Research on Networked Integration of Ideological and Political Education Teaching Resources in the Information Age

Ideological and political education is a kind of educational activity aimed at cultivating citizens’ political awareness, political literacy and political ability through schools, social organizations and other educational institutions [1-2]. It aims to guide citizens to form a correct worldview, outlook on life and values, to cultivate citizens’ civic morality, legal concept, national concept and sense of social responsibility, to raise the level of citizens’ understanding of the political system, political culture, political ideas and political participation [3-6], to guide citizens to actively participate in political activities in social life, and to maintain the stability and harmonious development of the state and society. And with the development of the information age, the integration of teaching resources of ideological and political education is of great significance to promote ideological and political education [7-10].

In recent years, the education department attaches great importance to the work of ideological and political education, and puts forward the fundamental task of “establishing morality and educating people”, and strengthens the construction of curriculum ideology and politics. In order to better achieve the goal of ideological and political education, based on the background of the era of information technology, China’s universities and colleges are actively carrying out the network integration construction of ideological and political teaching resources [11-14]. Civic and political teaching resources in colleges and universities are rich and diversified, including teaching materials, lesson plans, teaching methods, teaching cases, assessment tools, etc., which not only provide teachers with continuously updated teaching resources, but also provide students with more diversified learning experiences and challenges [15-17]. The network integration of Civics teaching resources has significant effects in improving teaching quality and promoting education and teaching reform [18-19].

This paper focuses on the networked integration of ideological and political education teaching resources. It analyzes the process of ontology-based teaching resources integration, constructs an ontology library of ideological and political teaching resources based on the seven-step development method, optimizes the semantic similarity calculation method and the semantic data retrieval strategy therein, and verifies its effectiveness through experiments. The ontology library of ideological and political teaching resources is applied to the teaching of ideological and political courses, and the effectiveness of the ontology library for teachers’ teaching and students’ learning is verified through controlled experiments, data analysis and satisfaction surveys.

2

Construction and optimization of the ontology library of ideological and political teaching resources

To integrate and study networked ideological and political education teaching resources, it is necessary to construct the corresponding ideological and political teaching resources ontology library according to the development steps, and optimize and iterate the ontology library to make it reach the level that can be used by the users to study and research. The following section will explain how to construct and optimize an ontology library of ideological and political teaching resources.

2.1

Ontology-based teaching resource integration process analysis

The process of integrating ontology-based instructional resources is an incremental and iterative development process, in which new concepts are continuously added and existing concepts are modified. Therefore, an important issue is how to ensure complete, consistent, and concise concept definitions and descriptions at the initial stage of ontology development, during the various stages of development, and between various development phases. Ontology-based integration of instructional resources has four main phases: requirements analysis of instructional resource ontologies, building a shared vocabulary of instructional resources (global ontology), building local ontologies, and implementation mapping. The basic process is shown in Figure 1. Each phase includes a number of tasks that must be accomplished.

Stage 1: Needs analysis of the teaching resource ontology. This stage clarifies the purpose, scope, use and users of the ontology construction in the teaching domain.

Phase 2: Building a shared vocabulary (global ontology). This phase consists of several main steps: analyzing data sources, finding terms (proto-languages) and defining the global ontology. The first step is a complete analysis of all data sources, e.g., what information is stored in the sources, how they are stored, the meaning of the information (semantics), etc. The second step is to find the terms (primitives) and select the terms or concepts that should be written into the shared vocabulary. The third stage involves defining the global ontology by creating a global ontology using the terms selected in the previous step.

Phase 3: Create local ontology. This phase contains two important steps: analyzing the data source and defining the local ontology. The first step is similar to the first step of the previous phase, where each data source is fully analyzed, but the difference is that this analysis is performed independently without considering other data sources.

Phase 4: Implementing the mapping. In this phase, the mapping (and relations) of concepts between the global ontology and the local ontology are defined. Semantic heterogeneity must be addressed during this phase.

2.2

Metadata-based ontology library construction of ideological and political teaching resources

After understanding the process of integrating teaching resources based on ontology, this part will introduce the current method of building teaching resources ontology based on CELTS specification and introducing metadata-related methods, and build a library of ideological and political teaching resources ontology based on this method.

2.2.1

Explanation of the Ontology Library of Ideological and Political Teaching Resources Hierarchy

The ontology repository is organized in different tiers, with differences in what each tier contains and who it serves, etc. The following section describes the different levels specifically in relation to the ontology library framework in Figure 2.

The fifth (bottom) layer is the content layer, which is the available resource entity, containing a large amount of webpage information, courseware material, video teaching, etc. It plays the role of providing basic resources, and is the most important component and basic entity of educational resources.

The fourth layer is the metadata layer. The establishment of this layer requires two processes of metadata collection and metadata integration. Metadata collection mainly involves extracting relevant data from the content layer, and implementing a mechanism to ensure the correctness and completeness of the extracted data. Metadata integration refers to mapping the extracted metadata to the normalized ontology framework by adding network identifiable descriptions according to the CELTS standard.

The third layer is the ontology layer. This layer provides a structured and systematized collection of data metadata for upper layer services to search and query. That is, this layer is an abstract representation layer of resources, which is used to provide the interface between upper layer services and underlying resources, formalizing the underlying data.

