English Multimedia Online Teaching Resource Processing System Based on Intelligent Cloud Platform

In order to improve the quality of online English teaching, starting from English online teaching resources, this paper applies cloud computing technology to the teaching resource integration process to design a comprehensive cloud storage control method based on user habits. Moreover, this paper uses the user habit extraction algorithm to extract the user’s habit information and then realizes the control of cloud storage with the help of these habit information. After that, the system can realize intelligent management of cloud storage in space by analyzing the information and converting it into corresponding cloud storage control instructions. In addition, this paper constructs an integrated system of English multimedia network teaching resources under the support of cloud storage technology. Finally, this paper uses simulation experiments to verify that the integrated system of English multimedia network teaching resources based on cloud storage has a good effect.


Introduction
e educational structure of a country depends on the structure of a country's industry. With the introduction of the concept of "Made in China 2025," China is advancing towards achieving its strategic goal of being a manufacturing power. Undoubtedly, in order to achieve this goal, higher education that cultivates skilled talents for the manufacturing industry is inevitably required to be connected with it. e Made in China 2.0 era interprets the level of modern higher education and expresses the level of national competitiveness. As a link between schools and society, professional construction is the focus and core of the construction of higher education institutions. It is related to the quality and e ciency of education, and professionalism is a prominent feature of higher education. erefore, the construction of a professional system is still the primary task of deepening the reform of the current higher education system and adapting to the transformation of the regional economic industrial structure. e construction of the modern higher education professional system should be closely linked to the local economy, realizing the docking of professional construction and industrial development needs and then serving the economic and social development. However, the current professional system of higher education institutions in China has many problems, such as high repetition rate of professional settings among colleges, weak adaptability of professional settings and industrial structure, scattered waste of resources and low competitiveness, unreasonable professional structure layout, and unable to meet the needs of the society for compound and comprehensive talents. is re ects the unreasonable structure of the traditional professional construction mechanism, which greatly reduces the "endogenous" development ability of higher education, and the contradiction between professional positions and professional training goals is also increasing. In order to break through the predicament, the concept of "cluster" gradually entered the eld of higher education from the economic eld, and professional group construction has become an e ective way to break through the predicament faced by higher education professional construction.
Literature [1] uses cloud storage in the compilation of distance education digital resources integrated with teaching materials. Under the premise of considering the organic connection with the context content, the cloud storage icon and cloud storage title are marked in the appropriate position in the manuscript. In literature [2], using cloud storage in the laboratory teaching resource management system, the system can meet the needs of students' independent learning after class, basically solve the problem of inconvenient access to resources for students, and meet the needs of daily laboratory management. In literature [3], sharing of learning resources through cloud storage, learners can directly scan with mobile phones, and then they can download complete related animations, videos, and other materials. Literature [4] combines MOOCs and cloud storage to establish a system suitable for various independent colleges. e MOOC learning platform facilitates students to learn anytime and anywhere through network resource sharing. Literature [5] analyzes the feasibility of cloud storage in education and teaching based on the technical advantages of cloud storage. is paper analyzes the teaching process design based on cloud storage and believes that the application of cloud storage in teaching is feasible and the effect is predictable. Literature [6] does make the use of course teaching resources more and more convenient by using cloud storage, which is more in line with the learning needs of learners.
Use virtual reality technology VR to realize the integration of teaching resources. Literature [7] introduces VR panoramic technology and its use in teaching, including VR panoramic campus, VR panoramic off-campus practice base, and so on, using VR panoramic technology to display teaching resources so that students can have an immersive experience. Literature [7] analyzed the advantages and disadvantages of applying VR technology to teaching resources. e advantages are mainly reflected in the following three aspects: (1) Improve the learning effect, break the boundaries of time and space, and give learners an immersive and intuitive experience; (2) Practice plus application to improve the experience of virtual interaction; (3) Intuitive experience, improve the actual experience of learners. Using virtual reality technology, students can have real and virtual experiences anytime and anywhere and stimulate students' brains to promote active learning. Literature [8] studies the application of helmet immersive VR technology in the construction of teaching resources in vocational colleges and believes that the use of VR technology can effectively avoid a series of problems such as low user concentration and difficulty in ensuring learning efficiency caused by flat education. Users can have an intuitive experience of accepting face-to-face teaching, which strengthens the vividness of teaching resources. It is an innovation for teaching resource construction and teaching resource sharing. At the same time, VR technology is based on the Internet, which can overcome the limitations of time and space and enhance the effectiveness of teaching resources. However, as a new technology, VR is still in its growth stage; and the related cloud storage price of VR is very expensive, which virtually blocks many learners, and it will take time to achieve widespread popularization [9].
