运行的应用 |
CN113902533B (zh) * | 2021-10-11 | 2023-08-25 | 税安科技(杭州)有限公司 | 一种适用于财税领域指标自定义和自动化运行的应用方法 |
CN113918677A (zh) * | 2021-10-18 | 2022-01-11 | 智联(无锡)信息技术有限公司 | 基于知识图谱自动化链路分层的数据处理方法、装置及计算机可读介质 |
WO2023088249A1 (zh) * | 2021-11-18 | 2023-05-25 | 华为技术有限公司 | 一种数据处理的合规性检测方法、装置和相关设备 |
CN114055451A (zh) * | 2021-11-24 | 2022-02-18 | 深圳大学 | 基于知识图谱的机器人操作技能表达方法 |
CN114055451B (zh) * | 2021-11-24 | 2023-07-07 | 深圳大学 | 基于知识图谱的机器人操作技能表达方法 |
CN114547324A (zh) * | 2021-12-31 | 2022-05-27 | 华中农业大学 | 食品安全国家标准领域本体的构建方法、系统和介质 |
CN114547324B (zh) * | 2021-12-31 | 2023-02-14 | 华中农业大学 | 食品安全国家标准领域本体的构建方法、系统和介质 |
CN114676840A (zh) * | 2022-01-27 | 2022-06-28 | 广州数说故事信息科技有限公司 | 一种基于海量数据的因果事理组提取方法、系统及计算机可读存储介质 |
CN114186759A (zh) * | 2022-02-16 | 2022-03-15 | 杭州杰牌传动科技有限公司 | 基于减速机知识图谱的物料调度控制方法及其系统 |
CN114201619A (zh) * | 2022-02-18 | 2022-03-18 | 中国电子技术标准化研究院 | 多层结构标准知识图谱构建、标准检索方法及装置 |
CN114580418A (zh) * | 2022-03-04 | 2022-06-03 | 河北师范大学 | 一种警察体能训练知识图谱系统 |
CN114580418B (zh) * | 2022-03-04 | 2024-03-26 | 河北师范大学 | 一种警察体能训练知识图谱系统 |
CN114417020A (zh) * | 2022-03-29 | 2022-04-29 | 浙江省标准化研究院(金砖国家标准化(浙江)研究中心 浙江省物品编码中心) | 一种产业链图谱构建系统及方法 |
CN114647743B (zh) * | 2022-05-20 | 2022-08-26 | 国网浙江省电力有限公司 | 电力营销全业务门禁规则图谱生成及处理方法、装置 |
CN114647743A (zh) * | 2022-05-20 | 2022-06-21 | 国网浙江省电力有限公司 | 电力营销全业务门禁规则图谱生成及处理方法、装置 |
CN114912637B (zh) * | 2022-05-21 | 2023-08-29 | 重庆大学 | 人机物知识图谱制造产线运维决策方法及系统、存储介质 |
CN114912637A (zh) * | 2022-05-21 | 2022-08-16 | 重庆大学 | 人机物知识图谱制造产线运维决策方法及系统、存储介质 |
WO2023225960A1 (en) * | 2022-05-26 | 2023-11-30 | Siemens Aktiengesellschaft | Industrial data processing method and apparatus for edge device |
CN114780083A (zh) * | 2022-06-17 | 2022-07-22 | 之江实验室 | 一种知识图谱系统的可视化构建方法及装置 |
US11907390B2 (en) | 2022-06-17 | 2024-02-20 | Zhejiang Lab | Method and apparatus for visual construction of knowledge graph system |
CN114969548A (zh) * | 2022-06-24 | 2022-08-30 | 天津联创科技发展有限公司 | 一种产业图谱数据智能获取方法及系统 |
CN114896426B (zh) * | 2022-07-14 | 2023-10-13 | 中国人民解放军国防科技大学 | 一种电子目标认知图谱的构建方法 |
CN114896426A (zh) * | 2022-07-14 | 2022-08-12 | 中国人民解放军国防科技大学 | 一种电子目标认知图谱的构建方法 |
CN115203436B (zh) * | 2022-07-15 | 2023-12-15 | 国网江苏省电力有限公司信息通信分公司 | 一种基于有向图数据融合的电力知识图谱构建方法和装置 |
CN115203436A (zh) * | 2022-07-15 | 2022-10-18 | 国网江苏省电力有限公司信息通信分公司 | 一种基于有向图数据融合的电力知识图谱构建方法和装置 |
CN114936295A (zh) * | 2022-07-21 | 2022-08-23 | 北斗天下卫星导航有限公司 | 一种基于领域知识图谱的通用热点分析方法和系统 |
CN115079588A (zh) * | 2022-08-05 | 2022-09-20 | 福州大学 | 基于时空数据驱动的智能家居服务管控方法 |
CN115114458A (zh) * | 2022-08-29 | 2022-09-27 | 北京北投智慧城市科技有限公司 | 一种基于智慧建筑知识图谱的设备控制方法及系统 |