The second layer is the resource service layer. This layer has two roles: to receive the upper layer query information and to provide feedback on the lower layer return information. This layer mainly implements semantic detection algorithms and provides interfaces to the upper layer, which can be defined for public or independent use.

The first layer is the application layer, mainly various user applications. This layer sends down information query requests and is the initiator and final recipient of information queries.

2.2.2

Construction of an ontology library of ideological and political teaching resources based on the seven-step development approach

The ontology library is at five different levels, and the construction of the ontology library needs to be combined with the seven-step development method. In the following section, the ontology library of ideological and political teaching resources will be developed according to the CELTS specification based on the seven-step development method, combined with the metadata of teaching resources. Figure 3 shows the process of developing the metadata and ontology library using the seven-step development method with reference to the CELTS standard.

1)

Determine the scope of the ontology domain

Determine the domain in which the ontology is applied, understand the knowledge composition and representation of the domain, and determine the basic program for the establishment of the ontology. The Ideological and Political Teaching Resources Ontology Library is mainly applied to teaching and research, so it mainly collects the relevant disciplinary resource data in the database and the relevant textual information on the page.

2)

Consider reuse of existing ontologies

In the process of metadata collection, if there is an ontology library that has been defined in the domain, or if there is a knowledge base that is related to the knowledge base that you want to construct, then you can consider reusing the existing knowledge base or establishing a mapping relationship with that knowledge base. In the process of building the ontology library of ideological and political teaching resources, knowledge bases such as the Knowledge Network Dictionary and the Dictionary of Common Word Synonyms are utilized.

3)

List the important terms of the ontology

List the description standards and meta-semantics required for the specification. This step completes the extraction, definition, and classification by structure of metadata related to the ideological and political boutique courses, and in the process of extraction and definition, it is executed in strict accordance with the CELTS standard, and metadata that cannot be clearly defined are manually defined and classified, and tagged for submission to form a standardized document.

4)

Defining Classes and Class Hierarchies

Classes are the core elements of most ontology libraries to describe the knowledge in the domain. The process of creating an ontology library of ideological and political teaching resources takes into account the usability and scalability of classes, as well as the structuring and importance hierarchy of resources.

5)

Defining attributes of classes

Attributes are the characteristics of classes, metadata, etc., describing the difference between metadata, classes, etc. and other metadata and classes. In the process of constructing the ontology library of ideological and political teaching resources, the scope of the definition of each class is clarified to ensure its standardization and normalization.

6)

Define the constraints of attributes

Attributes can describe metadata, and attributes also need to be defined and constrained. In the process of constructing the ontology library of ideological and political teaching resources, the value domain, default value, value type and other characteristics of the attributes are clearly defined, and the attributes are constrained by the database definitions.

7)

Generate instances

After the above steps, the ontology library of ideological and political teaching resources has been basically established successfully, but it still needs to be tested and evaluated in the near future, to evaluate the knowledge of the generated ontology library, to put forward the lack of knowledge and incomplete knowledge, to make supplementary modifications, and to finalize the establishment of the knowledge base ontology.

2.3

Improvement of Semantic Similarity Calculation Based on the Ontology Library of Ideological and Political Teaching Resources

After the ontology of ideological and political teaching resources has been successfully constructed, it is necessary to continuously improve the calculation method of semantic similarity, expand and enrich the knowledge points and related resources of the ontology, and improve the accuracy of the resource retrieval results for users. In the following section, we will explain the meaning of semantic similarity and the improvement method of calculating semantic similarity, and continuously optimize and iterate the ontology of ideological and political teaching resources, so as to make it better serve the teachers and students in ideological and political teaching and learning.

2.3.1

Explanation of the basic connotations of semantic similarity

Semantic similarity refers to the degree to which two concepts are semantically related. Semantic similarity computation is usually used in semantic query expansion to obtain the set of concepts that are semantically related to the user’s retrieval request from the ontology. In this paper, when a feature item of a resource cannot be directly queried in the Ideological and Political Teaching Resources ontology library, ontological reasoning is utilized to obtain the set of knowledge points related to the feature item. After calculating each knowledge point, the knowledge point with the strongest correlation is selected among the knowledge points exceeding the threshold range, and the feature item is added to the knowledge base in the ideological and political teaching resources ontology library with that knowledge point as the node.

There are knowledge point classes in the ontology library of ideological and political teaching resources constructed in this paper, as well as many object attributes and data attributes. To obtain the specific relationship between the feature items representing the resources and the inferred set of knowledge points, the semantic relatedness between the knowledge point concepts is calculated.

Through the investigation of related literature, it is found that the basic computation methods of ontology-based semantic relatedness computation and web-of-knowledge-based semantic relatedness computation are relatively mature, but they should be improved according to different domains as well as the situation of the constructed ontology knowledge base.

