Network online personnel management system based on online education entrepreneurship guidance

Abstract

The invention relates to the technical field of online education, in particular to a network online personnel management system based on online education entrepreneurship guidance, which comprises a vocabulary construction module, a student-tutor intelligent matching module and an entrepreneurship intention and teaching resource optimization module, according to the method, the word vectors corresponding to the entrepreneurial intention and the guidance experience are matched, the cosine similarity between the word vectors is calculated, and mutually adaptive tutors are recommended for different students, so that the students can efficiently and accurately find the tutor adaptive to the own entrepreneurial intention from the numerous tutors, the burden of processing a large number of invalid screening requests by the system is reduced, and the efficiency of screening the entrepreneurial intention is improved. The operation efficiency of the online education entrepreneurship guidance system is improved, and the entrepreneurship intention and teaching resource optimization module calculates the cosine similarity between the communication information and the teaching content by adopting the vocabulary again before a tutor selects an assistant, and determines the intention student who can take charge of the assistant.

Description

Technical Field

[0001] The present invention relates to the technical field of online education, and more specifically, to an online personnel management system based on online education entrepreneurship guidance. Background Art

[0002] Online education refers to a new type of education method that uses modern information technologies such as the Internet and artificial intelligence for teaching and learning interactions. It is an important part of modern education services. Online education provides a rich variety of learning resources for learners through Internet platforms, meeting the learning needs of different groups of people.

[0003] In the current online education entrepreneurship guidance system, before a student starts learning, the student needs to add personal information and entrepreneurial intentions and other information. Before a tutor conducts education, the tutor also needs to fill in personal information and teaching experience and other information. Then, based on the information filled in by multiple tutors, the student selects a tutor that suits their personal entrepreneurial intention for learning. The tutor understands the learning progress of the students through teaching different students and formulates learning plans for different students, so that the students can carry out online education.

[0004] However, with the continuous expansion of the scale of online education and the increasing number of tutors, during the process of students selecting tutors, in the face of the teaching experience of a large number of tutors, students often have difficulty accurately screening out tutors that suit their own entrepreneurial needs, resulting in students constantly making mistakes until they find a teacher that suits their own entrepreneurial needs. If multiple students are constantly making mistakes, it will lead to an increase in the processing volume of the existing online education entrepreneurship guidance system, resulting in a decrease in operating efficiency and other situations. In view of this, we propose an online personnel management system based on online education entrepreneurship guidance. Summary of the Invention

[0005] The purpose of the present invention is to solve the problem that students cannot promptly select a suitable tutor due to a large number of tutors during online learning.

[0006] To achieve the above purpose, the present invention provides an online personnel management system based on online education entrepreneurship guidance, including a vocabulary list construction module, a student-tutor intelligent matching module, and an entrepreneurial intention and teaching resource optimization module;

[0007] The vocabulary list construction module uses web crawlers to take well-known websites as the starting points for crawling, and obtains the vocabulary in the well-known websites as the vocabulary list vocabulary;

[0008] The student-tutor intelligent matching module is used to sense entrepreneurial intentions, teaching experience, and the vocabulary list constructed by the vocabulary list construction module, and uses the feature matching method to recommend suitable tutors for students;

[0009] The feature matching method compares the entrepreneurial intention and guiding experience with the words in the vocabulary. When the words are the same, the index value corresponding to the word in the vocabulary is retrieved, and the entrepreneurial intention and guiding experience are converted into word vectors;

[0010] The cosine similarity calculation method is used to calculate the cosine similarity between the student word vector and all tutor word vectors. Then, a recommendation threshold is set, and the tutors corresponding to the cosine similarity > the recommendation threshold are recommended to the student, and all the teachers are constructed into a tutor list;

[0011] The entrepreneurial intention and teaching resource optimization system is used to perceive the communication information of the student, the online teaching content of the tutor, and the vocabulary constructed in the vocabulary building module, and mark the students corresponding to the cosine similarity > the recommendation threshold in the student-tutor intelligent matching module. If the tutor for the current online teaching does not exist in the recommended tutor list of the intended student, it is determined that the entrepreneurial intention filled in by the student is incorrect. At this time, the entrepreneurial intention corresponding to the student is regenerated according to all the words in the communication information. Before the student with the incorrect entrepreneurial intention studies again, the student-tutor intelligent matching module recommends a suitable tutor to the student according to the regenerated entrepreneurial intention;

[0012] And the students without corresponding ones are excluded, and the non-excluded students are recommended to the tutors.

