【计算机外文翻译】A Courses Ontology System for Computer Science Education外文翻译

A Courses Ontology System for Computer Science Education
2019 IEEE International Conference on Computer Science and Educational Informatization (CSEI)

Yan Wang College of computer science and technology Jilin University Changchun, China
Tian Bai College of computer science and technology Jilin University Changchun, China
Zhao Wang College of computer science and technology Jilin University Changchun, China
Sen Yang* College of computer science and technology Jilin University Changchun, China yangsen18@mails.jlu.edu.cn
Xuemei Hu College of computer science and technology Jilin University Changchun, China
Lan Huang* College of computer science and technology Jilin University Changchun, China huanglan@jlu.edu.cn

Abstract—A computer knowledge system plays a crucial role in computer education owing to the diversification and speedy development of computer courses. There are many curriculum resources, especially in China, but most of them are relatively independent and cannot guide students to acquire more knowledge which is relevant to this course directly. To relieve these issues, in this paper, we propose an ontology system for the computer course architecture (OSCCA) to instruct, guide, and direct students efficient learning. OSCCA collects undergraduate course catalogs from the twenty famous Chinese universities and some network education websites. Then course entities and their interrelationships are culled to build Chinese course ontology. Based on these roles, OSCCA can provide effectively and contributed learning guidance to the fresh learner and OSCCA is developed by Java.
Keywords—computer education, Computer course architecture, ontology viewing, computer ontology system
I.INTRODUCTION
With the development of computer science and technology, numerous computer courses and curriculums have been spread on the internet. Most of them are relatively independent, which leads to problems that a fresh learner cannot obtain reasonable curriculums and courses easily immediately without other assistance.
To solve the problem, some previous systems have been developed. In 2015, Chayan Nuntawong, et al. presented a process model and a system for calculating the correspondence between the content of courses in Computer Science with the standard of The Thailand Qualifications Framework to improve the curriculum of universities in Thailand by meeting the standards and decreasing the duration of the operation to be more convenient [1]. In 2016, Karim Hadjar, et al. employed a case study to illustrate how the ontology assists web search engine to help user find useful information from huge amount of unstructured information [2]. In 2016, Antonio Maffei, et al., used of CONALI ontology as tool to represent the courses in an engineering program and evaluate the alignment of their activities [3]. In 2017, based on ontology system, A. Barbagallo, et al. developed a E-Learning for the Semantic ECM which allows the creation of e-learning courses which are customized according to the needs and learning preferences of the user [4]. In 2018, Hui Ma, et al. proposed a new approach for ontology-supported web-based HR recruitment systems which is facilitated by Formal Concept Analysis (FCA) for constructing domain-specific ontologies to model position requirements and applicants’ competences [5]. In 2019, Mohammad Aman Ullah, et al. presented and summarized different ways of reasoning the ontology and tried to fill the gap between existing works by including all the concepts and their related data and object properties [6].
In this paper, a computer course architectural ontology system (OSCCA) is built to associate computer fresh learners to select and explore a set of appropriate and reasonable course and curriculums. OSCCA employs some machine learning algorithms to extract and construct the relationship among the computer terms. These computer terms are utilized by Jena [7] to automatically establish the inner relationship of ontology. The OSCCA is develop by Java and provides web server for user.
II.THE OVERALL DESIGN