2.3.2

Improved ontology-based semantic similarity computation of concepts

Based on the above research, this paper proposes improvements to the ontology-based concept semantic similarity algorithm: 1)

Relationship type

Since most of the knowledge points in the ontology knowledge base of ideological and political teaching resources are relatively dense, the hierarchical relationship of the knowledge points is distinct, but the depth of the hierarchy will not be very large. Earlier distance-based semantic similarity computations set each edge to contribute equally to the semantic similarity, which proved to be unscientific. So in this paper, the semantic distance model is improved on the basis of the semantic distance model by adding the influence of relationship type to the process of semantic distance calculation. The shortest path between two concept nodes is defined as the semantic distance and influences the degree of semantic similarity, however, each edge contributes unequally to the semantic similarity, as shown in Fig. 4, to find the semantic distance between concept E and P, it goes through E − > B − > M − > P. E − > B (edge E1) is the containment relation, B − > M (edge E2) is the reference relation, and M − > P (edge E3) is the dependency relation, the procedure of the influence of the containment relation, the reference relation, and the dependency relation on the semantic similarity is definitely different. The influence procedures of inclusion, reference, and dependency relations on semantic similarity are definitely different. Different edges carry different types of relations, and different weights are assigned to different relationship types, so that the degree of similarity between the two concepts can be more accurately represented. The weight of each relation in a relation type is mainly determined by domain experts.

2)

Semantic overlap

The semantic overlap degree is used to indicate the ratio of the superordinate concepts shared by two concepts to the total superordinate concepts, indicating the degree of overlap of the concepts.

The formula for calculating the semantic overlap degree is referenced: (1) $S i m (C_{1}, C_{2}) = \frac{| C o u n t (U (C_{1}) \cap U (C_{2})) |}{| C o u n t (U (C_{1}) \cup U (C_{2})) |}$

where $U (C_{1})$ and $U (C_{2})$ denote the set of all ancestor nodes of concept nodes C₁ and C₂, respectively. The computational complexity of this formula is also not very high. 3)

Retain the distance-based semantic similarity, and make a simple screening of other factors affecting the calculation of semantic similarity. Combined with the previous section, it can be seen that the organization level of knowledge points in the knowledge base of teaching resources ontology is generally three to four layers, which has little influence on the semantic similarity, thus the node depth of concept nodes does not make too many requirements, and the density of nodes and the strength of edges are not taken into account in the calculation. Semantic distance and semantic similarity have an inverse relationship, that is, the larger the semantic distance, the smaller the semantic similarity, and vice versa. Distance based semantic similarity calculation formula: (2) $S i m (C_{1}, C_{2}) = \frac{α}{d i s \tan c e (C_{1}, C_{2}) + α}$

where $d i s \tan c e (C_{1}, C_{2})$ represents the semantic distance between concepts C₁, C₂, and α is the moderating factor.

When calculating the semantic distance between concepts C₁ and C₂, the weights of the edges corresponding to the relationship types are added to the distance calculation, taking into account the influence of the edge relationship types. The edge type (relationship type) is positively proportional to the semantic similarity, i.e., the closer the relationship type, the greater the semantic similarity. Since the above distance calculation is located in the denominator part, the weights should be opposite to it at this time, the smaller the weight value, the larger the semantic relationship. At the same time the Weighted Link method is introduced, then the semantic distance calculation formula is: (3) $d i s \tan c e (C_{1}, C_{2}) = {(\sum_{t = 1}^{n} l e n (e_{t}))}^{2}$

where $l e n (e_{i})$ denotes the ith edge through which concepts C₁ and C₂ are connected, and $l e n (e_{i})$ denotes the weight of this edge. The weights of the types of relations covered in this paper are set by the reference of the domain experts:. $l e n (e) = {\begin{matrix} 0 & Synonymous relationship & s a m e A s \\ 0.2 & inclusionary relationship & c o n t a i n 、 p a r t O f \\ 0.3 & Dependency & \begin{matrix} d e c i d e d B y 、 b e f o r e 、 \\ a f t e r 、 b a s e d O n \end{matrix} \\ 0.4 & Reference & r e f e r e n c e O f 、 r e p r e s e n t e d B y \\ 0.5 & Fraternization & h a s S i b l i n g \\ 0.7 & Separate Relationship & p a r a l l e l W i t h \\ 0.8 & Operational relationship & \begin{matrix} a t t r i b u t e O f 、 c o m p a r e d W i t h \\ h a s C o n d i t i o n 、 h a s O p e r a c t o r \\ h a s C o n v e r s i o n 、 h a s A t t r i b u t i o n \end{matrix} \end{matrix}$

Summarizing the above analysis, this paper proposes a formula for calculating the semantic similarity between two knowledge point concepts based on the domain ontology: (4) $S i m (C_{1}, C_{2}) = β \frac{α}{{(\sum_{i = 1}^{n} l e n (e_{i}))}^{2} + α} + γ \frac{| C o u n t (U (C_{1}) \cap U (C_{2})) |}{| C o u n t (U (C_{1}) \cup U (C_{2})) |}$

Where, $C o u n t (U (C_{1}) \cap U (C_{2}))$ denotes the common number of superordinate nodes of concepts C₁ and C₂, and $Count (U (C_{1}) \cup U (C_{2}))$ denotes the sum of the number of superordinate nodes owned by concepts C₁ and C₂ respectively. Because β and γ are the regulating factors of semantic distance, relationship type and semantic overlap, the above formula is improved a little bit, generally, β + γ = 1. According to the actual situation of the ontology knowledge base of ideological and political teaching resources, β and γ in this paper are taken as 0.6 and 0.4, respectively.

