CN111475629A - Knowledge graph construction method and system for math tutoring question-answering system - Google Patents
Knowledge graph construction method and system for math tutoring question-answering system Download PDFInfo
- Publication number
- CN111475629A CN111475629A CN202010243464.0A CN202010243464A CN111475629A CN 111475629 A CN111475629 A CN 111475629A CN 202010243464 A CN202010243464 A CN 202010243464A CN 111475629 A CN111475629 A CN 111475629A
- Authority
- CN
- China
- Prior art keywords
- entity
- knowledge
- question
- mathematical
- entities
- 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.)
- Pending
Links
- 238000010276 construction Methods 0.000 title claims abstract description 21
- 239000013598 vector Substances 0.000 claims abstract description 30
- 238000004364 calculation method Methods 0.000 claims abstract description 17
- 239000000463 material Substances 0.000 claims abstract description 12
- 238000000034 method Methods 0.000 claims description 19
- 238000000605 extraction Methods 0.000 claims description 11
- 230000011218 segmentation Effects 0.000 claims description 9
- 238000012216 screening Methods 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 7
- 230000002457 bidirectional effect Effects 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 2
- 239000000284 extract Substances 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 2
- 230000001902 propagating effect Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000012800 visualization Methods 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
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
Abstract
A knowledge graph construction method and a knowledge graph construction system for a mathematic tutoring question-answering system are characterized in that a web crawler program is used for obtaining a primary entity of mathematic knowledge, a mathematic teaching material and a tutoring book are electronized, the primary entity is screened, expanded and aligned by character matching, a TF-IDF algorithm, a TextRank algorithm and word vector similarity calculation to form an entity comprising mathematic concepts, knowledge points and question types, entity classification and entity relation recognition are carried out by combining Bi-L STM and Attenttion, knowledge contents, error-prone points, difficulties and tutoring words are used as attributes, entity attributes and attribute values are extracted by using a rule template, a graph database Neo4j is used for storing, inquiring, reasoning and visualizing the mathematic knowledge graph, and question types, difficulty in weight, error-prone points and tutoring words are established, and the constructed graph supports automatic reasoning of problems and can support automatic mathematic tutoring and automatic question-answering.
Description
Technical Field
The invention relates to the field of intelligent teaching, in particular to a knowledge graph construction method and a knowledge graph construction system for a math tutoring question-answering system.
Background
The knowledge graph is a core key technology for realizing automatic question answering and semantic search. The storage mode of the triples in the knowledge graph provides necessary and rich semantic information for information search and system automatic reasoning. The intelligent system based on the knowledge graph can better realize information search and relationship reasoning based on semantics, and brings better intelligent experience for users. At present, the knowledge graph becomes one of the core technologies for realizing an intelligent system.
In recent years, it has become a new trend to substitute question-answering systems for real teachers to guide students in learning. The mathematics tutoring question-answering system simulates an offline real teacher, can not only answer questions provided by students, but also can carry out knowledge explanation and question-answering, and can carry out correct and wrong judgment on the answers of the students. If the student answers wrongly, the reason for generating the mistake is analyzed so as to dynamically make the next teaching strategy.
Most of the existing knowledge maps of the mathematical disciplines are oriented to the query of mathematical knowledge or the automatic solution of mathematical problems. The knowledge contained in the knowledge map is mostly explicit mathematical knowledge, and lacks implicit knowledge in the aspects of mathematical problem solving skills, methods and mathematical ideas, mathematical teaching knowledge used by teachers in the process of mathematical tutoring and the associated content of the knowledge. Accordingly, semantic information in the knowledge-graph is missing, which is not enough to support the query and reasoning of the math-aided question-and-answer system.
Disclosure of Invention
The invention aims to solve the technical problem of providing a knowledge graph construction method and a knowledge graph construction system for a math tutoring question-and-answer system, which improve the efficiency and the knowledge coverage rate of constructing the knowledge graph of the subject question-and-answer system, and establish the association between subject knowledge and question types, difficult and important points, error-prone points and tutors.