CN115114458B (zh) * | 2022-08-29 | 2022-11-11 | 北京北投智慧城市科技有限公司 | 一种基于智慧建筑知识图谱的设备控制方法及系统 |
CN115510245B (zh) * | 2022-10-14 | 2024-05-14 | 北京理工大学 | 一种面向非结构化数据的领域知识抽取方法 |
CN115510245A (zh) * | 2022-10-14 | 2022-12-23 | 北京理工大学 | 一种面向非结构化数据的领域知识抽取方法 |
CN115545799A (zh) * | 2022-11-04 | 2022-12-30 | 北京赛西科技发展有限责任公司 | 信息技术服务质量评估方法、装置、设备及介质 |
CN115545799B (zh) * | 2022-11-04 | 2023-03-24 | 北京赛西科技发展有限责任公司 | 信息技术服务质量评估方法、装置、设备及介质 |
CN115759256A (zh) * | 2022-11-24 | 2023-03-07 | 中安华邦(北京)安全生产技术研究院股份有限公司 | 一种安全生产数字化知识库构建方法、系统、介质及设备 |
CN116701643A (zh) * | 2022-12-15 | 2023-09-05 | 重庆交通大学 | 基于社会传感数据的人群出行行为解析方法及存储介质 |
CN116010583A (zh) * | 2023-03-17 | 2023-04-25 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | 一种级联耦合的知识增强对话生成方法 |
CN116010583B (zh) * | 2023-03-17 | 2023-07-18 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | 一种级联耦合的知识增强对话生成方法 |
CN116010587A (zh) * | 2023-03-23 | 2023-04-25 | 中国人民解放军63921部队 | 航天测发保障性条件知识推送方法、装置、介质及设备 |
CN116204660B (zh) * | 2023-03-28 | 2024-06-11 | 北京航空航天大学 | 一种多源异构数据驱动的领域知识图谱构建方法 |
CN116136861A (zh) * | 2023-04-18 | 2023-05-19 | 中国电子科技集团公司第十研究所 | 基于知识图谱的分布式多源异构数据管理系统及方法 |
CN116136861B (zh) * | 2023-04-18 | 2023-08-15 | 中国电子科技集团公司第十研究所 | 基于知识图谱的分布式多源异构数据管理系统及方法 |
CN116842099A (zh) * | 2023-05-09 | 2023-10-03 | 中电科大数据研究院有限公司 | 一种多源异构数据处理方法和系统 |
CN116842099B (zh) * | 2023-05-09 | 2024-03-19 | 中电科大数据研究院有限公司 | 一种多源异构数据处理方法和系统 |
CN116992959B (zh) * | 2023-06-02 | 2024-03-15 | 广州数说故事信息科技有限公司 | 一种基于知识图谱的食品类产品创意概念生成验证方法 |
CN116992959A (zh) * | 2023-06-02 | 2023-11-03 | 广州数说故事信息科技有限公司 | 一种基于知识图谱的食品类产品创意概念生成验证方法 |
CN116756396B (zh) * | 2023-06-29 | 2023-12-22 | 广东齐峰信息科技有限公司 | 基于知识图谱的数字档案管理系统及方法 |
CN116756396A (zh) * | 2023-06-29 | 2023-09-15 | 广东齐峰信息科技有限公司 | 基于知识图谱的数字档案管理系统及方法 |
CN116611903A (zh) * | 2023-07-18 | 2023-08-18 | 太仓市律点信息技术有限公司 | 基于数字金融服务的大数据处理方法、服务器及存储介质 |
CN116611903B (zh) * | 2023-07-18 | 2023-09-22 | 太仓市律点信息技术有限公司 | 基于数字金融服务的大数据处理方法、服务器及存储介质 |
CN116702899B (zh) * | 2023-08-07 | 2023-11-28 | 上海银行股份有限公司 | 一种适用于公私联动场景的实体融合方法 |
CN116702899A (zh) * | 2023-08-07 | 2023-09-05 | 上海银行股份有限公司 | 一种适用于公私联动场景的实体融合方法 |
CN116701665A (zh) * | 2023-08-08 | 2023-09-05 | 滨州医学院 | 基于深度学习的中医古籍知识图谱构建方法 |
CN116719955A (zh) * | 2023-08-09 | 2023-09-08 | 北京国电通网络技术有限公司 | 标签标注信息生成方法、装置、电子设备和可读介质 |
CN116719955B (zh) * | 2023-08-09 | 2023-10-27 | 北京国电通网络技术有限公司 | 标签标注信息生成方法、装置、电子设备和可读介质 |
CN116821712A (zh) * | 2023-08-25 | 2023-09-29 | 中电科大数据研究院有限公司 | 非结构化文本与知识图谱的语义匹配方法及装置 |
