Abstract
With the deepening of economic globalization, the communication between countries is getting closer and closer. The common topic of communication between people of different cultures and environments is English. Since the country has established English as a subject, the importance of English has become one of the main factors for the all-round development of students. In recent years, higher vocational education in China has once again developed towards high quality, all-round diversification, and innovation. According to the continuous reform and innovation and independent innovation, college English has been proved to be an important part of higher vocational education and an important measure to promote the development trend of higher vocational education all over the country. With the progress of science and technology and the development of economy, the level of education in China is still improving. People have put forward stronger teaching standards and regulations for English, a course with strong adaptability and practicality. Part of Chinese translation in college English shows its comprehensive ability and high-level thinking ability. In order to integrate this trend, this paper proposes the formulation and use of college English translation teaching model. The experiment proves that the accuracy of the traditional fuzzy evaluation method in the comment module design of college English translation teaching mode is low, only 82% at most. In order to solve this problem, BP optimization algorithm is introduced. The precision and reliability of BP optimization algorithm are improved. Because of the learning effect of the college English translation teaching mode based on the fuzzy evaluation BP optimization algorithm proposed in the certification paper, the certification is carried out according to the questionnaire and the final evaluation method. The students’ satisfaction and favoritism at the assessment level and the final results of their classmates have proved the proposed college English translation teaching model. The article has excellent learning effect.
1. Introduction
College English is a comprehensive course integrating listening, speaking, reading, and writing. The goal of college English is not only to let students understand English as a language expression, but also to let them know how to use English to exchange information and cultural expressions. By applying this kind of college English teaching mode, we can build a high-quality talent team with all-round development for students to learn English. Finally, students’ Comprehensive English application ability is improved. The ultimate goal is to improve students’ Comprehensive English application ability and understanding of English culture and express their thinking in English as fluently as Chinese [1]. In essence, college English is an overall summary of junior middle school English, an improvement in the level of English application ability, and an organic unity of reading, writing, listening, speaking, and translation. College English translation is a major branch of college English classroom teaching and a major factor for students to learn English. Therefore, how to build a college English translation teaching model that integrates the development of the times and the current educational concept is the main situation and orientation that scholars must consider [2].
The scientific research on how to do a good job in the classroom teaching of Chinese translation in the direction of college English has never stopped. For example, some researchers have taken the college English translation teaching model as a breakthrough and put forward corresponding analysis and countermeasures for this breakthrough. Please put forward some innovative points on this basis.
At the end of the graduation thesis, it puts forward the proposal of university capital construction, emphasizing that the effect evaluation of English translation teaching model is a big problem. In addition, some experts also proposed to integrate the collaborative teaching model into the college English translation classroom teaching and put forward how to build the practical application of the collaborative college English translation teaching model from seven aspects [3]. In this article, the author first put forward four questions. These four questions are as follows: first, we considered the problem of what the current college English rotation teaching model should look like. Secondly, how to popularize the cooperative teaching model of college English translation. Third, how to choose materials suitable for students’ learning and training. Fourth, we pay attention to the selection of Chinese translation raw materials, which should not be oversystematized. Due to the excessive systematization of Chinese translation, the fourth problem is the crowd problem in the collaborative teaching model and emphasizes how to divide the crowd. Based on the discussion of the above four issues, this paper analyzes the advantages and disadvantages of the collaborative teaching model of college English translation and finally summarizes the basic theory of cooperative learning. In the brief introduction of the basic theory of cooperative learning, the creator first analyzed the meaning of cooperative learning and put forward that the core concept of this kind of teaching model is the division of responsibilities to achieve the goal. Finally, the author clearly put forward the application of such a collaborative teaching model to practical activities and put forward three stages, namely, the preparation stage, the management method stage, and the evaluation stage. In the evaluation stage, the author proposes to carry out mutual evaluation according to group cooperative learning and give full play to the guiding significance of teachers in teaching, so as to improve the motivation of students’ learning English and strengthen their comprehensive Chinese translation ability. There are many other scientific research papers, such as this one, but most of them have basic theoretical characteristics similar to the above [4].