In order to achieve the purpose, the invention adopts the following technical scheme:
A knowledge graph construction method for a math tutoring question-answering system comprises the following steps:
1. Acquiring a primary entity of mathematical knowledge from encyclopedia and Chinese Wikipedia by using a web crawler program;
2. Electronizing a mathematical teaching material and a tutor, and screening, expanding and aligning preliminary entities by utilizing character matching, a TF-IDF algorithm, a TextRank algorithm and word vector similarity calculation to form entities containing mathematical concepts, knowledge points and question types;
3. a bidirectional long-time and short-time memory network (Bi-L STM) is combined with an Attention mechanism (Attention) to carry out entity classification and entity relation identification;
4. Extracting entity attributes and attribute values by using a rule template by taking knowledge contents, error-prone points, difficulties and tutors as attributes;
5. The mathematical knowledge graph is stored, queried, inferred and visualized using graph database Neo4 j.
Preferably, the screening and expanding of the preliminary entities in step 2 are specifically performed as follows: the method comprises the steps of generating unstructured data by electronically processing a mathematical teaching material and a mathematical tutor, using contents in a < title > tag in a webpage acquired by a crawler program as an entity to be verified, using the unstructured data as a template, deleting entities which do not belong to mathematical knowledge in a primary entity by adopting a character matching method, acquiring keywords and key phrases by utilizing a TF-IDF algorithm and a TextRank algorithm, and further expanding the entity.
Preferably, the entity alignment is performed through word vector similarity calculation in step 2, and the specific operations are as follows:
Utilizing jieba to perform Word segmentation on unstructured data, forming a corpus together with the screened and expanded entities, and training through Word2vec to generate Word vectors;
Performing word similarity calculation based on the word vectors, and if the calculation result is less than 0.95, determining the word vectors as different entities; if the number is more than 0.95, the alias of the same entity is regarded as the alias of the same entity; if the result is equal to 1, the entity is a duplicate entity and is deleted directly.
Preferably, the entity classification and entity relationship identification in step 3 are specifically performed as follows:
Presetting entity relations, and determining three knowledge categories: mathematical concepts, knowledge points, question types;
and on the basis of word segmentation, performing dependency syntax analysis, labeling dependency syntax components, entity relations and knowledge categories of words in each sentence, converting the dependency syntax components, the entity relations and the knowledge categories into vectors, inputting the vectors and the word vectors corresponding to the words in the sentences into a Bi-L STM (Bi-L STM) model embedded with the Attention for training, generating an identification model, and extracting the categories and relations containing unknown entities.
More preferably, the specific operation of step 4 is as follows: on the basis of carrying out word segmentation and dependency syntactic analysis on unstructured data, a rule template is constructed according to characteristics of mathematical teaching materials and standard expression in a mathematical tutor, knowledge contents of mathematical concepts and knowledge points are extracted from unstructured data, error-prone points and difficulty points of the knowledge points are extracted, and a tutor of a question type is extracted.
A knowledge graph construction system for a math tutoring question-answering system comprises:
The entity identification module is used for acquiring a preliminary entity of mathematical knowledge from encyclopedia and Chinese Wikipedia by utilizing a web crawler program; electronizing a mathematical teaching material and a tutor, and screening, expanding and aligning preliminary entities by utilizing character matching, a TF-IDF algorithm, a TextRank algorithm and word vector similarity calculation to form entities containing mathematical concepts, knowledge points and question types; and transmitting to an entity relationship extraction module and an entity attribute and attribute value identification module;
the system comprises an entity relation extraction module, a knowledge graph storage and query module, a bidirectional long-time and short-time memory network (Bi-L STM) and an Attention mechanism (Attention) module, wherein the entity relation extraction module is used for extracting the relation of entities identified by the entity identification module by taking unstructured data as linguistic data;
The entity attribute and attribute value identification module is used for extracting entity attributes and attribute values of the entities identified in the entity identification module by adopting a rule template by taking knowledge content, error-prone points, difficulties and tutors as attributes; generating an entity, attribute and attribute value into a triple form of < entity, attribute and attribute value > and storing the triple form into a CSV file, and transmitting the triple form to a knowledge graph storage query module;
And the knowledge graph storage and query module is used for importing the CSV files generated by the entity relationship extraction module and the entity attribute and attribute value identification module into a graph database Neo4j, and storing, querying, reasoning and visualizing the mathematical knowledge graph by using the graph database Neo4 j.