2.3.3

Improved Web of Knowledge-based Semantic Similarity Calculation of Concepts

Upon the above analysis, it is known that another more used method of calculating semantic similarity is based on some kind of world knowledge. That is, the similarity of the whole is based on the similarity of the parts. The whole is divided into different parts, and if each corresponding part is similar, the whole is also similar. So the similarity of the whole can be obtained by calculating the sum of the similarity of each part. In KnowledgeNet, vocabulary is categorized into dummy words and real words. Usually, the semantics of dummy words do not have much influence on the semantics of a word or concept, and in this paper, the semantics of dummy words are not taken into account in the calculation. Consequently, in this paper, the first independent semantic primitive and relational semantic primitive descriptors will be utilized for the calculation of semantic similarity between two concepts. The first independent semantic principle’s similarity formula: (5) $S i m (S_{1}, S_{2}) = \frac{α}{d i s \tan c e (S_{1}, S_{2}) + α}$

where the justification principles S₁, S₂ are the first independent justification principles of concepts C₁, C₂ respectively. $d i s t a n c e (S_{1}, S_{2})$ is the semantic distance of S₁, S₂.

The relational justification principle and the relational symbolic justification principle together form the relational feature justification principle in KnowledgeNet. Generally speaking, the relational symbolic justification is the interpretation of the relational justification, here, we do not require the relational symbolic justification, but only calculate the similarity of the first attribute justification in the relational justification to represent the similarity of the relational justification, and the specific formula also refers to the above semantic distance formula, of course, we have to take the value of β in it. (6) $S i m_{t y p e} (S_{3}, S_{4}) = \frac{β}{d i s \tan c e (S_{3}, S_{4}) + β}$

where the justification principles S₃ and S₄ are the first relational justification principles of concepts C₁ and C₂, respectively. $d i s \tan c e (S_{3}, S_{4})$ is the semantic distance of S₃ and S₄.

Then the semantic similarity formula between two concepts is: (7) $S i m (C_{1}, C_{2}) = \frac{α}{d i s \tan c e (S_{1}, S_{2}) + α} + \frac{β}{d i s \tan c e (S_{3}, S_{4}) + β}$

By clarifying the basic connotation of semantic similarity and improving the calculation method of semantic similarity, it can effectively improve the resource query accuracy of the ontology knowledge base of ideological and political teaching resources, and expand and enrich the knowledge points and resources of the knowledge base.

2.4

Semantic Data Retrieval Based on the Ontology Library of Ideological and Political Teaching Resources

To further improve the accuracy of users’ query and retrieval of information in the ontological knowledge base of ideological and political teaching resources, in addition to the improvement of the calculation method for semantic similarity, it is still necessary to choose a scientific and effective semantic retrieval strategy. The semantic retrieval strategy and its steps adopted in this paper will be described below to improve users’ retrieval accuracy.

2.4.1

Description of Semantic Data Retrieval Strategy Adoption

Among the current semantic retrieval methods, the semanticization of retrieved information is the most difficult and strongly depends on natural language processing techniques. Because no matter how scientific and comprehensive the ontology is built to cover all known human knowledge, there is no method to deal with the user’s input, how to analyze lexically, syntactically and contextually to get which kind of association between two words belongs to the ontology, especially under the premise of the complexity of Chinese expressions.

After expanding the retrieval information entered by the user, a set of search terms that satisfy the user’s search intent is obtained. At this point, it needs to be sent to the retrieval component for data retrieval, and the retrieval results will be returned to the user. In this paper, we adopt a strategy that combines vector space modeling and ontology reasoning for learning object retrieval, obtaining preliminary resource retrieval results through similarity calculation, and then using teaching resource ontology to reason about the relationship between resources to obtain other reasonable retrieval results.

2.4.2

Step-by-step description of the semantic data retrieval strategy

This paper on semantic data retrieval strategy contains the following 6 steps:

Step1 Learning resource ontology sub-model extraction

According to the knowledge points and equivalent knowledge points found in the semantic query expansion process, find these resource instances in the ideological and political teaching resource ontology through the relationship between the knowledge points and the resources associated with them, and take out the metadata instances of the resources and the associated relationships for use in retrieval.

Step2 Construct Document Semantic Vector

Document semantic vector refers to analyzing the content of the resources under the domain ontology, extracting the feature words that can reflect the concepts or knowledge of the domain, searching for the relationship between these feature words, and using these feature words to build the document vector. However, the vocabulary owned by the domain ontology has limitations, so there will be some cases where some feature words with higher weights can be used as index items but do not belong to the knowledge in the domain ontology. For any document D_j ∈ D, its vector form can be expressed as: $D_{j} = (w_{t_{1}}, w_{t 2} \dots w_{t_{i}}, w_{t_{i + 1}}, w_{t_{i + 2}} \dots w_{t_{m}}) = (w_{c_{1}}, w_{c 2} \dots w_{c_{i}}, w_{t_{i + 1}}, w_{t_{i + 2}} \dots w_{t_{m}})$

where set ${t_{l} \dots t_{m}}$ represents the set of words as index items in document D_j, w_ti represents the weight of word t_i in document d, set ${c_{l} \dots c_{m}}$ represents the concepts or instances in the domain ontology corresponding to each feature word in set ${t_{l} \dots t_{m}}$ , and set ${t_{i + I} \dots t_{m}}$ indicates that any feature word in the set is not knowledge in the domain ontology.

