Online education intelligent management system and method thereof

Abstract

The invention discloses an online education intelligent management system and a method thereof, which are characterized in that after relevant information, course information and alternative teacher information of a target learner are collected, a semantic understanding technology is introduced at a rear end to perform semantic analysis and understanding of the information, so that the information can be analyzed and understood according to the characteristics and requirements of different courses; and according to characteristics and preferences of different learners, proper teachers are automatically matched, so that optimization and collaboration of resource allocation are realized. Therefore, the quality and efficiency of online education can be improved, the intelligent and personalized requirements of the online education are met, and the user experience and satisfaction of the online education are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent management, and particularly to an intelligent management system and method for online education. Background Art

[0002] With the development and popularization of Internet technology, online education has become an important branch in the field of education. Online education has advantages such as flexible time, free location, and rich resources, providing learners with convenient, efficient, and personalized learning services.

[0003] However, with the increase in the scale and complexity of online education, traditional education management methods have many defects and are difficult to meet the needs of online education. For example, online education usually involves a large number of learning resources, such as teachers, courses, teaching materials, etc., but traditional resource allocation methods often cannot meet the needs of online education. That is to say, each learner has his own learning characteristics, interests, and needs, but traditional education management methods often cannot achieve personalized management of learners. Learners may not be able to obtain learning resources and support suitable for themselves, resulting in poor learning effects, while some teachers may not be able to give full play to their teaching abilities.

[0004] Therefore, an intelligent management system for online education is desired. Summary of the Invention

[0005] Embodiments of the present invention provide an intelligent management system and method for online education. After collecting relevant information of target learners, course information, and alternative teacher information, semantic understanding technology is introduced at the backend to perform semantic analysis and understanding of these information, so as to automatically match suitable teachers according to the characteristics and requirements of different courses, as well as the characteristics and preferences of different learners, realizing the optimization and coordination of resource allocation. In this way, the quality and efficiency of online education can be improved, the intelligent and personalized needs of online education can be met, and the user experience and satisfaction of online education can be enhanced.

[0006] Embodiments of the present invention also provide an intelligent management system for online education, which includes: A learner and course information collection module, configured to obtain relevant information of target learners and course information; A teacher information collection module, configured to obtain relevant information of the first alternative teacher; An information semantic encoding module, configured to perform semantic encoding on the relevant information of the target learner, the course information, and the relevant information of the first alternative teacher respectively to obtain a target learner semantic encoding feature, a course information semantic encoding feature, and a first alternative teacher semantic encoding feature; A learner-course semantic fusion encoding module, which is used to fuse the target learner semantic encoding features and the course information semantic encoding features to obtain learner-course semantic encoding features; A semantic adaptation degree feature extraction module, which is used to perform semantic interaction fusion encoding on the learner-course semantic encoding features and the first alternative teacher semantic encoding features to obtain semantic adaptation features; A teacher matching module, which is used to determine whether to assign the first alternative teacher to the target learner based on the semantic adaptation features.

[0007] An embodiment of the present invention also provides an intelligent management method for online education, which includes: Obtaining relevant information of the target learner and course information; Obtaining relevant information of the first alternative teacher; Performing semantic encoding on the relevant information of the target learner, the course information and the relevant information of the first alternative teacher respectively to obtain target learner semantic encoding features, course information semantic encoding features and first alternative teacher semantic encoding features; Fusing the target learner semantic encoding features and the course information semantic encoding features to obtain learner-course semantic encoding features; Performing semantic interaction fusion encoding on the learner-course semantic encoding features and the first alternative teacher semantic encoding features to obtain semantic adaptation features; Based on the semantic adaptation features, determining whether to assign the first alternative teacher to the target learner. Description of the Drawings

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. In the drawings: Figure 1 It is a block diagram of an intelligent management system for online education provided in an embodiment of the present invention.

[0009] Figure 2 It is a flowchart of an intelligent management method for online education provided in an embodiment of the present invention.

[0010] Figure 3 It is a schematic diagram of the system architecture of an intelligent management method for online education provided in an embodiment of the present invention.

[0011] Figure 4This is an application scenario diagram of an intelligent management system for online education provided in the embodiments of the present invention. Detailed implementation manners

[0012] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer and more understandable, the following further elaborates on the embodiments of the present invention with reference to the accompanying drawings. Herein, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but do not limit the present invention.

