CN112464665B - Subject term extraction and application method - Google Patents
Subject term extraction and application method Download PDFInfo
- Publication number
- CN112464665B CN112464665B CN202011469630.5A CN202011469630A CN112464665B CN 112464665 B CN112464665 B CN 112464665B CN 202011469630 A CN202011469630 A CN 202011469630A CN 112464665 B CN112464665 B CN 112464665B
- Authority
- CN
- China
- Prior art keywords
- terms
- subject
- term
- candidate
- discipline
- 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
Images
Classifications
-
- 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
-
- 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
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Animal Behavior & Ethology (AREA)
- Probability & Statistics with Applications (AREA)
- Life Sciences & Earth Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a subject term extracting and applying method, which comprises the following steps: generating a universal discipline term seed based on a knowledge resource database; recalling candidate terms based on the generic term seed; generating generic discipline terms based on the generic term seeds and recalled candidate terms, and storing the generic discipline terms in an institutional discipline term database; associating the extracted institution course knowledge points with subject terms, and selectively adding subject terms associated with the institution course knowledge points to the institution subject term database. The invention can automatically extract general subject terms by utilizing terms including the Baidu and MOOCCube terms, and increase the subject terms of the organization according to the course content customization of the specific organization, thereby improving the method for automatically acquiring the subject terms on one hand; on the other hand, from the application perspective, the algorithm is combined with the actual application scene, and the method has practical application significance.
Description
Technical Field
The invention relates to the technical field of knowledge point extraction, in particular to a subject term extraction and application method.
Background
The Tangjie team, the university of Qinghua, proposes a document "Course Concept Extraction in MOOCs via Embedding-Based Graph development" to extract information about the Concept of a Course.
The accuracy rate of recalling the candidate entries by the scheme is low. Due to the complexity of the Chinese grammar, when the algorithm recalls the candidate terms according to the rules, the recalled candidate terms have low accuracy, for example, the algorithm is mistakenly identified as a candidate entry, and further the subsequent manual review cost is increased. The proposal has low recall rate of the candidate entries. In the scheme, the candidate terms are derived from the class outline, and the text of the class outline mostly presents the characteristics of high abstraction level and coarse granularity, so that the candidate terms are difficult to recall in a large amount, and particularly in the case of the absence of the class outline, the problem of incapability of extraction also occurs, which leads to the fact that the subject terms are more comprehensive and difficult to cover all the content taught by the subject. The study of this scheme has focused mainly on the algorithms for obtaining subject terms.
Disclosure of Invention
The embodiment of the specification provides a subject term extracting and applying method.
The subject term extracting and applying method provided by the embodiment of the specification comprises the following steps:
generating a universal discipline term seed based on a knowledge resource database;
recalling candidate terms based on the generic term seed;
generating generic discipline terms based on the generic term seeds and recalled candidate terms, and storing the generic discipline terms in an institutional discipline term database;
associating the extracted institution course knowledge points with subject terms, and selectively adding subject terms associated with the institution course knowledge points to the institution subject term database.
The embodiment of the invention can automatically extract general subject terms by utilizing the terms comprising the Baidu and MOOCCube terms, and increase the subject terms of the organization according to the course content customization of the specific organization, thereby improving the algorithm for automatically acquiring the subject terms on one hand; on the other hand, from the application perspective, the algorithm is combined with the actual application scene, and the method has practical application significance.
Drawings
FIG. 1 is a flow diagram of a discipline term extraction and application method in accordance with some embodiments of the present disclosure.
FIG. 2 is a detailed flow diagram of a discipline term extraction and application method according to some embodiments of the present disclosure.
Detailed Description
In order to make those skilled in the art better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only a part of the embodiments of the present specification, and not all of the embodiments. All other embodiments obtained by a person skilled in the art based on the embodiments in the present specification without any inventive step should fall within the scope of protection of the present specification.
As shown in FIG. 1, some embodiments of the present description provide a discipline term extraction and application method, including generating a universal discipline term seed based on a knowledge resource database; recalling candidate terms based on the generic term seed; generating generic discipline terms based on the generic term seeds and recalled candidate terms, and storing the generic discipline terms in an institutional discipline term database; associating the extracted institution course knowledge points with subject terms, and selectively adding subject terms associated with the institution course knowledge points to the institution subject term database.