[0013] As a further improvement of this technical solution, the working principle of the web crawler in the vocabulary building module is as follows:

[0014] The crawler takes a well-known website as the starting point for crawling, sends an HTTP request to the server corresponding to the well-known website, and obtains the HTML source code of the web page and the corresponding HTML syntax rules;

[0015] The server of the well-known website continuously listens on the specified port, waiting for the HTTP request sent by the crawler. After the well-known website server receives the request, it returns a response to the crawler. After the crawler receives the response, it judges whether the request is successful according to the status code of the response; if the response indicates that the request is successful, at this time the crawler obtains the HTML source code of the well-known website and the corresponding HTML syntax rules;

[0016] After obtaining the HTML source code, the crawler reads and analyzes character by character from the starting position of the source code according to the HTML syntax rules. When it encounters a start tag, it creates a corresponding node object and adds it to the document tree. When it encounters text content, it takes it as the text sub-node of the current node. When it encounters an end tag, it closes the current corresponding node, converting the linear HTML text into a tree-shaped structured data;

[0017] Based on the parsed structured data, traverse the document tree, ignore all tag nodes, and only collect the content of text nodes. For the script and style parts, since they are contained in specific tags, directly skip their internal content during the traversal process, remove irrelevant information, and only retain the words in the vocabulary list.

[0018] As a further improvement of this technical solution, when constructing the vocabulary list by the vocabulary list construction module, starting from the first word, assign an index value to it, assign an index value to the second word, and so on, until all words in the vocabulary list are assigned unique index values.

[0019] As a further improvement of this technical solution, the calculation formula for the word index value corresponding to the word in the vocabulary list retrieved by the feature matching method is as follows:

[0020] Perceived entrepreneurial intention text 、Guided experience text and vocabulary list ,The vocabulary list is a set containing words and their corresponding index values ;

[0021] Entrepreneurial intention text Each word in is , search in the vocabulary list to see if there is the same word. If there is, that is , then extract the index value of this word in the vocabulary list ;

[0022] Similarly, for each word in the number of words in, if , extract its index value .

[0023] As a further improvement of this technical solution, the cosine similarity calculation method measures the similarity between two vectors by calculating the cosine value of the angle between them. The smaller the angle, the more similar the information represented by the two vectors is semantically, indicating that the student and the tutor are more matched in entrepreneurial intention and guiding experience.

[0024] As a further improvement of this technical solution, the cosine similarity calculation method calculates the cosine similarity between the student word vector and all tutor word vectors. The cosine similarity calculation formula is as follows:

[0025] The word vectors corresponding to the perceived entrepreneurial intention and guiding experience are respectively and , and the cosine similarity calculation formula is:

[0026] ;

[0027] where is the dot product of word vectors and The calculation formula is , and are the and th dimension values of the word vectors respectively, is the dimension of the vector (i.e., the length of the word vector, determined by the number of words in the vocabulary);

[0028] and are the norms of the word vectors and respectively. The calculation formula is and .

[0029] As a further improvement of this technical solution, the calculation formula for the entrepreneurship intention and teaching resource optimization system to determine whether there is a current online teaching tutor in the list of recommended tutors for intended students is as follows:

[0030] Perceive the list of tutors recommended to the middle students , where is the th tutor object in the tutor list;

[0031] Define a judgment function . When the tutor object meets the "current online teaching" condition, ; otherwise, ;

[0032] Set a variable to record whether a tutor meeting the conditions is found, and the initial value is ;

[0033] Traverse the list . For each tutor object in the list, perform the following operations:

[0034] If , then , and stop traversing. Otherwise, continue traversing.

