Search

This study investigated the influence of CThink4CS2 Module on computational thinking (CT) skills of form four chemistry students. The CThink4CS[superscript]2 Module integrated CT with the Engineering Design Process (EDP) in chemistry class. This study utilized quantitative research methods and quasi-experimental design. Quantitative data were collected using the Computational Thinking Skill Test (CTST) which consisted of algorithmic reasoning, abstraction, decomposition, and pattern recognition constructs. A total of 73 students were in the treatment group (n=39) and control group (n=34). Experimental data were described by means of descriptive analysis and inferential analysis employing two-way MANOVA analysis. The results of the analysis indicated significant differences in CT skills between groups; students in the treatment group demonstrated better results compared to those in the control group. The paper provides insight into the integration of CT and EDP as effective pedagogical strategies for inculcating CT skills.

Computer science (CS) has become a critical part of K-12 education worldwide. Computational thinking (CT) skills are a key set of competencies in CS education that can solve problems and use computational design to create useful solutions. However, preservice and in-service teachers are not fully prepared to integrate CS and CT into their curricula. Furthermore, there are limited special topic courses and educational research on how to facilitate in-service teachers' professional learning of CS and CT, as well as their content specific integration. Therefore, this study investigated in service teachers' perceptions and development of CT skills in an online graduate emerging technologies course. Theoretically framed by the four cornerstones of CT (i.e., abstraction, algorithms, decomposition, and pattern recognition), participants perceived that they increased their CT problem-solving and creativity skills but decreased their collaborative learning and critical thinking skills. Additionally, teachers increased their CT test scores after taking the course. Most teachers used CT terminology correctly (i.e., algorithms and decomposition). However, only 59% correctly described abstraction and pattern recognition, while most teachers did not mention debugging. The authors call on teacher educators to address in-service teachers' CS knowledge gaps, increase their CT skills, and select appropriate strategies for CT professional learning.

"Using Sentiment Analysis to Ease Students toward or around Macroanalysis" introduces teachers to the mathematical analytical method currently being promoted by Matthew Jockers which he calls macroanalysis. This method uses an algorithm called "R," originally designed by researchers in sociology. In many ways, macroanalysis presents an approach to literary analysis that might seem to many traditionalists to run counter to our training as close readers or advocates of specific literary theories. Instead of focusing on specific elements in a limited number of texts, macroanalysis uses a computer to consume a huge number of literary works--more than a single human being could read in a lifetime--and collect a statistically significant amount of data, a search which has been programmed into the algorithm by the researcher. This use of a computer to do the reading for the researcher results in a new methodology which Jockers calls "distant reading" (a term he borrows from Franco Moretti). After providing an introductory discussion of macroanalysis, I turn my focus onto making a case for the advantages of using graphs to aid in classroom discussions of reading assignments. In the introductory discussion, I point out that one serious advantage that arises from Jocker's macroanalysis is how the "R" algorithm produces a variety of graphs that provide visual reinforcement for the statistical data it generates. Consequently, in the second section of the essay, I provide a justification for why it would be advantageous in a college or even high school discussion of a reading assignment to privilege visual presentations of student responses to the assignment. Once I've established the nature of macroanalysis, and how challenging it is for the average English major, I offer an alternative approach to creating graphs to clarify student responses. This alternative approach is a blend of "sentiment analysis" and reader-response theory. Essentially this approach can be used to ease students toward the more serious graphing features of macroanalysis, or it can function as a self-contained form of analysis without shifting to the difficult mathematics characteristic of macroanalysis. To prepare the readers for the rather lengthy discussion of a graph produced by a group of Virginia Governor's School students, I first provide a simple graph that shows my responses to the opening scene of Flannery O'Connor's short story "A Good Man Is Hard to Find." Having clarified how sentiment analysis, with the mechanical help of Microsoft Excel, can produce an engaging graph of emotional and intellectual responses that can be easily shared with the class, I then move to the last half of my essay--a detailed discussion of an actual student graph derived from a reading of Helen Klein Ross's "What Was Mine" which was much more sophisticated than the graph I created for the O'Connor story. Even though the students' graph was overly ambitious, the discussion of its visual dynamics demonstrate how even a faulty graph can teach students how to become close readers and appreciate the importance of analysis that is based on specific elements of the literature being analyzed.