the knowledge graph construction method has the advantages that the knowledge graph oriented to the mathematics tutoring question-answering system is constructed through the knowledge graph construction method, the efficiency and the knowledge coverage rate of the knowledge graph of the subject question-answering system are improved, the subject knowledge, the problem types, the difficult points, the error-prone points and the tutoring words are extracted from unstructured data through a bidirectional long-time memory network (Bi-L STM) and artificial rules method combined with an Attention mechanism (Attention), and the mutual association is established, and the knowledge graph constructed through the method not only supports the automatic reasoning of the mathematics problems, but also supports the automatic question-answering of the mathematics tutoring.
Drawings
FIG. 1 is a flow chart of a knowledge graph construction method for a math-assisted question-answering system;
FIG. 2 is a flow chart illustrating an entity identification process;
FIG. 3 is a diagram illustrating an example of generating word vector results;
FIG. 4 is a diagram showing a structure of a Bi-L STM model embedded in an Attention layer;
FIG. 5 is a diagram showing an example of the results of storing a knowledge-graph portion of neo4 j.
Fig. 6 is a functional block diagram of the system.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail below. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the examples given herein without any inventive step, are within the scope of the present invention.
As shown in fig. 1, the method for constructing a knowledge graph for a math-assisted question-answering system according to the present invention specifically includes the following steps:
1. And acquiring a preliminary entity of mathematical knowledge from the encyclopedia and the Chinese Wikipedia by utilizing a web crawler program.
2. The method comprises the steps of electronizing a mathematical teaching material and a tutor, and screening, expanding and aligning primary entities by utilizing character matching, a TF-IDF algorithm, a TextRank algorithm and word vector similarity calculation to form entities containing mathematical concepts, knowledge points and question types. As shown in fig. 2, the specific process includes:
2.1, electronically processing a mathematical teaching material and a mathematical tutor to generate unstructured data, taking contents in a < title > tag in a webpage acquired by a crawler program as an entity to be verified, taking the unstructured data as a template, and deleting entities which do not belong to mathematical knowledge in a preliminary entity by adopting a character matching method;
2.2 identifying keywords in unstructured data using the TF-IDF algorithm. The specific TF-IDF algorithm formula is as follows:
I.e. tf iThe number of times a word appears in a document/the total number of words in the document;
Wherein, tf iRepresenting the frequency of occurrence of terms in the text; n is iIndicates the number of times the word occurs, Representing the sum of the occurrences of all terms in the document.
idf is the inverse document frequency, i.e. the idf of a specific word, and can be obtained by dividing the total number of documents by the number of documents containing the word, and then taking the logarithm of the obtained quotient. | D | represents the total number of documents in the corpus. I { j: t i∈djDenotes the inclusion of the word t iThe number of documents.
the term TF-IDF value is TF-IDF which is TF multiplied by IDF;
And after the TF-IDF value is calculated according to the TF-IDF calculation formula, the words are sorted in a descending order according to the TF-IDF value, and the keywords are automatically obtained.
2.3 identifying key phrases in unstructured data using the TextRank algorithm. The method comprises the following specific steps: segmenting a given text T into complete sentences, i.e. T ═ S 1,S2,Λ,Sm](ii) a For each sentence S iThe method belongs to T, the jieba is utilized to carry out word segmentation and part-of-speech tagging, stop words are filtered, and only words with specified part-of-speech, such as nouns, verbs and adjectives, are reserved; and constructing a candidate keyword graph G which is (V, E), wherein V is a node set and consists of the generated candidate keywords, then constructing an edge between any two points by adopting a co-occurrence relationship (co-occurrence), and the edges between the two nodes are only co-occurred when the corresponding words co-occur in a window with the length of 7, namely, 7 words co-occur at most. Iteratively propagating the weight of each node until convergence; and (4) carrying out reverse sequencing on the node weights to obtain the most important T words, marking in the original text, and combining into a keyword phrase if adjacent phrases are formed.
