Research on Personalized Recommendation Method for Web Front End Teaching Courses Based on Implicit Feedback

PurposeThe purpose of this paper is to plan, implement, and deliver universal design for learning (UDL) benefits to a large class of undergraduate biology students.Design/methodology/approachThis pilot project used an online course site in the course management system (CMS) to establish a universal design (UD) and maintain it throughout a semester‐long class in undergraduate biology. Although the class is traditionally taught in a face‐to‐face lecture and lab format, this intervention is used to expand the representational, strategic, and affective aspects of the course.FindingsSurvey results and follow‐up interviews as well as student usage statistics and points earned on tests are collected and analysed, with the resulting data used to generate conclusions relating to the benefits of a UDL‐compliant supplemental site versus a supplemental site that does not apply UDL principles. While the students perceived‐added value in the UDL‐enhanced site, the overall intervention does not lead to improved grades leading to the possibility that there may be a “sweet spot” or optimal blend of tools and approaches.Research limitations/implicationsFurther research will be needed in order to establish a more reliable mode of using supplemental online course sites to deliver UDL. Specifically, it will be important to expand this pilot case to a wider range of courses including those that include more diverse populations of students.Practical implicationsThe CMS represents a potential means of delivering UD benefits in an optimal and replicable manner. It will be important to expand this pilot effort to identify an optimal balance of supports and also to increase its generalizability by including a wider range of courses, as well as those that include more diverse populations of students.Originality/valueThis paper uses a potentially replicable strategy to deliver the benefits of UD to college students. If an optimal balance of supports can be identified, this supplemental use of an online course site in the CMS could serve to deliver academic and student satisfaction benefits to a wide range of students, including those in the K‐12 setting.

Farfetch is the leading platform for online luxury fashion shopping. We have more than 3000 brands and high-end designers with the biggest catalog of luxury products available worldwide to more than 1 million customers.

E-commerce platforms usually train their recommender system models to achieve personalized recommendations based on user behavior data. User behavior can be categorized into implicit and explicit feedback. Explicit feedback data have been well studied. However, the implicit feedback data still have many issues, such as the multiple types of behavior data, lack of negative feedback, and lack of the ability to express the real user preference. Targeting these problems of implicit feedback, we propose a TDF-WNSP-WLFM (time decay factor-weight of negative sample possibility-weighted latent factor model) based on the latent factor model for product recommendation. Our method mainly focuses on reconstructing the implicit rating matrix to enable the algorithm to perform better. The TDF-WNSP-WLFM algorithm is tested on two public user behavior datasets from Taobao and REES46, two big e-commerce platforms. Our algorithm compares favorably with other known collaborative filtering methods.

In the recommendation system, compared with the explicit feedback, implicit feedback has the characteristics of large quantity, easier tracking and easier access. Even in many practical application scenarios of the recommendation system, there is only implicit feedback without explicit feedback Based on implicit feedback to recommend there has been a problem, that is usually implicit feedback provides only reflect the noise signal of user preferences, because although the observed interactions at least reflect the user’s interest in the item, but not observed the interaction may be missing data, and is not necessarily negative samples, so it’s natural lack of negative feedback In this work, this paper improves the neural colla-borative filtering algorithm (NCF) by extracting the popularity characteristics as the basis for extracting negative samples from implicit feedback data. To a certain extent, it solves the problem that negative sampling introduces a lot of noise in previous work. At the same time, we found that the gradient vanishing in the training process when using the traditional loss function Binary Cross Entropy (BCE). In order to solve this problem, a new loss function, BCE-Max, was proposed in this paper. Based on the above improvements, we proposed a neural collaborative filtering recommendation algorithm (PNCF) based on popularity characteristics. By comparing the performance of PNCF, NCF, traditional collaborative filtering and some matrix factorization recommendation algorithms in the evaluation indexes such as HR@10, NDCG@10 and avg_popularity@10, we find that PNCF has certain advantages over these recommendation algorithms. Meanwhile, We found that compared with NCF, the experimental results of PNCF improved by about 1.2% and 2% onHR@10 and NDCG@10 respectively, and even improved by about 2% on HR@1 and NDCG@I.