Many students persistently misinterpret histograms. Literature suggests that having students solve dotplot items may prepare for interpreting histograms, as interpreting dotplots can help students realize that the statistical variable is presented on the horizontal axis. In this study, we explore a special case of this suggestion, namely, how students' histogram interpretations alter during an assessment. The research question is: In what way do secondary school students' histogram interpretations change after solving dotplot items? Two histogram items were solved before solving dotplot items and two after. Students were asked to estimate or compare arithmetic means. Students' gaze data, answers, and cued retrospective verbal reports were collected. We used students' gaze data on four histogram items as inputs for a machine learning algorithm (MLA; random forest). Results show that the MLA can quite accurately classify whether students' gaze data belonged to an item solved before or after the dotplot items. Moreover, the direction (e.g., almost vertical) and length of students' saccades were different on the before and after items. These changes can indicate a change in strategies. A plausible explanation is that solving dotplot items creates readiness for learning and that reflecting on the solution strategy during recall then brings new insights. This study has implications for assessments and homework. Novel in the study is its use of spatial gaze data and its use of an MLA for finding differences in gazes that are relevant for changes in students' task-specific strategies.

Formative assessment is about providing and using feedback and diagnostic information. On this basis, further learning or further teaching should be adaptive and, in the best case, optimized. However, this aspect is difficult to implement in reality, as teachers work with a large number of students and the whole process of formative assessment, especially the evaluation of student performance takes a lot of time. To address this problem, this paper presents an approach in which student performance is collected through a concept map and quickly evaluated using Machine Learning techniques. For this purpose, a concept map on the topic of mechanics was developed and used in 14 physics classes in Germany. After the student maps were analysed by two human raters on the basis of a four-level feedback scheme, a supervised Machine Learning algorithm was trained on the data. The results show a very good agreement between the human and Machine Learning evaluation. Based on these results, an embedding in everyday school life is conceivable, especially as support for teachers. In this way, the teacher can use and interpret the automatic evaluation and use it in the classroom.

Data repositories currently constitute essential programs within institutions. In fact, universities are the primary institutions that promote the creation, management and storage for the safekeeping of a variety of documents, data and/or projects. This work is carried out within the framework of institutional need and the application of knowledge by the students and professors in the software degree program at a university in Ecuador. This project centers on a web application that makes it possible to manage, upload and download resources. This initial version focuses on the resources related to academic projects, data sets and algorithms. The scalability characteristics have been carefully introduced to ensure a progressive evolution of other types of resources and functionalities, in accordance with educational needs. Several aspects of the web interfaces are based on the characteristics of other existing repositories, and a series of tests related to functionality have been conducted in order to move beyond the introductory stage of use, to begin with, by the university community. This project is carried out according to a conception based on contributions, in which the agents, in principle, are members of the institution. The intensive and collaborative use of this utility will contribute to the growth of a data culture within the institution. Furthermore, the collection and organization of data for the purposes of study and research by the different university degree programs makes it possible to analyze and contrast them from different perspectives.

The teaching and learning of programming has proven to be a challenge for students of computer courses, since it presents challenges and requires complex skills for the good development of students. The traditional teaching model is not able to motivate students and arouse their interest in the topic. The tool proposed herein, the REA-LP, aims to facilitate the study and retention of content related to the discipline of programming logic at the technical level by presenting its content through various types of media, in addition to allowing students to actively participate in the construction of their knowledge, favoring engagement and motivation. From the results of an empirical study with 39 students, it can be concluded that the tool was very well accepted, being effective in facilitating and assisting participants in their learning, motivation, and interest in classes, mainly due to the way in which the content is presented by REA-LP.

The concept, development and assessment of computational thinking have increasingly become the focus of research in recent years. Most of this type of research focuses on older children or adults. Preschool age is a sensitive period when many skills develop intensively, so the development of computational thinking skills can already begin at this age. The increased interest in this field requires the development of appropriate assessments. Currently, there are only a limited number of computational thinking assessments for preschool children. Based on this shortcoming, an assessment tool, named AlgoPaint Unplugged Computational Thinking Assessment for Preschool, was created addressed for 4-7 years old children. It is a paper-pencil-based test, which examines the following computational thinking domains: algorithms and debugging. Regarding computational concepts, simple instructions, simple and nested loops, and conditionals are included in the test. For the preliminary testing, AlgoPaint test was applied by 11 preschool teachers with 56 preschool age children. The test was also evaluated by 6 experts in algorithmic thinking working at universities. Based on the feedback given by the teachers and the experts, and the results of the children, AlgoPaint Computational Thinking Test was revised and completed. The revised version of the test is included in the appendix of the paper.

