WO2022028249A1 - 一种面向在线学习社区的学习兴趣发现方法 - Google Patents
一种面向在线学习社区的学习兴趣发现方法 Download PDFInfo
- Publication number
- WO2022028249A1 WO2022028249A1 PCT/CN2021/107751 CN2021107751W WO2022028249A1 WO 2022028249 A1 WO2022028249 A1 WO 2022028249A1 CN 2021107751 W CN2021107751 W CN 2021107751W WO 2022028249 A1 WO2022028249 A1 WO 2022028249A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- learning
- emotion
- topic
- interest
- text
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000009826 distribution Methods 0.000 claims abstract description 26
- 230000008451 emotion Effects 0.000 claims abstract description 20
- 230000011218 segmentation Effects 0.000 claims abstract description 10
- 238000004364 calculation method Methods 0.000 claims abstract description 6
- 230000006399 behavior Effects 0.000 claims abstract description 5
- 239000013598 vector Substances 0.000 claims description 6
- 230000002996 emotional effect Effects 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 5
- 239000000463 material Substances 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000005065 mining Methods 0.000 abstract description 5
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000003058 natural language processing Methods 0.000 abstract description 2
- 230000003542 behavioural effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000000205 computational method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/247—Thesauruses; Synonyms
-
- 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
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- 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
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
-
- 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
- G06Q50/20—Education
Definitions
- the invention relates to a text mining technology in the field of computer technology natural language processing, in particular to a learning interest discovery method based on time series-emotion-topic modeling for online learning communities.
- the online learning community provides a place for collaborative learning and knowledge construction for learners in different spaces and times, and makes up for the lack of social-emotional and cognitive communication in the network interaction scenario. Among them, a large amount of learning interest information is contained in a large amount of unstructured text information generated by learners.
- the present invention provides a learning interest discovery method oriented to an online learning community, which is used to solve the problem that the existing learning interest discovery method deviates from the educational psychology theory and cannot effectively discover the learning interest.
- a learning interest discovery method for online learning community including the following steps:
- the specific method of performing text segmentation in step (2) includes:
- step (3) the "time sequence-emotion-topic-based text modeling algorithm" described in step (3) includes:
- step (4) "based on the semantic similarity calculation method, the learning interest and the non-learning interest of the learner are identified" described in step (4), including:
- (4-2) Calculate the semantic similarity between the interest word and the domain knowledge word vector, set a threshold, and use the threshold as a benchmark to identify learning interests and non-learning interests.
- the beneficial effect of the present invention is that: the present invention collects the text information and behavioral characteristics generated by learners in the learning community, and combines the time series-emotion-topic joint modeling to mine interest information in combination with educational psychology theory, and then uses the semantic similarity. Computational methods identify learning interests from non-learning interests. This method can effectively discover the learner's learning interest, significantly improve the interpretability and accuracy of the learning interest, and help to provide learners with personalized learning services.
- FIG. 1 is a flow chart of a method for discovering learning interests for an online learning community according to the present invention.
- FIG. 2 is a flowchart of the text modeling algorithm based on time sequence-emotion-topic of the present invention.
- an embodiment of the present invention provides a learning interest discovery method oriented to an online learning community, including the following steps:
- A000 Collect multi-dimensional behavioral and textual information generated by learners in online learning communities.
- the online learning community provides learners with rich dialogue expressions, including text content such as learners' posts, replies, and emoticons, as well as click behaviors such as disapproval, approval, and favorites. Among them, the click behavior will be further replaced with emotional words and post text splicing.
- A001 Integrate domain knowledge named entity words and learned emotional words for text segmentation. Import the initial named entity words through the textbook concept table, and use new word discovery methods such as word vector clustering, information entropy, and mutual information to expand domain knowledge-related entity words; obtain positive, perplexed and negative categories through manual screening and labeling After that, replace synonyms, remove stop words and low-frequency words to get word segmentation sequence.
- new word discovery methods such as word vector clustering, information entropy, and mutual information to expand domain knowledge-related entity words; obtain positive, perplexed and negative categories through manual screening and labeling After that, replace synonyms, remove stop words and low-frequency words to get word segmentation sequence.