Further, in some embodiments of the present specification, the step of generating a universal subject term seed based on the knowledge resource database includes downloading all the encyclopedia entries based on an encyclopedia API, filtering the encyclopedia entries based on preset subjects and internal labels of the encyclopedia, and sorting the encyclopedia entries according to a label key and a label value by using a string sorting algorithm; deleting the entries which do not belong to the preset subject based on the sorted encyclopedia entries, and then sequentially auditing the remaining entries to determine encyclopedia term seeds; and acquiring MOOCCube subject term seeds of the predetermined subject based on a MOOCCube database, combining the MOOCCube subject term seeds and the encyclopedia term seeds, and removing weight to generate the general subject term seeds. For the present embodiment, the following is specifically explained: the embodiment can download all the encyclopedia entries by using the encyclopedia API and filter the entries by using the internal tags of the encyclopedia. For example, when the predetermined subject is a computer subject, entries containing words such as "computer", "programming language", "database", "software" and the like in the tag information are recalled, deactivation information such as "first book", "second edition" and the like is filtered, and the entries are sorted according to tag keys and tag values by using a character string sorting algorithm, so that batch audit can be performed, audit efficiency is improved, and entries which are not computer subject terms can be deleted in batch to obtain encyclopedic terms. The auditing process can be divided into two rounds, and the first round can delete the fragments which do not belong to the computer science in batch; the second round may perform a review for each entry. In addition to the above-described method in which the discipline term seeds can be generated using the hundred-degree encyclopedia, MOOCCube can also expand the set of discipline term seeds. The MOOCCube is an educational resource database released by Qinghua university, and extracts concepts of relevant courses and the first-repair relationship of the concepts and the like from 706 courses and nearly 4 ten thousand videos through manual marking. The MOOCCube also contains the technical terms of more than 20 subjects. The embodiment can combine and deduplicate the subject term seeds and the encyclopedia term seeds in the MOOCCube database to serve as the general subject term seeds.
In some embodiments of the present specification, the step of recalling the candidate term includes, based on the generic term seed, recalling the subject term to be recalled as the candidate term when a proportion of the subject term to be recalled referred to in the content of the subject term to be recalled exceeds a first preset threshold, or when a proportion of the subject term to be recalled referred to by the generic term seed exceeds a second preset threshold. For the present embodiment, the following is specifically explained: the coverage of the generic discipline term seeds generated in the foregoing embodiment is still relatively narrow, and it is also necessary to recall more candidate terms based on the generated generic discipline term seeds, and after filtering, the coverage of the discipline terms becomes wider finally. In the encyclopedia database, the subject terms are not terms which exist in isolation, and have references to each other inside, and can be described as a directed network structure, for example, a binary tree term page comprises terms which include: "recursive," "complete binary tree," and "traversal of a tree," etc., so that candidate terms can be recalled based on the referenced and referenced relationships.
Although the number of entries is greatly reduced (for example, the number can be reduced from ten million to ten thousand), the above-mentioned recall process still has noise, so that an algorithm needs to be designed to further filter the recalled candidate terms to obtain accurate subject terms. Therefore, in some embodiments of the present specification, the step of generating generic discipline terms specifically is to use the generic discipline term seed, run a graph propagation algorithm on the recalled candidate terms, iteratively update the confidence of each candidate term, sort according to the confidence, and audit to obtain the generic discipline terms. Further, in some embodiments of the present specification, the operation graph propagation algorithm, specifically, sets an initial score of the candidate term; constructing a vector representation of the candidate term; constructing a graph network according to the similarity among the candidate terms, and carrying out belief propagation according to the initial scores of the candidate terms; and after the graph is propagated, obtaining the score of each candidate term, sorting the candidate terms based on the score, and auditing the general subject terms after sorting. Further, in some embodiments of the present description, an initial score of a candidate term is set to 1 if the candidate term appears in the generic discipline term seed, and is set to 0 otherwise; performing word segmentation on the candidate terms, and taking the mean value of word vectors of word segmentation results as vector representation of the candidate terms; normalizing the vector representation of the candidate terms to obtain a similarity matrix, and when the vector similarity of the two candidate terms is greater than a preset value, considering that an edge can be formed between the vertexes of the two candidate terms so as to construct a graph network; and (4) carrying out graph propagation by using the similarity matrix and the initial score, and finishing propagation when the total score value of all the general discipline terms seeds begins to increase.