Sixty (35 girls) ninth graders were assessed on measures of algebraic reasoning and usage of visual and symbolic representations (with a prompt for visual use) to solve equations and inequalities. The study grouped visual representations into two categories: arithmetic-visual, which entailed the use of real-world objects to represent specific values of variables, and algebraic-visual, which involved formal representations like the number line and the coordinate plane. Symbolic representations, on the other hand, encompassed the use of standard algorithms to solve equations, such as changing the place of terms in an equation. The results reveal that the use of algebraic visuals, as opposed to arithmetic visuals, was associated with enhanced algebraic reasoning. Further, although the students initially relied on standard algorithms to explain equations and inequalities, they could produce accurate algebraic-visual representations when prompted. These findings suggest that students have multiple representations of equations and inequalities but only express visual representations when asked to do so. In keeping with the general relationship between visuospatial abilities and mathematics, self-generated algebraic-visual representations partially mediated the relation between overall mathematics achievement and algebraic reasoning.

Traditional information literacy skills (e.g., effectively finding and evaluating information) need to be updated due to the rapidly changing information ecosystem and the growing dominance of online platforms that use algorithms to control and shape information. This article proposes additions to the current ACRL "Framework for Information Literacy for Higher Education" that relate to algorithmic literacy. The "Authority is Constructed and Contextual" frame can be applied to recognizing the need to question algorithmic authority (including algorithmic bias), the Information Has Value" frame can be used to acknowledge online platforms' use of proprietary algorithms allowing third parties to access personal data, and the "Searching as Strategic Exploration" frame can draw attention to search results in online platforms are mediated through algorithms. Classroom activities to teach the new knowledge practices and dispositions are also included.

The aim of this study is to classify the subjects and skills of middle school mathematics course in the context of MATH Taxonomy and to determine their relations. For this purpose, the questions and answers related to the mathematics subtest of a national exam were analyzed over the answers of 20154 students. The study continued with the analysis of the data set containing the answer patterns of the students according to the rules of association of data mining methods by using RapidMiner program and FP-Growth algorithm. The results show that students with any level of skill in MATH taxonomy related to mathematical skills could perform skills in subcategories at a high rate. Students who have attainment in the learning areas provided later in the mathematics program had a high rate of knowledge in the previously provided learning areas. However, this result is not valid for Probability of Simple Events.

Computational thinking (CT) is an important 21st-century skill. This paper aims at investigating predictors of CT self-efficacy among high-school students. The hypothesized predictors are grouped into three areas: (1) student characteristics, (2) home environment, and (3) learning opportunities. CT self-efficacy is measured with the Computational Thinking Scales (CTS) that comprises five dimensions: creativity, algorithmic thinking, cooperativity, critical thinking, and problem solving. N = 202 high-school students act as the sample, linear regression as the analysis method. The best prediction is possible for algorithmic thinking (R[superscript 2] = 0.511). For cooperativity, the explanatory power of our model it is weak (R[superscript 2] = 0.146). Across all five CTS dimensions, CT self-concept is the best predictor for CT self-efficacy. [For the full proceedings, see ED636095.]

The test view is on the Duolingo English Test (DET), an alternative online English proficiency test with a machine-driven characteristic. The review covers essential information of the DET such as test purpose, usage, score-mapping with CEFR scale, price, and publisher. Meanwhile, the test usefulness is discussed with focuses on reliability, content validity, and impact. Both good points and weaknesses are illustrated. Furthermore, the view tries to criticize the AI algorithm issues in test content and score rating.

