Investigating the use of fuzzy logical data for evaluating the teaching quality of college teachers

Abstract

The level of education in colleges is career and development-focused compared to that from high schools. Quality education relies on the teachers’ qualifications, knowledge, and experience over the years. However, the demand for technical and knowledge-based education is increasing with the world’s demands. Therefore, assessing the knowledge of teaching professionals to meet external demand becomes mandatory. This article introduces an Acceded Data Evaluation Method (ADEM) using Fuzzy Logic (FL) for teaching quality assessment. The proposed method inputs the teachers’ skills and students’ productivity for evaluation. The teachers’ knowledge and updated skills through training and self-learning are the key features for evaluating the independents’ performance. The impact of the above features on the student qualifying ratio and understandability (through examination) are analyzed periodically. Depending on the qualifications and performance, the teachers’ knowledge update is recommended with the new training programs. In this evaluation process, fuzzy logic is implied for balancing and identifying the maximum validation criteria that satisfy the quality requirements. The recommendations using partial and fulfilled quality constraints are identified using the logical truth over the varying assessments. The proposed method is analyzed using the metrics evaluation rate, quality detection, recommendations, evaluation time, and data balancing.

Keywords

Data balancing decision recommendations fuzzy logic teaching quality

1 Introduction

Teaching quality assessment (TQA) is a measure that organizes or plans certain strategies to improve the teaching process. TQA is taken by the head or higher faculties that examine the exact quality of teachers. TQA is a crucial task to perform in both colleges and schools [3]. TQA is conducted for every teacher to identify the skill set, quality, knowledge, and potential over the teaching process. TQA for college teachers detects teachers’ actual education, procedures, and examinations [8]. Various quality assessment methods are used in colleges to evaluate the studies and knowledge of teachers. The teacher evaluation form (TEF) is used to analyze effective characteristics and perceptions of teachers over students and teaching [31]. Characteristics such as communication skills, study skills, note distribution, problems solving skills, and frequent examination are calculated [26]. Certain questionnaires are provided to teachers based on the results TQA is evaluated. Key values and qualifications are also identified from college management systems which provide feasible data for the evaluation process [12]. TQA predicts the approximate qualities and potentials of teachers in the teaching field. TQA provides necessary information that maximizes the performance and efficiency of college teaching standards [12, 34]. The Acknowledged Data Evaluation Method (ADEM) is a tool for instructors in the field of quality evaluation. Management systems keep track of evaluations and provide useful information for the next steps. The effectiveness of both the teachers and the students is measured by this proposed method. Fuzzy logic is implied for balancing and identifying the maximum validation criteria that satisfy the quality requirements while assessing the quality of college instructors’ teaching.

Student assessments are gathered or collected from students in colleges. Assessments are stored in management systems which provide feasible data for further processes. TQA is evaluated from student assessments [10 , 16]. Students examine details such as communication, teaching, leadership, and teachers’ skills. The higher education quality assessment model (HEQAM) is used in colleges to examine teachers’ quality. HEQAM identified key factors and patterns for the evaluation process [1]. The hierarchical analytical process is also used in TQA that identifies the priorities and potentials of teachers in the teaching process. Various skills and qualities are evaluated based on patterns that improve performance and efficiency range teaching [27]. TQA provides relevant information which enhances the standard and effectiveness level of students. Students’ assessments contain the exact performance and potential of teachers who are provided to students [37]. Certain questionnaires are conducted among students that produce information for TQA. The quality assurance method is also used in TQA to get appropriate student assessment data. Student assessment reduces data identification latency, enhancing teaching feasibility [14].

ADEM is utilized in this fuzzy logic approach to finding evaluated results by associating and measuring information on targeted assessment. Decisions are made using valid student crisp data, which includes the number of students, bio data, session, and assessment, and reflects the teacher’s abilities. During the training, the information can be utilized to assess the trainees’ current levels of knowledge and proficiency. The periodic evaluation is performed using this information and expertise. Fuzzy logic is widely used for evaluation and identification processes in many fields. Fuzzy logic uses various tools to evaluate certain tasks in an application or system. Fuzzy logic is also used in TQA, which analyzes the exact qualities of teachers in schools and colleges [29]. TQA enhances the teaching process’s feasibility, efficiency, and flexibility ratio. The evaluation index is used in fuzzy logic to calculate the exact assessment indexes. The evaluation index produces the optimal information required for TQA evaluation and analysis processes [22]. Identifying an index is a complicated task in evaluating that requires proper datasets. A fuzzy logic model is also used in TQA that promotes and improves teachers’ quality and effectiveness range. The fuzzy logic model effectively evaluates the TQA of teachers in both schools and colleges [2]. The fuzzy logic model makes assessments based on indices and patterns which are provided by education management systems. Communication skills, leadership qualities, attitudes, commitments, and responsibilities are calculated based on the logic model. The logic model improves teachers’ quality and performance levels during teaching [20 , 38].

The research proposed a data-acceded assessment process using fuzzy logic to identify and enhance college instructors’ abilities. Teachers’ knowledge and present abilities, as acquired via training and self-learning, are crucial factors for evaluating the effectiveness of the independents.

Fuzzy logic is assumed to be at work in this evaluation procedure, with the goal of striking a good balance between and pinpointing the maximum validation criteria that meet the quality requirements. Defuzzification is performed when all of the rules for qualifying ratio and membership functions have been carefully considered.

In the presence of data uncertainty, the data balancing technique is employed to probe a large number of potential inputs. This approach uses a continuous study of the rules that depend on prior knowledge in conjunction with a fuzzy logic process to arrive at accurate conclusions.

2 Related works

Troussas et al. [4] designed a teaching strategy for computer programming. Adaptive learning activities are first analyzed using fuzzy logic. Learning activities provide relevant data for teaching. Teachers’ skills and knowledge are evaluated based on students’ assessments. Learning theory is also used here that produce appropriate information for the performance development process. Revised bloom taxonomy (RBT) is implemented in the teaching strategy that reduces the error range in the validation process. The proposed strategy enhances the efficiency and feasibility of teaching systems.

Yao and Deng [6] introduced fuzzy clustering and latent factor model-based course teacher recommendation model (FCTR-LFM) for both colleges and universities. Pedagogy theory is used here to analyze data that are relevant to courses. Recommendations are made based on certain patterns and features. An evaluation matrix that provides the necessary data for the recommendation process is identified here. LFM is mainly used here to predict teachers’ abilities based on certain skill sets. Experimental results show that the proposed FCTR-LFM increases accuracy in a recommendation which improves the education standards of students.