- A002 Text modeling algorithm based on time series-emotion-topic, mining the probability distribution of interest topics related to sentiment and time series information.
- time sequence, emotion, and topic are used as the generation variables of the probabilistic graphical model, so as to establish a formal generation model of the dialogue text of the learning community; after that, the interest in the time series dimension is calculated by the Gibbs sampling algorithm.
- Topic probability and distribution of topic words are used as the generation variables of the probabilistic graphical model.
- A003 Referring to formula (1), based on the semantic similarity calculation method, the learner's learning interest and non-learning interest are identified. Calculate the semantic similarity of word vectors between the distribution of interest subject words and domain knowledge words, and use a threshold to identify the learner's learning interest and non-learning interest, which is used to construct the learner's user interest profile and provide data for personalized learning services. Base.
- A004 According to the requirements of the application scenario, output learning interest labels and their weights. Based on the requirements of different application scenarios, learning interests are divided into persistent and transient learning interests, and different weights are marked.
- C000 Construct a probabilistic graphical model of multivariate associations such as time, emotion, subject, vocabulary, etc., based on educational psychology theory.
- open circles represent unknown variables
- solid circles represent known variables
- directional arrows represent conditional probabilities
- letters in the lower right corner of the box represent the number of repeated samplings.
- E, T, U are the number of sentiment categories, the number of topics, and the number of learners; t, w are the observable posting times and post words; e, z are the sentiments and topics implied by the posts; ⁇ , ⁇ , ⁇ , ⁇ are the hyperparameters of latent variables ⁇ djk , ⁇ jkw , ⁇ dj , and ⁇ jkh respectively, where ⁇ djk represents the probability distribution of the learner’s post d-emotion j-topic k, and ⁇ jkw represents the probability of vocabulary w-emotion j-topic k distribution, ⁇ dj denotes the probability distribution of learner posting d-emotion j, ⁇ jkh denotes the probability distribution of emotion j-topic k-time h.
- C001 The learner after reading the word segmentation generates a content and emotion dictionary, and sets the number of topics and other hyperparameters;
- C002 Initialize the sentiment category and topic number matrix of the sentence
- Equation (2) Estimate the sentiment and topic of each post through multiple iterations of Gibbs sampling, as shown in Equation (2).