Currently, the majority of e-learning lessons created and disseminated advocate a "one-size-fits-all" teaching philosophy. The e-learning environment, however, includes slow learners in a noticeable way, just like in traditional classroom settings. Learning analytics of educational data from a learning management system (LMS) have been considered by the researchers as a potential means of identifying slow e-learners and supporting, contesting, and altering present educational practices in e-learning. We used the students' rates of learning and grade points along with the total learning time, which is calculated from the time series log data, to cluster the learners. The rate at which a student learns determines whether he or she is a slow learner, an average learner, or a gifted learner. For classifying learners, we followed a step-by-step procedure that included instructional design to create a dataset, learning analytics of the dataset, and a machine learning strategy to cluster e-learners. The system has been adequately integrated with the methods for measuring student learning. A strategy based on the revised Bloom's Taxonomy is offered for the assessment of learners. The K-Means clustering approach is used to group learners who have similar performance without collecting a learner's previous academic records or demographic information. In the experimental evaluation, 7.7% of e-learners are grouped as slow learners, while advanced learners make up 61.3 percent of the student body and average learners make up 31 percent. According to the study, there is a correlation between learning rate and academic success, with fast learners having a lower learning rate.

Students organize and manage their own learning time, choosing when, what, and how to study due to the flexibility of online learning. Each person has unique learning habits that define their behaviours and distinguish them from others. To investigate the temporal behaviour of students in online learning environments, we seek to identify suitable time-windows that could be used to investigate their temporal behaviour. First, we present a novel perspective for identifying different types of sessions based on individual needs. The majority of previous works address this issue by establishing an arbitrary session timeout threshold. In this paper, we propose an algorithm for determining the optimal threshold for a given session. Second, we use data-driven methods to support investigators in determining time-windows based on the identified sessions. To this end, we developed a visual tool that assists data scientists and researchers to determine the optimal settings for session identification and locating suitable time-windows.

This study aims to conduct a comparative study of Bagging and Boosting algorithms among ensemble methods and to compare the classification performance of TreeNet and Random Forest methods using these algorithms on the data extracted from ABIDE application in education. The main factor in choosing them for analyses is that they are Ensemble methods combining decision trees via Bagging and Boosting algorithms and creating a single outcome by combining the outputs obtained from each of them. The data set consists of mathematics scores of ABIDE (Academic Skills Monitoring and Evaluation) 2016 implementation and various demographic variables regarding students. The study group involves 5000 students randomly recruited. On the deletion of loss data and assignment procedures, this number decreased to 4568. The analyses showed that the TreeNet method performed more successfully in terms of classification accuracy, sensitivity, F1-score and AUC value based on sample size, and the Random Forest method on specificity and accuracy. It can be alleged that the TreeNet method is more successful in all numerical estimation error rates for each sample size by producing lower values compared to the Random Forest method. When comparing both analysis methods based on ABIDE data, considering all the conditions, including sample size, cross validity and performance criteria following the analyses, TreeNet can be said to exhibit higher classification performance than Random Forest. Unlike a single classifier or predictive method, the classification or prediction of multiple methods by using Boosting and Bagging algorithms is considered important for the results obtained in education.

Cognitive Diagnosis Models (CDMs) are a special family of discrete latent variable models that are widely used in educational and psychological measurement. A key component of CDMs is the Q-matrix characterizing the dependence structure between the items and the latent attributes. Additionally, researchers also assume in many applications certain hierarchical structures among the latent attributes to characterize their dependence. In most CDM applications, the attribute--attribute hierarchical structures, the item-attribute Q-matrix, the item-level diagnostic models, as well as the number of latent attributes, need to be fully or partially pre-specified, which however may be subjective and misspecified as noted by many recent studies. This paper considers the problem of jointly learning these latent and hierarchical structures in CDMs from observed data with minimal model assumptions. Specifically, a penalized likelihood approach is proposed to select the number of attributes and estimate the latent and hierarchical structures simultaneously. An expectation-maximization (EM) algorithm is developed for efficient computation, and statistical consistency theory is also established under mild conditions. The good performance of the proposed method is illustrated by simulation studies and real data applications in educational assessment.