Litke et al. [9] developed a mathematics-specific observation framework for the teaching quality evaluation process. The framework’s main aim is to measure teachers’ teaching qualities based on some potentials and skills. Various instructions and advice are provided to students by teachers to improve understanding capabilities. Classroom observations, knowledge, interaction, and communication skills of teachers are analyzed that provide relevant data to further processes. The proposed framework enhances teachers’ efficiency and performance range during teaching progress.

Torbeck and Dunnington [19] analyzed the peer review of the operative teaching program in educational institutions. A multistep sequential model is used here that improves the efficacy of peer review systems. Communication skills, education knowledge, safety, understanding, and providing feedback over certain issues are calculated in peer review systems. Certain teaching programs are conducted among teachers to get important data related to teaching skills. The analyzed study improves education systems’ feasibility, robustness, and reliability.

Iqbal et al. [24] developed an effective theory for lesson plans in teaching and learning systems. The actual goal of the lesson plan is to improve the overall performance and efficiency of teaching systems. The lesson plan provides a strategic idea to the teacher that reduces latency in completing lessons for students’ classroom conditions, atmosphere, and understanding range are gathered from questionnaires. Some quality assessment and teaching techniques are used here to provide feasible lesson plan information.

Thommen et al. [5] evaluated teacher motivation using the integrated theoretical approach to teaching quality. Multiple characteristics and features of teachers’ profiles are evaluated based on patterns and functions. Latent profile analysis is used here that identify the motivational profiles of teachers. Profiles provide necessary data which are required for further processes in teaching systems. The discussed study increases efficiency and accuracy in the evaluation process, improving the systems’ effectiveness.

Sailer et al. [23] proposed a new validation method for scenario-based self-assessment instruments for teachers. Behaviors and relationships among tools and skill sets of teachers are evaluated, which provides optimal data for the quality assessment process. Student learning activities are mostly used here that produce students’ exact knowledge and potential. Improving students’ knowledge is the main task of teachers in both schools and colleges. The proposed method maximizes accuracy in the evaluation process, which enhances the efficiency of scenario-based assessment systems.

Lazarides and Schiefele [30] designed a multilevel analysis for teacher self-efficacy. The interest and self-efficacy of teachers are evaluated based on certain patterns and principles. Teachers’ qualities, such as communication skills, interaction ability, knowledge, classroom management, and problem-solving, are also analyzed by a multilevel analysis system. Unique ideas in presenting lessons and activities are identified that provide data for the evaluation process. The designed multilevel analysis enhances the effectiveness and feasibility of teaching systems.

Cotronei-Baird [35] introduced academic hindrances for teachers during teaching and assessment practices. A teacher’s main aim is to improve students’ knowledge, employability skills, and potential. Academic development is also provided to every student, which reduces the error ratio in curriculum management systems. Interviews and questionnaires are conducted among students to get the exact teaching abilities of teachers in universities. The proposed method maximizes the overall employability skills among students.

Yun et al. [13] proposed a new deep learning (DL) based innovative ideological political education platform for students. Strategic assessment and policies are implemented in education platforms that get feasible performance among students. Quality analysis is used here to analyze the necessary aspects required for a teacher. Teachers require a certain set of skills and abilities before teaching students. The proposed method achieves high accuracy in evaluation and prediction, improving teaching quality efficiency.

Bao and Yu [18] developed a new online and offline hybrid teaching quality evaluation method for physical education systems. Mobile edge computing (MEC) is used here that maximizes efficiency and accuracy evaluation method. A fuzzy comprehensive evaluation model identifies grade indexes and patterns of physical education, which reduces identification latency. Compared with other evaluation methods, the proposed hybrid method reduces the cost and complexity of computation, enhancing the performance and mobility of physical education systems.

Alvarez-Blanco et al. [17] designed an intelligent analysis system-based education quality evaluation on a virtual campus. The main aim of the proposed system is to analyze the impacts and effects of education on the virtual campus. Students’ perceptive and skills are gathered from the student management system, which provides necessary information for the evaluation process. Experimental results show that the proposed analysis improves education systems’ overall efficiency and feasibility.

Xu et al. [32] proposed a multi-criteria decision-making method using a fuzzy analytical hierarchy process (Fuzzy AHP). Important elements and components which are related to teaching are gathered from management systems that reduce latency in computation. The proposed method provides the actual aspects and curriculum that provide better teaching skills to teachers. Fuzzy AHP increases the accuracy of the decision-making and evaluation process. The proposed method enhances teaching systems’ effectiveness, robustness, and reliability.

Abd-Elwahed and Al-Bahi [25] introduced a new teaching and assessment framework for engineering curricula. The main goal of the proposed framework is to provide awareness to increase the sustainability range in teaching systems. A questionnaire is conducted among students to get relevant information for analysis and curriculum development processes. Both time and energy consumption range in the computation are reduced, enhancing the systems’ energy efficiency. The introduced framework improves the sustainability and efficiency of teaching systems.

Zhang et al. [11] analyzed the multilevel approach that examines the self-concept, quality teaching, and relatedness of teaching. Science development and skills among students are analyzed, which produces necessary information for further processes. Certain teaching qualities and skills are required for science teachers that improve education standards among students. The proposed approach achieves high accuracy in the evaluation process, enhancing the systems’ performance and feasibility.

Hii et al. [28] proposed a post-adoption model of e-learning for Malaysian educational institutions. Information was gathered using online surveys filled out by 36 instructors at public and private Malaysian universities who are considered to be authorities on e-learning. Based on the results of this research, the quality of the instructor service is the most essential component in e-success learning once it has been adopted. The results are helpful for university administrators’ long-term strategic planning, and the model can be used as a basis for creating a rating system to evaluate the effectiveness of e-learning once it has been adopted.

Rong [21] investigate the possibility of employing a fast interval monotonic decision tree method to construct and design a quality rating system for distant education. With the help of relational database technology, we’ve centralized the administration of different pieces of knowledge in the VIoT system. Knowledge transparency for designers is improved, and intervals are introduced, thanks to the VIoT system’s ability to do actions like browsing, deleting, adding, and modifying knowledge. The suggested method uses the full weighing procedure to generate the evaluation matrix for the decision, summarize it, and get the evaluation conclusion.