[0035] As a further improvement of this technical solution, the entrepreneurship intention and teaching resource optimization system regenerates the corresponding entrepreneurship intention of the student according to all the words in the communication information. The specific working steps are as follows:

[0036] Step 1: Use the web crawler technology in the vocabulary building module to obtain the existing semantic knowledge base, classify and hierarchically organize different words in the vocabulary through the existing semantic knowledge base, and define the relationships and attribute constraints between different words.

[0037] Step 2: Use the words in the communication information as query terms to find exactly the same words in the existing semantic knowledge base.

[0038] Since in the existing semantic knowledge base, each word has a specific semantic relationship with other words, determine the semantic relationship of the words in the communication information in the knowledge base, and construct a local semantic network for the words in the communication information.

[0039] Step 3: Generate the corresponding entrepreneurial intentions of the students according to the relationships and classifications of the words in the semantic knowledge base.

[0040] Compared with the prior art, the beneficial effects of the present invention are:

[0041] In the student-tutor intelligent matching module of the online personnel management system based on online education entrepreneurship guidance, word vectors corresponding to entrepreneurial intentions and guidance experience are matched through different words in the vocabulary, and the cosine similarity between the word vectors is calculated to recommend suitable tutors for different students, enabling students to efficiently and accurately find tutors that match their own entrepreneurial intentions from among many tutors, avoiding blind trial and error, and at the same time reducing the trial and error behavior of students in the process of selecting tutors, reducing the burden on the system to process a large number of invalid screening requests, and improving the operation efficiency of the online education entrepreneurship guidance system.

[0042] Moreover, before the tutor selects teaching assistants, the entrepreneurship intention and teaching resource optimization module again calculates the cosine similarity between the communication information and the teaching content using the vocabulary to determine the intended students who can serve as teaching assistants. If the tutor at this time is the tutor recommended by the student-tutor intelligent matching module to the student, then the students corresponding to the calculated cosine similarity between the communication information and the teaching content are retrieved and recommended to the tutor. The accurately selected teaching assistants can better assist the tutor in teaching, improve the teaching effect, and reduce the repeated handling of teaching problems caused by inappropriate teaching assistants, such as frequently replacing teaching assistants and the tutor spending extra energy guiding the teaching assistants. There is no need to invest resources in these problems caused by mismatched teaching assistants, thereby optimizing the allocation of teaching resources and further reducing the overall burden on the system;

[0043] When the tutor at this time is not the tutor recommended by the student-tutor intelligent matching module to the student, the entrepreneurial intention and teaching resource optimization module reconstructs the student's entrepreneurial intention again according to the vocabulary list and the existing semantic knowledge base. After reconstructing the accurate entrepreneurial intention, the student-tutor intelligent matching module recommends a suitable tutor for the student again based on the new entrepreneurial intention, avoiding the situation that the student repeatedly searches for a tutor due to the deviation of the entrepreneurial intention. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is the overall module schematic diagram of the present invention;

[0045] Figure 2 It is the working principle flowchart of the present invention.

[0046] The meanings of the various reference numerals in the figure are as follows:

[0047] 100, vocabulary list construction module; 200, student-tutor intelligent matching module; 300, entrepreneurial intention and teaching resource optimization module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] Hereinafter, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0049] Embodiment 1

[0050] An online personnel management system for online education entrepreneurship guidance includes a vocabulary list construction module 100, a student-tutor intelligent matching module 200, and an entrepreneurial intention and teaching resource optimization module 300;

[0051] Online education refers to a new type of education method that uses modern information technologies such as the Internet and artificial intelligence for teaching and learning interactions. It is an important part of modern education services. Online education provides a rich variety of learning resources for a large number of learners through the Internet platform, meeting the learning needs of different groups of people.