In general, according to the analysis of relevant papers, it can be seen that most of today’s papers are about the theory and practice of how to build a college English translation teaching model, or about the theoretical research on the evaluation of the teaching content of the college English translation teaching model. However, this kind of theoretical research has not been well applied to practical activities. Practice is the only standard to test true knowledge. In order to better build the college English translation teaching mode and, at the same time, combine the educational concept with the teaching mode to make it more suitable for the fashion trend and development trend of the times, on this basis, combine the fuzzy comprehensive evaluation of independent innovation with the construction of the college English translation teaching mode, and finally build a college English translation teaching mode suitable for the characteristics of students at this stage. According to the questionnaire, the traditional college English translation teaching model based on fuzzy comprehensive evaluation method is compared with the technological innovation college English translation teaching model mentioned in this paper. Then, the fuzzy comprehensive evaluation and BP optimization algorithm are applied to objectively evaluate the college English translation teaching model mentioned in this paper. Finally, according to the analysis of subjective factors, the author constructs all college English translation teaching models.
2. Research Background
2.1. Analysis of Problems Existing in English Translation Teaching
From the first section, it can be seen that college English teaching is a unified organism integrating listening, speaking, reading, writing, and translation, and college English translation teaching is an indispensable part of English teaching at this stage, which is directly related to students’ comprehensive English ability [5]. However, due to the harm of the traditional concept of examination taking, English teaching has always regarded Chinese translation teaching as a testing tool in China. Since the harm of English translation teaching to the actual effect of teaching is not significant, there is no test centered on “eating without distinguishing students.” Therefore, in the case of teaching, the purpose of English translation teaching is to test all teaching [6]. More importantly, in order to be able to pass the English test, some colleges and universities feel that “readers win the world” and then spend a lot of time reading articles, thus reducing the length of other parts of English, such as college English translation, which generally reduces the length of classroom teaching. In addition, more importantly, because of the diversified development trend of understanding at this stage, the traditional teaching mode is no longer suitable for the characteristics of students at this stage. At this stage, college English translation teaching modes do not fully use students’ initiative, and students’ overall Chinese translation quality has not been greatly improved. In order to better build a college English translation teaching model suitable for the diversified development trend of today’s students, this paper first analyzes in detail the deficiencies in the current college English translation teaching link and then builds a college English translation teaching model in line with the current teaching concept and the current situation of students’ development according to these problems [7].
Firstly, college English translation has not been paid enough attention, which has seriously endangered the development trend of English translation teaching and scientific research [8]. Translation is a kind of comprehensive ability, which provides a bridge for people with different cultural characteristics and language expressions. However, the teaching of English translation has been neglected, which has already caused many students to neglect the learning and training of English translation. For example, under the pressure of relativism, many English students learn English every day just to brush their teeth and gargle and recite English words. Just because they choose the correct answer, there is not much professional knowledge and specific content related to Chinese translation. Because there are no such professional examination questions and norms as oral English, Chinese translation in college students’ English teaching has not been paid attention to.
Second, there is lack of mastery of Chinese traditional culture [9]. In China, Chinese is used to communicate and express feelings. A person’s historical and cultural heritage is closely related to his Chinese. China has a long, extensive, and profound history. Chinese has experienced thousands of years of progress. The same word can have different meanings in different contexts, but it is different from English. English is simpler than Chinese in terms of vocabulary. Therefore, many students may misunderstand differences in different contexts when learning English. Especially under the test-oriented education mode in China, teachers are usually utilitarian in teaching, pay insufficient attention to language and culture teaching, and then ignore the differences in ideological status and cultural expression between the east and the west, resulting in the unclear language expression ability [10]. Therefore, based on establishing college English translation teaching methods, we should first pay attention to English translation, then pay attention to English language expression and culture, and pay attention to the differences between Chinese and English cultures.
In the case of teaching, it can not only shape students’ ability to use English, but also improve their understanding of English.
Thirdly, in the traditional English translation classroom teaching, the teaching strategies are out of date, and the educational ideas do not conform to the characteristics of students [11]. Many teachers still adopt a step-by-step teaching method in the teaching process, blindly following the trend in writing scripts, thus failing to effectively translate words into certain technical progress and broaden professional knowledge. More importantly, because the course content and its source are from daily life, students are also interested in English translation [12]. In addition, because the teachers did not interpret the use of professional skills and professional knowledge development related to translation in their teaching courses, many students did not accept the learning, training, and learning of technical professional translation knowledge and skills. As a result, many students’ translation ability is slowly improved, their participation is very low, and their professional learning is boring.