The findings demonstrated that the chosen indicators of teaching practice quality were both specific enough to be used in the internal evaluation and general enough to be used in the quality assessment of any teaching practice as well as in the evaluation of the effectiveness of different university-level teacher education programs. However, after reviewing the relevant literature, it became clear that teacher evaluation frequently imposed limits on professional behaviors and instead rated teachers only on the basis of their teaching.

3 Proposed method

Accessed Data Evaluation Method (ADEM) using Fuzzy Logic (FL) is introduced for teaching quality assessment. The level of education in colleges is career and development-focused compared to that from high schools. A college education, as opposed to one from a high school, is more focused on the student’s future job and personal growth. Quality education relies on the teacher’s qualifications, knowledge, and experience over the years. ADEM is used in the process of associating and measuring information on targeted assessment in this fuzzy logic method to find the evaluated outcomes. Valid data are used to make subsequent decisions based on the teacher’s skills reflected in the results. This procedure provides a starting point from which to identify and, in some cases, determine what to improve to conduct a successful assessment. The input includes the database of the students, session information, and training sessions, students should be able to rely on the instructor’s credentials, experience, and expertise so the input equation is not specified. The periodic evaluation is performed using this information and expertise. Students’ understanding is evaluated based on what they’ve learned in class and other information gained throughout instruction. At the evaluation stage, their efficiency can be judged based on performance. Here, fuzzy logic in ADEM is employed, which is based on the observation that decisions are influenced by imprecise data. This fuzzy model can recognize, represent, manipulate, interpret, and employ vague and uncertain data and information. Here in this proposed system, the fuzzy logic in ADEM is used to find the teacher’s skills in their teaching session to provide a successful assessment. The education level in colleges is at a high level as it is necessary to improve the student’s career compared to high schools. Also, the intensity of the education depends on the teacher’s qualification and their knowledge of the teaching. Their skills take a major part in the student’s life during the training session. Figure 1 illustrates the ADEM-FL process.

Fig. 1

ADEM-FL process.

This proposed method evaluates the teacher’s qualifications, experience, and knowledge during the training sessions. During this session, the quality of the teaching is also identified, and the data will be collected. The data can be used to find their skills and knowledge during the training session. This knowledge and skills are used to do the periodic assessment. From the teaching sessions, the student’s knowledge is also calculated, like how much they understand from this class. Their output can be verified during the assessment process by their performance. The Qualifying Ratio can be calculated from the performance, i.e., how many students qualified in the assessment. Then the output of the teacher’s knowledge assessment and student’s Qualifying Ratio assessment is given to the fuzzy logic method to recommend the decision. This decision recommendation concludes whether the teacher needs training or if the teacher is fit for the sessions. This proposed method provides the exact decision recommendation for improving students’ skills and performance. The major factors in decision-making to determine the independents are the teachers’ knowledge and the frequency with which they have updated their abilities through training and self-learning. Utilizing fuzzy logic data in a data-balancing decision-making process, the accessible data evaluation approach produces reliable results regarding the quality of college instructors.

The teacher’s data balancing on their skills and knowledge helps them make decisions without counterfeits. These can be given as the input to the fuzzy logic to identify the perfect decision. The student’s assessment also helps in this decision-recommendation process with a qualifying ratio as the input. This decision can tell about their training skills and their updating of their knowledge which is to improve for the other upcoming training sessions. The teacher’s qualifications, knowledge, and experience over the years are primarily calculated for the training sessions. Their skills can also be more important in the student’s training class. The student’s performance will be calculated from their skills and knowledge in their training sessions. The demand for technical and knowledge-based education is increasing with the world’s demand. Therefore, assessing the knowledge of teaching professionals to meet external demands becomes mandatory, so ADEM using FL is introduced for teaching quality assessment. From the output of the teaching session, the data are collected. The teaching sessions may contain the teacher’s performance using their skills and knowledge. Their experience in the teaching field is to perform well in the training session to provide the efficacious outcome. Their performance speaks a lot in the student’s assessment, like their own performance. Their efficiency during the training sessions is calculated by identifying the data. The following Equation (1) given below explains the process of extracting the data from the training session.

$A ⩽_{y}^{x} B \Leftrightarrow a_{i} ⩽ b_{i} where i = 1 \dots .$ (1)

Whereas (A) is denoted as the teacher’s performance, (B) is denoted as their efficiency, (a_i) is denoted as their skills and (b_i) is denoted as data. Now the data can be divided into their skills and knowledge. The teacher’s skill is given as input in the training session to train the students. Their skill and knowledge are the essential part of the training sessions. The input they give in this session will be seen as the output of the student’s performance in their assessment. The teacher’s knowledge and updated skills through training and self-learning are the key features for evaluating independent performance. This helps to improve the student’s knowledge in their training sessions. The teacher’s skills and knowledge play an important role in sculpting the student’s performance. Their skills are also improved according to the teacher’s knowledge during the training session. As they are at the college level, their career must be a very important means of development. So, they must be demanded technical and knowledge-based education from the teachers. Teachers must also be prepared with these skills according to the student’s demands that they should improve their skills and knowledge to make the students perform well in their assessment process. The skills of the teachers are well managed to provide the appropriate input in the training sessions. The teachers’ skills and knowledge are associated with providing an effective outcome for the students in the assessment performance.

The teachers must provide their cent percent hard work and skills to the students in the training sessions. They should improve their knowledge as the world’s demand for technical education is increasing. The demand goes high as the world develops. So the teacher’s responsibility is to improve their skills and knowledge according to it. It helps in the student’s career development process with technical-based education. They can enhance their knowledge with the teacher’s input in the training session. The teachers must provide education according to the requirements of the students, so they should enhance and upgrade their skills to provide an efficacious education to the students during the training session. The data accumulating process extracts their skills and knowledge from these training sessions. This helps in the performance of product assessment. This enables them to find the knowledge and skills in every training session as they assess regularly. For every training session, the teacher’s skills and knowledge are evaluated periodically to improve the student’s knowledge. As the world demands are increasing, teachers can also enhance or develop their skills and knowledge according to it. The process of extracting the data of skills and knowledge of the teachers from the training session is elucidated by the following Equations (3) given below.

$A \lor_{y}^{x} B = ({a_{1}, b_{1}} \dots \dots . {a_{n}, b_{n}})$ (2)

$A \land_{y}^{x} B = ({a_{1}, b_{1}} \dots \dots . {a_{n}, b_{n}})$ (3)

Now from the training sessions, the student’s performance is evaluated. Students are enabling to improve their skills and knowledge in the session. During the training period, they can develop the activity towards decision-making and problem-solving skills. This skill is very useful to develop their career in the future. The student’s skills are based on the input given by the teachers to make them efficient in their process. The intensity of the teaching can be verified clearly in the outcome of the students. Their performance in the assessment process can explain their skills and knowledge provided in the training sessions. These assessments, such as exams and oral assessments, can distinctly explain the intensity of the training session. In Fig. 2, the teachers’ skill assessment process is illustrated.