Currently, the computation of each component of vector d is mainly by using the TF-IDF method, whose basic formula is: (8) $w_{i k} = \frac{t f_{i k} \times \log (N / d f_{k})}{\sqrt{\sum_{t_{k} \in D_{i}} {[t f_{i k} \times \log (N / d f_{k})]}^{2}}}$

tf_ik indicates the frequency of occurrence of word t_k in document D_i, df_k is the number of documents containing word t_k, N is the number of documents, and $i d f_{k} = \log (N / d f_{k})$ is the frequency of inverse documents. The larger tf_ik indicates that the word t_k contributes more to the document D_i; the larger df_k indicates that the word t_k shares less in the whole document set; the larger idf_k indicates that the word t_k has a higher role in the document. Comprehensively speaking, using TF-IDF for word weighting can give higher weights to words that often appear in fewer documents and less frequently in the whole document set. However, through researching related literature, it is found that the same words appearing in different positions in Web documents have different degrees of representation of the documents in terms of topics. Based on such research results, combined with the definition of multiple attributes in the teaching resource ontology, different weights can be assigned to words appearing in different metadata items.

Let the data type attribute set $I T E M = {i t e m_{1}, i t e m_{2} \dots i t e m_{m}}$ of the metadata of resource instance R, and the formula for improving the meta TF-IDF is: (9) $\begin{matrix} w_{i t e m_{i} k} = \sum_{i = 1}^{n} \frac{t f_{i t e m_{i} k} \times \log (N_{i t e m_{i}} / d f_{i t e m, k})}{\sqrt{\sum_{t_{k} \in i t e m_{i}} {[t f_{i t e m_{i} k} \times \log (N_{i t e m_{i}} / d f_{i t e m_{i} k})]}^{2}}} \times τ_{i t e m_{i}} \\ τ_{i t e m_{i}} (0, 1) A n d \sum_{i = 1}^{n} τ_{i t e m_{i}} = 1 \end{matrix}$

Equation (9) makes a hierarchical treatment on the original TF-IDF method, where tf_itemik denotes the frequency of words t_k in metadata attributes item_i, df_itemik denotes the number of resources containing words t_k in the item_i attributes of the set of resource instances, N_itemi denotes the number of resources with item_i attributes in the set of resource instances, and τ_i is an adjustable parameter that denotes the contribution rate of the metadata item item_i to the resource R in terms of content/topic.

According to Equation (9), the weight vector on metadata item_i in resource R can be expressed as: $w (i t e m_{i}) = (w_{i t e m_{i} k_{1}}, w_{i t e m_{i} k_{2}} \dots w_{i t e m_{i} k_{m}})$

By building vectors for multiple metadata attributes, the semantic vector about resource instance R is partitioned into a vector space, $w (R) = {w (i t e m_{1}), w (i t e m_{2}) \dots w (i t e m_{n})}$ .

Step3 Constructing the query semantic vector

After constructing the semantic vector of resource instance R, it is necessary to construct the retrieval vector of the user input retrieval information. Let the search terms expanded by the semantic query be the set Q′, and arrange the search terms in the order of the defined priority set to form the retrieval vectors without weights $Q' = (q_{l}, q_{2} \dots q_{m})$ , q_i representing the search terms in the query. Its weighted form is $Q' = (w_{q}, w_{q 2} \dots w_{q m})$ , w_qi denotes the weight of the search term q_i in the query vector. Regarding component w_qi can be calculated using equation (10): (10) $w_{q_{i}} = {\begin{matrix} 0 & I f q_{i} \notin i t e m_{i} \land q_{i} \in Q^{'} \\ \sum_{i = 1}^{n} \log (N_{i t e m_{i}} / d f_{i t e m, k}) τ_{i t e m}^{i} & I f q_{i} \in i t e m_{i} \land q_{i} \in Q^{'} \end{matrix}$

When the search term q_i in Q′ is present in the data attributes, to facilitate the evaluation of the weight of q_i in the set of search terms, the hierarchical idf calculation is used, which is expressed as the global contribution of the word index term in the repository.