[0013] Unless otherwise specified, all the technical and scientific terms used in the embodiments of this application have the same meanings as those commonly understood by those skilled in the technical field of this application. The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit the scope of this application.

[0014] In the description of the embodiments of this application, it should be noted that unless otherwise specified and defined, the term "connection" should be understood in a broad sense. For example, it can be an electrical connection or the connection inside two components. It can be directly connected or indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to specific situations.

[0015] It should be noted that the terms "first", "second", and "third" involved in the embodiments of this application are only used to distinguish similar objects and do not represent a specific order for the objects. It can be understood that "first", "second", and "third" can be interchanged in a specific order or sequence when permitted. It should be understood that the objects distinguished by "first", "second", and "third" can be interchanged appropriately so that the embodiments of this application described here can be implemented in an order other than those illustrated or described here.

[0016] Online education refers to an educational form that provides educational and learning services through Internet technology and online platforms. Compared with traditional face-to-face education, online education has the following characteristics and advantages: Flexible time: Online education can be carried out at any time and place. Learners can schedule their learning according to their own time and are not restricted by time and place, which is very convenient for those learners with work, family, or other time constraints.

[0017] Free location: Online education can be carried out at any location. As long as there is a device with an Internet connection, such as a computer, tablet, or mobile phone, learners can obtain educational resources and participate in learning activities at any time and place.

[0018] Rich resources: Online education provides a rich variety of learning resources, including video courses recorded by teachers, e-textbooks, online discussion forums, assignments, and quizzes. Learners can select and obtain the required learning resources according to their own needs.

[0019] Personalized learning: Online education can provide a personalized learning experience according to the characteristics and needs of learners. Through learners' learning data and feedback, the system can make intelligent recommendations and customize learning content based on learners' progress, interests, and learning styles, providing a learning experience that better meets learners' needs.

[0020] Interactivity and collaboration: Online education offers various interactive and collaborative learning methods. Learners can have online discussions, exchanges, and cooperation with teachers and other learners, jointly learn and solve problems, and promote the improvement of learning effects.

[0021] Lifelong learning: Online education provides learners with the opportunity for lifelong learning. Learners can choose different courses and learning content according to their interests and needs, without being restricted by time and location, and achieve continuous learning and self-improvement.

[0022] The development of online education benefits from the progress and popularization of Internet technology and has become an important branch in the field of education. With the continuous innovation and improvement of technology, online education will further enhance the quality and efficiency of education, providing learners with a better learning experience and personalized learning services.

[0023] With the rapid development of online education, some deficiencies in traditional education management methods have gradually emerged and are difficult to meet the needs of online education. The traditional resource allocation method often fails to meet the needs of online education. Since online education involves a large number of learning resources such as teachers, courses, teaching materials, etc., traditional education management methods are difficult to accurately match learners and resources, resulting in unbalanced resource allocation. Some popular courses may be in short supply, while some other courses may not be fully utilized due to overabundant resources.

[0024] Each learner has their own learning characteristics, interests, and needs, but traditional education management methods often fail to achieve personalized management of learners. Learners may not obtain learning resources and support suitable for themselves, resulting in poor learning effects. In the traditional face-to-face education model, teachers can better understand students' needs and provide personalized guidance, but in online education, this personalized management faces certain challenges.

[0025] In online education, the underutilization of teachers' teaching capabilities is also a problem. In traditional face-to-face education, teachers can provide a better teaching experience through real-time interaction and face-to-face guidance. However, in online education, teachers may not be able to directly perceive students' feedback and confusion, nor can they adjust teaching strategies in real time, thus affecting the teaching effect.

[0026] To solve these problems, it is necessary to introduce an intelligent management system to optimize the resource allocation and personalized management of online education. By using semantic understanding technology to analyze and match learner, course, and teacher information, the optimal allocation of resources and a personalized learning experience can be achieved. In this way, learners can obtain learning resources and support suitable for themselves, and teachers can also better exert their teaching abilities, improving the quality and efficiency of online education.