Fig. 1. The overall design of the ontology system of Computer course architecture. There are five steps to construct the ontology system of Computer course architecture, Step1(S1): collect course datasets, Step2(S2): analyse and define the relationship among course, catalogs and knowledge points Step3(S3): construct ontology system, Step4(S4): establish reasoning rules, Step5(S5): Implement the system by Java
Figure 1 shows the overall design of OSCCA. Five basic steps are needed, including data collection, data preprocessing, constructing Ontology, establishing reasonable rules and implementing the system. Firstly, some course, units and knowledge points corpus are collected from the raw data which includes curriculums arrangement of some famous universities, internet course, and some computer forum. The second step is crucial step, at first, all of the raw data is processed by natural language processing (NLP) [8]. And then, based on some NLP methods such as word segmentation, entity Recognition and semantic analysis et al., a series of key and master terms would be found. After getting these useful information, we use Apriori algorithm [9] to extract and analyze the interrelationship of items. Thirdly, the interrelationships of these items are applied to construct an ontology by Jena which employs Protege OWL API to automatedly construct based courses ontology system for computer science. To enable the OSCCA has inference ability, we define four reasoning rules which can direct inference processing. Finally, we build a website by Java which can provide available services for other fresh learner in computer science.
III.COLLECT COURSE DATASETS
Course datasets are collected from some college curriculums and internet information. College curriculums source from some university official websites and the internet course information are fetched by Scrapy (a Python spider package) [10]. The Course datasets consist of courses, units and knowledge points. The relationship among the three categories data is shown in Figure 2. According Figure 2 Courses include Units and Units include knowledge points. knowledge points are smallest meta items. User can find a set of curriculums and course by the meta information.

Fig. 2. The hierarchical structure of Courses, Units and Knowledge points.
IV.THE ANALYSIS OF TERMS RELATIONSHIP
To analyze the relationship of terms, the relationship of each item should be defined. In this paper, we employ NLP to detect and extract useful and contributed items. In data processing, words segmentation technology (a kind of NLP technology) is used. A continuous sentence will be segmented, filtered, and counted to get key terms. The selected key items are analyzed by the Apriori algorithm. Apriori algorithm aims to solve frequent items set problems and can assist rule learning on transactional databases. Apriori algorithm can be described as follow:

Based on the rules and relationship of key items, we can construct courses ontology system by three steps: defining the relationship of items, constructing courses ontology system by Protégé OWL API [11] and building the reasoning rules system. There are four relationships which include leading, trailing, inclusion and ownership relationship.
There are certain learning orders in the chapters of course. Taking a certain subject knowledge for example, we can find the precursor knowledge infers to the concept of subject knowledge when we cannot understand some concepts clearly. The precursor knowledge of a certain knowledge point is relatively clear and easy to obtain. Because the precursor relationship is transitive, which can transfer to the next concepts. Herein, we take the three chapters of A, B and C for example. If you want to learn C course, you need to learn B content. As the same time, if you want to learn B content, you must master A content firstly. If A course or unit lead to B content, and the interrelationship is transitive, we can reckon that A course is the process of B content. Based on the interrelationship, when the learning of the course chapters reaches a certain knowledge point, its