Fig. 2

Skill assessment process.

The teacher’s skill in the session can make the student perform better in the assessment process. This educational assessment is the process of registering and using the teacher’s data on the knowledge, skill, and aptitude to refine the process and improve student learning from the training sessions. This assessment can focus on the individual student and their learning system as a whole in the training period. It is the outcome of the students from what they learned in the training period with the help of skilled teachers. This assessment helps to identify the teacher’s skills and knowledge in the teaching sessions (Fig. 2). The teaching intensity can be identified through this assessment process. The quality of the teachers’ teaching can be verified with the students’ output of the assessment process. The student’s output as their performance in the examination will be recorded to find the skills of the teachers in the training session. The teacher’s knowledge and updated skills through training and self-learning are the key features for evaluating independence performance. This assessment is used to find the performance of the students with their enhanced skills and knowledge. The input given by the teacher in the training sessions to improve the skills of the students can also be identified. If the teacher lacks skill and knowledge or they are fit for a session, it can also be found in the student’s output. The teacher’s performance in the training period and the performance of the students with the knowledge given by the teacher can be identified during the process. The student’s output in the assessment process can be elucidated by the following Equation (4) given below.

$\begin{matrix} (c, d) ⩽ (A, B) \\ \Leftrightarrow {\begin{matrix} a ⩽ c and d ⩽ b \\ a < c and d < b \end{matrix} or \end{matrix}$ (4)

Where (c) is denoted as the performance of the students, (d) is denoted as the skills gained from the training sessions by the students, (a, b) is denoted as the output of the process. Now from the assessment process, the performance is calculated. The qualifying ratio shows how many students qualified in their examination assessment. The output of the student’s performance in the examination process can be recorded in this qualifying ratio. The improvement in the performance of the students helps to qualify for the next process in the training sessions. This qualifying ratio helps to find the skills and knowledge of the teachers during the training session. The key features for evaluating the students’ performance have an impact on the student qualifying ratio, and understandability is analyzed periodically depending on the qualifications and performance, and the teachers’ knowledge update is recommended with the new training process. The recommendation of the system include whether the instructor is qualified to teach can be determined from the advice for making a decision. With the help of fuzzy logic data, a recommendation for an instructor’s quality can be found. The students must improve their skills and knowledge during their training sessions to qualify for the examination assessment. From the session output of the students, the qualifying ratio is identified to improve their present performance and to identify the knowledge given by the teachers. This qualifying ratio helps improve the teachers’ skills if necessary and helps them find their level of knowledge. The number of qualified students can tell the teacher’s knowledge in the assessment process. This qualifying ratio of the students makes them enhance their performance for the upcoming assessment process. From this output, the teacher can also put their efficacious input during the teaching sessions with the upgraded skills and knowledge. The student’s performance from the assessment process can be calculated using the following Equation (5) below.

$\begin{matrix} A ⩽_{α}^{α, G} B \\ \Rightarrow {\begin{matrix} G (a) < G (b), \\ \begin{matrix} G (x) = G (b) ⩽ α and | b | < | a | or \\ G (x) = G (b) > α and | a | < | b | or \\ G (x) = G (b), | a | = | y | = n and or a ⩽_{n} b \end{matrix} \end{matrix} \end{matrix}$ (5)

Where (G) is denoted as the evaluation of the student’s performance in the assessment process, (α) is denoted as the calculation of the gained knowledge. Now from the performance of the students, the qualifying ratio is calculated. Their performance in the examination assessment helps to find the number of students qualified for the next assessment. This qualifying ratio has information about the student’s skills and knowledge in performing the assessment, which is gained from the teachers in the training session. The student’s performance can tell the intensity of the training sessions and the input of the teachers. The teacher’s performance in the training session can also be found in the student’s performance in the assessment process. Here in this performance, they can also improve their skills and knowledge according to the assessment to perform better in the examination. The teachers help improve their performance at a high level by providing upgraded skills. The input of the teachers can make the students perform better in their assessment process. Today, fuzzy logic information is governed by the rule of the knowledge base. To ensure that knowledge base systems generate consistent hypotheses, this foundational rule primarily embodies the facts of the ADEM to determine instructor quality via fuzzy set theory. To ensure that knowledge base systems generate consistent hypotheses, this foundational rule primarily represents the facts of the ADEM to determine the quality of the teachers, which is based on the fuzzy set theory.

The efficacious performance of the students makes the high qualifying ratio. This performance can increase the qualifying rate of the assessment with the help of enhanced skills and knowledge. The qualifying ratio not only aids in identifying the teacher’s quality but also helps in improving their knowledge to perform better in another session. This will be recorded periodically for the complete successful assessment. The performance of the students in the assessment process is also analyzed periodically to find the drastic change in the teacher’s knowledge and skills. This helps the teachers improve their efficiency in developing the training sessions and helps the students perform better in the assessment process. Thus the knowledge skills of both teachers and the students can be enhanced by the qualifying ratio from the student’s performance during an assessment. The qualifying ratio is evaluated by the following Equation (6) given below.

$V (a) = V (b) \Rightarrow V (A V_{n} B) = V (a)$ (6)

Where (V) is denoted as the calculation of the qualifying ratio from the student’s performance in the assessment process. Now fuzzy logic is used for the evaluation process. The proposed method inputs the teachers’ skills and students’ productivity for evaluation. The teachers’ knowledge and updated skills through training and self-learning are the key features for evaluating the independents’ performance. The evaluation process from the student’s perspective is illustrated in Fig. 3.

Fig. 3

Evaluation process.

ADEM employs fuzzy logic, which is founded on the notion that judgments are influenced by imprecise data despite the absence of fuzzy data. This fuzzy theory is capable of recognizing, representing, manipulating, interpreting, and employing ambiguous and uncertain data and information. In this proposed system, ADEM’s fuzzy logic is employed to identify the teacher’s teaching qualities in order to deliver a satisfactory assessment. In this evaluation process, fuzzy logic is implied for balancing and identifying the maximum validation criteria that satisfy the quality requirements. The recommendations using partial and fulfilled quality constraints are identified using the logical truth over the varying assessments. Fuzzy logical data is used for evaluating the teaching quality of college teachers.