In some embodiments of the present specification, the step of associating the extracted institution course knowledge points with the subject terms specifically includes, based on the extracted institution course knowledge points, obtaining a subject term ranking result after invoking an encyclopedia search API; acquiring text contents of the first-ranked subject terms, classifying and filtering the texts based on a pre-trained naive Bayesian model to detect whether the text contents are related to the preset subjects or not, and further obtaining the association between the institution course knowledge points and the subject terms. For the present embodiment, the following is specifically exemplified: taking a knowledge point 'CSS box model' of a course 'Web front-end development foundation' as an example, calling an encyclopedia API to sort, taking a term of Top1 as the 'CSS box model', acquiring text content in the term, classifying the text by using a pre-trained naive Bayes model, detecting whether the content is subject-related, and finally further filtering the classification result of naive Bayes to finally obtain the association of the knowledge point and the encyclopedia term.
In some embodiments of the present disclosure, the step of selectively adding the subject term associated with the institution course knowledge point to the institution subject term database includes, after associating the extracted institution course knowledge point with the subject term, if the associated subject term is not in the institution subject term database, checking whether the associated subject term exists in all the subject terms of the encyclopedic, if not, judging that the associated subject term is a newly added subject term of the encyclopedic, adding the associated subject term to the institution subject term database, and storing the association of the knowledge point and the subject term. For the present embodiment, the following is specifically explained: the mechanism subject terms are subject terms more closely combined with a certain mechanism subject, the ability of providing data services can be enhanced by adding the mechanism subject terms, and the logic for adding the mechanism subject terms in the embodiment can be: and obtaining an encyclopedia term corresponding to a certain knowledge point according to a method for associating the encyclopedia terms with the knowledge points, if the encyclopedia term is not in the mechanism subject term, checking whether all entries of the encyclopedia exist, if the encyclopedia term does not exist, indicating that the entry is a newly added entry of the encyclopedia, acquiring the entry from the encyclopedia, adding the entry into the mechanism subject term, and finally storing the association of the knowledge points and the encyclopedia terms into a database.
In some embodiments of the present disclosure, the subject term extracting and applying method may further include obtaining a precise question and answer related to the subject term, specifically, obtaining a most relevant precise question and answer from a hundred degree knowledge according to the subject term, as an external extension of the subject term.
At the application level, the method of the embodiments of the present invention can be used for 4 application directions including, but not limited to: the method comprises the steps of lesson preparation and teaching of teachers, question book related recommendation, video page related recommendation and subject knowledge map construction. Aiming at the application of preparing lessons and giving lessons for teachers, the knowledge graph of the subject can be referred in the course of preparing lessons for teachers, so that the efficiency of preparing lessons is improved; and the extended data of the knowledge points in the courseware can be combined to help teachers enrich the contents of the courseware. In the teacher stage of giving lessons, can be to the visual coverage of this course in this subject knowledge map of show of student, help the whole main points of holding the course of student, promote teacher's the effect of giving lessons. For the application of the related recommendation of the wrong questions, when the students make the wrong questions, the students can be shown the related knowledge points of the questions, and the students can jump to the corresponding subject terms by clicking and provide the related data (concepts, synonyms, essences, etc.) of the subject terms to help the students to further understand the concepts. And the positions of the knowledge points in the disciplinary knowledge map can be shown to students, the contents of the father knowledge points and the son knowledge points of the knowledge points are provided, and the students are helped to understand and grasp the knowledge points integrally. Aiming at the application of the video page related recommendation, at the moment when a video is played to a certain knowledge point, a subject term corresponding to the knowledge point is called and displayed on the video in a suspension mode, and if a student is interested, the student can click to look up detailed information of the subject term, wherein the detailed information comprises concept explanation, synonyms, positions in a knowledge graph, essential questions and answers and the like, so that the knowledge plane of the student is expanded, and the interest of learning is stimulated. Aiming at the application of the discipline knowledge graph in the construction, the acquisition of discipline terms is an indispensable link for constructing the discipline knowledge graph, the discipline terms exist in the discipline knowledge graph as nodes, and more relationships can be established based on the nodes, such as the relationship correction, the relationship inclusion and the like, so that the basis that the knowledge graph provides more data services is improved, and data support is provided for intelligent education.