Associate artificial intelligence technology with teaching resources to realize the integration of teaching resources. Literature [10] studies the integration and application of artificial intelligence in education, and the research believes that artificial intelligence technology is widely used in education, which can present teaching content in multimedia and networked manner, and implement the teaching process in a personalized and interactive manner. It has brought profound changes to the mode of education and teaching, the process of education and teaching, teaching design, teaching content, and educational evaluation. Literature [11] explored the educational application of machine learning in the field of artificial intelligence and found that the current educational application of machine learning. It mainly focuses on six aspects: student modeling, student behavior modeling, predicting learning behavior, early warning of the risk of dropping out of school, learning support and evaluation, and resource recommendation. Literature [12] studies the "Internet + education" in the context of artificial intelligence. is paper introduces the new mode of combining artificial intelligence technology with education industry, including automatic homework correction, online question answering, intelligent evaluation and so on. Literature [13] believes that with the development of artificial intelligence technology, traditional education methods and learning methods Earth-shaking changes will take place, and artificial intelligence will inject new impetus into the development of "Internet + education" from various aspects such as improving teaching efficiency, satisfying personalized learning, and enhancing interaction and feedback. Literature [14] proposes the construction of "artificial intelligence + education." e core elements of this ecosystem are the application form, technical structure, and business trend. "Artificial intelligence + education" can effectively promote the sustainable development of information technology and education and teaching. Artificial intelligence technology can effectively help the construction and integration of teaching resources. Although the threshold of artificial intelligence technology is relatively high, when using teaching resources based on artificial intelligence technology, it does not need to spend a lot of money, and even only needs a camera. It reduces the learning threshold for learners [15].
With the emergence of resource scarcity, the contradiction between the increasing demand for higher education and the limited supply of teaching resources in our country has become more and more prominent. e total shortage of teaching resources in the absolute sense, the inefficiency of macro resource allocation within the existing education system, and the inefficiency of microresource utilization have become the bottleneck restricting the development of higher education institutions. As higher education is most closely linked to the economy and society, its reform and development over the past 30 years can also be seen as a process of achieving "Pareto Optimality" through "Pareto Improvement" in the field of higher education. In other words, in a country or region, under the condition of limited educational resources, how to achieve the maximum benefit and maximum value of education as a whole is the focus of current exploration. e problem of resource shortage is a frequently talked-about but difficult problem to get rid of. e problem of resource shortage has always been shortcoming of colleges and universities. At present, not only the overall input resources of higher education institutions are insufficient, but also the allocation of stock resources is unreasonable and wasteful. e concrete manifestation is the prominent contradiction between the insufficient investment of financial resources and material resources and the irrational allocation of financial resources and material resources. ese not only affect the training and output efficiency of talents but also affect the overall efficiency of running a school. In addition, not only the healthy development of colleges and universities has been affected, but their sustainable development will also face severe tests. How to rationally allocate, integrate, and optimize teaching resources to improve the efficiency and value of resources has become a hot issue that needs to be solved urgently in the "connotative" development process of Chinese colleges and universities today. erefore, strengthening the resource integration of colleges and universities is an inevitable choice to break through the bottleneck of the development of colleges and universities, and it is also an objective requirement to strengthen the connotation management of colleges and universities, improve the quality and level of running a school, and promote the development of connotation. Under the guidance of the scientific system, only when resources are integrated and optimized can Chinese education realize the transformation from an extensional to an intentional development path.
is paper combines cloud storage technology to study the integration of English multimedia networked teaching resources, builds an intelligent system, and promotes the effective integration of English multimedia networked teaching resources.