[0052] In the current online education entrepreneurship guidance system, before a trainee starts learning, the trainee needs to add personal information and entrepreneurship intentions (entrepreneurship intentions include, but are not limited to, information such as industry direction, market size, and target market, etc.). Before a tutor conducts education, the tutor also needs to fill in personal information and teaching experience (teaching experience includes, but is not limited to, teaching experience, guidance project experience, and successful cases, etc.). Then, based on the information filled in by multiple tutors, the trainee selects a tutor suitable for their personal entrepreneurship intention to study. The tutor understands the learning progress of the trainee by teaching different trainees and formulates a learning plan for different trainees, so that the trainee can carry out online education;

[0053] However, with the continuous expansion of the scale of online education and the increasing number of tutors, during the process of trainees selecting tutors, faced with the teaching experience of a large number of tutors, trainees often have difficulty accurately screening out tutors that meet their own entrepreneurship needs, resulting in trainees constantly making mistakes until they find a teacher that meets their own entrepreneurship needs. If multiple trainees constantly make mistakes, it will lead to an increase in the processing volume of the existing online education entrepreneurship guidance system, causing situations such as a decrease in operating efficiency;

[0054] The vocabulary building module 100 collects a large number of words using a web crawler, constructs the collected words into a vocabulary according to different industries, and assigns a unique index value (the index value is a continuous integer starting from 0) to each word in the vocabulary. The specific working principle is as follows:

[0055] The crawler uses well-known websites as the starting point for crawling, sends an HTTP request to the server corresponding to the well-known websites (well-known websites are websites containing a large number of individual words and are websites in different industry fields), and obtains the HTML source code of the web page and the corresponding HTML syntax rules;

[0056] The server of the well-known website continuously listens on the specified port, waiting for the HTTP request sent by the crawler. After the well-known website server receives the request, it returns a response to the crawler. After the crawler receives the response, it determines whether the request is successful according to the status code of the response;

[0057] If the response indicates that the request is successful, at this time, the crawler obtains the HTML source code of the well-known website and the corresponding HTML syntax rules;

[0058] After obtaining the HTML source code, the crawler reads and analyzes character by character from the starting position of the source code according to the HTML syntax rules. When encountering a start tag, it creates a corresponding node object and adds it to the document tree. When encountering text content, it will use it as the text sub-node of the current node. When encountering an end tag, it will close the current corresponding node, converting the linear HTML text into a tree-shaped structured data;

[0059] Based on the parsed structured data, traverse the document tree, ignoring all tag nodes and only collecting the content of text nodes. For the script and style parts, since the script and style parts are included in specific tags (such as <script>和<style>）中，遍历过程中遇到特定标签时直接跳过其内部内容，从而去除无关信息，只保留词汇为词汇表中的词汇。

[0060] 学员-导师智能匹配模块200用于感知创业意向、指导经验和词汇表构建模块100构建的词汇表，并采用特征匹配法为学生推荐适配的导师；

[0061] 学员-导师智能匹配模块200中特征匹配法的工作步骤如下：

[0062] S1、将创业意向和指导经验与词汇表中的词汇对比，词汇相同时，则调出词汇表中词汇对应词汇的索引值，索引值按照创业意向和指导经验中的词汇的先后顺序排序，排序后的索引值即为创业意向和指导经验对应的词向量，对应的计算公式如下：