The fourth reason is that the course content is old, and the translation raw materials are not updated. In the teaching of English translation in colleges and universities, many colleges and universities do not attach importance to English translation, and the content of teaching materials on English translation has not been updated and filled. The content of old teaching materials is difficult to closely integrate the basic theory of the concept of sustainable development with the times, and it is difficult to connect with students’ daily life and technical majors. Therefore, students’ interest can not be enhanced, and the development concept of today’s students is reversed [13]. In addition, there is a key problem; that is, nowadays translation software is more excellent, and many students and even teachers rely on translation software. Overreliance on translation software causes teachers and students to pay more attention to English translation and even lose thinking about the translation process from English to Chinese and from Chinese to English.
Applicability to intelligent system: in addition, another problem is that the current translation software is only a software after all, and intelligent translation is not omnipotent. Under the complex semantic and historical background, translation software has been unable to surpass human logical thinking, especially in the ability of human thinking and manipulating difficult language expression [14].
2.2. The Role and Significance of Translation for English Language Teaching
This paper definitely puts forward a more detailed teaching method of college English translation in order to build a more detailed teaching method of college English translation.
Firstly, there is a connection between the characteristics of college students’ English and translation. The social development attribute and historical and humanistic attribute of college English teaching have been known to all over the years. The attribute of social development refers to the use of English in daily life for mutual communication and leisure activities, while the attribute of people is to express ideas and emotions in English. It happens that these two attributes are closely related to the whole process of English translation, either explicitly or implicitly. When college students learn English in colleges and universities, through long-term learning and training, they have had a rich and colorful foundation and formed their own learning methods. It happens that this kind of relatively solid foundation for English words, reading, and writing enables them to skillfully apply English grammar and relevant sentence patterns as the basis for English translation [15].
Secondly, English translation can further improve students’ English learning and training ability and enable them to have a more in-depth and correct understanding of oral English and a more comprehensive cognitive ability. Compared with college students, they can improve vocabulary memory, reading, and understanding, English grammar use and creative expression ability through many translation practice activities. In addition, English translation requires not only a certain vocabulary foundation, but also a more solid English grammar foundation and reading ability. A certain amount of English translation training can comprehensively improve college students’ English dictation, reading and writing ability, and translation ability, reading time, and the ability to understand the theme of the article content or the meaning of some paragraphs of the article content [16].
Translation classroom teaching can reasonably improve college students’ creative ability and quickly improve their creative ability and English listening skills. In a word, English translation is an indispensable foundation and an important part of college English. It can be said that it is at the core of relativity in all college English teaching situations.
Thirdly, the relationship between English translation and western culture learning was discussed. English learning is not only the learning of language expression, but also the learning and understanding of culture. Language expression and culture are closely related. Therefore, cultural concepts are particularly important in English learning, and mastering western culture is one of the necessary ways to learn English. As the creative thinking of Chinese-English transformation, Chinese translation is the core area of cultural communication between China and the West [17]. As a road bridge for cross-cultural interpersonal communication, college English students learn western culture according to Chinese translation and comparison between Chinese and Western culture. On the one hand, they can deepen their understanding of local culture and external culture, and on the other hand, they can improve their working ability in cross-cultural interpersonal communication.
In general, English translation learning and classroom teaching not only conform to the characteristics of actual English, but also enhance college students’ grasp of English at the level of vocabulary and reading. More importantly, according to the classroom teaching of Chinese translation, students can deepen their understanding of the cultural differences between China and the west, deepen their understanding of their own culture and foreign culture, and better grasp the vocabulary English, the grammatical application of English, and their writing methods in different situations, so as to further improve students’ English level.
3. Materials and Methods
3.1. Fuzzy Integrated Assessment
Comprehensive assessment can be distinguished between subjective and objective levels. The subjective level is essentially a qualitative analysis, often called the subjective empowerment evaluation method, and the specific methods commonly used are the hierarchical analysis method and the fuzzy integrated assessment method. From the objective level, it is essential from the correlation or the coefficient of variation of each index, and the commonly used methods are the gray correlation method, TOPSIS method, and principal component analysis. The specific relationships can be shown in Figure 1.