Data Type Analysis and Algorithm Selection.
is paper designs a comprehensive cloud storage control method based on user habits. e core idea is that we use user habit extraction algorithms to extract user habit information and then use this habit information to achieve cloud storage control.
Among the three types of data in a single operation information table, the relevant operation data is composed of multiple independent discrete attribute states. erefore, statistics-based data clustering methods can be used for processing. However, time data and various sensor data are in the one-dimensional data structure so that each record can correspond to a point in the four-dimensional data space erefore, this paper can use the SCLCHV algorithm proposed by the CLIQUE algorithm to investigate the distribution of points in the high-dimensional space to obtain its common characteristics and finally obtain possible user habits.
When extracting habit information, two main data clustering methods need to be used, so the two clustering methods need to be explained first [16]: (1) Statistical clustering algorithm: the main ideas are as follows. e algorithm counts the related operation information in the single operation information table and obtains the occurrence frequency Y of each related operation. e calculation method is the ratio of the number of occurrences of the operation to the total number of related operations. en, the habit information is obtained by comparing the value of Y, and the specific process will be described below.
(2) SCLCHV algorithm: the traditional CLIQUE algorithm divides each attribute of the high-dimensional space equally and divides the entire data space into a super cuboid collection. However, due to the complexity of habit information, simple attribute divisions cannot achieve the purpose of searching user habits. Because only when the projection of points in any dimension is relatively dense, there will be possible user habits in that dimension. erefore, the grid cannot be divided equally, and one-dimensional data clustering must be performed on the projection of the points in each dimension.

Habit Information Extraction Process.
One-dimensional data clustering refer to the adjacent data in a single dimension and the difference is less than the critical value La to form a data interval. e interval whose total length does not exceed the maximum boundary length L max is regarded as the clustering interval that meets the requirements, and the average value of its data is regarded as the effective value of this interval, as shown in Figure 1 [17]. is article uses a two-dimensional subspace as an example to illustrate the correlation degree investigation process (as shown in Figure 2). We assume that there are clustering intervals X 1 , X 2 , Y 1 , Y 2 in two dimensions, and the effective values of the intervals are x 1 ′ , x 2 ′ , y 1 ′ , y 2 ′ ; then a total of four grids (A, B, C, and D) are formed in the two-dimensional subspace. e projection point of the data in the subspace can be expressed as P 2 (x, y); then the correlation degree of the grid formed by the clustering space X h , Y k is as follows [18]: where N(P and ) is the number of P and points and N(P or ) is the number of P or points.
After obtaining the grid correlation degree, the correlation degree is compared with the correlation degree threshold Ψ. If it exceeds Ψ, there is a correlation between the two; otherwise, there is no correlation. Among them, the correlation threshold is the ratio of the total number of points in the grid to the total number of points. From Figure 2, it can be concluded that Ψ � 10/19 � 0.53, the interval X 1 and Y 2 corresponding to grid C are related, and the other three do not exist. erefore, the possible habit trigger conditions are x 1 ′ &y 1 ′ , x 2 ′ , y 2 ′ [19].
After completing the data clustering, the three types of data are compared. Time data and environmental information data are characterized by the clustering interval, and the interval is represented by its e ective value, that is, the e ective value is used as the trigger condition corresponding to this interval. erefore, there is a possibility that the trigger condition may deviate from the actual expected trigger value, which a ects the accuracy of the habit information. However, because the relevant operation information is an independent discrete quantity, there is no problem of numerical deviation. Objectively speaking, its credibility is higher. In view of the above reasons, in order to improve the accuracy of the acquired habit information as much as possible, in the process of information extraction, relevant operational information should be treated rst.
For the high reference frequency c H and the low reference frequency c L as the critical value to measure the frequency of related operations c, the algorithm calculates the c value of each related operation. For related operations that exceed c H , it is directly used as a trigger condition for the user's habit. However, between c H and c L , we further examine whether it is related to other conditions. If all related operations do not exceed c L , then the algorithm checks whether there is habit information in other conditions. Among them, in order to reduce the occurrence of errors, c H should choose a larger value (such as 0.85 or more), and c L can choose a smaller value (such as 0.4), so as to reduce the occurrence of missed detection. e speci c process is shown in Figure 3 [20].