[0027] In one embodiment of the present invention, Figure 1 is a block diagram of an intelligent management system for online education provided in an embodiment of the present invention. As Figure 1 shown, the online education intelligent management system 100 according to an embodiment of the present invention includes: a learner and course information collection module 110, configured to obtain relevant information of a target learner and course information; a teacher information collection module 120, configured to obtain relevant information of a first alternative teacher; an information semantic encoding module 130, configured to perform semantic encoding on the relevant information of the target learner, the course information, and the relevant information of the first alternative teacher respectively to obtain a target learner semantic encoding feature, a course information semantic encoding feature, and a first alternative teacher semantic encoding feature; a learner-course semantic fusion encoding module 140, configured to fuse the target learner semantic encoding feature and the course information semantic encoding feature to obtain a learner-course semantic encoding feature; a semantic fitness feature extraction module 150, configured to perform semantic interaction fusion encoding on the learner-course semantic encoding feature and the first alternative teacher semantic encoding feature to obtain a semantic fitness feature; and a teacher matching module 160, configured to determine whether to assign the first alternative teacher to the target learner based on the semantic fitness feature.

[0028] In the learner and course information collection module 110, when collecting relevant information of the target learner and course information, it is necessary to ensure the accuracy and integrity of the information. At the same time, the privacy and personal information security of the learner should be protected. By obtaining the relevant information of the target learner and course information, it can provide basic data for subsequent personalized management and resource allocation. In the teacher information collection module 120, when collecting relevant information of the first alternative teacher, it is necessary to ensure the authenticity and reliability of the teacher information. Information can be collected through aspects such as the teacher's educational background, professional field, teaching experience, etc. By obtaining the relevant information of the teacher, it can provide basic data for subsequent teacher matching and personalized teaching. In the information semantic encoding module 130, when performing semantic encoding on the relevant information of the target learner, course information, and the relevant information of the first alternative teacher, appropriate natural language processing and semantic understanding technologies are used to ensure the accuracy and effectiveness of the encoding. By performing semantic encoding on the information, the information can be transformed into a feature representation that can be understood by a computer, providing a basis for subsequent semantic fusion and matching. In the learner-course semantic fusion encoding module 140, when fusing the semantic encoding features of the target learner and the semantic encoding features of the course information, an appropriate fusion method needs to be selected to retain important feature information and reduce information redundancy. Through learner-course semantic fusion encoding, a comprehensive feature representation of the learner and the course can be obtained, providing a basis for subsequent personalized matching and recommendation. In the semantic fitness feature extraction module 150, when performing semantic interaction fusion encoding on the learner-course semantic encoding features and the first alternative teacher semantic encoding features, an appropriate method is selected to extract semantic fitness features. These features should be able to reflect the matching degree between the learner and the teacher. By extracting semantic fitness features, the matching degree between the learner and the teacher can be evaluated, providing a basis for teacher allocation and personalized teaching. In the teacher matching module 160, when determining whether to allocate the first alternative teacher to the target learner based on the semantic fitness features, appropriate thresholds or rules are set to ensure the accuracy and reasonableness of the matching. Through the teacher matching module, it can be determined whether to allocate the first alternative teacher to the target learner according to the semantic fitness features between the learner and the teacher, thereby realizing personalized teaching services.

[0029] To address the above technical problems, the technical concept of this application is that after collecting the relevant information of the target learner, course information, and alternative teacher information, semantic understanding technology is introduced at the backend to perform semantic analysis and understanding of this information, so as to automatically match appropriate teachers according to the characteristics and requirements of different courses, as well as the characteristics and preferences of different learners, achieving optimization and coordination of resource allocation. In this way, the quality and efficiency of online education can be improved, the intelligent and personalized needs of online education can be met, and the user experience and satisfaction of online education can be enhanced.

[0030] Specifically, in the technical solution of the present application, first, relevant information of the target learner, as well as course information, and relevant information of the first alternative teacher are obtained. It should be understood that an online education management system needs to understand the needs of learners, the characteristics of courses, and the capabilities of teachers for appropriate matching. Semantic encoding can extract the key features of the target learner, course, and teacher for subsequent matching and recommendation processes. Therefore, in the technical solution of the present application, the relevant information of the target learner, the course information, and the relevant information of the first alternative teacher are further semantically encoded respectively to obtain a target learner semantic encoding feature vector, a course information semantic encoding feature vector, and a first alternative teacher semantic encoding feature vector. By semantically encoding each piece of relevant information, important features such as the interests of learners, the content characteristics of courses, and the professional capabilities of teachers can be captured for more accurate matching.