2.4 utilizing jieba to divide words of unstructured data, forming a corpus together with the screened and expanded entities, and training through Word2vec to generate Word vectors; an example of the generated word vector is shown in fig. 3.
2.5, calculating word similarity based on the word vectors to realize entity alignment. If the calculation result is less than 0.95, the entity is regarded as a different entity; if the number is more than 0.95, the alias of the same entity is regarded as the alias of the same entity; if the result is equal to 1, the entity is a duplicate entity and is deleted directly.
3. the method is characterized in that a bidirectional long-time and short-time memory network (Bi-L STM) and an Attention mechanism (Attention) are applied to entity classification and entity relationship identification, and the method comprises the following specific steps:
3.1 presetting entity relations, determining three knowledge categories: mathematical concepts, knowledge points, question types.
3.2, carrying out word segmentation on the unstructured data by using jieba, and then carrying out dependency syntax analysis. For example, for "a unary equation of one degree refers to an equation that contains only one unknown, the highest degree of the unknown is 1, and both sides are integers", the partial result after performing the dependency syntax analysis is:
< means, equation of once a unary, noun subject >
< means, equation, direct object >
< content, frequency, correlation >
And 3.3, carrying out entity relation annotation. The entities and their relationships in each sentence are labeled. For example, the sentence in 3.2 is labeled with entity relationship:
< equation of once a unary, unknown >
< unknown number, frequency, include >
< equation of the first order of unity, is-kind-of >
< equation, integer, correlation >
3.4 vectorizing the dependency syntax components, the entity relations and the knowledge categories;
3.5 combines the word vector for each word in the sentence, and its corresponding vector in the previous step 3.4 into a new vector as input to the input layer of the Bi-L STM model.
an Attention layer is embedded in a Bi-L STM model to realize the extraction of the relationship between entity types and entities, and the structural diagram of the Bi-L STM model embedded in the Attention layer is shown in FIG. 4.
The weight of each time sequence is calculated through the Attention, then the vectors of all the time sequences are weighted and summed to be used as the characteristic vector, and then the softmax classification is carried out. The weight calculation formula of Attention is:
Where i denotes the time of day, j denotes the jth element in the sequence, T xDenotes the length of the sequence, f (x) j) Representing element x jThe coding of (2). a is ijIs probability, reflects C iOf importance of a ijThe calculation formula of (2) is as follows:
Wherein e ijRepresenting the degree of match between the element to be encoded and the other elements.
4. The method comprises the following steps of taking knowledge content, error-prone points, difficulty points and tutors as attributes, and adopting a rule template to extract entity attributes and attribute values, wherein the specific implementation method comprises the following steps:
4.1 on the basis of carrying out word segmentation and dependency syntactic analysis on the unstructured data, constructing a rule template according to characteristics of standard expression in a mathematical textbook and a mathematical tutor. The rule template is composed of entity names, core verbs, knowledge contents, error prone points, difficulties and guide words in the expression order of the entity names, the core verbs, the knowledge contents, the error prone points, the difficulties and the guide words. For example, in the textbook, the definition content of the unary equation is: "unary linear equation" refers to an equation that contains only one unknown, the highest degree of the unknown is 1, and both sides are integers, and the matched rule template is: entity name-core verb-knowledge content. If the sentence is matched with the template, the attribute value of the attribute of the entity 'knowledge content' is 'an equation which only contains one unknown number, the highest degree of the unknown number is 1, and both sides of the unknown number are integer';
4.2 extracting the knowledge content of the mathematical concept and the knowledge points from the unstructured data according to the rule template, extracting error-prone points and difficulty points of the knowledge points, and extracting the auxiliary guide words of the question type.
5. The mathematical knowledge-graph is stored, queried, inferred and visualized using Neo4 j. Importing the attributes and attribute values of the entities, and CSV files of the entities, the relationships and the entities stored in the form of triples into a Neo4j database. The partial visualization of the knowledge-graph is shown in fig. 5.