- each post d consists of one or more sentences s;
- C004 Referring to formula (3), calculate the emotion-topic distribution ⁇ djk , emotion-topic-word distribution ⁇ jkw , emotion distribution ⁇ dj , emotion-topic-temporal distribution ⁇ jkh of each learner’s posts, and get the topic of interest Content;
- C005 Referring to formula (4), calculate the subject quantity evaluation curve PS 2 EK, and the minimum value within the range of subject quantity is the optimal subject quantity.
- Perplexity represents the fitting performance of the training data
- Similarity A represents the average correlation of different topic distributions
- Similarity E represents the average correlation between different emotional distributions among topics
- Entropy represents the average coherence of each topic word distribution
- KL distance Indicates the average difference in the distribution of each subject heading.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Data Mining & Analysis (AREA)
- Tourism & Hospitality (AREA)
- Evolutionary Biology (AREA)
- Primary Health Care (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Business, Economics & Management (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Evolutionary Computation (AREA)
- Strategic Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Electrically Operated Instructional Devices (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
一种面向在线学习社区的学习兴趣发现方法,涉及计算机技术自然语言处理领域的文本挖掘技术,包括:采集在线学习社区中学习者生成的多维度行为和文本信息(A000);融合领域知识命名实体词和学习情绪词进行文本分词(A001);基于时序-情绪-主题的文本建模算法,挖掘与情绪和时序信息相关的兴趣主题概率分布(A002);基于语义相似度计算方法,鉴别学习者的学习兴趣和非学习兴趣(A003);根据应用场景,输出学习兴趣标签及其权重(A004)。该方法能有效发现学习者的学习兴趣,并显著提高学习兴趣的可解释性和准确性,有助于为学习者提供个性化的学习服务。
Description
本发明涉及计算机技术自然语言处理领域的文本挖掘技术,尤其涉及一种面向在线学习社区的基于时序-情绪-主题建模的学习兴趣发现方法。
在线学习社区为不同空间和时间的学习者提供了协作学习和知识建构的场所,弥补了网络互动场景下社会情绪和认知交流的缺失。其中,学习者产生的大量非结构化文本信息中蕴藏着大量的学习兴趣信息。
然而,由于学习兴趣作为一个教育心理学的概念,其与领域知识、学习者情绪、时序演化等因素密切相关,常用的点击流日志分析方法和关键词挖掘方法并不能有效地发现和追踪与学习内容相关的兴趣。
【发明内容】
针对现有技术的以上缺陷或改进需求,本发明提供了一种面向在线学习社区的学习兴趣发现方法,用于解决现有学习兴趣发现方法偏离教育心理学理论,无法有效发现学习兴趣的问题。
本发明的目的是通过以下技术措施实现的。
一种面向在线学习社区的学习兴趣发现方法,包括以下步骤:
(1)采集在线学习社区中学习者生成的多维度行为和文本信 息;
(2)融合领域知识命名实体词和学习情绪词进行文本分词;
(3)基于时序-情绪-主题的文本建模算法,挖掘与情绪和时序信息相关的兴趣主题概率分布;
(4)基于语义相似度计算方法,鉴别学习者的学习兴趣和非学习兴趣;
(5)根据应用场景,输出学习兴趣标签及其权重。
在上述技术方案中,步骤(2)中进行文本分词的具体方法包括:
(2-1)通过筛选情绪词典的方法和人工标注的方法,获得学习情绪词典;
(2-2)使用新词发现方法从网络课程的学习材料(例如:课件和习题等)中获取领域知识相关的命名实体词,获得领域知识命名实体词典;
(2-3)基于领域知识命名实体词典和学习情绪词典,对学习者生成内容进行分词,同时去除停用词和替换同义词。
在上述技术方案中,步骤(3)中所述“基于时序-情绪-主题的文本建模算法”,包括:
(3-1)根据教育心理学理论建构时间、情绪、主题、词汇等多变量关联的概率图模型;
(3-2)读入分词后的学习者生成内容和学习情绪词典,设置主题数量和其他超参数;
(3-3)初始化句子的情绪类别和主题编号矩阵;
(3-4)通过吉布斯采样的多次迭代,估计每个句子的情绪和主题;
(3-5)计算每个学习者发帖的情绪-主题分布、情绪-主题-词分布、情绪分布、情感-主题-时序分布;
(3-6)计算主题数量评估曲线,选取最优主题数量。
在上述技术方案中,步骤(4)中所述“基于语义相似度计算方法,鉴别学习者的学习兴趣和非学习兴趣”,包括:
(4-1)使用维基百科和教学材料文本训练词向量;
(4-2)计算兴趣词与领域知识词向量的语义相似度,设置阈值,以该阈值为基准,鉴别学习兴趣和非学习兴趣。