[0031] In a specific embodiment of the present application, the information semantic encoding module includes: a word segmentation processing unit, configured to perform word segmentation processing on the relevant information of the target learner, the course information, and the relevant information of the first alternative teacher respectively to convert the relevant information of the target learner, the course information, and the relevant information of the first alternative teacher into a target learner word sequence, a course information word sequence, and a relevant information word sequence of the alternative teacher respectively, each of which is composed of multiple words; a context encoding unit, configured to map each word in the target learner word sequence, the course information word sequence, and the relevant information word sequence of the alternative teacher to a word vector by using the embedding layer of the context semantic encoder including the embedding layer to obtain a target learner semantic encoding feature vector, a course information semantic encoding feature vector, and a first alternative teacher semantic encoding feature vector; and a semantic encoding feature generation unit, configured to use the obtained target learner semantic encoding feature vector as the target learner semantic encoding feature, the course information semantic encoding feature vector as the course information semantic encoding feature, and the first alternative teacher semantic encoding feature vector as the first alternative teacher semantic encoding feature.

[0032] Then, considering that each learner has different characteristics and needs, and each course also has different characteristics and requirements, matching the semantics of learner needs and the semantics of course requirements is an important task in the online education management system. In other words, the individual needs of learners and the characteristics of the course have an impact on the resource matching results. Therefore, in the technical solution of the present application, the target learner semantic coding feature vector and the course information semantic coding feature vector are spliced ​​to obtain the learner-course semantic coding feature vector. By splicing the semantic coding feature vectors of learners and courses, the information of learners and courses can be comprehensively considered to more comprehensively describe the learner-course matching relationship, so that the subsequent automatic matching of candidate teachers can be more accurately performed to achieve optimization and coordination of resource allocation.

[0033] In a specific embodiment of the present application, the learner-course semantic fusion coding module is used to: concatenate the target learner semantic coding feature vector and the course information semantic coding feature vector to obtain a learner-course semantic coding feature vector as the learner-course semantic coding feature.

[0034] In an online education management system, the adaptation between teachers and learners is crucial, and in the process of resource matching, factors such as the teacher's teaching style and professional ability, the learner's needs and learning style, and the characteristics and needs of the course will affect the teaching effect. Therefore, in the technical solution of the present application, a dual-stream interactive fusion module is further used to fuse the first candidate teacher semantic encoding feature vector and the learner-course semantic encoding feature vector, so as to capture the association and mutual influence between the semantic features of the first candidate teacher information and the learner-course semantic fusion features, thereby obtaining a semantic adaptation feature vector to better match a suitable teacher. It should be understood that the semantic features of the teacher information and the fusion semantic features of the learner-course can influence each other during the interaction process. By using the dual-stream interactive fusion module, the semantic features of the teacher information and the fusion semantic features of the learner-course can be interacted to obtain a richer and more accurate feature representation. This interaction can capture the association and interaction between teachers and learner-courses, improve the accuracy of the matching algorithm, and thus provide better matching and recommendation results to better meet the personalized needs of learners and provide more suitable teacher recommendations.

[0035] In a specific embodiment of the present application, the semantic adaptation feature extraction module is used to: use a dual-stream interactive fusion module to fuse the first candidate teacher semantic encoding feature vector and the learner-course semantic encoding feature vector to obtain a semantic adaptation feature vector as the semantic adaptation feature.

[0036] Among them, the semantic adaptation degree feature extraction module includes: a semantic correlation calculation unit, configured to calculate the correlation between the first alternative teacher semantic coding feature vector and the learner-course semantic coding feature vector to obtain a teacher-learner and course semantic association feature matrix; a feature interaction attention enhancement unit, configured to perform feature interaction attention coding on the first alternative teacher semantic coding feature vector and the learner-course semantic coding feature vector based on the teacher-learner and course semantic association feature matrix to obtain an attention-enhanced first alternative teacher semantic coding feature vector and an attention-enhanced learner-course semantic coding feature vector; an attention-enhanced feature fusion unit, configured to fuse the first alternative teacher semantic coding feature vector and the attention-enhanced first alternative teacher semantic coding feature vector to obtain a first alternative teacher semantic fusion local temporal feature vector, and fuse the learner-course semantic coding feature vector and the attention-enhanced learner-course semantic coding feature vector to obtain a learner-course semantic fusion local temporal feature vector; a semantic feature fusion unit, configured to fuse the first alternative teacher semantic fusion local temporal feature vector and the learner-course semantic fusion local temporal feature vector to obtain the semantic adaptation feature vector.