As shown in fig. 6, the knowledge graph construction system for the math-aided question-and-answer system according to the present invention specifically includes:
An entity identification module, which acquires a preliminary entity of mathematical knowledge from encyclopedia and Chinese Wikipedia by using a web crawler program; electronizing a mathematical teaching material and a tutor, and screening, expanding and aligning preliminary entities by utilizing character matching, a TF-IDF algorithm, a TextRank algorithm and word vector similarity calculation to form entities containing mathematical concepts, knowledge points and question types; and respectively transmitting the data to an entity relationship extraction module and an entity attribute and attribute value identification module;
the entity relation extraction module takes unstructured data as a corpus, and performs entity classification and entity relation identification on entities identified by the entity identification module by applying a bidirectional long-and-short time memory network (Bi-L STM) in combination with an Attention mechanism (Attention), stores entity relations in a form of < entity, relation and entity > triple into a CSV file, and transmits the CSV file to the knowledge map storage query module;
The entity attribute and attribute value identification module takes knowledge content, error-prone points, difficulties and tutors as attributes and adopts a rule template to extract entity attributes and attribute values of entities identified in the entity identification module; generating an entity, attribute and attribute value into a triple form of < entity, attribute and attribute value > and storing the triple form into a CSV file, and transmitting the triple form to a knowledge graph storage query module;
And the knowledge map storage and query module is used for importing the CSV files generated by the entity relationship extraction module and the entity attribute and attribute value identification module into a graph database Neo4j, and storing, querying, reasoning and visualizing the mathematical knowledge map by using the graph database Neo4 j.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.
Claims (6)
1. A knowledge graph construction method for a math tutoring question-answering system is characterized by comprising the following steps:
(1) Acquiring a primary entity of mathematical knowledge from encyclopedia and Chinese Wikipedia by using a web crawler program;
(2) Electronizing a mathematical teaching material and a tutor, and screening, expanding and aligning preliminary entities by utilizing character matching, a TF-IDF algorithm, a TextRank algorithm and word vector similarity calculation to form entities containing mathematical concepts, knowledge points and question types;
(3) a bidirectional long-time and short-time memory network (Bi-L STM) is combined with an Attention mechanism (Attention) to carry out entity classification and entity relation identification;
(4) Extracting entity attributes and attribute values by using a rule template by taking knowledge contents, error-prone points, difficulties and tutors as attributes;
(5) The mathematical knowledge graph is stored, queried, inferred and visualized using graph database Neo4 j.
2. The knowledge graph construction method for the math-aided question-and-answer system according to claim 1, wherein the specific operations of screening and expanding the preliminary entities in the step (2) are as follows: the method comprises the steps of generating unstructured data by electronically processing a mathematical teaching material and a mathematical tutor, using contents in a < title > tag in a webpage acquired by a crawler program as an entity to be verified, using the unstructured data as a template, deleting entities which do not belong to mathematical knowledge in a primary entity by adopting a character matching method, acquiring keywords and key phrases by utilizing a TF-IDF algorithm and a TextRank algorithm, and further expanding the entity.
3. The knowledge graph construction method for the math-oriented tutoring question-answering system according to claim 2, wherein in the step (2), entity alignment is performed through word vector similarity calculation, and the specific operations are as follows:
Utilizing jieba to perform Word segmentation on unstructured data, forming a corpus together with the screened and expanded entities, and training through Word2vec to generate Word vectors;
Performing word similarity calculation based on the word vectors, and if the calculation result is less than 0.95, determining the word vectors as different entities; if the number is more than 0.95, the alias of the same entity is regarded as the alias of the same entity; if the result is equal to 1, the entity is a duplicate entity and is deleted directly.
4. The knowledge graph construction method for the math-oriented tutoring question-answering system according to claim 3, wherein the entity classification and entity relationship identification in step (3) are specifically performed as follows:
Presetting entity relations, and determining three knowledge categories: mathematical concepts, knowledge points, question types;
and on the basis of word segmentation, performing dependency syntax analysis, labeling dependency syntax components, entity relations and knowledge categories of words in each sentence, converting the dependency syntax components, the entity relations and the knowledge categories into vectors, inputting the vectors and the word vectors corresponding to the words in the sentences into a Bi-L STM (Bi-L STM) model embedded with the Attention for training, generating an identification model, and extracting the categories and relations containing unknown entities.