Step4 Query-Resource Similarity Calculation

Similarity occupies a very important position in semantic data retrieval methods that mix vector spaces and ontologies, because the size of the similarity directly determines whether the resource should be put into the retrieval result or not. After the retrieved information and resources are transformed into vectors, the similarity can be calculated in various ways, including vector inner product, vector cosine, Dice coefficient and Jaccard coefficient. In this paper, the most commonly used vector cosine is used to solve the similarity $S i m (Q, R)$ , which is calculated by the formula: (11) $\begin{array}{rcl} S i m (Q, R) & = & \frac{1}{n} \sum_{i = 1}^{n} S i m (Q, w (i t e m_{i})) \\ = & \frac{1}{n} \sum_{i = 1}^{n} \frac{\sum_{j = 1}^{m} w_{i t e m_{i} j} w_{q_{j}}}{\sqrt{(\sum_{j = 1}^{m} w_{i t e m_{i} j}^{2}) (\sum_{j = 1}^{m} w_{q_{j}}^{3})}} \end{array}$

Thus in Eq. (11), n is the upper limit of predefined metadata attributes that can be used as constructed vectors, $S i m (Q, w (i t e m_{i}))$ is the similarity of Q to metadata item item_i, and m is the dimension of an individual vector. The resource instance R is transformed into a vector space $w (R) = {w (i t e m_{1}), w (i t e m_{2}) \dots w (i t e m_{n})}$ consisting of multiple metadata attribute vectors, and thus the similarity of individual metadata attributes to the retrieval vector needs to be summed and then averaged.

After completing the query-resource similarity computation, the computation results are sorted by the size of the similarity value, and a threshold value is set to put the resources larger than the threshold value into the retrieval result collection Result.

Step5 Ontology Reasoning

In the retrieval result Result, the resources are obtained by the way of calculation, however, in the ideological and political teaching resources ontology library, there are many kinds of relationships between learning objects, only through the semantic query extension to enrich the original retrieval information, in the similarity calculation combined with the vector space model to make more relevant resources fall into the retrieval results, but ignored the relationship between the resources on the retrieval results.

After obtaining the retrieval results, the ontology reasoning machine is activated to reason about the ontology of ideological and political teaching resources and obtain the reasoned ontology. Let the set of relations about the resource be ${r_{1}, r_{2} \dots r_{m}}$ , and for each relation, the relation weight value is predefined by the domain expert ${α_{l}, α_{2} \dots α_{m}}$ . Let a certain instance of learning object R and R′ that has fallen into the retrieval result have the relations ${r_{i}, r_{i + 1} \dots r_{j}}$ , Sim(Q, R) = V, then R and R′ are computed on the semantic association as: (12) ${Sim}_{r} (R, R') = V * \max (r_{i}), i \leq l \leq j$

If the value of $S i m_{r} (R, R')$ is greater than the threshold value, then it can be put into the retrieval result Result and sorted, and if R′ is already in the retrieval result, then there is no need to calculate $S i m_{r} (R, R')$ .

Step6 Output retrieval result

The final retrieval results are obtained through Step4 and Step5, sorted by similarity size and output, and fed back to the user.

Using the semantic data retrieval strategy selected above to search in the ideological and political teaching resources ontology library, after the above six steps, the user can get a more scientific and reasonable retrieval results to improve the efficiency of use.

3

Specific applications of the ontology library of ideological and political teaching resources

After basically completing the construction and optimization of the ontology library of ideological and political teaching resources, the ontology library is applied in actual teaching, which effectively improves the teaching level of teachers and students’ enthusiasm for learning. The following is a detailed description of the use of the ontology library.

3.1

Validation of Semantic Retrieval of Ideological and Political Teaching Resources

In order to verify the feasibility and effectiveness of the semantic retrieval strategy proposed above, this paper builds a prototype of semantic retrieval of ontology-based educational resources using relevant development tools, and analyzes the retrieval results by checking the completeness rate, checking the accuracy rate, and metrics to verify the effectiveness of the model system.

The experiment uses various discrete knowledge points in the course “Ethics and Rule of Law”, which includes a wide variety of materials such as text, audio, video, web pages and so on. The experiment selected more than 100 relatively large knowledge points and more than 1,000 articles in terms of the “ideology, morality and the rule of law” course. Among the knowledge points include the outlook on life, morality, Marxism, traditional Chinese virtues, reform and innovation, etc. This paper conducted a multi-group query test on these complex knowledge points.

In this test, this paper adopts the method of manual evaluation to evaluate the relevant literature in the experimental test results, for the final search results through the artificial relevance to judge, taking into account the ease of operation, etc., this experiment is only the search results are divided into two categories: relevant and irrelevant.

The results of the experiment are mainly examined in three aspects: the search rate, the search accuracy rate and the metrics. The comparison of the results is made between the traditional keyword-based search, the traditional vector space model search, and the search that combines vector space model and ontological reasoning proposed in this paper. By setting each parameter in the retrieval model, the main settings of each parameter provided in the previous chapter of this paper are as follows: 1)

In the semantic relatedness algorithm, α=0.65, β=0.25, γ=0.15, υ=0.05, ν=0.05. n=0.05 is taken here.

2)

In the semantic relatedness algorithm, the base threshold for concept-to-concept similarity is set to 0.65.

3)

The weighted moderator in the F-value was set to 0.7.

4)

The threshold value is set to 0.4 when the concept-to-resource query similarity calculation is performed.

Table 1 represents the experimental data obtained for the three retrieval methods after the parameter settings above. Analyzing the experimental data in Table 1, it can be seen that the retrieval method combining vector space model and ontology reasoning chosen in this paper can more effectively improve the users’ retrieval efficiency and retrieval accuracy compared with the traditional single retrieval method.