[0037] In a specific embodiment of the present application, the teacher matching module includes: a feature distribution optimization unit, configured to optimize the feature distribution of the semantic adaptation feature vector to obtain an optimized semantic adaptation feature vector; a teacher matching classification unit, configured to pass the optimized semantic adaptation feature vector through a classifier to obtain a classification result, and the classification result is used to indicate whether to assign the first alternative teacher to the target learner.

[0038] In the feature distribution optimization unit, by optimizing the feature distribution of the semantic adaptation feature vector, the discrimination and expression ability of the features can be improved. The optimized feature vector can better capture the semantic matching relationship between the learner and the teacher, thereby improving the accuracy and reliability of the matching. In the teacher matching classification unit, by inputting the optimized semantic adaptation feature vector into the classifier, a classification result can be obtained, which is used to indicate whether to assign the first alternative teacher to the target learner. Such a classification result can help decision-makers or systems automatically judge the matching degree between the learner and the teacher, thereby achieving more accurate and efficient teacher matching.

[0039] The feature distribution optimization unit and the teacher matching and classification unit play a key role in the intelligent management system. The feature distribution optimization unit improves the discrimination and expression ability of features by optimizing the distribution of semantic adaptation feature vectors. The teacher matching and classification unit inputs the optimized semantic adaptation feature vectors into a classifier to obtain classification results, thereby achieving an accurate judgment of the matching degree between learners and teachers. The beneficial effects of these units can improve the accuracy of teacher matching and the effect of personalized management, thus enhancing the quality of online education and the learning experience of learners.

[0040] Among them, the feature distribution optimization unit includes: a feature fusion and correction sub-unit, configured to fuse and correct the first alternative teacher semantic encoding feature vector and the learner-course semantic encoding feature vector to obtain a corrected feature vector; a semantic matching feature optimization sub-unit, configured to fuse the corrected feature vector with the semantic adaptation feature vector to obtain the optimized semantic adaptation feature vector.

[0041] In the technical solution of this application, the first alternative teacher semantic encoding feature vector and the learner-course semantic encoding feature vector are respectively used to express the text semantic encoding features of the relevant information of the first alternative teacher and the relevant information and course information of the target learner. Moreover, by using a two-stream interaction fusion module to fuse the first alternative teacher semantic encoding feature vector and the learner-course semantic encoding feature vector, the interactive fusion of the semantic encoding features of the relevant information of the first alternative teacher and the relevant information and course information of the target learner can be realized.

[0042] However, this application takes into account the difference in the text semantic feature distribution intensity between the first alternative teacher semantic encoding feature vector and the learner-course semantic encoding feature vector, that is, the text semantic feature distribution intensity of the learner-course semantic encoding feature vector is greater than that of the first alternative teacher semantic encoding feature vector. Therefore, when obtaining the semantic adaptation feature vector through interactive fusion, it may lead to unbalanced expression of the semantic adaptation feature vector, affecting the expression effect of the semantic adaptation feature vector.