5. The knowledge graph construction method for the math-aided question-answering system according to claim 1, wherein the specific operation of the step (4) is as follows: on the basis of carrying out word segmentation and dependency syntactic analysis on unstructured data, a rule template is constructed according to characteristics of mathematical teaching materials and standard expression in a mathematical tutor, knowledge contents of mathematical concepts and knowledge points are extracted from unstructured data, error-prone points and difficulty points of the knowledge points are extracted, and a tutor of a question type is extracted.
6. A construction system for implementing the math-assisted question-answering system-oriented knowledge graph construction method according to any one of claims 1 to 5, characterized by comprising:
The entity identification module is used for acquiring a preliminary entity of mathematical knowledge from encyclopedia and Chinese Wikipedia by utilizing a web crawler program; electronizing a mathematical teaching material and a tutor, and screening, expanding and aligning preliminary entities by utilizing character matching, a TF-IDF algorithm, a TextRank algorithm and word vector similarity calculation to form entities containing mathematical concepts, knowledge points and question types; and transmitting to an entity relationship extraction module and an entity attribute and attribute value identification module;
the entity relation extraction module extracts the relation of the entity identified by the entity identification module by taking unstructured data as a corpus, classifies the entity and identifies the entity relation by applying a bidirectional long-and-short time memory network (Bi-L STM) and combining an Attention mechanism (Attention), stores the entity relation in a form of < entity, relation and entity > triple into a CSV file, and transmits the CSV file to the knowledge map storage query module;
The entity attribute and attribute value identification module is used for extracting entity attributes and attribute values of the entities identified in the entity identification module by adopting a rule template by taking knowledge contents, error-prone points, difficulties and tutors as attributes; generating an entity, attribute and attribute value into a triple form of < entity, attribute and attribute value > and storing the triple form into a CSV file, and transmitting the triple form to a knowledge graph storage query module;
The knowledge graph storage and query module is used for importing the CSV files generated by the entity relationship extraction module and the entity attribute and attribute value identification module into a graph database Neo4j, and storing, querying, reasoning and visualizing the mathematical knowledge graph by using a graph database Neo4 j.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010243464.0A CN111475629A (en) | 2020-03-31 | 2020-03-31 | Knowledge graph construction method and system for math tutoring question-answering system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010243464.0A CN111475629A (en) | 2020-03-31 | 2020-03-31 | Knowledge graph construction method and system for math tutoring question-answering system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111475629A true CN111475629A (en) | 2020-07-31 |
Family
ID=71749344
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010243464.0A Pending CN111475629A (en) | 2020-03-31 | 2020-03-31 | Knowledge graph construction method and system for math tutoring question-answering system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111475629A (en) |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112015907A (en) * | 2020-08-18 | 2020-12-01 | 大连东软教育科技集团有限公司 | Method and device for quickly constructing discipline knowledge graph and storage medium |
CN112115275A (en) * | 2020-09-18 | 2020-12-22 | 湖南工学院 | Knowledge graph construction method and system for math tutoring question-answering system |
CN112131401A (en) * | 2020-09-14 | 2020-12-25 | 腾讯科技(深圳)有限公司 | Method and device for constructing concept knowledge graph |
CN112559765A (en) * | 2020-12-11 | 2021-03-26 | 中电科大数据研究院有限公司 | Multi-source heterogeneous database semantic integration method |