following learning knowledge points may not be unique, so the best candidate needs to be selected under the condition when the following concepts is a uncertain choices. The more relevant following concepts content of the concept is determined according to the correlation obtained from the previous analysis and statistics. Similarly, to making the internal system more closely linked. we take the contents of three chapters of A, B and C for example. After mastering the content of a knowledge, we can learn the content of Chapter B or the content of Chapter C. That is to say, there is a following relationship between C content and A content, which is reversible. Inclusive relationship mainly refers to the content of chapters included in the course, and the content of chapters also includes the content of knowledge points. Ownership and inclusion are reciprocal, which mainly manifests itself in the case of concepts and concepts. Relevance refer to the contents of the two chapters which belongs to the same field of knowledge but cannot be determined whether there is a clear following relationship. Although relevance is not defined in the body, but it is the implicit relationship derived from the curriculums. There are no unambiguous definition of the interrelationship. Based on the defined interrelationship we can construct rules inference system.
After defining several conceptual relationships and conceptual instances, these relationships need to be defined in the ontology. First of all, we need to define the orientation of the chapters with the relationship between the leading and the following. According to the order in the curriculum arrangement of the twenty computer colleges, we can refer to the chronological order of the curriculum arrangement of colleges and universities. Moreover, because the selected data sources cover a wide range, there are still individual courses that cannot verify the relevance when selecting the relevant keyword pairs. Therefore, in order to make the system has better performance and expand the coverage, it is also necessary to refer to the interrelation between school content courses and improve the ontology.
After preparing all the data information to construct the course ontology, we will construct the course ontology. Because the data of the courses, chapters and knowledge points included in the computer course ontology in this paper are very large, if we construct it manually by Protégé tool, it will not only take time and effort but also cannot avoid some undetectable errors, so this paper chooses to construct the course ontology. Protege OWL API is used to complete the automatic construction of ontology. Finally, the created ontology file is imported into Protege for subsequent adjustment and improvement.
V.ESTABLISH REASONING RULES
To make OSCCA can reason automatically, we defined the following five based rules.
Rule1: if a includes a1, a1 includes a2, a2 is related to b2, b2 belongs to b1 and b1 belongs to b, we can reason a is related to b.
Rule2: if a includes a1, a1 includes a2, a2 leads to b2, b2 belongs to b1, b1 belongs to b, we can reason a lead to b directly.
Rule3: if a includes a1, a1 includes a2, a2 follows up b2, b2 belongs to b1 and b1 belongs to b, we can reason a directly follows up b.
Rule4: if a leads to b and b leads to c, we can reason a can lead to c.
Rule5: if a follows up b and b follows to c, we can reason a can follow to c.
Based on above rules, we use Protégé to show the interrelationship among some course name, units names and knowledge points. Figure 3 is an example of Java course to show the five basic rules. From the relationships between items and knowledge point of Java course, we can see that some item or knowledge point is related and some item which can lead to multiple knowledge points. The relationships of each item are the basic elements for inference. The inference of courses ontology system for computer science can assist user to find their interesting knowledge points.
We show some based functions of OSCCA, including home page, course query page, course information page and recorded page.