In course design, topic outline organization encompasses the structuring and sequencing of topics to be delivered in a learning environment. Recent studies in topic outline optimization revolve around massive open online courses (MOOCs) due to their abundance but not much has been studied on the traditional courses. This study investigates the organization of topic outlines in traditional introductory programming courses across Philippine higher education institutions (HEIs) and evaluates the viability of the synthesized alternative topic outlines. Course syllabi were collected from 16 HEIs. A topic precedence graph (TPG) model that provides a structured overview of the introductory programming was created via a semi-supervised pairwise constrained k-means (PCK-Means) clustering to structure the topics which produced 20 topic clusters with strong topic cohesion within the clusters. The TPG showed that HEIs tend to start the outline similarly, followed by core programming topics with varied sequences, and divergent ways of ending the outline. Two anomaly clusters were identified as having topic titles grouped that do not seem to have a unifying topic. Limitations of the clustering algorithm are identified where it cannot identify semantic meaning between words which may affect its applicability in situations where topic titles are named inconsistently. From the TPG, alternative optimal and comprehensive topic outlines were synthesized via greedy and DFS graph traversal algorithms. However, these alternative outlines performed very poorly when compared with the evaluators' (n=19) arrangement of topic outlines due to some prerequisite topics being discussed in the latter part already. Overall, this study introduces a method to incorporate computer science technologies in structuring topics across HEIs and aiding educators in topic outline design but more research is needed before it can be implemented in a real classroom setting.

Online discussion research has mainly been conducted using case methods. This article proposes a method for comparative analysis based on network metrics such as information entropy and global network efficiency as more holistic measures characterizing social learning group dynamics. We applied social learning optimization analysis of networks (SLOAN) to a data set consisting of Coursera courses from a range of disciplines. We examined the relationship of discussion forum uses and measures of network efficiency, characterized by the information flow through the network. Discussion forums vary greatly in size and in use. Courses with a greater prevalence of subject-related versus procedural talk differed significantly in seeking but not disseminating behaviors in massive open online course discussion forums. Subject-related talk was related to higher network efficiency and had higher seeking and disseminating scores overall. We discuss the value of SLOAN for social learning and argue for the experimental study of online discussion optimization using a discussion post recommendation system for maximizing social learning.

The COVID-19 pandemic induced a digital transformation of education and inspired both instructors and learners to adopt and leverage technology for learning. This led to online learning becoming an important component of the new normal, with home-based virtual learning an essential aspect for learners on various levels. This, in turn, has caused learners of varying levels to interact more frequently with virtual resources to supplement their learning. Even though virtual learning environments provide basic resources to help monitor the learners' online behaviour, there is room for more insights to be derived concerning individual learner performance. In this study, we propose a framework for visualising learners' online behaviour and use the data obtained to predict whether the learners would clear a course. We explored a variety of binary classifiers from which we achieved an overall accuracy of 80%-85%, thereby indicating the effectiveness of our approach and that learners' online behaviour had a significant effect on their academic performance. Further analysis showed that common patterns of behaviour among learners and/or anomalies in online behaviour could cause incorrect interpretations of a learner's performance, which gave us a better understanding of how our approach could be modified in the future.

Rough set theory (RST) was proposed by Zdzistaw Pawlak (Pawlak,1982) as a methodology for data analysis using the notion of discernibility of objects based on their attribute values. The main advantage of using RST approach is that it does not need additional assumptions--like data distribution in statistical analysis. Besides, it provides efficient algorithms and tools for finding hidden patterns. Despite its advantages, the adoption of RST in educational research is still limited, which may be due to limited quality software available for RST. Recently, the "RoughSets" package in R statistical system has been developed (Riza et al., 2014), providing various utilities of the RST methods. In the paper, we will first describe the basic RST concepts and steps of data analysis under RST. We will then apply RST to a study, which aimed to determine the relative significance of various SRL strategies used by student participants, so as to illustrate the steps of data analysis using the "RoughSets" R package. From the illustration, it is expected that more researchers in the field of education will be encouraged to try RST methods in their own studies.

Academic integrity fosters a culture of honesty, trust, and respect within the educational community. Evidence indicates that manual plagiarism checks through human judgment remain prevalent in undergraduate theses, terminal assignments, and group projects in developing countries. To fill this gap, we engaged with students and staff of the Kwame Nkrumah University of Science and Technology, Ghana, to propose a new plagiarism detection architecture, a novel open-source algorithm, and a web-based application for plagiarism detection and prevention. We demonstrate the performance of our algorithms, discuss the results, and suggest avenues for future research to advance this field.