CN112650821A (en) * | 2021-01-20 | 2021-04-13 | 济南浪潮高新科技投资发展有限公司 | Entity alignment method fusing Wikidata |
CN112988953A (en) * | 2021-04-26 | 2021-06-18 | 成都索贝数码科技股份有限公司 | Adaptive broadcast television news keyword standardization method |
CN113012482A (en) * | 2021-03-02 | 2021-06-22 | 甘肃有色冶金职业技术学院 | Cross-border e-commerce spoken English conversation interactive system |
CN113254659A (en) * | 2021-02-04 | 2021-08-13 | 天津德尔塔科技有限公司 | File studying and judging method and system based on knowledge graph technology |
CN113360678A (en) * | 2021-07-08 | 2021-09-07 | 电子科技大学 | Elementary mathematic knowledge graph construction method based on Neo4j and big data |
CN113688256A (en) * | 2021-10-27 | 2021-11-23 | 上海柯林布瑞信息技术有限公司 | Construction method and device of clinical knowledge base |
CN113704499A (en) * | 2020-09-24 | 2021-11-26 | 广东昭阳信息技术有限公司 | Accurate and efficient intelligent education knowledge map construction method |
CN115438624A (en) * | 2022-11-07 | 2022-12-06 | 江西风向标智能科技有限公司 | Identification method, system, storage medium and equipment for question setting intention of mathematical subjects |
TWI817092B (en) * | 2021-03-11 | 2023-10-01 | 宏碁股份有限公司 | Method for searching frequently asked questions |
CN116910250A (en) * | 2023-06-28 | 2023-10-20 | 北京百度网讯科技有限公司 | Knowledge processing method, knowledge processing device, electronic equipment and storage medium |
CN117272992A (en) * | 2023-08-21 | 2023-12-22 | 华中师范大学 | Mathematical application problem machine answering system and method integrating physical property knowledge prompt |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050086222A1 (en) * | 2003-10-16 | 2005-04-21 | Wang Ji H. | Semi-automatic construction method for knowledge base of encyclopedia question answering system |
CN108875051A (en) * | 2018-06-28 | 2018-11-23 | 中译语通科技股份有限公司 | Knowledge mapping method for auto constructing and system towards magnanimity non-structured text |
CN110147436A (en) * | 2019-03-18 | 2019-08-20 | 清华大学 | A kind of mixing automatic question-answering method based on padagogical knowledge map and text |
CN110807084A (en) * | 2019-05-15 | 2020-02-18 | 北京信息科技大学 | Attention mechanism-based patent term relationship extraction method for Bi-LSTM and keyword strategy |
CN110825721A (en) * | 2019-11-06 | 2020-02-21 | 武汉大学 | Hypertension knowledge base construction and system integration method under big data environment |
-
2020
- 2020-03-31 CN CN202010243464.0A patent/CN111475629A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050086222A1 (en) * | 2003-10-16 | 2005-04-21 | Wang Ji H. | Semi-automatic construction method for knowledge base of encyclopedia question answering system |
CN108875051A (en) * | 2018-06-28 | 2018-11-23 | 中译语通科技股份有限公司 | Knowledge mapping method for auto constructing and system towards magnanimity non-structured text |
CN110147436A (en) * | 2019-03-18 | 2019-08-20 | 清华大学 | A kind of mixing automatic question-answering method based on padagogical knowledge map and text |
CN110807084A (en) * | 2019-05-15 | 2020-02-18 | 北京信息科技大学 | Attention mechanism-based patent term relationship extraction method for Bi-LSTM and keyword strategy |
CN110825721A (en) * | 2019-11-06 | 2020-02-21 | 武汉大学 | Hypertension knowledge base construction and system integration method under big data environment |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112015907A (en) * | 2020-08-18 | 2020-12-01 | 大连东软教育科技集团有限公司 | Method and device for quickly constructing discipline knowledge graph and storage medium |
CN112131401A (en) * | 2020-09-14 | 2020-12-25 | 腾讯科技(深圳)有限公司 | Method and device for constructing concept knowledge graph |
CN112131401B (en) * | 2020-09-14 | 2024-02-13 | 腾讯科技(深圳)有限公司 | Concept knowledge graph construction method and device |
CN112115275A (en) * | 2020-09-18 | 2020-12-22 | 湖南工学院 | Knowledge graph construction method and system for math tutoring question-answering system |
CN113704499A (en) * | 2020-09-24 | 2021-11-26 | 广东昭阳信息技术有限公司 | Accurate and efficient intelligent education knowledge map construction method |
CN112559765A (en) * | 2020-12-11 | 2021-03-26 | 中电科大数据研究院有限公司 | Multi-source heterogeneous database semantic integration method |
CN112559765B (en) * | 2020-12-11 | 2023-06-16 | 中电科大数据研究院有限公司 | Semantic integration method for multi-source heterogeneous database |