[0063] 感知创业意向文本、指导经验文本和词汇表，词汇表是一个包含词汇及其对应索引值的集合；

[0064] 创业意向文本中的每一个词汇为，在词汇表中查找是否存在与之相同的词汇，若存在，即，则提取该词汇在词汇表中的索引值；

[0065] 同理，对于中的每一个词汇中词汇的数量，若，提取其索引值；

[0066] S2、采用余弦相似度计算法计算学员词向量和所有导师词向量之间的余弦相似度，工作原理如下：

[0067] 感知创业意向和指导经验对应的词向量分别为和，余弦相似度计算公式为：

[0068] ；

[0069] 其中是词向量和的点积，计算公式为，和分别为词向量和的第个维度的值，是向量的维度（即词向量的长度，由词汇表中词汇数量决定）；

[0070] 和分别是词向量和的模，计算公式为和；

[0071] S3、设定推荐阈值，将余弦相似度＞推荐阈值内对应的导师推荐至学员。

[0072] 实施例二

[0073] 在当前的在线教育创业指导系统里，导师致力于实现更优质的线上教学，学员也期望能够更高效地学习，所以在导师线上教学时，搭建沟通平台就成了关键举措。这个平台能让学员在学习过程中顺畅地向导师发送沟通信息，极大地促进了教与学之间的互动交流；

[0074] 不过，线上教学存在天然的局限性，难以像线下教学那样方便地考核学生的学习情况，随着导师所教授学员数量不断攀升，考核难度更是呈指数级增长，在这种情况下，导师需要借助其他方式来了解学员的学习状况，而学员在学习过程中的沟通信息就成了重要参考，导师会依据沟通信息挑选助教，希望借助助教的力量更好地辅助教学，同时也能更全面地掌握学员的学习动态；

[0075] 然而，仅仅依靠导师记忆来精准判断哪些学员适合担任助教变得极为困难。一方面，学员数量众多，每个学员的沟通信息繁杂多样，包含提问内容、观点表达、疑惑点等多个方面，导师难以将所有学员的信息都清晰且准确地记在脑海中，另一方面，随着教学时间的推进，新的沟通信息不断产生，旧信息容易被覆盖或遗忘，导致导师对学员的印象逐渐模糊；

[0076] 创业意向与教学资源优化系统300用于感知学员的沟通信息、导师线上教学内容和词汇表构建模块100中构建的词汇表；

[0077] 将词汇表中的词汇与沟通信息和教学内容中的文字对比，标记沟通信息和教学内容中的所有词汇，并调出词汇表中词汇对应的索引值，将索引值按照所有词汇的先后顺序构建为所有词汇向量，采用学员-导师智能匹配模块200中余弦相似度计算法计算沟通信息和教学内容之间的余弦相似度，将余弦相似度＞推荐阈值的学员标记为意向学员；

[0078] 意向学员在学员-导师智能匹配模块200匹配导师时，意向学员推荐导师的列表中是否存在当前线上教学的导师，若不存在时，则判定学员的创业意向填写错误，此时根据沟通信息中的所有词汇重新生成学员对应创业意向，创业意向填写错误的学员再次学习前，学员-导师智能匹配模块200根据重新生成的创业意向向学员推荐适配的导师；

[0079] 并剔除不存在对应的学员，将未剔除的学员推荐至导师。

[0080] 创业意向与教学资源优化系统300判断意向学员推荐导师列表中是否存在当前线上教学导师的计算公式如下：

[0081] 感知学员-导师智能匹配模块200中向学员推荐的导师列表，其中为导师列表中第个导师对象；

[0082] 定义一个判断函数，当导师对象满足"当前线上教学”条件时，；否则，；

[0083] 设变量用于记录是否找到满足条件的导师，初始值为；

[0084] 遍历列表，对于列表中的每一个导师对象，执行以下操作：

[0085] 如果，则，并停止遍历，反之则继续遍历。

[0086] 创业意向与教学资源优化系统300根据沟通信息中的所有词汇重新生成学员对应创业意向，具体的工作原理如下：

[0087] 利用词汇表构建模块100中的网络爬虫技术获取已有的语义知识库，通过已有语义知识库对词汇表中不同词汇进行分类和层次化组织，定义各个不同词汇之间的关系和属性约束。比如，将创业领域划分为不同的行业类别，每个行业类别下又包含各种创业项目、创业者等实体，明确它们之间的上下位关系、关联关系等，将沟通信息中的所有词汇输入至已有的语义知识库；

[0088] 将沟通信息中的词汇作为查询项，在已有语义知识库中查找完全相同的词汇通过学员-导师智能匹配模块200中的余弦相似度计算法计算，例如，沟通信息中有"直播带货”一词，在语义知识库中直接搜索"直播带货”，若找到，则确定其在知识库中的位置，并获取与之相关的类别、关系等信息；