This fuzzy logic method has four steps for the data balancing process. At first, the membership process takes place in the fuzzy logic method to provide the data recommendations. The membership function has information about the quality of the teaching sessions according to the teacher’s skills and upgraded knowledge. It has all the information in the fuzzy logic data. It can be used as a technique to identify decisions based on the teacher’s characteristics and skills on the students’ experience rather than their knowledge. Data balancing stabilizes performance between periodic evaluations identifies knowledge-dependent rules and repeatedly analyses them using the fuzzy process for precise conclusions. Fuzzy logic data balancing comprises four steps. First, fuzzy logic membership provides data recommendations. The membership function provides information on the teacher’s skills and knowledge. It contains all fuzzy logic data. It can be used to detect teacher-based decisions based on student experience rather than knowledge. Equation 7 explains fuzzy logic data membership functions in data balancing and decision suggestion. The membership functions of the fuzzy logic data in the process of data balancing and decision recommendation can be explained by the following Equation (7) given below:

$L \lor U = {\begin{matrix} u, if l ⩽_{\infty}^{α, V} u \\ l, if u ⩽_{\infty}^{α, V} l \\ x \forall_{n} u if V (x) = V (y), \\ | l | = | u | = n and l | | ⩽_{n} b \end{matrix}$ (7)

Where (L ∨ U) is denoted as the evaluation of the membership functions in the fuzzy logic data, (V) is denoted as the calculation of the total information which is held by the membership functions. Now the knowledge base rule takes place in the fuzzy logic data. Primarily, this base rule represents the facts of the ADEM to identify the quality of the teachers, which is based on the fuzzy set theory so that the knowledge base systems will produce a comparable hypothesis. Here it is used to identify the facts of the teaching sessions from where the quality of the teachers can be found, and the student’s assessment performance can be calculated. This can help find the intensity of the training sessions and the efficiency of the teachers’ skills and knowledge. Thus it helps provide approximate decision recommendations by using the fuzzy logic data in evaluating the teaching quality of the college teachers. The process of the knowledge base rule in the uncertain data can be explained by the following Equations (9) given below:

$M (a \lor_{n} b) < V (H)$ (8)

$M (a) = M (b) \Rightarrow M (a \land_{n}) = M (a)$ (9)

Where M is denoted as the calculation of the knowledge base rules in the fuzzy logic data, (H) is denoted as the calculation of the information about the process. Now the fuzzifier and the defuzzifier process take place. The fuzzy process for evaluation is diagrammatically represented in Fig. 4.

Fig. 4

Fuzzy process.

Fuzzification is used to convert the correct input of the teachers’ skills and the students’ productivity into fuzzy output, which provides the exact decision recommendation with the use of the information in the knowledge base. Defuzzification is used to provide the napped results without any counterfeits. Fuzzification is utilized to transform the precise input of teacher expertise and student achievement into fuzzy output, which then yields a precise decision recommendation by drawing on the data stored in the knowledge base. With the use of defuzzification the finest results without any forgeries. As the data-balancing and a recommendation for a potentially risky option, the results are credible. It produces reckonable results as the fragile decision recommendation and the data balancing (Fig. 4). The process of fuzzification and defuzzification in the fuzzy logic data can be explained by the following Equations (11) given below:

$e \land f = {\begin{matrix} x, if a ⩽_{\infty}^{α, V} b \\ y, if b ⩽_{\infty}^{α, V} a \\ x \land_{n} b, if V (a) = V (b), \\ | a | = | b | = n and a | | ⩽_{n} b \end{matrix}$ (10)

$P_{j} (a) = P_{j} (b) \forall_{j} \in N \Leftrightarrow a = b$ (11)

Where (e ∧ f) is denoted as the results after the fuzzifier and the defuzzifier process, P (_j (a)) is denoted as the calculation of the data balancing. Now from the outputs of all the processes made in fuzzy logic data, the data balancing and the decision recommendations are made. The decision recommendation can clearly say whether the teacher lacks training or is fit for it. The fuzzy logic data helps in finding the decision recommendation to evaluate the teaching quality of college teachers. The decision recommendation and the data balancing process using the fuzzy logic data in the accessed data evaluation method is explained by the following Equations (13) given below:

$P_{j} (a) = {\begin{matrix} 0, if | a | < j \\ 1, if | a | > j \\ h_{j} (σ (a)), otherwise \end{matrix}$ (12)

$\begin{matrix} L^{V} (a_{1}, \dots a_{n}) = & (L_{1} (V (a_{1}) \dots, V (a_{n})) \\ \dots L_{n} ((F (a_{1}) \dots V (a_{n}))) \end{matrix}$ (13)

Where (L^V (a₁, … a_n)) is denoted as the concluded decision recommendation after the process of uncertain data in ADEM. The recommendation process based on defuzzification is illustrated in Fig. 5.

Fig. 5

Recommendation process.

Teachers’ efficacy as educators is measured using FL in this proposed approach. It is also useful for locating balanced data and precise decision recommendations. The rational truth over different evaluations is used to determine which suggestions make use of partial or met quality limitations. Metrics such as evaluation rate, quality detection, suggestions, assessment time, and data balancing are used to assess the suggested technique.

4 Results and discussion

The discussion of the proposed method using [7] data set is presented in this sub-section. The dataset provides 21 fields, out of which 18 fields are specific for teacher evaluation metrics. This data set presents the evaluation for English and non-English (29) input. https://www.kaggle.com/datasets/johnmantios/teaching-assistant-evaluation-dataset. The proposed method uses a teaching Assistant evaluation dataset. The statistics come from 151 TA evaluations in the Statistics Department at the University of Wisconsin–Madison, spanning three academic years and two summer terms. The class variable was created by classifying the scores into three groups of nearly equal size: low, medium, and high. We counted 4171 views and 269 downloads. From this input, the assessment features are represented with their values as in Table 1.