[0043] Based on this, preferably, for the first alternative teacher semantic encoding feature vector, denoted as and the learner-course semantic encoding feature vector, denoted as perform fusion correction to obtain a corrected feature vector, denoted as , specifically including the following steps: Respectively, the first alternative teacher semantic encoding feature vector and the learner-course semantic encoding feature vector are passed through a common learnable weight matrix With Function activation to obtain the first alternative teacher semantic coding multi-dimensional constraint feature vector And the learner-course semantic coding multi-dimensional constraint feature vector : ; ; Wherein, Denotes dot product by position, Denotes vector subtraction, Denotes unit feature vector; Calculate the squares of the two-norms of the first alternative teacher semantic coding multi-dimensional constraint calibration feature vector And the learner-course semantic coding multi-dimensional constraint calibration feature vector To obtain the first alternative teacher semantic coding multi-dimensional constraint calibration regularization coefficient And the learner-course semantic coding multi-dimensional constraint calibration regularization coefficient ; Dot multiply the first alternative teacher semantic coding multi-dimensional constraint calibration regularization coefficient With the learner-course semantic coding multi-dimensional constraint calibration feature vector And dot multiply the learner-course semantic coding multi-dimensional constraint calibration regularization coefficient With the first alternative teacher semantic coding multi-dimensional constraint calibration feature vector After dot multiplication, then add the dot product feature vectors to obtain the corrected feature vector : ; Wherein, Denotes vector addition; That is, when integrating the dynamic interaction integration mechanism based on the language expression pattern of the first alternative teacher semantic coding feature vector And the learner-course semantic coding feature vector If the first alternative teacher semantic coding feature vector to be fused And the learner-course semantic coding feature vector Are regarded as text sequence association reinforcement signal inputs, it may weaken the first alternative teacher semantic coding feature vector And the learner-course semantic coding feature vector The inherent distribution pattern of the core representation dimension in the feature distribution difference space, resulting in the failure of prediction error compensation. Therefore, by implementing the multi-dimensional constraint calibration strategy, in the first alternative teacher semantic coding feature vector And the learner-course semantic coding feature vector Under the linear feature combination framework, a joint optimization mechanism for characterizing robustness and suppressing error propagation is constructed, which can ensure the first alternative teacher semantic encoding feature vector and the learner-course semantic encoding feature vector While enhancing the fusion features, the self-supervised balance of prediction errors is realized, and finally the effectiveness of the overall performance index is improved. In this way, by fusing the corrected feature vector With the semantic adaptation feature vector, the expression effect of the semantic adaptation feature vector can be improved, so as to improve the accuracy of the classification result obtained by the classifier. In this way, according to the characteristics and requirements of different courses and the characteristics and preferences of different learners, suitable teachers can be automatically matched, realizing the optimization and coordination of resource allocation. In this way, the quality and efficiency of online education can be improved, the intelligent and personalized needs of online education can be met, and the user experience and satisfaction of online education can be improved.

[0044] Subsequently, the optimized semantic adaptation feature vector is passed through the classifier to obtain a classification result, which is used to indicate whether the first alternative teacher is assigned to the target learner. In this way, according to the characteristics and requirements of different courses and the characteristics and preferences of different learners, suitable teachers can be automatically matched, realizing the optimization and coordination of resource allocation to improve the quality and efficiency of online education.

[0045] In a specific embodiment of the present application, the teacher matching classification unit includes: a fully connected encoding subunit, which is used to perform fully connected encoding on the optimized semantic adaptation feature vector using multiple fully connected layers of the classifier to obtain an encoded classification feature vector; and a classification subunit, which is used to pass the encoded classification feature vector through the Softmax classification function of the classifier to obtain the classification result.

[0046] In summary, the intelligent management system 100 for online education based on the embodiments of the present invention is clarified, which can improve the quality and efficiency of online education, meet the intelligent and personalized needs of online education, and improve the user experience and satisfaction of online education.

[0047] As described above, the intelligent management system 100 for online education according to an embodiment of the present invention can be implemented in various terminal devices, such as a server for intelligent management of online education. In one example, the intelligent management system 100 for online education according to an embodiment of the present invention can be integrated into the terminal device as a software module and / or a hardware module. For example, the intelligent management system 100 for online education can be a software module in the operating system of the terminal device, or can be an application program developed for the terminal device; of course, the intelligent management system 100 for online education can also be one of many hardware modules of the terminal device.

[0048] Alternatively, in another example, the intelligent management system 100 for online education and the terminal device can also be separate devices, and the intelligent management system 100 for online education can be connected to the terminal device through a wired and / or wireless network, and transmit interaction information according to a predefined data format.

[0049] Figure 2 It is a flowchart of an intelligent management method for online education provided in an embodiment of the present invention. Figure 3 It is a schematic diagram of the system architecture of an intelligent management method for online education provided in an embodiment of the present invention. As Figure 2 and Figure 3 shown, an intelligent management method for online education includes: 210, obtaining relevant information of a target learner and course information; 220, obtaining relevant information of a first alternative teacher; 230, respectively performing semantic encoding on the relevant information of the target learner, the course information, and the relevant information of the first alternative teacher to obtain a target learner semantic encoding feature, a course information semantic encoding feature, and a first alternative teacher semantic encoding feature; 240, fusing the target learner semantic encoding feature and the course information semantic encoding feature to obtain a learner-course semantic encoding feature; 250, performing semantic interaction fusion encoding on the learner-course semantic encoding feature and the first alternative teacher semantic encoding feature to obtain a semantic adaptation feature; 260, based on the semantic adaptation feature, determining whether to assign the first alternative teacher to the target learner.