In summary, as shown in fig. 2, in the algorithm layer, on one hand, the embodiment of the present invention may first obtain encyclopedia term seeds and MOOC term seeds based on the encyclopedia and MOOC database, combine and generate general subject term seeds, then recall candidate terms based on the general subject term seeds and a preset recall algorithm, filter the candidate terms by using a graph propagation algorithm, and finally generate general subject terms and store the general subject terms in the organization subject term database; on the other hand, the knowledge points extracted from the curriculum can be sorted and filtered by combining with the encyclopedia API, the knowledge points are associated with the encyclopedia terms, the association between the knowledge points and the encyclopedia terms is stored in the knowledge point association subject term database, and if the associated encyclopedia terms are newly added encyclopedia terms, the associated encyclopedia terms are stored in the institution subject term database. Further, the elite quiz may be periodically obtained and updated incrementally on a periodic basis based on institutional discipline terminology. In the application level, the embodiment of the invention can be applied to but not limited to the preparation of lessons and teaching of teachers, the related recommendation of error textbooks, the related recommendation of video pages and the building of discipline knowledge maps.
In summary, the invention can utilize the entries including the Baidu and MOOCCube entries to automatically extract the general subject terms, and customize and increase the subject terms of the institution according to the course content of the specific institution, thereby improving the method for automatically acquiring the subject terms on one hand; on the other hand, from the application perspective, the algorithm is combined with the actual application scene, and the method has practical application significance.
While the process flows described above include operations that occur in a particular order, it should be appreciated that the processes may include more or less operations that are performed sequentially or in parallel (e.g., using parallel processors or a multi-threaded environment). The present invention is described with reference to flowchart illustrations and/or block diagrams of methods according to embodiments of the invention.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method or device comprising the element.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the method embodiment, since it is substantially similar to the apparatus embodiment, the description is simple, and the relevant points can be referred to the partial description of the apparatus embodiment. The above description is only an example of the present specification, and is not intended to limit the present specification. Various modifications and alterations to this description will become apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present specification should be included in the scope of the claims of the present specification.
Claims (9)
1. A discipline term extraction and application method, characterized in that the method comprises:
generating a universal discipline term seed based on a knowledge resource database;
recalling candidate terms based on the generic discipline term seed;
generating general subject terms based on the general subject term seeds and recalled candidate terms, and storing the general subject terms in an institutional subject term database;
associating the extracted institution course knowledge points with subject terms and selectively adding the subject terms associated with the institution course knowledge points to the institution subject term database;
wherein the step of generating the universal discipline term seeds based on the knowledge resource database is specifically,
downloading all the encyclopedia entries based on an encyclopedia API, filtering the encyclopedia entries based on preset subject and internal labels of the encyclopedia, and sequencing the encyclopedia entries according to label keys and label values by using a character string sequencing algorithm;
deleting the entries which do not belong to the preset subject based on the sorted encyclopedia entries, and then sequentially auditing the remaining entries to determine encyclopedia term seeds;
and acquiring MOOCCube subject term seeds of the preset subjects based on a MOOCCube database, combining the MOOCCube subject term seeds and the encyclopedia term seeds, and removing weight to generate the universal subject term seeds.
2. The discipline term extraction and application method as claimed in claim 1,
the step of recalling the candidate term may, in particular,
and based on the general subject term seeds, when the proportion of the general subject term seeds referred to in the content of the subject terms to be recalled exceeds a first preset threshold value, or when the proportion of the subject terms to be recalled referred to by the general subject term seeds exceeds a second preset threshold value, recalling the subject terms to be recalled as candidate terms.