Fig. 3. An example of Java course to show five basic rules
VI.SHOWED BASIC FUNCTIONS OF OSCCA
Figure 4 shows the home page of OSCCA, users could type a course or a key word of the course, then OSCCA could show the query results. These query results are calculated by inference rules and algorithms automatically.

Fig. 4. The home page of OSCCA
When user wants to some detail information of some knowledge points, user can click detail information link. Figure 5 shows course information. If the user selects a course, OSCCA can show all knowledge points and the order to guide the user study.

Fig. 5. The course information page of OSCCA
Figure 6 is relative courses page. When user types an interesting course name, units name or a knowledge point, OSCCA can find and give some relevant contents which are listed orderly by degree of correlation.

Fig. 6. The relevant courses page of OSCCA

Fig. 7. The recording page of query of OSCCA
OSCCA can record the query information which can assist the user know what has been queried. Figure 7 shows the record page.
VII.CONCLUSION
With the number of computer course rapidly increasing, most computer science courses are quite independent. A fresh learner of computer science may be confused by numerous guidelines from internet and experiential suggestion. To solved this problem, a course ontology system for computer science education is developed.
In this paper, we develop an ontology system for computer course architecture (OSCCA) based on data mining technology OSCCA can assist students to choose a useful and beneficial learning curriculum. By extracting the relationship between chapters and knowledge points of computer courses, we propose ontology construction method by analyzing the computer course system and the ontology hierarchy, extracting the course entities, and defining the inference rules between courses. Based on fifty computer courses, we use data mining method and protégé software to automatically construct the computer course system ontology. Finally, we implement the OSCCA on web servers and shows some basic functions. In a word, OSCCA is a useful and knowledgeable courses ontology system for computer science education.
ACKNOWLEDGMENT
This research was funded by the Science-Technology Development Project from Jilin Province(20180414012GH, 20170520063JH, 20170101006JC, 20180101050JC), Jilin Provincial Key Laboratory of Big Date Intelligent Computing(20180622002JC), Guangdong Premier KeyDiscipline Enhancement Scheme (Grant 2016GDYSZDXK036) and Guangdong Key Project for Applied Fundamental Research (Grant 2018KZDXM076).
REFERENCES
[1] C. Nuntawong, C.S. Namahoot, M. Brückner, A semantic similarity assessment tool for computer science subjects using extended Wu & Palmer’s algorithm and ontology, Information Science and Applications, 2015, pp. 989-996.
[2] K. Hadjar, University Ontology: A Case Study at Ahlia University, Semantic Web, 2016, pp. 173-183.
[3] A. Maffei, L. Daghini, A. Archenti, N.J.P.C. Lohse, CONALI ontology. A framework for design and evaluation of constructively aligned courses in higher education: putting in focus the educational goal verbs, 2016, pp. 765-772.
[4] A. Barbagallo, A.J.I.L.E. Formica, ELSE: an ontology-based system integrating semantic search and e-learning technologies, Interactive Learning Environments, 2017, pp. 650-666.
[5] H. Ma, S. Hartmann, P.J.E.M. Vechsamutvaree, I.S. Architectures, Towards FCA-facilitated Ontology-supported Recruitment Systems, 2018, pp. 182-189.
[6] M.A. Ullah, S.A. Hossain, Ontology-Based Information Retrieval System for University: Methods and Reasoning, Emerging Technologies in Data Mining and Information Security, 2019, pp. 119- 128.
[7] J.J. Carroll, I. Dickinson, C. Dollin, D. Reynolds, A. Seaborne, K. Wilkinson, Jena: implementing the semantic web recommendations, Proceedings of the 13th international World Wide Web conference on Alternate track papers & posters, 2004, pp. 74-83.
[8] E. Cambria, B.J.C.I.M.I. White, Jumping NLP Curves: A Review of Natural Language Processing Research, IEEE Computational Intelligence Magazine, 2014, pp. 48-57.
[9] Xindong Wu, Vipin Kumar, J. Ross Quinlan, Top 10 algorithms in data mining, Knowledge and Information Systems, 2008, pp. 1-37.
[10] D. Myers, J.W.J.J.o.C.S.i.C. McGuffee, Choosing Scrapy, Journal of Computing Sciences in Colleges, 2015, pp. 83-89.
[11] H. Zhao, S. Zhang, J. Zhao, Research of Using Protégé to Build Ontology, IEEE/ACIS International Conference on Computer & Information Science, 2012, pp. 697-700.

一种面向计算机科学教育的课程本体论系统
2019年IEEE计算机科学与教育信息化国际会议(CSEI)

吉林大学计算机科技学院长春
吉林大学计算机科技学院长春
吉林大学计算机科技学院长春
吉林大学计算机科技学院长春yangsen18@mails.jlu.edu.cn
胡雪梅吉林大学计算机科技学院长春
黄兰裕*中国吉林大学计算机科技学院,长春huanglan@jlu.edu.cn