CN116821712B (zh) * | 2023-08-25 | 2023-12-19 | 中电科大数据研究院有限公司 | 非结构化文本与知识图谱的语义匹配方法及装置 |
CN116821376A (zh) * | 2023-08-30 | 2023-09-29 | 北京华琦远航国际咨询有限公司 | 煤矿安全生产领域的知识图谱构建方法及系统 |
CN116821376B (zh) * | 2023-08-30 | 2024-03-08 | 北京华琦远航国际咨询有限公司 | 煤矿安全生产领域的知识图谱构建方法及系统 |
CN117150049B (zh) * | 2023-09-15 | 2024-03-22 | 上海峻思寰宇数据科技有限公司 | 一种个案图谱架构系统 |
CN117150049A (zh) * | 2023-09-15 | 2023-12-01 | 上海歆广数据科技有限公司 | 一种个案图谱架构系统 |
CN116955589B (zh) * | 2023-09-19 | 2024-01-30 | 山东山大鸥玛软件股份有限公司 | 一种基于教材知识图谱的智能命题方法、系统、命题终端及存储介质 |
CN116955589A (zh) * | 2023-09-19 | 2023-10-27 | 山东山大鸥玛软件股份有限公司 | 一种基于教材知识图谱的智能命题方法、系统、命题终端及存储介质 |
CN117236432A (zh) * | 2023-09-26 | 2023-12-15 | 中国科学院沈阳自动化研究所 | 一种面向多模态数据的制造工艺知识图谱构建方法及系统 |
CN117112806B (zh) * | 2023-10-12 | 2024-01-26 | 北京大学深圳研究生院 | 一种基于知识图谱的信息结构化方法和装置 |
CN117112806A (zh) * | 2023-10-12 | 2023-11-24 | 北京大学深圳研究生院 | 一种基于知识图谱的信息结构化方法和装置 |
CN117131245A (zh) * | 2023-10-27 | 2023-11-28 | 江苏数兑科技有限公司 | 一种运用知识图谱技术实现目录资源推荐机制的方法 |
CN117131245B (zh) * | 2023-10-27 | 2024-03-05 | 江苏数兑科技有限公司 | 一种运用知识图谱技术实现目录资源推荐机制的方法 |
CN117151122B (zh) * | 2023-10-30 | 2024-03-22 | 湖南三湘银行股份有限公司 | 基于自然语言处理的银行客服会话问答处理方法及系统 |
CN117151122A (zh) * | 2023-10-30 | 2023-12-01 | 湖南三湘银行股份有限公司 | 一种基于自然语言处理的银行客服会话问答处理方法及系统 |
CN117520796A (zh) * | 2023-11-23 | 2024-02-06 | 交通运输部规划研究院 | 一种基于知识图谱的公路路基水毁评估方法及系统 |
CN117312565A (zh) * | 2023-11-28 | 2023-12-29 | 山东科技大学 | 一种基于关系融合与表示学习的文献作者姓名消歧方法 |
CN117313855B (zh) * | 2023-11-28 | 2024-03-15 | 支付宝(杭州)信息技术有限公司 | 规则决策方法和装置 |
CN117313855A (zh) * | 2023-11-28 | 2023-12-29 | 支付宝(杭州)信息技术有限公司 | 规则决策方法和装置 |
CN117312565B (zh) * | 2023-11-28 | 2024-02-06 | 山东科技大学 | 一种基于关系融合与表示学习的文献作者姓名消歧方法 |
CN117391313A (zh) * | 2023-12-12 | 2024-01-12 | 广东正迪科技股份有限公司 | 基于ai的智能决策方法、系统、设备以及介质 |
CN117391313B (zh) * | 2023-12-12 | 2024-04-30 | 广东正迪科技股份有限公司 | 基于ai的智能决策方法、系统、设备以及介质 |
CN117473431A (zh) * | 2023-12-22 | 2024-01-30 | 青岛民航凯亚系统集成有限公司 | 一种基于知识图谱的机场数据分类分级方法及系统 |
CN117633518A (zh) * | 2024-01-25 | 2024-03-01 | 北京大学 | 一种产业链构建方法及系统 |
CN117633518B (zh) * | 2024-01-25 | 2024-04-26 | 北京大学 | 一种产业链构建方法及系统 |
CN117708680A (zh) * | 2024-02-06 | 2024-03-15 | 青岛海尔科技有限公司 | 一种用于提升分类模型准确度的方法及装置、存储介质、电子装置 |
CN117743558B (zh) * | 2024-02-20 | 2024-05-24 | 青岛海尔科技有限公司 | 基于大模型的知识加工、知识问答方法、装置及介质 |
CN117743558A (zh) * | 2024-02-20 | 2024-03-22 | 青岛海尔科技有限公司 | 基于大模型的知识加工、知识问答方法、装置及介质 |
CN117909492A (zh) * | 2024-03-19 | 2024-04-19 | 国网山东省电力公司信息通信公司 | 一种电网非结构化信息抽取方法、系统、设备及介质 |
CN117973520B (zh) * | 2024-03-29 | 2024-06-07 | 山东云力信息科技有限公司 | 基于大数据可视化的智慧社区知识图谱的构建方法 |
