We have proposed design of soft computing techniques for college English lecture writing and translation based on 5g technology. The proposed work flow is described in this section. The schematic representation of the proposed method is depicted in figure 1.
Dataset description:
We created a testing data set of 7,500,000 documents in the movie area, with 3,750,000 positive and negative documents in English. All of the documents in our testing data set were gathered automatically from English Facebook, websites, and social networks, and then categorized as positive or negative [18].
Preprocessing using Normalization:
The incoming information is raw and shall contain duplicate packet and incomplete information. It's been cleaned up and preprocessed to remove repetitive and duplicate instances as well as incomplete information. Because the datasets for the system of education is huge, sample size reduction approaches must be used. Since there are so many features in this database, feature extraction tools are required to filter out those that aren't important. During the pre-processing phase, the database might be normalized. The z-score, which is given by equation (1), is generated in the first step of the normalization procedure.
$$Z=\left[\right(R-\alpha )/\omega ]$$
1
Here, \(\alpha\) denotes the mean of the data and \(\omega\) indicates the standard deviation. And Z is expressed as,
$$Z=\frac{R-\stackrel{-}{R}}{SD}$$
2
Here ̅\(\stackrel{-}{R}\)denotes the mean of the sample, and \(SD\) denotes the standard deviation of the samples.
The random sample is in the form of,
$${Z}_{k}={\beta }_{0}+{\beta }_{1}{R}_{k}+{ℇ}_{k}$$
3
Here \({ℇ}_{k}\) denotes the errors that is dependent on the\({\omega }^{2}\)
Following that, the errors must not rely on each other, as provided below.
Here r denotes a random variable.
Thereafter, the standard deviation is utilized to normalize the movements of the variable.
The below expression is utilized to estimate the moment scale deviation.
$$MS=\frac{{\lambda }^{ms}}{{\varnothing }^{ms}}$$
5
Here \(ms\) denotes the moment scale.
$${\lambda }^{ms}=E\left(R-\alpha \right)^MS$$
6
Here \(R\) indicates a random variable and \(E\) denotes the expected value.
$${\varnothing }^{ms}=\left(\sqrt{E\left(R-\alpha \right)^MS}\right)^2$$
7
$${r}_{w}=\frac{ms}{\stackrel{-}{R}}$$
8
Here \({r}_{w}\) represents the coefficient of the variance.
By adjusting all of the variables to 0 or 1, the feature scaling method will be terminated. This process is known as the unison-based normalizing method. The normalized equation would be written as follows:
$$R\text{'}=\frac{(r-{r}_{min})}{({r}_{max}-{r}_{min})}$$
9
The data could be maintained once the data has been normalized, and the data's range and inconsistency may remain consistent. The goal of this phase is to reduce or eliminate data delay. The normalized information could then be fed into the future stages as an input.
English materials feature recognition model:
The recognition and evaluation rules are determined based on the experimental samples, and the feature parameters of English features are filtered by rules before the formal experiment of this recognition model. This study uses five categorization approaches in its English feature recognition model, ranging from excellent to unqualified. These grades are given a score once the grading has been determined. Excellent grade score interval [8, 10], decent grade score interval [6, 8], qualified grade score interval [4, 6], basic qualified grade score interval [2, 4], and unqualified grade score interval [0, 2] are the five grades for which a score of 10 points is assigned. The following is how its grade evaluation function is expressed:
$$q\left(x\right)=\frac{x-{e}^{x}}{{e}^{x-1}+x{e}^{x} }$$
10
The following is the absolute function expression for the evaluation function:
$$V\left(x\right)=\left|\frac{1-{e}^{x-1}}{x{e}^{x}+\left(x+1\right){e}^{x+1}}\right|$$
11
Adaptive Optimization of English teaching:
The expert system analysis model for assessing the impact of the innovation reform of college English teaching is constructed, and the hierarchical grey association analysis technique is employed for adaptive optimization and decision control of the innovation reform impact of college English teaching. The innovation effect's descriptive statistical sequence is: \(\left\{x\left({t}_{0}+i\varDelta t\right)\right\},i-0,\dots ,N-1\). The best iterative method for improving college English teaching innovation is:
$$X=[{{S}_{1 }, {S}_{2 },\dots {S}_{K }]}_{n }=({x}_{n }, {x}_{n-\pi },\dots {x}_{n}-\left(m-1\right)\pi )$$
12
The fuzzy parameter detection of the influence of college English teaching innovation and reform is assessed using a combination of parameter examination and panel parameter examination approaches. The formula for building a statistical examination design to assess the impact of innovation in college English teaching reform is:
$$\frac{dz\left(t\right)}{dt}=F\left(z\right)$$
13
Set \(f\left({S}_{i}\right)=(f\left({x}_{1}\right),f\left({x}_{2}\right),\dots ,f\left({x}_{n}\right))\). The evaluation of college English teaching innovation reform is built using a fuzzy subspace scheduling model expressed as, \(P\left({n}_{i}\right)=\left\{{p}_{k }\right|{pr}_{kj }=1,K-1,,\dots ,m\}\). Adaptive assessment is carried out on the assessment information of the effect of innovation reform of college English teaching, depending on the correlation scheduling and fuzziness examination of the effect assessment of teaching innovation reform, corresponding to the fuzzy feature distribution set of the effect assessment. The following are the new conditions for course ideological and political adjustment:
$$\phi =\frac{1}{1+\alpha (\frac{\partial s }{\partial t}{)}^{2}}$$
14
Classification using Genetic Algorithm (GA):
We use a supervised learning approach whose fitness function is the TE recognition method's count of erroneous entailment judgments. The algorithmic population consists of a set of candidate weight vectors and the associated thresholds. Every person depicts a candidate equality operation and the threshold judgment that goes with it. The data is split into two sections for training and testing. Based on the training data, the algorithm determines the best equality function and judgment threshold. The generated similarity function and threshold are evaluated through a series of experiments based on the testing data.
Selection, crossover, and mutation are all common genetic operators used by the GA. The following are the characteristics of these genetic operators:
Selection Operator: To generate a parent pool for the crossover function, a random selection method is used.
Crossover Operator: The following three offspring are produced using two crossover strategies.
Two offspring are produced using the one-point crossover procedure.
By estimating the mean of the parent's gene values, the arithmetic crossover procedure creates a new offspring.
Mutation Operator: By employing a crossover function with a 15% risk of mutation into a random value in the interval [0-1], every gene is allocated to the created individual.
To prepare the future generation, the algorithm chooses the fittest individual. From one generation to the next, the count of people in the population remains constant. To bring the GA to a close, two criteria are applied.
There is a person in the population who has a fitness value lesser than an empirically set threshold.
The maximum number of generations has been achieved.