Despite problem solving being a core skill in chemistry, students often struggle to solve chemistry problems. This difficulty may arise from students trying to solve problems through memorising algorithms. Goldilocks Help serves as a problem-solving scaffold that supports students through structured problem solving and its elements, such as planning and evaluation. In this study, we investigated how first-year chemistry students solved problems, when taught with Goldilocks Help, and whether their problem-solving success and approaches changed over the course of one semester. The data comprised of student written problem-solving work, and was analysed using frequency analysis and grouped based on the problem-solving success and the extent of the demonstrated problem-solving elements. Throughout the course of semester, students exhibited increasingly consistent demonstration of structured problem solving. Nonetheless, they encountered difficulties in fully demonstrating such aspects of problem solving as understanding and evaluating concepts, which demand critical thinking and a firm grasp of chemistry principles. Overall, the study indicated progress in successful and structured problem solving, with a growing proportion of students demonstrating an exploratory approach as time progressed. These findings imply the need for incorporation of metacognitive problem-solving scaffolding, exposure to expert solutions, reflective assignments, and rubric-based feedback into wide teaching practice. Further research is required to extend the exploration of the effectiveness of metacognitive scaffolding, in particular "via" think-aloud interviews, which should help identify productive and unproductive uses of the problem-solving elements.

Multiple-choice questions have become ubiquitous in educational measurement because the format allows for efficient and accurate scoring. Nonetheless, there remains continued interest in constructed-response formats. This interest has driven efforts to develop computer-based scoring procedures that can accurately and efficiently score these items. Early procedures were typically based on surface features of the responses or simple matching procedures, but recent developments in natural language processing have allowed for much more sophisticated approaches. This paper reports on a state-of-the-art methodology for scoring short answer questions supported by a large language model. Responses were collected in the context of a high-stakes test for medical students. More than 35,000 responses were collected across 71 studied items. Aggregated across all responses the proportion of agreement with human scores ranged from 0.97 to 0.99 (depending on specifics such as training sample size). In addition to reporting detailed results, the paper discusses practical issues that require consideration when adopting this type of scoring system.

At present, the construction of online courses is an important content in the process of educational informatization. At present, there are two problems in the construction of online courses: from the teaching point of view, the design of teaching activities and teaching processes in online courses is not enough, and the teaching effect is not good. Technically, the development of most courses is basically one-off and cannot be updated quickly. Due to the small scale, limited access and strong pertinence of SPOC, it has obvious advantages to implement the online and offline mixed teaching mode of college English based on SPOC in college English teaching. This paper mainly studies the research and analysis of English online courses based on the hybrid real-time synchronization algorithm. This paper combines OFDM synchronization algorithm with SPOC hybrid algorithm teaching to analyze and study it. According to the research, the influence of English online teaching course is very great, and the effect is more obvious than that of traditional teaching.

In this study, we analyzed student participation and success in a college-level Calculus I course that utilized standards-based grading. By measuring the level to which students participate in this class structure, we were able to use a clustering algorithm that revealed multiple groupings of students that were distinct based on activity throughout the semester. Additionally, we analyzed student progress, defined as the number of graded activities successfully completed each week. We found that students who progressed steadily throughout the semester, and thus had lower variability in the number of completed activities per week, tended to receive a higher overall grade. Students whose progress was less consistent, and thus exhibited higher variability in weekly activities completed, tended to receive a lower grade. Overall, this shows a relationship between understanding the expectations in standards-based grading and succeeding in a course.

This study combines link grammar (LG) detector with N-grammar model to analyze and evaluate grammar in compositions. And then the composition level is judged through information entropy. Finally, the composition score is calculated based on the overall composition level and grammar weight. The experimental results show that the combined weight of recall and accuracy of the proposed method in this study is 89.9%, which is 26.6% higher than LG and 9.7% higher than Grammarly. In the performance test of scoring the entire essay, the proportion of error between the proposed method and manual evaluation is 87.29%, with a lower overall mean square error of only 3.08 and a shorter average running time of only 22.69 seconds. The method proposed in this study has a high accuracy and strong applicability in the evaluation of English compositions for middle school students, providing a new approach for teaching English writing for middle school students.

In order to explore the maturity of online concerts and the digital content of music resources, this article analyzes the role of artificial intelligence in music education, discusses the application of artificial intelligence in music education and the development trend of artificial intelligence in education, and studies the quality of vocal music teaching based on artificial intelligence technology. In this paper, ARM and SA algorithms, as well as internal and external probability algorithms, are combined for research and analysis. Through this study, the authors show that human intelligence skills have a certain impact on vocal music teaching, with an impact rate of 56.42%. It can be seen that artificial intelligence can directly optimize the level of music teachers and promote the improvement of teachers' teaching quality and efficiency. This article improves the effective understanding of artificial intelligence in music education and strengthens the scientific and rational application of artificial intelligence in education.