Table 1
Assessment features and values

Field Score Credit Rank Evaluation % Density Type

Tenure Experience Teaching

Value 1-Very Unsatisfactory <2.5 √ 52.38 25 Single

2-Average 2.5-3.0 √ √ 66.36 19 Multi

3-Good 3.0-3.8 √ √ 79.78 69 Single

4-Fair 3.8-4.5 √ √ √ 91.69 71 Multi

5-Excellent >4.5 √ √ √ 95.83 84 Multi

Field	Score	Credit	Rank	Evaluation %	Density	Type
Value	1-Very Unsatisfactory	<2.5			√	52.38	25	Single
	2-Average	2.5-3.0		√	√	66.36	19	Multi
	3-Good	3.0-3.8	√		√	79.78	69	Single
	4-Fair	3.8-4.5	√	√	√	91.69	71	Multi
	5-Excellent	>4.5	√	√	√	95.83	84	Multi

The assessment features are classified as low and high depending on the availability and (G, α). The ranges of values specified in the above table are extracted from the non-English “inputs”. This evaluation value is either single (one subject) or multi (different subjects) for which a cumulative score value is specific. First, the analysis for ${AV}_{y}^{x} B$ and $A \land_{y}^{x} B$ is presented (before fuzzification) in Fig. 6. The analyses revalues around the score obtained from single and multi-types.

Fig. 6

${AV}_{y}^{x} B$ & $A \land_{y}^{x} B$ analysis.

The above analyses show that the score validation is cumulative/ a constant in both $({AV}_{y}^{x} B$ or $(A \land_{y}^{x} B)$ assessments. The first case (i.e.) $({AV}_{y}^{x} B)$ maximizes single score compared to $(A \land_{y}^{x} B)$ as the LUV functions are not implied. Therefore, the assessment is estimated as Q/density based on which ${AV}_{y}^{x} B$ or $(A \land_{y}^{x} B)$ is computed. This is analyzed for n = 16 sessions. Therefore the constant links in the above representation require optimization based on [M, H, and (c, d)]. Therefore, the same optimization post the fuzzification based on LVU is presented in Fig. 7.

Fig. 7

LVU-based analyses.

Post the fuzzification process, the ${AV}_{y}^{x} B$ and $(A \land_{y}^{x} B)$ are tuned using LVU. Here the membership function rules are analyzed based on the credit observed. In this process, the evaluation % for the varying type is required in identifying individual assessments. In this assessment, different sessions are required to identify the varying score. However, the final (e ∧ f) based on P_j (α) is required to prevent failures in assessment (Fig. 7). Students’ session output is used to calculate their qualifying ratio, which can be used to gauge the efficacy of classroom instruction and motivate students to do better. By establishing a benchmark for teachers to achieve, this ratio allows them to assess and, if necessary, enhance their own skill set. Quantity of trained educators whose expertise can be used to evaluate pupils’ learning. Now, based on the P_j (d), its precise requirements for teachers and student perspective analysis are presented in Fig. 8.

Fig. 8

P _j ( α) analysis.

The P_j (α) the analysis is presented in Fig. 8 for (LUV) such that x, y, and n availability is high. The fuzzy constraints for single evaluation (teacher perspective) and (x, y, n) (student perspective) are analyzed. Depending on the available P_j (α), the two distinct processes are evaluated. Using the M (α) ∀ (x, y, n) (e ∧ f) is improved for which the rating is provided. The contrary part is the variation between (x, y, n) ∀ H availability for L^V (α₁, α₂, …, α_n) (Refer to Fig. 8).

5 Comparative analysis

The comparative analysis is performed for the metrics evaluation rate, quality detection, recommendation, evaluation time, and data balancing factor. The assessments are varied from 2 to 16, and the periodic variations are observed from 30 to 240. Along the proposed ADEM-FL, the existing FCTR-LFM [6], QE-OOHT [18], and Fuzzy AHP [32] methods are considered for comparison. The above performance metrics as the essential aspects for gauging the independencies’ performance are the teachers’ knowledge and current skills through training and self-learning. In addition to helping determine whether or not a teacher is effective, the qualifying ratio can also be used to help that instructor prepare for future lessons. The approach takes less time overall to evaluate the effectiveness of college professors’ classroom instruction. This external demand necessitates a systematic approach to evaluating the expertise of education professionals.

5.1 Evaluation rate

The evaluation rate in this proposed method is better by using fuzzy logic in the accessed data evaluation method. The proposed method inputs the teachers’ skills and students’ productivity for evaluation. The teachers’ knowledge and updated skills through training and self-learning are the key features for evaluating the independents’ performance. The impact of the above features on the student qualifying ratio and understandability (through examination) are analyzed periodically. Depending on the qualifications and performance, the teachers’ knowledge update is recommended with the new training programs. In this evaluation process, fuzzy logic is implied for balancing and identifying the maximum validation criteria that satisfy the quality requirements. The teaching sessions may contain the teacher’s performance using their skills and knowledge. Their experience in the teaching field is to perform well in the training session to provide the efficacious outcome. The uncertain data helps in finding the decision recommendation to evaluate the teaching quality of college teachers (Refer to Fig. 9).

Fig. 9

Evaluation rate.

5.2 Quality fetection

The efficacious performance of the students makes the high qualifying ratio. This performance can increase the qualifying rate of the assessment with the help of enhanced skills and knowledge. The qualifying ratio not only aids in identifying the teacher’s quality but also helps in improving their knowledge to perform better in another session. This will be recorded periodically for the complete successful assessment. The quality of the teacher’s skills and knowledge is detected by using the FL in ADEM. The key features for evaluating the students’ performance have an impact on the student qualifying ratio, and understandability is analyzed periodically depending on the qualifications and performance. The teachers’ knowledge update is recommended with the new training process. The efficacious performance of the students makes the high qualifying ratio. This performance can increase the qualifying rate of the assessment with the help of enhanced skills and knowledge. The qualifying ratio not only aids in identifying the teacher’s quality but also helps in improving their knowledge to perform better in another session (Refer to Fig. 10).

Fig. 10

Quality detection.

5.3 Recommendation

Depending on the qualifications and performance, the teachers’ knowledge update is recommended with the new training programs. The recommendations using partial and fulfilled quality constraints are identified using the logical truth over the varying assessments. In this fuzzy logic method, four steps are done for the decision recommendation process: membership, knowledge base rules, fuzzifier, and defuzzifier. The membership function has information about the quality of the teaching sessions according to the teacher’s skills and upgraded knowledge. Primarily, this base rule represents the facts of the ADEM to identify the quality of the teachers, which is based on the fuzzy set theory so that the knowledge base systems will produce a comparable hypothesis. Fuzzification in the fuzzy logic data is used to convert the correct input of the teachers’ skills and the students’ productivity into fuzzy output. Defuzzification in the fuzzy logic data provides the napped results without any counterfeits. The decision recommendation can clearly say whether the teacher lacks training or they are fit for it. The fuzzy logic data helps in finding the decision recommendation to evaluate the teaching quality of college teachers (Refer Fig. 11).