本发明的有益效果在于:本发明通过采集学习者在学习社区中生成的文本信息及行为特征,并结合教育心理学理论将时序-情绪-主题进行联合建模挖掘兴趣信息,之后使用语义相似度计算方法鉴别学习兴趣与非学习兴趣。该方法能有效发现学习者的学习兴趣,并显著提高学习兴趣的可解释性和准确性,有助于为学习者提供个性化的学习服务。
为了更清晰明确地说明本发明实施例的技术方案,下面将对实施例的实现流程附图作简要介绍。
图1为本发明面向在线学习社区的学习兴趣发现方法的流程图。
图2为本发明的基于时序-情绪-主题的文本建模算法流程图。
为了更清晰具体地说明本发明的目的和技术方案,以下结合附图及实施例,对本发明的具体细节做详细说明。应当理解,此处描述的具体实施例仅用于解释本发明,并不限定于本发明。
请参阅图1所示,本发明实施例提供一种面向在线学习社区的学习兴趣发现方法,包括以下步骤:
A000:采集在线学习社区中学习者生成的多维度行为与文本信息。在线学习社区为学习者提供了丰富的对话表达方式,包括学习者的发帖、回复和表情等文本内容,以及反对、赞同和收藏等点击行为。其中,点击行为会被进一步替换为情绪词与发帖文本拼接。
A001:融合领域知识命名实体词和学习情绪词进行文本分词。通过教材概念表导入初始的命名实体词,并使用词向量聚类、信息熵和互信息等新词发现方法扩充领域知识相关实体词;通过人工筛选和标注的方法获得包含积极、困惑和消极类别的学习情绪词典;之后,替换同义词,去除停用词和低频词得到分词序列。
A002:基于时序-情绪-主题的文本建模算法,挖掘情绪和时序信息相关的兴趣主题概率分布。通过教育心理学对学习兴趣的假设,将时序、情绪、主题作为概率图模型的生成变量,从而建立学习社区对话文本的形式化生成模型;之后,通过吉布斯采样算法计算出时序维度上兴趣主题概率及主题词的分布。
A003:参阅公式(1)所示,基于语义相似度计算方法,鉴别学习者的学习兴趣和非学习兴趣。将兴趣主题词分布与领域知识词进行 词向量的语义相似度计算,采用一个阈值鉴别出学习者的学习兴趣和非学习兴趣,用于建构学习者的用户兴趣画像,为个性化学习服务提供数据基础。
A004:根据应用场景的要求,输出学习兴趣标签及其权重。基于不同的应用场景要求,将学习兴趣划分为持续型和短暂型学习兴趣,并标记不同权值。
请参阅下表所示,为本发明的兴趣发现方法的输入与输出示例。
请参阅图2所示,所述基于时序-情绪-主题的文本建模算法的步骤如下:
C000:根据教育心理学理论建构时间、情绪、主题、词汇等多变量关联的概率图模型。该模型中,空心圆代表未知变量,实心圆代表已知变量,有向箭头代表条件概率,方框右下角字母代表重复采样次数。E、T、U是情绪类别数量、主题数量、学习者数量;t、w是可观察到的发帖时间和帖子词;e、z是帖子隐含的情绪和主题;α、β、γ、μ分别为潜在变量θ
djk、φ
jkw、π
dj、ψ
jkh的超参数,其中θ
djk表示学习者发帖d-情绪j-主题k的概率分布、φ
jkw表示词汇w-情绪j-主题k的概率分布、π
dj表示学习者发帖d-情绪j的概率分布、ψ
jkh表示情绪j-主题k-时间h的概率分布。
C001:读入分词后的学习者生成内容和情绪词典,设置主题数 量和其他超参数;
C002:初始化句子的情绪类别和主题编号矩阵;
C003:参阅公式(2)所示,通过多次吉布斯采样迭代,估计每个帖子的情绪和主题。其中每个发帖d由一个或者多个句子s组成;
C004:参阅公式(3)所示,计算每个学习者发帖的情绪-主题分布θ
djk、情绪-主题-词分布φ
jkw、情绪分布π
dj、情感-主题-时序分布ψ
jkh,得到兴趣主题的内容;
C005:参阅公式(4)所示,计算主题数量评估曲线PS
2EK,主题数量范围内的最小值为最优主题数量。其中,Perplexity表示训练数据的拟合性能,Similarity
A表示不同主题分布的平均相关性;Similarity
E表示不同情绪分布在各主题间的平均相关性;Entropy表示各主题词分布的平均相干性;KL距离表示各主题词分布的平均差异。
PS
2EK=Perplexity·Similarity
A·Similarity
E·Entropy/KL (4)
本说明书中未作详细描述的内容,属于本专业技术人员公知的 现有技术。
以上所述仅为本发明的较佳实施例,并不用以限制本发明,凡在本发明的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本发明的保护范围之内。
Claims (4)
- 一种面向在线学习社区的学习兴趣发现方法,其特征在于该方法包括以下步骤:(1)采集在线学习社区中学习者生成的多维度行为和文本信息;(2)融合领域知识命名实体词和学习情绪词进行文本分词;(3)基于时序-情绪-主题的文本建模算法,挖掘与情绪和时序信息相关的兴趣主题概率分布;(4)基于语义相似度计算方法,鉴别学习者的学习兴趣和非学习兴趣;(5)根据应用场景,输出学习兴趣标签及其权重。
- 根据权利要求1所述的面向在线学习社区的学习兴趣发现方法,其特征在于步骤(2)中进行文本分词的具体方法包括:(2-1)通过筛选情绪词典的方法和人工标注的方法,获得学习情绪词典;(2-2)使用新词发现方法从网络课程的学习材料中获取领域知识相关的命名实体词,获得领域知识命名实体词典;(2-3)基于领域知识命名实体词典和学习情绪词典,对学习者生成内容进行分词,同时去除停用词和替换同义词。
- 根据权利要求1所述的面向在线学习社区的学习兴趣发现方法,其特征在于步骤(3)中所述“基于时序-情绪-主题的文本建模算法”,包括:(3-1)建构时间、情绪、主题、词汇多变量关联的概率图模型;(3-2)读入分词后的学习者生成内容和学习情绪词典,设置主 题数量和其他超参数;(3-3)初始化句子的情绪类别和主题编号矩阵;(3-4)通过吉布斯采样的多次迭代,估计每个句子的情绪和主题;(3-5)计算每个学习者发帖的情绪-主题分布、情绪-主题-词分布、情绪分布、情绪-主题-时序分布;(3-6)计算主题数量评估曲线,选取最优主题数量。
- 根据权利要求1所述的面向在线学习社区的学习兴趣发现方法,其特征在于步骤(4)中所述“基于语义相似度计算方法,鉴别学习者的学习兴趣和非学习兴趣”,包括:(4-1)使用维基百科和教学材料文本训练词向量;(4-2)计算兴趣词与领域知识词向量的语义相似度,根据场景需求设置阈值,以该阈值为基准,鉴别学习兴趣和非学习兴趣。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010776809.9 | 2020-08-05 | ||