[0050] In the intelligent management method for online education, semantic encoding is respectively performed on the relevant information of the target learner, the course information, and the relevant information of the first alternative teacher to obtain the semantic encoding features of the target learner, the semantic encoding features of the course information, and the semantic encoding features of the first alternative teacher, including: performing word segmentation on the relevant information of the target learner, the course information, and the relevant information of the first alternative teacher respectively to convert the relevant information of the target learner, the course information, and the relevant information of the first alternative teacher into a target learner word sequence, a course information word sequence, and a relevant information word sequence of the alternative teacher respectively composed of multiple words; using the embedding layer of the context semantic encoder including the embedding layer to map each word in the target learner word sequence, the course information word sequence, and the relevant information word sequence of the alternative teacher to a word vector to obtain a target learner semantic encoding feature vector, a course information semantic encoding feature vector, and a first alternative teacher semantic encoding feature vector; and using the obtained target learner semantic encoding feature vector as the semantic encoding features of the target learner, the course information semantic encoding feature vector as the semantic encoding features of the course information, and the first alternative teacher semantic encoding feature vector as the semantic encoding features of the first alternative teacher.

[0051] Those skilled in the art can understand that the specific operations of each step in the above intelligent management method for online education have been described in detail in the description of the Figure 1 intelligent management system for online education above, and therefore, its repeated description will be omitted.

[0052] Figure 4 This is an application scenario diagram of an intelligent management system for online education provided in an embodiment of the present invention. As Figure 4 shown, in this application scenario, first, relevant information of the target learner (for example, C1 as shown in Figure 4 ) is obtained, and course information (for example, C2 as shown in Figure 4 ) is obtained; and relevant information of the first alternative teacher (for example, C3 as shown in Figure 4 ) is obtained; then, the obtained relevant information of the target learner, the course information, and the relevant information of the first alternative teacher are input into a server (for example, S as shown in Figure 4 ) deployed with an intelligent management algorithm for online education, where the server can process the relevant information of the target learner, the course information, and the relevant information of the first alternative teacher based on the intelligent management algorithm for online education to determine whether to assign the first alternative teacher to the target learner.

[0053] The specific embodiments described above further elaborate on the objective, technical solution and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent online education management system, characterized in that: include: The learner and course information collection module is used to obtain relevant information of target learners and course information; A teacher information collection module is used to obtain relevant information of the first candidate teacher; An information semantic coding module, used for semantically coding the relevant information of the target learner, the course information and the relevant information of the first candidate teacher respectively to obtain semantic coding features of the target learner, semantic coding features of the course information and semantic coding features of the first candidate teacher; A learner-course semantic fusion coding module, used for fusing the target learner semantic coding features and the course information semantic coding features to obtain learner-course semantic coding features; A semantic adaptation feature extraction module, used for performing semantic interaction fusion encoding on the learner-course semantic coding feature and the first candidate teacher semantic coding feature to obtain a semantic adaptation feature; The teacher matching module is used to determine whether to assign a first candidate teacher to the target learner based on the semantic adaptation feature.

2. The online education intelligent management system according to claim 1, characterized in that: The information semantic encoding module comprises: A word segmentation processing unit is used to perform word segmentation processing on the relevant information of the target learner, the course information and the relevant information of the first candidate teacher respectively, so as to convert the relevant information of the target learner, the course information and the relevant information of the first candidate teacher into a target learner word sequence, a course information word sequence and a candidate teacher related information word sequence respectively consisting of a plurality of words; a context encoding unit, configured to map each word in the target learner word sequence, the course information word sequence, and the candidate teacher's related information word sequence to a word vector using the embedding layer of the context semantic encoder including the embedding layer to obtain a target learner semantic encoding feature vector, a course information semantic encoding feature vector, and a first candidate teacher semantic encoding feature vector; and A semantic coding feature generating unit is used to use the obtained target learner semantic coding feature vector as the target learner semantic coding feature, the course information semantic coding feature vector as the course information semantic coding feature and the first candidate teacher semantic coding feature vector as the first candidate teacher semantic coding feature.

3. The online education intelligent management system according to claim 2 is characterized in that: The learner-course semantic fusion coding module is used to: concatenate the target learner semantic coding feature vector and the course information semantic coding feature vector to obtain a learner-course semantic coding feature vector as the learner-course semantic coding feature.

4. The online education intelligent management system according to claim 3 is characterized in that: The semantic adaptation feature extraction module is used to: use a dual-stream interactive fusion module to fuse the first candidate teacher semantic encoding feature vector and the learner-course semantic encoding feature vector to obtain a semantic adaptation feature vector as the semantic adaptation feature.