CN112650821A (en) * | 2021-01-20 | 2021-04-13 | 济南浪潮高新科技投资发展有限公司 | Entity alignment method fusing Wikidata |
CN113254659A (en) * | 2021-02-04 | 2021-08-13 | 天津德尔塔科技有限公司 | File studying and judging method and system based on knowledge graph technology |
CN113012482A (en) * | 2021-03-02 | 2021-06-22 | 甘肃有色冶金职业技术学院 | Cross-border e-commerce spoken English conversation interactive system |
TWI817092B (en) * | 2021-03-11 | 2023-10-01 | 宏碁股份有限公司 | Method for searching frequently asked questions |
CN112988953B (en) * | 2021-04-26 | 2021-09-03 | 成都索贝数码科技股份有限公司 | Adaptive broadcast television news keyword standardization method |
CN112988953A (en) * | 2021-04-26 | 2021-06-18 | 成都索贝数码科技股份有限公司 | Adaptive broadcast television news keyword standardization method |
CN113360678B (en) * | 2021-07-08 | 2022-07-15 | 电子科技大学 | Elementary mathematic knowledge graph construction method based on Neo4j and big data |
CN113360678A (en) * | 2021-07-08 | 2021-09-07 | 电子科技大学 | Elementary mathematic knowledge graph construction method based on Neo4j and big data |
CN113688256A (en) * | 2021-10-27 | 2021-11-23 | 上海柯林布瑞信息技术有限公司 | Construction method and device of clinical knowledge base |
CN115438624A (en) * | 2022-11-07 | 2022-12-06 | 江西风向标智能科技有限公司 | Identification method, system, storage medium and equipment for question setting intention of mathematical subjects |
CN116910250A (en) * | 2023-06-28 | 2023-10-20 | 北京百度网讯科技有限公司 | Knowledge processing method, knowledge processing device, electronic equipment and storage medium |
CN117272992A (en) * | 2023-08-21 | 2023-12-22 | 华中师范大学 | Mathematical application problem machine answering system and method integrating physical property knowledge prompt |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111475629A (en) | Knowledge graph construction method and system for math tutoring question-answering system | |
CN111475623B (en) | Case Information Semantic Retrieval Method and Device Based on Knowledge Graph | |
CN108363743B (en) | Intelligent problem generation method and device and computer readable storage medium | |
Hirst et al. | Lexical chains as representations of context for the detection and correction of malapropisms | |
CN101079026B (en) | Text similarity, acceptation similarity calculating method and system and application system | |
US20050027664A1 (en) | Interactive machine learning system for automated annotation of information in text | |
CN110851599A (en) | Automatic scoring method and teaching and assisting system for Chinese composition | |
CN108182275A (en) | A kind of mathematics variant training topic supplying system and correlating method | |
CN113196277A (en) | System for retrieving natural language documents | |
CN113168499A (en) | Method for searching patent document | |
CN110968708A (en) | Method and system for labeling education information resource attributes | |
CN110781681A (en) | Translation model-based elementary mathematic application problem automatic solving method and system | |
CN112966117A (en) | Entity linking method | |
CN111091002B (en) | Chinese named entity recognition method | |
CN112765983A (en) | Entity disambiguation method based on neural network combined with knowledge description | |
CN110888989B (en) | Intelligent learning platform and construction method thereof | |
Saha et al. | Development of a practical system for computerized evaluation of descriptive answers of middle school level students | |
CN115309910A (en) | Language piece element and element relation combined extraction method and knowledge graph construction method | |
Atapattu et al. | Automated extraction of semantic concepts from semi-structured data: Supporting computer-based education through the analysis of lecture notes | |
Oka et al. | Fully automated short answer scoring of the trial tests for common entrance examinations for Japanese university | |
CN113157885B (en) | Efficient intelligent question-answering system oriented to knowledge in artificial intelligence field | |
CN114611520A (en) | Text abstract generating method | |
CN114840685A (en) | Emergency plan knowledge graph construction method | |
CN116562265B (en) | Information intelligent analysis method, system and storage medium | |
CN111274354B (en) | Referee document structuring method and referee document structuring device |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200731 |