Reinforcement and Elimination of Habitual Information.
Corresponding to the principle of strengthening and elimination of conditioned re ex, this part is to further verify the possible habit information extracted from all single operation information, realize information strengthening and elimination processing, and nally obtain e ective habit information. In order to facilitate the description of the veri cation process, two concepts of accuracy need to be proposed: e accuracy of the habit information refers to the credibility of the habit information, which is expressed as to whether the corresponding cloud storage operation will occur after the trigger condition of the habit information is met. e accuracy of the trigger condition refers to the accuracy of each trigger condition, which is expressed as to whether the corresponding cloud storage operation will occur after the trigger condition is met. e former is for the entire habit information, while the latter is for each individual trigger condition. For single trigger habit information, the meaning of the two is the same. However, for multitrigger habit information, the former is for the whole, while the latter is for each individual conditional among them. In terms of numerical value, for single trigger information, the accuracy of the habit information is equal to the accuracy of the trigger condition. However, for multitrigger information, the accuracy of the habit information takes the smallest value among the accuracy of each trigger condition.
In order to assess habit information based on accuracy, this paper proposes the concepts of execution probability and elimination probability: (1) e execution probability (Φ exe ) is for the accuracy of the habit information, that is, when the accuracy of certain habit information reaches the execution probability, the habit information can be regarded as the real habit information; otherwise, it can only be regarded as the quasihabit information. (2) e elimination probability (Φ quit ) refers to the accuracy of the trigger condition, which means that when the accuracy of a trigger condition is lower than the elimination probability, it will be eliminated.
(3) For single trigger habit information, when the habit information accuracy (or trigger condition accuracy) exceeds Φ exe , the habit information will be used as e ective habit information. However, when it is lower than Φ quit , the habit information will be eliminated.
(4) For multitrigger habit information, only when the accuracy of the habit information exceeds Φ exe , the habit information will be regarded as e ective habit information. However, when the accuracy of a certain trigger condition is lower than Φ quit , only the trigger condition is eliminated, that is, the habit information only reduces one trigger condition and will not be eliminated. Only when all trigger conditions are eliminated, the habit information will be eliminated.
It should be noted that the calculation of accuracy in this article is mainly for the accuracy of the trigger condition. If there is no explanation, the accuracy mentioned below refers to the accuracy of the trigger condition.
After obtaining quasihabit information through data clustering, the system will calculate the initial accuracy of Less-than L max Valid internal Greater than L max Invalid interval Less-than L a Greater than L a Less-than L max Valid internal Figure 1: One-dimensional data clustering. Figure 2: Two-dimensional subspace grid.

Calculate the frequency of all related operations
Get a single operation information table Is there any operation above the high baseline frequency Is there operation between high and low base frequency Additional information from this single operation information table is extracted using the SCLCHV algorithm Use the operation information directly as the habit information and the operation as a trigger condition The obtained effective clustering information is used as the habit information, and the corresponding parameters are the trigger condition Obtain all the habit information and its trigger conditions that may be implied in this single operation information table Additional information associated with this operation is extracted by clustering using the SCLCHV algorithm The obtained clustering information is used with the relevant operation information as habit information and the joint trigger condition e calculation formula of initial accuracy is shown in the following formulae: where Φ 0 is the initial accuracy, n 0 is the number of occurrences of the trigger condition in a single operation information table (information such as time is the number of points in the same clustering interval), N is the amount of data in the single operation information table, and Ω is the con dence parameter (Ω > 1), which re ects the degree of accuracy required for the amount of data. e smaller the Ω,   the larger the amount of data required to obtain the same accuracy. is article selects the Ω value to be 1.2. It can be obtained from the above formula that the greater the amount of information in the single operation information table, the higher the credibility of the trigger condition. At the same time, the higher the frequency of the trigger condition, the higher its credibility. is article uses result feedback to evaluate the accuracy of the trigger condition, that is, when the condition meets a certain trigger condition, it will examine whether the user's next behavior is consistent with the cloud storage operation corresponding to the trigger condition. If they match, the result of the veri cation is true; if they do not match, the result is false. e information accuracy changes with the number of veri cations and the veri cation results. erefore, the text rst gives the relationship between the accuracy Φ and the number of times the veri cation result is true i as follows:   Mobile Information Systems According to formula (4), if the i + 1-th time is true, the new accuracy Φ new is It can be deduced by formulas (4) and (5) that when the veri cation result is true, the calculation formula of the new accuracy Φ new is On the other hand, in formula (4), when the accuracy Φ is small, it is sensitive to changes as i increases, and when it is large, it is not sensitive to changes in value as i decreases (as shown in Figure 4). erefore, when the veri cation result is false, if the new accuracy is obtained by reducing i in formula (4), the result cannot be corrected in time.  In view of this, when the veri cation result is false, in order to correct the error message in time, this article uses a linear method to calculate the new accuracy, as shown in the following equation: where β is a single attenuation, and this paper takes 0.05. Combining formulas (7) and (8), we can get a new accuracy calculation formula as follows [21]: After obtaining user habits information, the system analyzes the information and converts it into corresponding cloud storage control instructions to realize intelligent management of cloud storage in the space.