3. The discipline term extraction and application method as claimed in claim 1,
the step of generating general subject terms, specifically,
and using the general discipline term seeds, running a graph propagation algorithm on the recalled candidate terms, iteratively updating the confidence of each candidate term, sequencing according to the confidence, and auditing to obtain the general discipline terms.
4. The discipline term extraction and application method as claimed in claim 3,
the operation graph propagation algorithm, specifically,
setting an initial score of the candidate term;
constructing a vector representation of the candidate term;
constructing a graph network according to the similarity among the candidate terms, and carrying out belief propagation according to the initial scores of the candidate terms;
and after the graph is propagated, obtaining the score of each candidate term, sorting the candidate terms based on the score, and auditing the general subject terms after sorting.
5. The discipline term extraction and application method as claimed in claim 4,
setting an initial score of a candidate term to 1 if the candidate term appears in the generic discipline term seed, and to 0 otherwise;
performing word segmentation on the candidate terms, and taking the mean value of word vectors of word segmentation results as vector representation of the candidate terms;
normalizing the vector representation of the candidate terms to obtain a similarity matrix, and when the vector similarity of the two candidate terms is greater than a preset value, considering that an edge can be formed between the vertexes of the two candidate terms so as to construct a graph network;
and (5) carrying out graph propagation by using the similarity matrix and the initial scores, and finishing propagation when the total score values of all the seeds start to increase.
6. The discipline term extraction and application method as claimed in claim 1,
the step of associating the extracted institution course knowledge points with subject terms, specifically,
calling an Baidu encyclopedia search Application Program Interface (API) based on the extracted institution course knowledge points to obtain subject term sequencing results;
acquiring text contents of the first-ranked subject terms, classifying and filtering the texts based on a pre-trained naive Bayesian model to detect whether the text contents are related to the preset subjects or not, and further obtaining the association between the institution course knowledge points and the subject terms.
7. The subject term extraction and application method of claim 1 or 6,
the step of selectively adding subject terms associated with the institution course knowledge points to the institution subject term database may include, in particular,
after associating the extracted knowledge points of the institution courses with the subject terms, if the associated subject terms are not in the institution subject term database, checking whether the associated subject terms exist in all the subject terms of the encyclopedia, if not, judging that the associated subject terms are new subject terms of the encyclopedia, adding the associated subject terms into the institution subject term database, and storing the association of the knowledge points and the subject terms.
8. The discipline term extraction and application method of claim 1, wherein the method further comprises,
acquiring the essence questions and answers related to the subject terms, specifically, acquiring the most related essence questions and answers from hundredths regularly according to the subject terms as external extension data of the subject terms.
9. The discipline term extraction and application method of claim 1, wherein the method further comprises,
and applying the extracted subject terms to the teacher lesson preparation and teaching process and/or the wrong exercise book related recommendation and/or the video page related recommendation and/or the subject knowledge map construction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011469630.5A CN112464665B (en) | 2020-12-14 | 2020-12-14 | Subject term extraction and application method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011469630.5A CN112464665B (en) | 2020-12-14 | 2020-12-14 | Subject term extraction and application method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112464665A CN112464665A (en) | 2021-03-09 |
| CN112464665B true CN112464665B (en) | 2021-07-13 |
Family
ID=74804091
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011469630.5A Active CN112464665B (en) | 2020-12-14 | 2020-12-14 | Subject term extraction and application method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112464665B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113221543B (en) * | 2021-05-07 | 2023-10-10 | 中国医学科学院医学信息研究所 | Medical term integration method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106156335A (en) * | 2016-07-07 | 2016-11-23 | 苏州大学 | A kind of discovery and arrangement method and system of teaching material knowledge point |