CN202010776809.9A CN112101039A (zh) | 2020-08-05 | 2020-08-05 | 一种面向在线学习社区的学习兴趣发现方法 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022028249A1 true WO2022028249A1 (zh) | 2022-02-10 |
Family
ID=73750354
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2021/107751 WO2022028249A1 (zh) | 2020-08-05 | 2021-07-22 | 一种面向在线学习社区的学习兴趣发现方法 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN112101039A (zh) |
WO (1) | WO2022028249A1 (zh) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115600945A (zh) * | 2022-09-07 | 2023-01-13 | 淮阴工学院(Cn) | 基于多粒度的冷链配载用户画像构建方法及装置 |
CN115964626A (zh) * | 2022-10-27 | 2023-04-14 | 河南大学 | 一种基于动态多尺度特征融合网络的社区检测方法 |
CN116307792A (zh) * | 2022-10-12 | 2023-06-23 | 广州市阿尔法软件信息技术有限公司 | 一种面向城市体检主题场景的评估方法及装置 |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112101039A (zh) * | 2020-08-05 | 2020-12-18 | 华中师范大学 | 一种面向在线学习社区的学习兴趣发现方法 |
CN114429281B (zh) * | 2021-12-30 | 2022-11-15 | 华中师范大学 | 一种基于深度聚类算法的在线学习者活跃度测评方法 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942340A (zh) * | 2014-05-09 | 2014-07-23 | 电子科技大学 | 一种基于文本挖掘的微博用户兴趣识别方法 |
CN105677873A (zh) * | 2016-01-11 | 2016-06-15 | 中国电子科技集团公司第十研究所 | 基于领域知识模型的文本情报关联聚类汇集处理方法 |
KR101781458B1 (ko) * | 2016-04-29 | 2017-09-26 | (주)웅진컴퍼스 | 언어 학습 서비스 제공 시스템 및 언어 학습 서비스 제공 방법 |
CN109033255A (zh) * | 2018-07-06 | 2018-12-18 | 合肥明高软件技术有限公司 | 一种在线学习兴趣点分析方法及系统 |
CN112101039A (zh) * | 2020-08-05 | 2020-12-18 | 华中师范大学 | 一种面向在线学习社区的学习兴趣发现方法 |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102737120B (zh) * | 2012-06-01 | 2015-05-27 | 西安交通大学 | 一种个性化网络学习资源推荐方法 |
-
2020
- 2020-08-05 CN CN202010776809.9A patent/CN112101039A/zh active Pending
-
2021
- 2021-07-22 WO PCT/CN2021/107751 patent/WO2022028249A1/zh active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942340A (zh) * | 2014-05-09 | 2014-07-23 | 电子科技大学 | 一种基于文本挖掘的微博用户兴趣识别方法 |
CN105677873A (zh) * | 2016-01-11 | 2016-06-15 | 中国电子科技集团公司第十研究所 | 基于领域知识模型的文本情报关联聚类汇集处理方法 |
KR101781458B1 (ko) * | 2016-04-29 | 2017-09-26 | (주)웅진컴퍼스 | 언어 학습 서비스 제공 시스템 및 언어 학습 서비스 제공 방법 |
CN109033255A (zh) * | 2018-07-06 | 2018-12-18 | 合肥明高软件技术有限公司 | 一种在线学习兴趣点分析方法及系统 |
CN112101039A (zh) * | 2020-08-05 | 2020-12-18 | 华中师范大学 | 一种面向在线学习社区的学习兴趣发现方法 |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115600945A (zh) * | 2022-09-07 | 2023-01-13 | 淮阴工学院(Cn) | 基于多粒度的冷链配载用户画像构建方法及装置 |
CN116307792A (zh) * | 2022-10-12 | 2023-06-23 | 广州市阿尔法软件信息技术有限公司 | 一种面向城市体检主题场景的评估方法及装置 |
CN116307792B (zh) * | 2022-10-12 | 2024-03-12 | 广州市阿尔法软件信息技术有限公司 | 一种面向城市体检主题场景的评估方法及装置 |
CN115964626A (zh) * | 2022-10-27 | 2023-04-14 | 河南大学 | 一种基于动态多尺度特征融合网络的社区检测方法 |
Also Published As
Publication number | Publication date |
---|---|
CN112101039A (zh) | 2020-12-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2022028249A1 (zh) | 一种面向在线学习社区的学习兴趣发现方法 | |
Liu et al. | Finding similar exercises in online education systems | |