Fig. 11

Recommendation.

5.4 Evaluation time

The total evaluation time for evaluating the teaching quality of the college teachers is less in this method. Assessing the knowledge of teaching professionals to meet the external demand becomes mandatory. This article introduces an Acceded Data Evaluation Method (ADEM) using Fuzzy Logic (FL) for teaching quality assessment. The proposed method inputs the teachers’ skills and students’ productivity for evaluation. In this evaluation process, fuzzy logic is implied for balancing and identifying the maximum validation criteria that satisfy the quality requirements. The teachers’ skills and their potential to teach are calculated by using FL in ADEM. Here fuzzy logic in ADEM is used, based on the observation that the decisions depend on imprecise information. Teachers must also be prepared with these skills according to the student’s demands that they improve their skills and knowledge to make the students perform well in their assessment process. The evaluation process is done in a limited period from the performance of the teachers and students (Refer to Fig. 12).

Fig. 12

Evaluation time.

5.5 Data balancing factor

In evaluating the teaching quality of the college teachers’ evaluation process, fuzzy logic is implied for balancing and identifying the maximum validation criteria that satisfy the quality requirements. Fuzzy logic data is used for the data balancing from the outputs of the teachers’ data and the productivity of the students. The decision recommendation and the data balancing process using the fuzzy logic data in the accessed data evaluation method help to provide the correct output about the teaching quality of the college teachers. The teachers’ skills can be examined by extracting the data from the teaching sessions, and the productivity of the students can be examined by the assessment process where the qualifying ratio takes place. These are the outputs given to the fuzzy logic method, which helps in balancing the data and also helps in providing the decision recommendation. These can be given as the input to the fuzzy logic to identify the perfect data balancing factor (Refer Fig. 13). These data can be examined in every training session periodically of teachers for the data balancing factor with the output of the student’s performance in the examination. Tables 2 3 present the comparative analysis summary for the varying assessments and periodic variations.

Fig. 13

Data balancing factor.

Table 2

Comparative analysis summary (assessments)

Metrics	FCTR-LFM	QE-OOHT	Fuzzy AHP	ADEM-FL
Evaluation Rate	0.587	0.715	0.841	0.9522
Quality Detection (%)	69.88	75.34	84.86	94.791
Recommendation	0.767	0.836	0.889	0.9278
Evaluation Time (Variation)	14.23	11.16	8.96	7.163
Data Balancing Factor (Recommendation/ Assessment)	0.655	0.739	0.802	0.8769

Table 3

Comparative analysis summary (periodic variations)

Metrics	FCTR-LFM	QE-OOHT	Fuzzy AHP	ADEM-FL
Evaluation Rate	0.592	0.726	0.844	0.9464
Quality Detection (%)	69.72	75.29	83.69	94.915
Recommendation	0.798	0.832	0.896	0.9252
Evaluation Time (Variation)	14.05	11.39	9.02	7.066
Data Balancing Factor (Recommendation/ Assessment)	0.638	0.698	0.809	0.8853

The proposed method improves the evaluation rate, quality detection, recommendation, and data balancing factor by 11.89%, 9.05%, 9.7%, and 14.49%, respectively. The proposed method also reduces the evaluation time by 6.24%.

The proposed method improves the evaluation rate, quality detection, recommendation, and data balancing factor by 11.29%, 9.34%, 8.32%, and 17.03%, respectively. The proposed method also reduces the evaluation time by 6.41%.

The findings demonstrated that the chosen indicators of teaching practice quality were both specific enough to be used in the internal evaluation and general enough to be used in the quality assessment of any teaching practice as well as in the evaluation of the effectiveness of different university-level teacher education programs. However, after reviewing the relevant literature, it became clear that teacher evaluation frequently imposed limits on professional behavior and instead rated teachers only on the basis of their teaching.

6 Conclusion

This article introduced an acceded data evaluation method using fuzzy logic for identifying and improving the teaching quality of professionals in colleges. The teaching skills of the professionals are extracted from different sessions based on knowledge and experience. This is periodically performed using different observed data from teachers’ and students’ perspectives. The fuzzification process is instigated depending on the independent evaluation for skills, student qualifying ratio, and knowledge. After considering the different rules for qualifying ratio and membership functions, defuzzification is performed. This defuzzification analyzes different combinations of inputs using data balance. The data balance is required for stabilizing the performance across different periodic evaluations. In this process, the knowledge-dependent rules are identified and are recurrently analyzed using the fuzzy process for precise decisions. The proposed method improves the evaluation rate, quality detection, recommendation, and data balancing factor by 11.89%, 9.05%, 9.7%, and 14.49%, respectively. The proposed method also reduces the evaluation time by 6.24%.

Footnotes

Acknowledgment

This research was supported by the key project of education and teaching reform research of Jiamusi University, Research and Practice on the Training Mode of Normal Professionals in Local Comprehensive Universities under the Background of Professional Certification, project number is 2020JY1-05 and the 2017 National Science of Education “13th Five-Year” Plan Project Titled A Study of the Relationship between Teachers’ Professional Ability Composition and Students’ Learning Quality in Applied Technology Colleges and Universities (the Key lssues of Ministry of Education, project number is DIA170357).

References

Aditomo and Köhler

, Do student ratings provide reliable and valid information about teaching quality at the school level? Evaluating measures of science teaching in PISA, Educational Assessment, Evaluation and Accountability 32 (2020), 275–310.

Panayiotou , Herbert

, Sammons

and Kyriakides

, Conceptualizing and exploring the quality of teaching using generic frameworks: A way forward, Studies in Educational Evaluation 70 (2021), 101028.

Fauth , Atlay

, Dumont

and Decristan

, Does what you get depend on who you are with? Effects of student composition on teaching quality, Learning and Instruction 71 (2021), 101355.

Troussas , Krouska

and Sgouropoulou

, A novel teaching strategy through adaptive learning activities for computer programming, IEEE Transactions on Education 64 (2020), 103–109.

Thommen , Sieber

, Grob

and Praetorius

A.K.

, Teachers’ motivational profiles and their longitudinal associations with teaching quality, Learning and Instruction 76 (2021), 101514.

Yao and Deng

, Teaching teacher recommendation method based on fuzzy clustering and latent factor model, IEEE Access 8 (2020), 210868–210885.

Data Sets, https://www.openintro.org/book/statdata/?data=evals

Carter , Onwuegbuzie

, Singal

and van der Velde

, Perceptions of teaching quality in Rwandan secondary schools: A contextual analysis, International Journal of Educational Research 109 (2021), 101843.