The English Multimedia Networked Teaching
Resource Integration System Based on Cloud Storage is article combines the third part of the algorithm to build an English multimedia networked teaching resource integration system with the support of cloud storage technology. e integration process can generally be divided into three  steps: select cloud servers to be o ine, sort cloud instances to be migrated, and lter cloud servers to be migrated. e cloud server integration process is shown in Figure 5. From the perspective of the lower level of the computer, this set of cloud server integration processes is also completed in a speci c online migration mechanism. e simpli ed structure of the system proposed in this paper can be roughly divided into three submodules logically, as shown in Figure 6. e process of building an integrated index library for English multimedia networked teaching resources based on cloud storage is: the index node reads documents from HDFS, extracts index items from them, represents the document, and generates an index table for the document collection. e detailed process is shown in Figure 7. e query request submitted by the user is usually a phrase or short sentence. After the search module accepts the user's visit, it will perform a series of processing and nally submit the results to the user. After the user enters a keyword to search, the search program nds all relevant documents that match the keyword from each index database. Because all documents have a good relevance for the keyword, they only need to be sorted according to the existing relevance values. e higher the relevance is, the higher the ranking is. Finally, the search results found in each index library are collected and reordered, and the link addresses of the search results and page content abstracts are organized and returned to the user. is process is the basic process of a search query, including query keyword preprocessing, text database matching, similarity and ranking calculation, document sorting, document collection, and generation. e overall processing ow is shown in Figure 8.
As shown in Figure 9, the target case is analyzed by the user inquiry behavior analysis module, so as to guide the case query process, that is, according to the analysis of the target case, it selects a subset of cases with high similarity from the case database, modeling user query behavior based on user feedback. e modeling and analysis process of user query behavior can also be regarded as the process that the cBR system uses user query information to train and learn.
e English multimedia networked teaching resource integration system based on cloud storage is constructed above, and the performance of the system proposed in this paper is analyzed through simulation experiments. e English multimedia networked teaching resource integration system based on cloud storage is veri ed and studied through multiple sets of experiments, and the results shown in Table 1 and Figure 10 are obtained.
From the above analysis, we can see that the English multimedia networked teaching resource integration system based on cloud storage proposed in this article has a better cloud storage e ect. On this basis, the teaching resource integration e ect of the English multimedia networked teaching resource integration system based on cloud storage is evaluated, and the results shown in Table 2 and Figure 11 are obtained.
From the above analysis, it can be seen that the English multimedia networked teaching resource integration system based on cloud storage proposed in this article has good results and meets the needs of modern English networked teaching.

Conclusion
Combining the characteristics of teaching resources including tangible and intangible resources, this research summarizes the teaching resources as the research object into three main resources: human resources (teacher resources), information resources (course resources), and hardware resources (training base resources). However, this is not to ignore the role of intangible resources such as system and management but to use it as a factor that depends on the carrier of teacher resources as a technical resource. Moreover, it is an important means and tool to improve tangible resources in this research. erefore, the professional group teaching resources are the collection of human resources (teacher resources), information resources (course resources), and hardware resources (training base resources) of each profession within the professional group. is article combines cloud storage technology to study the integration of English multimedia networked teaching resources, constructs an intelligent system, and promotes the e ective integration of English multimedia networked teaching resources. e analysis shows that the English multimedia networked teaching resource integration system based on cloud storage proposed in this paper has good results and meets the needs of modern English networked teaching.
Data Availability e labeled data set used to support the ndings of this study is available from the corresponding author upon request.

Conflicts of Interest
e author declares that there are no con icts of interest.
Acknowledgments is study was sponsored by Xinyang Vocational and Technical College.  Figure 11: e statistical diagram of the English multimedia networked teaching resource integration system based on cloud storage's teaching resource integration e ect.