| CN107092675A (en) * | 2017-04-12 | 2017-08-25 | 新疆大学 | A kind of Uighur semanteme string abstracting method based on statistics and shallow-layer language analysis |
| CN107622051A (en) * | 2017-09-14 | 2018-01-23 | 马上消费金融股份有限公司 | New word screening method and device |
| CN111079419A (en) * | 2019-11-28 | 2020-04-28 | 中国人民解放军军事科学院军事科学信息研究中心 | Big data-based national defense science and technology hot word discovery method and system |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104572758B (en) * | 2013-10-24 | 2017-10-24 | 山东大学 | A kind of automatic abstracting method of power domain specialized vocabulary and system |
| CN104933026A (en) * | 2015-06-11 | 2015-09-23 | 福建工程学院 | Method for automatically extracting knowledge in the field of traditional Chinese medicine acupuncture and moxibustion |
| CN106294320B (en) * | 2016-08-04 | 2019-04-12 | 武汉数为科技有限公司 | A kind of terminology extraction method and system towards academic paper |
| CN107544958B (en) * | 2017-07-12 | 2020-02-18 | 清华大学 | Term extraction method and device |
| US11163952B2 (en) * | 2018-07-11 | 2021-11-02 | International Business Machines Corporation | Linked data seeded multi-lingual lexicon extraction |
| CN109522338B (en) * | 2018-11-09 | 2021-01-29 | 天津开心生活科技有限公司 | Clinical term mining method, device, electronic equipment and computer readable medium |
-
2020
- 2020-12-14 CN CN202011469630.5A patent/CN112464665B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106156335A (en) * | 2016-07-07 | 2016-11-23 | 苏州大学 | A kind of discovery and arrangement method and system of teaching material knowledge point |
| CN107092675A (en) * | 2017-04-12 | 2017-08-25 | 新疆大学 | A kind of Uighur semanteme string abstracting method based on statistics and shallow-layer language analysis |
| CN107622051A (en) * | 2017-09-14 | 2018-01-23 | 马上消费金融股份有限公司 | New word screening method and device |
| CN111079419A (en) * | 2019-11-28 | 2020-04-28 | 中国人民解放军军事科学院军事科学信息研究中心 | Big data-based national defense science and technology hot word discovery method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112464665A (en) | 2021-03-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108021616B (en) | Community question-answer expert recommendation method based on recurrent neural network | |
| US10325517B2 (en) | Systems and methods for extracting keywords in language learning | |
| CN108363743B (en) | Intelligent problem generation method and device and computer readable storage medium | |
| US20160133148A1 (en) | Intelligent content analysis and creation | |
| CN113590956B (en) | Knowledge point recommendation method, knowledge point recommendation device, knowledge point recommendation terminal and computer readable storage medium | |
| Sukkarieh et al. | Automarking: using computational linguistics to score short ‚free− text responses | |
| CN112784608B (en) | Test question recommending method and device, electronic equipment and storage medium | |
| CN109947952B (en) | Retrieval method, device, equipment and storage medium based on English knowledge graph | |
| CN115329200B (en) | Teaching resource recommendation method based on knowledge graph and user similarity | |
| CN111324692B (en) | Automatic subjective question scoring method and device based on artificial intelligence | |
| US20150026184A1 (en) | Methods and systems for content management | |
| CN111783428A (en) | Emergency management type objective question automatic generation system based on deep learning | |
| CN110929045A (en) | Construction method and system of poetry-semantic knowledge map | |
| CN112464665B (en) | Subject term extraction and application method | |
| CN110110218A (en) | A kind of Identity Association method and terminal | |
| JP2002072841A (en) | Word learning system | |
| Badrujaman et al. | CIPP vs kirkpatrick in the evaluation of physics teacher competency development programs: literature study | |
| CN117435689A (en) | Teaching plan generation method, teaching plan generation device, integrated machine, electronic equipment and storage medium | |
| CN110705306B (en) | Evaluation method for consistency of written and written texts | |
| KR100872595B1 (en) | On-line study-aid book system and management method thereof | |
| Nalawati et al. | Ratcliff/obershelp algorithm as an automatic assessment on e-learning | |
| CN116720509A (en) | Construction method of emotion dictionary in student teaching evaluation field | |
| Aliyanto et al. | Supervised probabilistic latent semantic analysis (sPLSA) for estimating technology readiness level | |
| CN111159360B (en) | Method and device for obtaining query topic classification model and query topic classification | |
| CN118020080A (en) | Method for establishing knowledge base for online courses |
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 |