CN117973520A (zh) * | 2024-03-29 | 2024-05-03 | 山东云力信息科技有限公司 | 基于大数据可视化的智慧社区知识图谱的构建方法 |
CN118075267A (zh) * | 2024-04-22 | 2024-05-24 | 青岛青软晶尊微电子科技有限公司 | 一种基于设计电路的数字后端服务平台构建方法 |
Also Published As
Publication number | Publication date |
---|---|
CN111428053A (zh) | 2020-07-17 |
CN111428053B (zh) | 2023-10-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021196520A1 (zh) | 一种面向税务领域知识图谱的构建方法及系统 | |
US20190354544A1 (en) | Machine learning-based relationship association and related discovery and search engines | |
CN111488465A (zh) | 一种知识图谱构建方法及相关装置 | |
CN112559766B (zh) | 一种法律知识图谱构建系统 | |
CN101710343A (zh) | 一种基于文本挖掘的本体自动构建系统及方法 | |
CN111967761A (zh) | 一种基于知识图谱的监控预警方法、装置及电子设备 | |
KR102121901B1 (ko) | 정책자금 온라인 심사평가 시스템 | |
CN112183059B (zh) | 一种中文结构化事件抽取方法 | |
CN114661914A (zh) | 一种基于深度学习和知识图谱的合同审查方法、装置、设备和存储介质 | |
CN113850570A (zh) | 基于ai的专业方案辅助决策专家系统构建方法 | |
CN115640406A (zh) | 一种基于多源异构大数据分析处理与知识图谱构建方法 | |
Zhang et al. | A hybrid neural network approach for fine-grained emotion classification and computing | |
Bai et al. | FTMF: Few-shot temporal knowledge graph completion based on meta-optimization and fault-tolerant mechanism | |
Zhao | RETRACTED ARTICLE: Application of deep learning algorithm in college English teaching process evaluation | |
Sowa | Future directions for semantic systems | |
CN114911893A (zh) | 基于知识图谱的自动化构建知识库的方法及系统 | |
CN112632223B (zh) | 案事件知识图谱构建方法及相关设备 | |
Pang et al. | Methodology and mechanisms for federation of heterogeneous metadata sources and ontology development in emerging collaborative environment | |
Zhang et al. | SAKA: an intelligent platform for semi-automated knowledge graph construction and application | |
Soussi | Querying and extracting heterogeneous graphs from structured data and unstrutured content | |
CN117473054A (zh) | 基于知识图谱的通用智能问答方法及装置 | |
Ataeva et al. | Ontological approach: Knowledge representation and knowledge extraction | |
CN117033654A (zh) | 一种面向科技迷雾识别的科技事件图谱构建方法 | |
CN116595191A (zh) | 一种交互式低代码知识图谱的构建方法及装置 | |
CN116467291A (zh) | 一种知识图谱存储与搜索方法及系统 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20929476 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20929476 Country of ref document: EP Kind code of ref document: A1 |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20929476 Country of ref document: EP Kind code of ref document: A1 |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 17.05.2023) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20929476 Country of ref document: EP Kind code of ref document: A1 |