[0089] 因在已有语义知识库中，每个词汇都与其他词汇存在特定的语义关系，如上下位关系（"水果”是"苹果”的上位词、因果关系"技术创新”可能导致"产品升级”）、部分-整体关系（"发动机”是"汽车”的一部分）等，确定沟通信息中词汇在知识库中的语义关系，将沟通信息中词汇构建一个局部的语义网络，例如，若"直播带货”与"电商营销手段”存在上下位关系，"电商营销手段”又与"电商行业”相关联；

[0090] 根据语义知识库中词汇的关系和分类，生成学员对应创业意向，例如，当沟通信息中同时出现"直播带货”"农产品”"供应链优化”等词汇时，根据语义知识库中词汇的关系和分类，发现"直播带货”与"农产品销售”存在业务关联，"供应链优化”对"农产品直播带货业务”有支持作用，推断创业意向可能是农产品电商直播带货领域，涉及供应链优化的业务。

[0091] 以上显示和描述了本发明的基本原理、主要特征和本发明的优点。本行业的技术人员应该了解，本发明不受上述实施例的限制，上述实施例和说明书中描述的仅为本发明的优选例，并不用来限制本发明，在不脱离本发明精神和范围的前提下，本发明还会有各种变化和改进，这些变化和改进都落入要求保护的本发明范围内。本发明要求保护范围由所附的权利要求书及其等效物界定。< / script>

Claims

1. A network-based online personnel management system based on online education and entrepreneurship guidance, characterized in that: It includes a vocabulary building module (100), a student-mentor intelligent matching module (200), and an entrepreneurial intention and teaching resource optimization module (300); The vocabulary building module (100) uses a web crawler to take a well-known website as a starting point for crawling, and obtains vocabulary in the well-known website as vocabulary; the student-mentor intelligent matching module (200) is used to perceive the entrepreneurial intention, guidance experience and the vocabulary built by the vocabulary building module (100), and uses a feature matching method to recommend a suitable mentor to the student, and compares the entrepreneurial intention and guidance experience with the vocabulary in the vocabulary; If the words are the same, the index value of the corresponding word in the vocabulary table is called up, the entrepreneurial intention and guidance experience are converted into word vectors, and the cosine similarity calculation method is used to calculate the cosine similarity between the student's word vector and the word vectors of all mentors. The recommendation threshold is set, and the corresponding mentors within the cosine similarity> recommendation threshold are recommended to the students, and all teachers are constructed into a mentor list; The entrepreneurial intention and teaching resource optimization system (300) is used to perceive the communication information of the students, the online teaching content and vocabulary of the tutor, and mark the students corresponding to the cosine similarity greater than the recommendation threshold in the student-tutor intelligent matching module (200). If the tutor currently teaching online does not exist in the tutor recommended list of the intended student, it is determined that the student's entrepreneurial intention is filled in incorrectly, and the corresponding entrepreneurial intention of the student is regenerated. Before the student who filled in the entrepreneurial intention incorrectly studies again, the student-tutor intelligent matching module (200) recommends a suitable tutor to the student based on the regenerated entrepreneurial intention.

2. The online personnel management system based on online education and entrepreneurship guidance according to claim 1 is characterized by: The working principle of the web crawler in the vocabulary building module (100) is as follows: The crawler uses the well-known website as the starting point for crawling, sends HTTP requests to the server corresponding to the well-known website, and obtains the HTML source code of the web page and the corresponding HTML syntax rules; The server of the well-known website keeps listening on the specified port, waiting for the HTTP request sent by the crawler. After receiving the request, the server of the well-known website returns a response to the crawler. After receiving the response, the crawler determines whether the request is successful based on the status code of the response. If the response is successful, it means that the request is successful. At this time, the crawler obtains the HTML source code and corresponding HTML syntax rules of the well-known website; After obtaining the HTML source code, the crawler reads and analyzes the source code character by character according to the HTML syntax rules, starting from the beginning of the source code. When encountering a start tag, it creates a corresponding node object and adds it to the document tree. When encountering text content, it uses it as the text child node of the current node. When encountering an end tag, it closes the current corresponding node and converts the linear HTML text into tree-shaped structured data. Based on the parsed structured data, the document tree is traversed, all tag nodes are ignored, and only the content of the text nodes is collected. For the script and style parts, since the script and style parts are contained in specific tags, when encountering specific tags during the traversal process, the internal content is directly skipped, irrelevant information is removed, and only the words in the vocabulary are retained.