Table 1.

Experimental data

Term	Retrieval mode	Precision ratio	Recall factor	Balance degree
Outlook on life	Traditional keyword search	16/25=0.64	16/65=0.246	0.421
	Traditional vector search	40/63=0.635	35/65=0.538	0.624
	Textual method search	52/66=0.788	48/65=0.738	0.778
Moral outlook	Traditional keyword search	31/42=0.738	33/105=0.314	0.542
	Traditional vector search	50/93=0.537	56/105=0.533	0.576
	Textual method search	81/111=0.729	79/105=0.752	0.728
Marxism	Traditional keyword search	70/82=0.853	65/182=0.357	0.572
	Traditional vector search	110/163=0.675	111/182=0.610	0.664
	Textual method search	161/175=0.920	156/182=0.857	0.896
Traditional Chinese virtue	Traditional keyword search	95/149=0.638	91/265=0.343	0.512
	Traditional vector search	149/241=0.618	185/265=0.698	0.719
	Textual method search	222/331=0.671	203/265=0.766	0.738
Reform and innovation	Traditional keyword search	52/75=0.693	52/360=0.144	0.315
	Traditional vector search	160/240=0.667	163/360=0.453	0.597
	Textual method search	312/380=0.821	310/360=0.861	0.825

3.2

Controlled Experiments on Teaching and Learning with an Ontology of Teaching Resources

After verifying the effectiveness of the semantic retrieval system, this paper continues to set up experiments to verify the role of the ontology library of ideological and political teaching resources in improving the level of students’ ideological and political quality. Two classes, the experimental class (using Ideological and Political Teaching Resource Ontology Library) and the control class (using traditional offline learning resources) are set up respectively, and the data on the level of Civic and Political quality of the two classes before and after the experiment is conducted are collected and analyzed in a comparative manner. The specific data are shown in Tables 2 and 3. Table 2 shows the differences in the level of ideological quality between the pre-test experimental class and the control class group.

Table 2.

Pre-test data of ideological and political quality of two classes

Dimensionality	Comparison class(N=40)	Experimental class(N=40)	T	P
patriotism	2.723±0.413	2.791±0.353	0.861	0.392
Collectivism spirit	2.791±0.531	2.815±0.479	0.873	0.382
Team spirit	3.035±0.583	2.791±0.512	-1.925	0.059
Rule consciousness	2.833±0.657	2.816±0.529	-0.141	0.069
Equity and justice	2.822±0.583	2.702±0.580	-0.878	0.079
Sense of responsibility	2.921±0.549	2.715±0.566	-2.487	0.088
Correct view of outcome	2.811±0.689	2.710±0.593	-0.646	0.115
Enterprising spirit	2.751±0.502	2.661±0.443	-0.849	0.398
Will quality	2.811±0.590	2.791±0.326	-0.461	0.645

Table 3.

Post-test data of ideological and political quality of two classes

Dimensionality	Comparison class(N=40)	Experimental class(N=40)	T	P
Patriotism	2.812±0.429	3.796±0.507	3.793±0.504	0.000
Collectivism spirit	2.886±0.528	3.814±0.532	7.826	0.000
Team spirit	2.859±0.551	3.876±0.584	7.983	0.001
Rule consciousness	3.016±0.502	3.951±0.564	7.886	0.000
Equity and justice	2.855±0.451	3.873±0.556	9.021	0.000
Sense of responsibility	2.901±0.382	3.943±0.627	3.921±0.625	0.000
Correct view of outcome	2.943±0.488	3.918±0.549	8.355	0.002
Enterprising spirit	2.980±0.432	3.890±0.530	3.891±0.501	0.000
Will quality	3.016±0.513	3.961±0.542	3.963±0.528	0.000

The students in the experimental class and the control class were tested on the level of Civic and Political Quality before the beginning of the experiment, and the data obtained were analyzed by data analysis. As can be seen from Table 2, the comparison results of the data of the two groups of students in the nine dimensions of patriotism and other data show that P>0.05, there is no significant difference, indicating that the Civic and Political Quality of the two groups of students before the start of the experiment is at the same level, with the conditions of the experiment, and can be carried out in the next step of the experiment. Table 3 displays the disparity in Civic and Political quality between the post-test experimental class and the control class group.

Through the three-month practice of using the ideological and political teaching resource ontology library, the questionnaire scale statistics of the ideological and political qualities of the two groups of students and the independent samples t-test through SPSS26.0 statistical software, the following results were obtained: from Table 2, it can be intuitively seen that the P-value of nine dimensions, such as the spirit of patriotism, are all less than 0.05, which is a significant difference, and the mean values of the students in the experimental class are higher than those of the control class students, the difference in mean values are 0.982, 0.929, 1.027, 0.945, 1.015, 1.037, 0.966, 0.902, 0.955, which indicates that compared with traditional teaching resources, the use of Ideological and Political Teaching Resource Ontology Library can make the level of the students’ quality of ideological and political affairs be significantly improved.