Due to the growth of China's social economy and culture, the demand for psychology is becoming more and more urgent. Environmental health is the unity of people's behavioral health that constitutes the environment and environmental conditions that meet people's basic needs. In this paper, the improved a priori algorithm is combined with English-Chinese bilingual classroom of university students' psychology. It is based on the collected data and basic information of university students' psychology, which provides ideas for the construction of English-Chinese bilingual classroom and teaching platform of psychology under the background of environmental health. The consistency test results show that the consistency between the proposed algorithm and the manual processing method can reach over 96.5%, that is, the accuracy of the algorithm is over 96.5%. Compared with the traditional a priori algorithm, the improved a priori algorithm has a higher recognition accuracy, with the accuracy increased by 21.36% and the recall increased by 13.99%.

This research designed an improved collaborative filtering algorithm to be responsible for music recommendation tasks in the online music teaching platform. This algorithm integrates the user's social trust into the similarity calculation formula. Then, the algorithm uses behavioral feature data driven by preferences, music tags, and popularity as the basis for recommendation calculation. It adopts user data testing on an online music teaching platform. The results showed that when the number of recommended music was eight, the recommended recall rates of XCF, CTR, TSR, and UB-CF recommendation models reached their maximum, reaching 97.82%, 95.26%, 93.95%, and 88.72%, respectively. The AUC and average computational time of the ROC curves for XCF, CTR, TSR, and UB-CF recommended models are 0.7, 0.68, 0.64, 0.57, and 160ms, 136ms, 114ms, and 88ms, respectively. The experimental data shows that the recommendation accuracy of the music recommendation model designed in this study is significantly higher than that of traditional recommendation models.

At present, many schools implement teaching quality evaluation systems with student performance and practical activities as the core data of evaluation. Based on the introduction of the training objectives and discipline nature of the information management and information system specialty, this paper analyzes the construction principles of the multi-disciplinary curriculum system of the specialty and puts forward relevant countermeasures for the construction of the multi-disciplinary education system. This paper introduces the definition of practical teaching evaluation system and teaching evaluation algorithm. Among them, naive Bayesian algorithm can improve the sample accuracy by 95%, so it is necessary to build an application-oriented undergraduate practice teaching evaluation system.

Higher education is becoming increasingly competitive and all educational institutions are concentrating on improving quality and changing traditional higher education teaching methods. New-type classroom instruction has embraced a unique advancement opportunity with the arrival of the fifth generation (5G) era. It is critical to develop a teaching assistance system that makes use of high-speed network methodology and new-type display methodology. For the innovation and reform of higher education, this article combines soft computing techniques, artificial intelligence (AI), and 5G networks. This paper outlines the exact processes and measures for incorporating "5G" technology into higher education. Finally, conduct a comparative experiment to see how good the system is at learning AI knowledge when compared to standard learning methods. The outcomes of the experiments are examined to show that employing this approach to gain AI knowledge is successful and improves students' enthusiasm in learning as well as their hands-on abilities.

In recent years, China has accelerated the process of internationalization and made more and more achievements in transnational communication and cooperation. English learning is very important for contemporary college students. And English reading is an important means to acquire English language knowledge, understand external information and improve English language practice ability. Therefore, cultivating college students' English reading ability is a pivotal element in college English teaching. This study will focus on the application of clustering algorithm based on big data framework in the evaluation of college students' English reading ability, aiming at exploring and establishing an effective evaluation model to promote the improvement of college students' English reading ability and academic achievement.

Kruskal algorithm, as a minimum spanning tree optimization analysis algorithm, has been well used in the field of industrial production to find the optimal parameters. In this study, an interactive teaching platform based on Kruskal algorithm is built based on the ecological aesthetic education theory. Secondly, based on multiple internet of things systems, through the analysis of the problems existing in traditional classroom teaching, combined with the characteristics of ecological aesthetic education, the interactive platform can automatically generate more efficient teaching courses and related methods. Finally, this study verifies the reliability and stability of the above platform through experiments and practices. The experimental results show that the application of the theory and method of ecological aesthetic education in the interactive teaching platform based on Kruskal algorithm achieves the combination of traditional classroom teaching and ecological aesthetic education and gives full play to the main role of students in the classroom.