It can be seen from Figure 1 that the fuzzy comprehensive evaluation method belongs to the subjective value evaluation method, that is, the method of using doctors’ experience and professional knowledge to award a certain weight value to new projects, and the weight value reflects the quality level of evaluation. Fuzzy evaluation should be based on fuzzy set. Under the condition of mathematical model, the comprehensive basic principle of fuzzy relation should be applied to carry out quantitative analysis on some elements with unclear boundaries and not easy to quantify analysis and then carry out comprehensive evaluation on the membership of the evaluated things from several elements.
In the process of English translation teaching, traditional translation teaching is to gradually teach translation professional knowledge or skills according to the content of teaching materials or translation raw materials. These methods are not conducive to shaping students’ logical thinking ability and sufficient English translation initiative. However, there are many uncertain factors in the process of English translation teaching, such as students’ acceptance of basic English knowledge, or students’ different situations in English teaching. From the perspective of practical activities, simple English translation teaching is not suitable for the teaching orientation of polymorphism and the specific acceptability of students [18]. In contrast, the fuzziness assessment method can quantitatively analyze this uncertainty and generate a fuzziness drainage matrix. Finally, the weight values of uncertainty are all over the world, and the uncertainty is limited in a quantitative way, which can give some practical significance for English translation classroom teaching, which can further improve students’ Chinese translation ability and their vocabulary, reading, and writing abilities. More importantly, according to this kind of comparative method, more importantly, according to this kind of comparative method, students’ understanding of the differences between Chinese and Western cultures can be further improved. Finally, students’ consciousness can be fully used to produce a closed-loop control learning method of active learning.
3.2. Neural Network under Fuzzy Comprehensive Judgment
In this paper, firstly, the uncertainties of English translation will be determined by fuzzy evaluation method, and then the softmax layer of neural network will be used to classify these uncertainties, and then the classified results will provide teachers with certain reference in the teaching process of English translation, and finally a mixed-mode university English translation teaching model will be formed. Therefore, the next part of this paper will introduce the neural network under fuzzy evaluation.
Neural network is abbreviated as NN, which can also be called artificial neural network, and the basic units of neural network are all neurons, in which classification, adaptive noise filtering, and system identification can be implemented [19]. The basic structure of one of the neurons is shown in Figure 2.
From Figure 2, it can be seen that the basic structure of neuron under deep learning can be composed of basic linear function and excitation function. Among them, the equations of the basic linear functions are generally shown as follows:
From (1), W is the weight value of all neural network models, where B is the error function formula or error value, which is set as an adjustable error value in this article [20]. In the neural network model, input x is first used for SVM algorithm according to the input layer, and then it is immediately sent to the linear function formula together with the input according to the obtained weight value. However, if it is only solved according to the linear function formula, the accuracy of this kind of linear neural network model is very low, and the convergence performance is also poor, which is not conducive to convergence. Therefore, generally, nonlinear solutions will be added after linear solutions. That is, the part of F shown in the Figure is usually added after the linear solution. This part is generally called the activation function. After the nonlinear solution, it realizes the derivation of Y according to the derivation layer, in which the activation function can be regarded as a nonlinear projection again. The introduction of activation function makes all neural networks look richer and more complicated, thus enhancing the language expression ability of all the Internet. If only one linear relation is applied, the neural network model obtained from linear accumulation only has the effect of linear projection, which is not very practical in real life.
It can be seen from Figure 2 that, in terms of network architecture, the neural network can be divided into input layer, export layer, and pooling layer, as well as convolution layer and pooling layer in the neural network. The key method in the work of neural network is to gather the difference data signals in the input layer according to the linear combination entity model. Each data signal has different weight values, and then determine whether to export according to the activation function. In the neural network model, the final value is the weight value, because the selection of the weight value determines the orientation.
Neural network model, on the contrary, determines the convergence performance of neural network and other main factors that can finally achieve the most advantage. The common activation functions are generally sigmoid function formula and linear rectifier function formula ReLU function. The actual formula calculation is shown in (2) and (3).
The relationship between the activation function and the linear combination can be shown in (4), where b denotes the bias function, and denotes the weight. From (4), it can be seen that the relationship between the linear and nonlinear functions in the neural network is nested. And it is because of this nested relationship that the neural network has a strong learning ability and self-adaptive ability, with a certain degree of complexity of the nonlinear system making the whole neural network model fully imitate the human thinking activities and then in the learning aspect to show a certain intelligence and intelligent thinking.