摘要:由于计算机课程的多元化和快速发展,计算机知识系统在计算机教育中起着至关重要的作用。课程资源很多,尤其是在中国,但它们大多是相对独立的,不能引导学生获得更多与本课程直接相关的知识。为了解决这些问题,本文提出了一个计算机课程体系结构(OSCCA)的本体系统来指导、指导和指导学生的高效学习。OSCCA从中国二十所著名大学和一些网络教育网站上收集本科课程目录。然后筛选课程实体及其相互关系,构建汉语课程本体。基于这些角色,OSCCA可以为新的学习者提供有效和有效的学习指导,OSCCA是由Java开发的。
关键词:计算机教育、计算机课程架构、本体观、计算机本体系统
I.介绍
随着计算机科学和技术的发展,许多计算机课程和课程已经在互联网上传播。其中大多数都是相对独立的,这导致了一个新的学习者在没有其他帮助的情况下无法立即获得合理的课程和课程的问题。
为了解决这个问题,我们已经开发了一些以前的系统。2015年,中国农庄,等提出了一个过程模型和系统计算课程的对应内容在计算机科学的标准泰国资格框架提高泰国大学的课程会议标准和减少操作的持续时间是更方便[1]。2016年,Karim Hadjar等人利用一个案例研究来说明本体如何帮助网络搜索引擎帮助用户从大量的非结构化信息[2]中找到有用的信息。2016年,安东尼奥·Maffei等人使用CONALI本体作为工具来表示一个工程项目中的课程,并评估其活动[3]的对齐。2017年,a.Barbagalalo等基于本体系统开发了语义ECM电子学习,允许根据用户[4]的需求和学习偏好定制电子学习课程。2018年,许马等人。提出了一种基于本体支持的基于网络的人力资源招聘系统的新方法,该方法由形式概念分析(FCA)促进,用于构建特定领域的本体,以建模职位要求和申请人的能力[5]。2019年,穆罕默德·阿曼·Ullah等人提出并总结了不同的本体推理方法,并试图通过包含所有概念及其相关数据和对象属性[6]来填补现有作品之间的空白。
本文建立了一个计算机课程建筑本体系统(OSCCA),结合计算机新学习者,选择和探索一套合适和合理的课程和课程。OSCCA采用了一些机器学习算法来提取和构建计算机术语之间的关系。Jena [7]利用这些计算机术语,自动建立了本体论的内部关系。OSCCA由Java开发,为用户提供web服务器。
II.整体设计

图1。计算机课程架构本体系统的总体设计。构建计算机课程体系结构的本体系统有五个步骤,第1(S1):收集课程数据集,第2(S2):分析和定义课程、目录和知识点之间的关系。第3(S3):构建本体系统,第4(S4):建立推理规则,第5(S5):通过Java实现系统
图1显示了OSCCA的总体设计。需要进行数据收集、数据预处理、构建本体、建立合理的规则和实现系统等五个基本步骤。首先,从原始数据中收集了一些课程、单元和知识点语料库,其中包括一些著名大学的课程安排、互联网课程和一些计算机论坛。第二步是关键的步骤,首先,所有的原始数据都由自然语言处理(NLP)[8]进行处理。然后,基于分词、实体识别、语义分析等NLP方法,可以找到一系列的关键术语和主术语。在得到这些有用的信息后,我们使用先验算法[9]来提取和分析项目之间的相互关系。第三,利用这些项目之间的相互关系,由Jena构建本体,利用Protege OWL API自动构建基于课程的计算机科学课程本体系统。为了使OSCCA具有推理能力,我们定义了四个可以指导推理处理的推理规则。最后,我们通过Java建立了一个网站,可以为其他计算机科学领域的新学习者提供可用的服务。
III.收集课程数据集
课程数据集收集自一些大学课程和互联网信息。大学课程来源从一些大学官方网站和互联网课程信息由Scrapy(一个Python蜘蛛包)[10]获取。课程数据集包括课程、单元和知识点。这三个类别的数据之间的关系如图2所示。根据图2,课程包括单位,单位包括知识点。知识点是最小的元项目。用户可以通过元信息找到一组课程和课程。

图2。课程、单元和知识点的层次结构。
IV.术语关系的分析
为了分析术语之间的关系,应该定义每个项目之间的关系。在本文中,我们使用NLP来检测和提取有用的和有贡献的项目。在数据处理中,采用了分词技术(一种NLP技术)。一个连续的句子将被分割、过滤和计数,以得到关键术语。采用先验算法进行分析。先验算法旨在解决频繁项集问题,可以帮助事务数据库的规则学习。先验算法可以描述如下:

基于关键项目之间的规则和关系,我们可以通过三个步骤来构建课程本体系统:定义项目之间的关系、通过Protege OWL API [11]构建课程本体系统和构建推理规则系统。有四种关系,包括领先、跟踪、包含和所有权关系。
当然,在这些章节中有一定的学习顺序。以某一学科知识为例,当我们不能清楚地理解某些概念时,我们可以找到推断出学科知识概念的前体知识。某一知识的前驱知识相对清晰,容易获得。因为前体关系是传递的,它可以转移到下一个概念。在此,我们以A、B、C的三章为例。如果你想学习C的课程,你需要学习B的内容。同时,如果你想学习B内容,你必须首先掌握A内容。如果一个课程或单元导致B内容,并且这种相互关系是传递的,我们可以认为A课程是B内容的过程。基于相互关系,当课程章节的学习达到某个知识点时,其接下来的学习知识点可能不是唯一的,因此在以下概念是不确定选择的情况下,需要选择最好的候选人。以下更相关的概念内容是根据之前的分析和统计数据所得到的相关性来确定的。同样地,使内部系统更紧密地联系起来。我们以A、B、C三章的内容为例。在掌握了一种知识的内容后,我们就可以学习B章的内容或C章的内容。也就是说,C内容和a内容之间存在以下关系,这是可逆的。包容性关系主要是指课程中包含的章节的内容,章节的内容也包括知识点的内容。所有权和包容是相互的,主要表现在概念和概念的情况下。关联性是指这两章的内容,属于同一知识领域,但不能确定是否存在明确的后续关系。虽然相关性在身体中没有定义,但它是源自课程的隐性关系。对这种相互关系并没有明确的定义。基于所定义的相互关系,我们可以构建规则推理系统。
在定义了多个概念关系和概念实例之后,需要在本体中定义这些关系。首先,我们需要定义章节的方向和下面的关系之间的领导。根据二十所计算机学院的课程安排的顺序,我们可以参考高校课程安排的时间顺序。此外,由于所选的数据源覆盖的范围很广,所以在选择相关的关键字对时,仍然有个别的课程无法验证相关性。因此,为了使系统具有更好的性能,扩大覆盖范围,也有必要参考学校内容课程之间的相互关系,改进本体论。
在准备好所有的数据信息来构建课程本体之后,我们将会构建课程本体。因为课程的数据,章节和知识点包含在计算机课程本体非常大,如果我们手动构建它的Protege工具,它不仅需要时间和精力,也不能避免一些无法察觉的错误,所以本文选择构建课程本体。使用Protege OWL API来完成本体论的自动构建。最后,将创建的本体文件导入到Protege中,以便进行后续的调整和改进。
V.建立推理规则
为了使OSCCA能够自动推理,我们定义了以下五个基于推理的规则。
规则1:如果a包含a1,a1包含a2,a2与b2相关,b2属于b1,b1属于b,我们可以推断a与b相关。
规则2:如果a包含a1,a1包含a2,a2引b2,b2属于b1,b1属于b,我们可以直接推理引bb。
规则3:如果a包含a1,a1包含a2,a2跟随b2,b2属于b1,b1属于b,我们可以直接推理a直接跟踪b。
规则4:如果a导致b,b导致c,我们可以推断a可以导致c。
规则5:如果a遵循b和b遵循c,我们可以推断a可以遵循c。
基于上述规则,我们使用Protege来显示一些课程名称、单位名称和知识点之间的相互关系。图3是一个显示五个基本规则的Java课程示例。从Java课程的项目与知识点之间的关系可以看出,一些项目或知识点是相关的,而一些项目可能会导致多个知识点。每个项目之间的关系都是推理的基本元素。计算机科学课程本体系统的推理可以帮助用户找到自己有趣的知识点。
我们展示了OSCCA的一些基础功能,包括主页、课程查询页面、课程信息页面和记录页面。

图3。一个Java课程的例子,显示了五个基本规则
VI.显示了OSCCA的基本功能
图4显示了OSCCA的主页,用户可以输入一个课程或一个课程的关键字,然后OSCCA可以显示查询结果。这些查询结果是由推理规则和算法自动计算出来的。

图4。OSCCA的主页
当用户想要了解一些知识点的详细信息时,用户可以单击详细信息链接。图5显示了课程信息。如果用户选择了一门课程,OSCCA可以显示所有的知识点和指导用户学习的顺序。