3. The online personnel management system based on online education and entrepreneurship guidance according to claim 2 is characterized by: When the vocabulary building module (100) builds the vocabulary, it starts from the first word and assigns it an index value of 0, the second word is assigned an index value of 1, and so on, until all words in the vocabulary are assigned unique index values.

4. The online personnel management system based on online education and entrepreneurship guidance according to claim 3 is characterized by: The calculation formula for the vocabulary index value corresponding to the vocabulary in the vocabulary table called out by the feature matching method is as follows: Perceiving entrepreneurial intention text , Guidance Experience Text and Glossary , Glossary is a word that contains and its corresponding index value A collection of; Entrepreneurial intention text Every word in for , in the vocabulary Find whether there is the same word in it, if so, , then extract the index value of the word in the vocabulary ; Similarly, for Every word in The number of words in , extract its index value .

5. The online personnel management system based on online education and entrepreneurship guidance according to claim 4 is characterized by: The cosine similarity calculation method measures the similarity between two vectors by calculating the cosine value of the angle between them. The smaller the angle, the more semantically similar the information represented by the two vectors is, indicating that the student and the mentor are more matched in terms of entrepreneurial intention and guidance experience.

6. The online personnel management system based on online education and entrepreneurship guidance according to claim 5 is characterized by: The cosine similarity calculation method calculates the cosine similarity between the student word vector and all tutor word vectors. The cosine similarity calculation formula is as follows: The word vectors corresponding to perceived entrepreneurial intention and guidance experience are and , the cosine similarity calculation formula is: ； in is the word vector and The dot product of is calculated as , and The word vectors are and No. The value of the dimension, is the dimension of the vector (i.e. the length of the word vector, determined by the number of words in the vocabulary); and They are word vectors and The calculation formula is and .

7. The online personnel management system based on online education and entrepreneurship guidance according to claim 6 is characterized by: The calculation formula for the entrepreneurial intention and teaching resource optimization system (300) to determine whether the current online teaching instructor is in the list of instructors recommended by the intended student is as follows: List of recommended mentors for students in perception (200) ,in List of tutors mentor subjects; Define a judgment function , when the mentor object When the "current online teaching" condition is met, ; otherwise, ; Set variable Used to record whether a mentor who meets the conditions is found. The initial value is ; Iterating over a list , for each tutor object in the list , do the following: if ,but , and stop traversal, otherwise continue traversal.

8. The online personnel management system based on online education and entrepreneurship guidance according to claim 7 is characterized by: The entrepreneurial intention and teaching resource optimization system (300) regenerates the student's corresponding entrepreneurial intention according to all the words in the communication information. The specific working steps are as follows: Step 1: using the web crawler technology in the vocabulary building module (100) to obtain an existing semantic knowledge base, classifying and hierarchically organizing different words in the vocabulary through the existing semantic knowledge base, and defining the relationship and attribute constraints between different words; Step 2: Use the words in the communication information as query items and search for exactly the same words in the existing semantic knowledge base; Because each word in the existing semantic knowledge base has a specific semantic relationship with other words, the semantic relationship of the words in the communication information in the knowledge base is determined, and a local semantic network is constructed for the words in the communication information; Step 3: Generate the students’ corresponding entrepreneurial intentions based on the relationships and classifications of the vocabulary in the semantic knowledge base.

Images