The role of fuzzy judgment in the neural network model is not only to determine the judgment criteria, but also to derive the fuzzy matrix, which is used as the input of the feature matrix of the neural network after the feature extraction is completed.
4. Results and Discussion
4.1. General Framework
In order to better reflect the advantages and innovative ability of the college English translation teaching model mentioned in the article, the paper integrates the basic knowledge of in-depth learning to formulate and build the teaching model. Deep learning is one of the best ways for learners to learn independently and improve their professional skills of independent innovation. Deep learning not only cares about learning, but also cares about the depth of learning. Today’s students spend very little time learning the professional knowledge in the teaching materials, let alone increasing their understanding, but according to the in-depth learning, they can further improve the shaping of high-level thinking, promote students to actively carry out classroom teaching and learning theme activities, and deal with learning problems. On the premise of solving problems, students can increase their understanding of knowledge, so as to optimize all college English translation learning, produce a closed-loop control learning method of “continuous breakthrough and continuous expansion,” gradually improve students’ college English translation learning management system under the condition of pioneering learning, and finally improve students’ subjective learning motivation. In this article, we have studied the deep learning method scientifically and applied it flexibly. Therefore, on the premise of comprehensive scientific research on the theory of in-depth learning knowledge, the paper puts forward a basic framework for building a college English translation teaching model, as shown in Figure 3.
In Figure 3, the text is carried out under the premise of machine learning. The college students’ English translation teaching plan is designed as a mixed teaching mode, which selects online learning and offline self-study, respectively, comprehensively considers the normative situation of the epidemic situation, integrates the teaching mode of smart classroom mobile students’ autonomous learning, and divides it into pretranslation personal behavior, translation personal behavior, and posttranslation behavior. According to the above methods of online learning and offline learning, promote and implement the mixed teaching mode and cultivate students’ innovative thinking ability, students’ autonomous learning ability, and problem-solving ability. The key points in this paper are to conduct personal behavior before translation according to online learning, including video online prelearning training, defining daily tasks in Translation Classroom Teaching, and conducting classroom students’ autonomous learning. Change the translation teaching mode into a “creative cooperation” translation teaching mode. Personal behavior after translation is the key to this thesis. At the level of personal behavior after translation, the key points in this paper can be divided into three parts: preparation for examination and work, comments, and retranslation. In the comments, the paper selects the BP optimization algorithm of fuzzy evaluation neural network to distinguish the variability of primary and secondary school students and teachers in college English translation teaching and carry out horizontal zoning, so that teachers and students can better learn, train, and teach translation professional knowledge, improve college students’ English translation ability, and promote the learning effect together.
In general, in order to build a more comprehensive and detailed English translation teaching model, due to the subjectivity of fuzzy evaluation, when the index value set u is relatively large, the relative membership power index is usually small, resulting in the mismatch between the power space vector and the fuzzy drainage matrix R.
In more serious cases, it may also happen that the screen is too fuzzy. On the contrary, it will cause poor screen resolution, unable to distinguish the membership of different matters, and even may cause unsuccessful discrimination. In addition, when there are too many index values, there are no effective data statistics, and the weight value is difficult to be clear. More importantly, there are too few quantitative data for pure fuzzy evaluation, there are too many judgment components, and the evaluation conclusion is usually unreliable. To solve the above problems, a neural network model based on ambiguity resolution is introduced, as shown in Figure 4.
In Figure 4, the neural network entity model is completed as fuzzy judgment, and a BP algorithm fuzzy judgment entity model with fuzzy judgment and neural network embedded each other is generated. It can be seen from the overall framework steps in Figure 4 that the text will first define the element set and comment set according to the evaluation and discussion of doctors and students on English translation teaching methods, crawl according to the evaluation of the online teaching platform by the web crawler, then define the weight value of the index value according to the evaluation of subjective values, then define their attribution to the spatial vector, and then carry out the fuzzy judgment matrix of over, Finally, fuzzy solution is carried out for the fuzzy judgment matrix. Finally, the obtained data is normalized to generate samples with suitable sample format. Such samples can be transmitted to the introduction layer of the neural network as the input of the central nervous system transmission layer. Subsequently, learning, training, and practice are carried out for the BP algorithm of the neural network, and finally the results are output.