图5。OSCCA的课程信息页面
图6是相关课程页面。当用户键入一个有趣的课程名称、单元名称或一个知识点时,OSCCA可以找到并给出一些按相关程度按顺序列出的相关内容。

图6。OSCCA的相关课程页面

图7。OSCCA查询的记录页面
OSCCA可以记录查询信息,从而帮助用户知道已查询的内容。图7显示了记录页面。
VII.结论
随着计算机课程数量的迅速增加,大多数计算机科学课程都是相当独立的。一个计算机科学的新学习者可能会被来自互联网和经验建议的众多指导方针所混淆。为了解决这一问题,开发了一个计算机科学教育课程本体系统。
本文开发了一种基于数据挖掘技术OSCCA的计算机课程体系结构本体系统(OSCCA),可以帮助学生选择有益的学习课程。通过提取计算机课程章节与知识点之间的关系,通过分析计算机课程系统和本体层次,提取课程实体,定义课程之间的推理规则,提出了本体构建方法。以50门计算机课程为基础,采用数据挖掘方法和门生软件自动构建计算机课程系统本体。最后,我们在web服务器上实现了OSCCA,并展示了一些基本的功能。总之,OSCCA是一个有用的、知识渊博的计算机科学教育课程本体论系统。
确认
本研究由吉林省科技发展项目(201804140102GH、2010170520063JH、20170101006JC、20180101050JC)、吉林省大日期智能计算重点实验室(20180622002JC)、广东省重点学科增强计划(2016GDYSZDXK036)、广东省应用基础研究重点项目(2018KZDXM076)资助。
参考文献
[1] C.Nuntawong, C.S.利用扩展算法与本体的语义相似性评估工具,信息科学与应用,2015,页。989-996.
[2] K. Hadjar,《大学本体论:Ahlia大学的案例研究》,语义网,2016年,页。173-183.
[3] A. Maffei,L。达吉尼,A。原型,N.J.P.C. Lohse,CONALI本体论。《高等教育中建设性协调课程的设计与评价框架:重点关注教育目标动词》,2016年,第3页。765-772.
[4] A. Barbagallo,A.J.I.L.E. Formica,ELSE:一个基于集成语义搜索和电子学习技术的本体的系统,交互式学习环境,2017,第650-666页。
[5] H. Ma,S.哈特曼,P.J.E.M.《面向fca促进的本体论支持的招聘系统》,2018年,第3页。182-189.
[6] M. A. Ullah,S.。 A. 侯赛因,基于本体的大学信息检索系统:方法与推理,数据挖掘与信息安全中的新兴技术,2019年,第119128页。
[7] J.J.卡罗尔,I.狄金森, C. Dollin, D. 雷诺兹 A. 《耶:实施语义网建议》,第13届国际万维网会议论文集》,2004年,第74-83页。
[8] E.坎布里亚,B.J.C.I.M.I. White,《跳跃自然语言处理曲线:自然语言处理研究综述》,《IEEE计算智能》杂志,2014年,第3页。48-57.
吴新东,库马尔,昆兰,数据挖掘,知识与信息系统十大算法,2008,页。1-37.
[10]D.Myers,J.W.J.J.o.C.S.i.C…《选择粗糙》,《大学计算机科学杂志》,2015年,页。83-89.
[11] H. Zhao,张S.,Zhao,利用原生代构建本体的研究,IEEE/ACIS国际计算机与信息科学会议,2012,页。697-700.

  • 13
    点赞
  • 8
    收藏
    觉得还不错? 一键收藏
  • 打赏
    打赏
  • 0
    评论
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包

打赏作者

写JAVA代码的人

你的鼓励将是我创作的最大动力

¥1 ¥2 ¥4 ¥6 ¥10 ¥20
扫码支付:¥1
获取中
扫码支付

您的余额不足,请更换扫码支付或充值

打赏作者

实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值