To analyze the reason, the use of ideological and political teaching resources ontology library can enable students to quickly retrieve the need to master the relevant learning resources, in-depth understanding of the connotation of the relevant knowledge points, so that the students are subjected to subtle positive influence, and gradually cultivate students’ love of community, teamwork, compliance with the rules, striving to forge ahead, tenacity, fair competition, dare to take responsibility and other good ideological and political qualities.

3.3

Analysis of student learning data

At the end of the experiment in 3.2, this paper verifies the effectiveness of the Ontology Library of Ideological and Political Teaching Resources in students’ learning. By analyzing the learning data of students using the Ontology of Ideological and Political Teaching Resources, it can be clearly seen that students are more enthusiastic about using the Ontology for learning, which also provides relevant data support for teachers on how to use the Ontology for teaching. The following is an analysis of the relevant learning data.

The background of the Ideological and Political Teaching Resource Ontology Library stores all kinds of learning data of students, and this study obtains the number of times of searching chapter resources, the number of times of querying the connotation of knowledge points, the number of times of learning chapter contents and the number of times of searching course-related articles. For the number of chapter content learning times, since the use of the ideological and political teaching resources ontology library, the total number of students’ chapter content learning times amounted to 2011 times, and the number of chapter content learning times is shown in Figure 5. Taking the learning status in October 2023 as an example, the number of daily learning times is more than 30 times, among which the lowest number of learning times is 30 times on October 11th, and the highest number of learning times is 138 times on October 8th. Synthesizing other learning data, it can be concluded that students are more interested in using the ideological and political teaching resources ontology library for learning and have a higher degree of learning commitment, according to which teachers can guide students more to further explore the use of this ontology library and improve their learning ability.

3.4

Survey on Students’ Satisfaction in Using the Ontology Library of Ideological and Political Teaching Resources

On the basis of verifying the effectiveness of the ontology library of ideological and political teaching resources, a questionnaire survey was conducted to investigate the students’ satisfaction with its use. In this paper, the questionnaire was conducted on 1580 students in the school of Chinese Marxism, and a total of 1389 completed questionnaires were retrieved, of which 1253 were valid questionnaires. Table 4 shows the student satisfaction data obtained by organizing and analyzing the valid questionnaires afterwards.

Table 4.

Students’ satisfaction with ontology library

		frequency	percent	Effective percentage	Cumulative percentage
effective	Very satisfied	473	37.7	37.7	37.7
	Satisfied	596	47.6	47.6	85.3
	normal	184	14.7	14.7	100.0
	total	1253	100.0	100.0

It can be seen from Table 4 that 37.7% of the students were “very satisfied”, 47.6% were satisfied, 14.7% were average, and no student chose “dissatisfied” or “very dissatisfied”. After using the ontology of ideological and political teaching resources, more than 85% of the students expressed “satisfaction” and “great satisfaction” with the application of the ontology of ideological and political teaching resources as a teaching auxiliary resource, and the use effect was remarkable.

4

Conclusion

This paper focuses on the networked integration of ideological and political education and teaching, the construction of the ideological and political teaching resources ontology library, the optimization of the related retrieval accuracy, and the application of the ontology library as a teaching aid resource in the teaching of ideological and political courses. The ontology library of ideological and political teaching resources constructed based on the seven-step development method has the advantages of clear content hierarchy, high semantic retrieval efficiency, high semantic retrieval accuracy, etc., which can effectively help students and teachers search for the knowledge and learning resources of ideological and political related courses.

Students using the ideological and political teaching resources ontology library for ideological and political course learning can quickly retrieve the need to master the relevant learning resources, in-depth understanding of the connotation of the relevant knowledge points, and improve the ability to use the learning tools and relevant ideological and political literacy. The number of times students searched with the Ideological and Political Teaching Resources Ontology Library on the same day exceeded 30 times, which verified that the use of the Ontology Library could improve students’ enthusiasm for learning, and in the questionnaire survey on students’ satisfaction with the use of the Ontology Library, more than 85% of the students expressed their “very satisfied” and “satisfied”.

Combined with the research in this paper, it can be clearly concluded that the use of ideological and political teaching resources ontology library can enhance students’ enthusiasm and learning ability in ideological and political courses, which also provides necessary support for teachers to further optimize the ideological and political teaching resources ontology library, make full use of the ontology library to assist teaching and promote the effective implementation of ideological and political education teaching.

Język:: Angielski

Częstotliwość wydawania:: 1 razy w roku
Dziedziny czasopisma:: Nauki biologiczne, Nauki biologiczne, inne, Matematyka, Matematyka stosowana, Matematyka ogólna, Fizyka, Fizyka, inne

Kanał RSS czasopisma

Research on Networked Integration of Ideological and Political Education Teaching Resources in the Information Age

Li Ma

Data publikacji: 24 mar 2025

Otrzymano: 26 paź 2024

Przyjęty: 07 lut 2025

DOI: https://doi.org/10.2478/amns-2025-0734

Słowa kluczoweSeven-step development method, Semantic similarity, Semantic data retrieval, Ideological and political teaching resources ontology library

© 2025 Li Ma, published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Słowa kluczowe
Seven-step development method, Semantic similarity, Semantic data retrieval, Ideological and political teaching resources ontology library