4.2. Implementation of BP Algorithm under Fuzzy Judging
It can be seen from Section 3 that the neural network is mainly composed of three parts, namely, network architecture, activation function, and main parameter learning and training algorithm, so as to find the best weight value. In the third part, BP algorithm is a more common main parameter learning and training algorithm, which is a double-layer feedforward control neural network practiced according to the error back propagation algorithm. The general whole process of BP algorithm is to calculate FP first, that is, the spread of sperm motility, type in the sample, taking the above fuzzy drainage matrix obtained after fuzzy evaluation as the original weight W and error item B, and calculating the error between the actual value and the derived value, that is, the error difference. If the damage value is not within the given range, it is the second step of BP algorithm, that is, backpropagation; otherwise, the upgrade of weight W and error value B will be terminated. In the case of backpropagation, it is mainly based on the backpropagation of the error. The error will be backpropagated from the derived layer to the imported layer in a certain way, and the error will be spread to all nerve cells in each layer, so as to obtain the error data signal of each layer module, and then take it as the basis for calibrating the weight size of each nerve cell module.
In general, the purpose of BP algorithm is to continuously adjust the correct direction of all neural network models according to the upgraded weight value and error value, so that it finally converges to the optimal value. In this article, as shown in Figure 4, we will first get the corresponding evaluation based on the crawling technology and then get the corresponding data based on some data preprocessing technology. Finally, the index value set and comment set are obtained according to the arrangement, and then fuzzy processing is carried out to obtain the ambiguity.
Then, the fuzzy and relevant overall planning is carried out for the Y drainage matrix to make it suitable for the original weight and error function formula of the neural network clear. Finally, the weight and error value are continuously innovated according to the BP algorithm and backpropagation, and finally the accuracy of its evaluation is obtained. In order to obtain faster precision, the ReLU function formula is used as the activation function in this paper. Because the standard evaluation interval in this paper is [0100], there can be no evaluation lower than 0. In addition, when the gradient direction exceeds 1, the ReLU function formula can ensure stability, and it is not easy to fade in the gradient direction. Because of the effectiveness and accuracy of part of the evaluation of college English translation teaching methods based on the fuzzy evaluation method clearly proposed in this article, the simple fuzzy evaluation method is first used to comment on all matters. According to the statistical analysis of data, the accuracy chart of fuzzy prediction analysis is shown in Figure 5.
It can be seen from Figure 5 that if the traditional fuzzy evaluation method is used to predict and analyze the comments on college English translation teaching mode, its accuracy is in the middle of [0.65, 0.83], but this accuracy is not high enough. Therefore, in this prediction and analysis case, the frequency with a success rate of more than 80% is far less than the frequency in the middle of 60%–70%, mainly because in the traditional fuzzy identification method, the fundamental reason for this is that, among the traditional ambiguity resolution methods, the purpose of the assessment method is to be subjective, which is also very subjective, which will improve the error rate of the assessment prediction and analysis. Therefore, the accuracy of the assessment prediction and analysis of the college English translation teaching model based on the traditional ambiguity assessment method needs to be improved.
According to the above analysis, it can be seen that the traditional fuzzy evaluation method mainly carries out comments on college English translation because of the harm of subjective reasons. In order to improve the problem of low accuracy of the traditional fuzzy evaluation method, we must reduce the harm of subjective reasons. Therefore, this paper introduces the BP optimization algorithm into the traditional fuzzy evaluation method and adjusts the weight value and error according to the back propagation to improve the objective reasons, so as to achieve the goal of reduction. Therefore, this paper introduces BP optimization algorithm into the traditional fuzzy evaluation method, adjusts the weight value and error according to the backpropagation, improves the objective reasons, reduces the subjective reasons, and finally improves the effectiveness of the teaching mode. The precision of comment prediction analysis of college English translation teaching mode based on the improved fuzzy evaluation method is shown in the Figure, in which the dark blue curve shows the precision of comment prediction analysis of college English translation teaching mode based on the traditional fuzzy evaluation method.