Saha et al. | Sentence level or token level features for automatic short answer grading?: Use both | |
Hasibuan et al. | Model Detecting Learning Styles with Artificial Neural Network. | |
Misuraca et al. | Using Opinion Mining as an educational analytic: An integrated strategy for the analysis of students’ feedback | |
García et al. | Drawbacks and solutions of applying association rule mining in learning management systems | |
Pilán et al. | Predicting proficiency levels in learner writings by transferring a linguistic complexity model from expert-written coursebooks | |
Chen et al. | Building e‐rater® scoring models using machine learning methods | |
Lee et al. | Use of training, validation, and test sets for developing automated classifiers in quantitative ethnography | |
Qi et al. | An English teaching quality evaluation model based on Gaussian process machine learning | |
Hastings et al. | Active learning for improving machine learning of student explanatory essays | |
Ren et al. | Automatic scoring of student feedback for teaching evaluation based on aspect-level sentiment analysis | |
Berdanier et al. | Opportunities for natural language processing in qualitative engineering education research: Two examples | |
Kubsch et al. | Distributing epistemic functions and tasks—A framework for augmenting human analytic power with machine learning in science education research | |
Ravichandran et al. | Intelligent topical sentiment analysis for the classification of e-learners and their topics of interest | |
Ibrahim et al. | Mining unit feedback to explore students’ learning experiences | |
Muollo et al. | Understanding Students’ online reviews to improve college experience and graduation rates of STEM programs at the largest post-secondary institution: A learner-centered study | |
CN116720509A (zh) | 一种学生教学评价领域情感词典的构建方法 | |
Yuan | Data mining techniques with its application to the dataset of mental health of college students | |
Li et al. | Discovering student models with a clustering algorithm using problem content | |
Lin et al. | Robust educational dialogue act classifiers with low-resource and imbalanced datasets | |
Tenriawaru et al. | A new model of students participation measurement in e-learning systems based on meaningful learning characteristics: An initial investigation | |
Kumari et al. | Domain-Specific Chatbot Development Using the Deep Learning-Based RASA Framework | |
Duran et al. | Inter-annotator agreement using the conversation analysis modelling schema, for dialogue | |
Li | An English Writing Grammar Error Correction Technology Based on Similarity Algorithm |
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: 21854175 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: 21854175 Country of ref document: EP Kind code of ref document: A1 |