With the development of track and field, people pay more and more attention to the quality of classroom teaching of track and field technology, and the evaluation of teaching quality plays a key role in it. In today's educational reform, teaching evaluation plays an important role as an important method to test teachers' teaching and students' learning. With the rapid development of machine learning, especially deep learning, image-based individual abnormal behavior detection technology is becoming more and more mature, but there are still many difficulties to be solved in video-based group abnormal behavior detection technology. Therefore, it is necessary to study the detection algorithm. Based on machine vision theory, image processing theory and video analysis technology, this paper studies three key technologies involved in human abnormal behavior detection in video: moving target detection, moving target tracking and abnormal behavior detection.

The large scale expansion of online courses has led to the crisis of course quality issues. In this study, we first established an evaluation index system for online courses using factor analysis, encompassing three key constructs: course resource construction, course implementation, and teaching effectiveness. Subsequently, we employed factor analysis and entropy weight TOPSIS multi-attribute decision analysis methods to comprehensively evaluate 541 courses. Later on, we conducted correlation and regression analyses between the evaluation results of these two methods and that of experts. The results reveal that the comprehensive evaluation scores derived from both factor analysis and entropy weight TOPSIS models exhibit a significant positive correlation with the experts'. Furthermore, the factor analysis model outperforms the entropy weight TOPSIS model in predicting experts' evaluation scores. The application of this model will help achieve timely and accurate course evaluation, and provide novel idea and approach for real-time monitoring of online courses quality.

Based on a brief analysis of the current situation of university education management and research on intelligent algorithms, this article constructs a university education management system based on big data. For the clustering and prediction modules in higher education management, corresponding algorithms are used for optimization design. A fuzzy clustering algorithm based on entropy weight is proposed to address the shortcomings of the C-means clustering algorithm. This algorithm adds weighting coefficients on the basis of improvement and continuously updates the clustering centers. The prediction module uses Apriori algorithm to map and compress the target transaction database, reduces the number of candidate item sets through pruning process, and designs simulation experiments to measure the performance of the algorithm. The simulation results show that the clustering results of this algorithm are closer to the actual clustering situation, with shorter running time and better algorithm performance.

In order to promote the growth of contemporary music and the reform of music, this article designs an improved collaborative filtering (CF) algorithm to solve the problem of sparse matrix in traditional recommendation algorithms. The data matrix is dimensionally reduced to find the nearest neighbor, so as to realize personalized recommendation of music teaching resources in colleges and universities. The test results show that the accuracy of the proposed teaching resource recommendation algorithm is improved by 22.56% compared with the traditional CF algorithm. The improved CF algorithm can provide more accurate prediction, and the recommendation effect of the improved algorithm is better than the original algorithm, which can effectively avoid the sparse matrix problem faced by the CF algorithm, and provide technical support for the development of contemporary music discipline and the reform of music discipline.

The objective of this work is to predict the employment rate of students based on the information in the SSM (student status management) in colleges and universities. Firstly, the relevant content of SSM is introduced. Secondly, the BP (Back Propagation) neural network, the LM (Levenberg Marquardt) algorithm, and the BR (Bayesian Regularization) algorithm are introduced. In addition, the LM algorithm is combined with the BR algorithm to optimize the BP neural network, so as to establish a prediction model of the employment rate of college graduates based on the LM-BP neural network (the established prediction model). Finally, the established prediction model is verified after the historical data of college students in the SSM are managed and processed using the big data analysis technology. It suggests that the established prediction model shows higher prediction accuracy, more stable prediction performance, more ideal prediction effect, and higher practical application value.

This article introduces a novel method that integrates collaborative filtering into the naive Bayes model to enhance predicting student academic performance. The combined approach leverages collaborative user behavior analysis and probabilistic modeling, showing promising results in improved prediction precision. Collaborative Filtering explores user behavior patterns, while Naive Bayes employs Bayes' theorem for probabilistic data classification. Focused on predicting academic success, the integration incorporates collaborative patterns from student data for increased accuracy. The method considers similar students' performance and behavior for nuanced, personalized predictions. Starting with diverse data collection, including collaborative patterns among students, Collaborative Filtering identifies relationships and patterns among those with similar academic histories. These insights enrich the naive Bayes algorithm, creating a holistic approach for more accurate predictions, and contributing to ongoing machine learning initiatives in education.