ZY resolution method and orange curve chart illustrate the accuracy of the evaluation and prediction analysis of the college English translation teaching model based on Fuzzy ambiguity resolution method after introducing BP optimization algorithm.
As can be seen from Figure 6, the success rate of the improved fuzzy evaluation method is significantly better than the traditional fuzzy evaluation method for the evaluation and prediction of college English translation teaching mode. From the comparison of precision value, the average precision in this paper is increased by 5% after introducing BP optimization algorithm. In terms of the main production range of accuracy, the success rate above 80% is significantly higher than the traditional fuzzy evaluation method, and only one value is less than 80%. In the smoothness level of the curve, the improved fuzzy evaluation method in this paper has smoother prediction effect than the traditional fuzzy evaluation method, and the neural network model of the improved fuzzy evaluation method in the explanatory text has scalability and robustness as shown in Figure 6.
4.3. Analysis of Teaching Indexes Based on the Traditional and the Improved Fuzzy Judgment Method
In order to better reflect the actual effect of the improved fuzzy evaluation control module mentioned in the article on the college English translation teaching model, the traditional fuzzy evaluation method and the improved fuzzy evaluation method are selected for the college English translation teaching model, and the teaching practice is carried out in two classes of a university. The effect of the questionnaire is shown in the figure. Class 1 is a college English translation teaching mode based on the traditional fuzzy evaluation, and class 2 is a college English translation teaching mode using BP optimization algorithm to introduce fuzzy evaluation. The number of students in both classes is 60. The English level of the two classes is basically the same, and there is no significant difference in the number of students with different degrees.
It can be seen from Figure 7 that, under the evaluation module of the two ambiguous evaluation methods, the survey report on the degree of satisfaction and favor of college English translation teaching methods is as follows. For class 1, the number of satisfactory students is significantly lower than that of class 2. The number of satisfactory students is basically the same, but the relative height of the column chart is still slightly lower than that of class 2. The number of dissatisfied students and disliked students is much higher than that of class 2. It can be seen that, according to the application questionnaire, the conclusion is drawn that, compared with the evaluation module under the traditional fuzzy evaluation method, students prefer to choose the evaluation module of post-fuzzy evaluation method integrating BP optimization algorithm. See Figure 7.
Figure 8 shows the comparison of the final research results of the evaluation control module of college English translation teaching methods by two different ambiguous evaluation methods. In order to ensure the fairness of the test and the universality of the experimental results, all the papers in this round of test are one paper, and the class level of students is basically the same. It can be seen from the figure that the number of students in class 2 with scores above 80 is significantly better than that in class 1, indicating that the fuzzy evaluation method with BP optimization algorithm is more effective and easier for students to recognize; However, in the score of [70, 80], within the score range of [70, 80], the number of students in class 1 is higher than that in class 2, which means that the improved fuzzy discrimination method in the text should be improved for junior students as shown in Figure 8.
5. Conclusion
This paper takes the college English translation teaching mode as the research object, and the selected implementation method is the fuzzy evaluation BP optimization algorithm. In order to get a faster learning effect, and whether college students’ acceptance of the college English translation teaching model and their translation level have been improved after scientific research to improve the fuzzy evaluation method, this paper puts forward some solutions from the shortcomings of the traditional college English translation teaching model. It also proposes to apply fuzzy evaluation method and neural network together in terms of materials and analysis methods to complete the basic theory based on in-depth learning in this paper. According to the deep application of this basic theory, the online and offline teaching modes are fully integrated, and finally, the overall structure of college English translation teaching mode is obtained. Introduce BP optimization algorithm based on this structure and more traditional fuzzy evaluation methods and fuzzy evaluation methods, The comment control module is improved. On the premise of comprehensive consideration of human factors, the satisfaction and liking of students are investigated according to the method of questionnaire. At the same time, in order to fully verify the effectiveness of the college English teaching model proposed in this paper in enhancing the students’ foreign language level, the graduation thesis is finally certified according to the final examination conclusion.
In general, the above test results show that the accuracy of the improved comment module is about 3%–5 points higher than the traditional fuzzy evaluation module, and the overall performance is more stable. In terms of final examination results, the improved fuzzy evaluation method is more reasonable in enhancing students’ learning results and has a good learning effect.
Data Availability
The dataset can be accessed upon request.
Conflicts of Interest
The author declares no conflicts of interest.