Litke

, Boston

and Walkowiak

T.A.

, Affordances and constraints of mathematics-specific observation frameworks and general elements of teaching quality, Studies in Educational Evaluation 68 (2021), 100956.

10.

Meng , Xiao

and Wang

, Rating the crisis of online public opinion using a multi-level index system, The International Arab Journal of Information Technology 19 (2022), 597–608.

11.

Zhang

, Bae

C.L.

and Broda

, Science self-concept, relatedness, and teaching quality: A multilevel approach to examining factors that predict science achievement, International Journal of Science and Mathematics Education 20 (2022), 503–529.

12.

Skedsmo

and Huber

S.G.

, Measuring teaching quality, designing tests, and transforming feedback targeting various education actors, Educational Assessment, Evaluation and Accountability 32 (2020), 271–273.

13.

Yun

, Ravi

R.V.

and Jumani

A.K.

, Analysis of the teaching quality on deep learning-based innovative ideological political education platform, Progress in Artificial Intelligence (2022), https://doi.org/10.1007/s13748-021-00272-0.

14.

Romel , Tadesse

and Jibat

, Teacher quality, self-efficacy, and quality teaching in Ethiopian primary schools: An integrated sociological and psychological perspective, Studies in Educational Evaluation 70 (2021), 101029.

15.

Wang

, Cui

, Liu

, Fang

and Sha

, A Multi-type transferable method for missing link prediction in heterogeneous social networks, IEEE Transactions on Knowledge and Data Engineering (2023), https://doi.org/10.1109/TKDE.2022.3233481

16.

Zaky

, Peer assessment for teaching quality of online writing classes in higher education: investigating learning styles’ impact, SN Social Sciences 1 (2021), 1–33.

17.

Alvarez-Blanco

, Castro-Lopez

and Cervero

, Intelligent analysis of the quality of education through teaching practices on virtual campuses, European Journal of Psychology of Education, (2022), https://doi.org/10.1007/s10212-022-00649-2

18.

Bao

and Yu

, Evaluation method of online and offline hybrid teaching quality of physical education based on mobile edge computing, Mobile Networks and Applications 26 (2021), 2188–2198.

19.

Torbeck

and Dunnington

, Development of a peer review of the operative teaching process and assessment tool, The American Journal of Surgery 221 (2021), 263–269.

20.

Yang

, Chun

, Liu

, Wang

and Yang

, A novel quality evaluation method for standardized experiment teaching, Soft Computing 26 (2022), 6889–6906.

21.

Rong

L.J.

, Distance learning quality assessment of universities based on interval monotonic decision tree algorithm, Computers and Electrical Engineering 102 (2022), 108116.

22.

Nojavan

, Heidari

and Mohammaditabar

, A fuzzy service quality based approach for performance evaluation of educational units, Socio-Economic Planning Sciences 73 (2021), 100816.

23.

Sailer

, Stadler

, Schultz-Pernice

, Franke

, Schöffmann

, Paniotova

, Husagic

and Fischer

, Technology-related teaching skills and attitudes: Validation of a scenario-based self-assessment instrument for teachers, Computers in Human Behavior 115 (2021), 106625.

24.

Iqbal

M.H.

, Siddiqie

S.A.

and Mazid

M.A.

, Rethinking theories of lesson plan for effective teaching and learning, Social Sciences & Humanities Open 4 (2021), 100172.

25.

Abd-Elwahed

M.S.

and Al-Bahi

A.M.

, Sustainability awareness in engineering curriculum through a proposed teaching and assessment framework, International Journal of Technology and Design Education 31 (2021), 633–651.

26.

Alscher

, Ludewig

and McElvany

, Civic education, teaching quality and students’ willingness to participate in political and civic life: Political interest and knowledge as mediators, Journal of Youth and Adolescence 51 (2022), 1886–1900.

27.

Beuchel

, Ophoff

J.G.

, Cramer

and Hautzinger

, Promoting occupational health and teaching quality: The impact of a mindfulness intervention in teacher training, Teaching and Teacher Education 114 (2022), 103703.

28.

Hii

P.K.

, Goh

C.F.

, Tan

O.K.

, Amran

and Ong

C.H.

, An information system success model for e-learning postadoption using the fuzzy analytic network process, Education and Information Technologies (2023), https://doi.org/10.1007/s10639-023-11621-y

29.

Sun

, Wu

and Liu

, Toward understanding students’ learning performance in an object-oriented programming course: The perspective of program quality, IEEE Access 8 (2020), 37505–37517.

30.

Lazarides

and Schiefele

, The relative strength of relations between different facets of teacher motivation and core dimensions of teaching quality in mathematics-A multilevel analysis, Learning and Instruction 76 (2021), 101489.

31.

Lazarides

, Fauth

, Gaspard

and Göllner

, Teacher self-efficacy and enthusiasm: Relations to changes in student-perceived teaching quality at the beginning of secondary education, Learning and Instruction 73 (2021), 101435.

32.

S.L.

, Yeyao

and Shabaz

, Multi-criteria decision-making for determining the best teaching method using fuzzy analytical hierarchy process, Soft Computing (2022), https://doi.org/10.1007/s00500-022-07554-2

33.

, Li

, Hoque

M.A.

, Xia

, Tarkoma

and Hui

, To what extent we repeat ourselves? Discovering daily activity patterns across mobile app usage, IEEE Transactions on Mobile Computing 21 (2022), 1492–1507.

34.

Voss

, Zachrich

, Fauth

and Wittwer

, The same yet different? Teaching quality differs across a teacher’s classes, but teachers with higher knowledge make teaching quality more similar, Learning and Instruction 80 (2022), 101614.

35.

Cotronei-Baird

V.S.

, Academic hindrances in the integration of employability skills development in teaching and assessment practice, Higher Education 79 (2020), 203–223.

36.

Qin

, Liu

, Yang

, Yin

, Liu

and Zheng

, User OCEAN personality model construction method using a BP neural network, Electronics 11 (2022), 3022.

37.

Xin

, Shu-Jiang

, Nan

, ChenXu

and Dan

, Review on A big data-based innovative knowledge teaching evaluation system in universities, Journal of Innovation & Knowledge 7 (2022), 100197.

38.

Xiong

, Liu

and Huang

, The influence of digital educationalgames on preschool Children’s creative thinking, Computers & Education 189 (2022), 104578.