5. The online education intelligent management system according to claim 4, characterized in that: The semantic adaptation feature extraction module comprises: A semantic relevance calculation unit, used for calculating the relevance between the first candidate teacher semantic coding feature vector and the learner-course semantic coding feature vector to obtain a teacher-learner and course semantic association feature matrix; A feature interaction attention enhancement unit, configured to perform feature interaction attention encoding on the first candidate teacher semantic encoding feature vector and the learner-course semantic encoding feature vector based on the teacher-learner and course semantic association feature matrix to obtain an attention-enhanced first candidate teacher semantic encoding feature vector and an attention-enhanced learner-course semantic encoding feature vector; an attention-enhanced feature fusion unit, configured to fuse the first candidate teacher semantic coding feature vector and the attention-enhanced first candidate teacher semantic coding feature vector to obtain a first candidate teacher semantic fusion local temporal feature vector, and to fuse the learner-course semantic coding feature vector and the attention-enhanced learner-course semantic coding feature vector to obtain a learner-course semantic fusion local temporal feature vector; A semantic feature fusion unit is used to fuse the first candidate teacher semantic fusion local temporal feature vector and the learner-course semantic fusion local temporal feature vector to obtain the semantic adaptation feature vector.

6. The online education intelligent management system according to claim 5, characterized in that: The teacher matching module includes: A feature distribution optimization unit, used for performing feature distribution optimization on the semantic adaptation feature vector to obtain an optimized semantic adaptation feature vector; The teacher matching classification unit is used to pass the optimized semantic adaptation feature vector through a classifier to obtain a classification result, and the classification result is used to indicate whether to assign the first candidate teacher to the target learner.

7. The online education intelligent management system according to claim 6, characterized in that: The feature distribution optimization unit comprises: A feature fusion correction subunit, used for fusing and correcting the first candidate teacher semantic encoding feature vector and the learner-course semantic encoding feature vector to obtain a corrected feature vector; The semantic matching feature optimization subunit is used to fuse the correction feature vector with the semantic adaptation feature vector to obtain the optimized semantic adaptation feature vector.

8. The online education intelligent management system according to claim 7, characterized in that: The teacher matching classification unit comprises: a fully connected encoding subunit, configured to perform fully connected encoding on the optimized semantic adaptation feature vector using a plurality of fully connected layers of the classifier to obtain an encoded classification feature vector; and The classification subunit is used to pass the encoded classification feature vector through the Softmax classification function of the classifier to obtain the classification result.

9. An intelligent management method for online education, characterized in that: include: Obtain relevant information of target learners and course information; Obtain relevant information of the first candidate teacher; Semantically encoding the relevant information of the target learner, the course information, and the relevant information of the first candidate teacher respectively to obtain semantic encoding features of the target learner, semantic encoding features of the course information, and semantic encoding features of the first candidate teacher; Fusion of the target learner semantic coding features and the course information semantic coding features to obtain learner-course semantic coding features; Performing semantic interactive fusion coding on the learner-course semantic coding feature and the first candidate teacher semantic coding feature to obtain a semantic adaptation feature; Based on the semantic adaptation feature, it is determined whether to assign a first candidate teacher to the target learner.

10. The online education intelligent management method according to claim 9, characterized in that: Semantically encoding the relevant information of the target learner, the course information, and the relevant information of the first candidate teacher respectively to obtain semantic encoding features of the target learner, semantic encoding features of the course information, and semantic encoding features of the first candidate teacher, including: Performing word segmentation processing on the relevant information of the target learner, the course information and the relevant information of the first candidate teacher respectively to convert the relevant information of the target learner, the course information and the relevant information of the first candidate teacher into a target learner word sequence, a course information word sequence and a candidate teacher related information word sequence respectively consisting of multiple words; Map each word in the target learner word sequence, the course information word sequence and the candidate teacher's related information word sequence to a word vector using the embedding layer of the context semantic encoder including the embedding layer to obtain a target learner semantic encoding feature vector, a course information semantic encoding feature vector and a first candidate teacher semantic encoding feature vector; and The obtained target learner semantic coding feature vector is used as the target learner semantic coding feature, the course information semantic coding feature vector is used as the course information semantic coding feature, and the first candidate teacher semantic coding feature vector is used as the first candidate teacher semantic coding feature.

Images