Due to the inherent challenges in acquiring explicit feedback, graph collaborative filtering (GCF) models often resort to implicit feedback. However, there exists noise in implicit feedback that may not accurately reflect users’ preferences. This noise will be amplified by the aggregating and propagating operations of GCF, thereby affecting the performance of GCF. Existing noise mitigation methods attempt to filter noisy samples from implicit feedback data, yet they face limitations such as dependency on side information for sample selection, neglect of false-negative noise, and disregard for the impact of previous selections on current iteration. To overcome these challenges, we propose a <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">d</u>ynamic <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">g</u>raph <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">l</u>earning framework to <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">d</u>enoise implicit feedback for GCF (DGLD). DGLD comprises a graph learning module and a reinforcement learning module. The graph learning module evaluates the confidence degrees of interactions by leveraging user–item cosine similarity and global user preferences, updating user–item interaction graph without relying on side information. Meanwhile, the reinforcement learning module employs a policy network to select “false-positive denoising” and “false-negative denoising” actions based on performance of the recommendation model and state of interaction graph obtained from previous iterations. These modules work synergistically to dynamically denoise implicit feedback. Experimental results on three benchmark datasets underscore the superiority of our approach over state-of-the-art general and denoising recommendation models.

A novel context-aware recommendation algorithm with two-level SVD in social networks

Ranking algorithms in recommender systems influence people to make decisions. Conventional ranking algorithms based on implicit feedback data aim to maximize the utility to users by capturing users’ preferences over items. However, these utility-focused algorithms tend to cause fairness issues that require careful consideration in online platforms. Existing fairness-focused studies does not explicitly consider the problem of lacking negative feedback in implicit feedback data, while previous utility-focused methods ignore the importance of fairness in recommendations. To fill this gap, we propose a Generative Adversarial Networks (GANs) based learning algorithm FairGAN mapping the exposure fairness issue to the problem of negative preferences in implicit feedback data. FairGAN does not explicitly treat unobserved interactions as negative, but instead, adopts a novel fairness-aware learning strategy to dynamically generate fairness signals. This optimizes the search direction to make FairGAN capable of searching the space of the optimal ranking that can fairly allocate exposure to individual items while preserving users’ utilities as high as possible.

Participation in heart surgery requires procedural and factual knowledge and intensive preparation. There is evidence in the literature that multimedia driven learning has advantages in medical fields where an understanding of complex temporal and spatial events plays an important role. This work describes the development and evaluation of a multimedia driven, online teaching course on aortic valve replacement for students and residents. The instructional model followed a methodological approach with scalable information for different user groups. Various interactive multimedia development tools were employed. In a prospective study, 20 students and 10 residents were exposed to the program in a standardized environment. Both groups completed a 20-item multiple choice pre- and post-test. Psychometric evaluation with HILVE (Heidelberg inventory for the evaluation of teaching courses, 50 items) was performed. The multimedia course integrates more than 200 high quality surgical video and audio sequences, interactive 2 D and 3 D models, as well as illustrations and text. It is available at www.lamedica.de. Study time in the student group was 103 +/- 11 min and 70 +/- 10 min in the resident group. Mean number of correct responses to the knowledge pre-test was 6.23 +/- 2.71 in the student group and 12.2 +/- 2.66 in the resident group. Mean number of correct responses to the knowledge post-test was 15.24 +/- 2.06 in the student group ( p < 0.0001 vs. pre-test) and 17 +/- 2.98 in the resident group ( p = 0.004 vs. pre-test). The HILVE test showed positive results for teaching conditions, the program's instructional competence, student motivation and individual learning. Our data demonstrate that multimedia driven training can adapt to the individual needs of learners and improves procedural knowledge required for open heart surgery. Consequently, the whole course forms part of the training of residents and students.

The outlook on the roles and needs of the centre players in the cycle of production and delivery of e-learning materials is investigated. The perspectives of e-lecturers, e-tutors and e-students on use and design of multimedia learning objects are presented in this paper. Their various cultures, skill levels and learning styles necessitate integration of flexibility into learning environment. Flexible use of multimedia learning objects and how the scalable design principles can be applied in practice are demonstrated. This adaptable framework relies on the learning style and background knowledge of the tutor in selecting, organizing, and presenting learning material produced by e-lecturers. It adopts a flexible approach to educational content organization and delivery based on multimedia components that can be integrated into a wide range of courses. The flexible framework leverages XML metadata for describing multimedia learning objects and XSL for their seamless integration into the existing e-learning systems

The Malaysian education system offers a wide range of courses to school leavers in public and private universities. Medical course is one of the courses chosen by outstanding students. Several universities across the country offer such courses. This study is going to determine the most favorable comprehensive universities in Malaysia that offer medical courses and examine the ranking of criteria that have an impact on choosing universities. Specifically, this study applied fuzzy triangular numbers (TFN) and fuzzy analytic hierarchy process (FAHP) to multi-criteria decision-making (MCDM). A set of questionnaires was distributed to the students of matriculation session 2020/2021 which were from Module 1. Responses were analyzed using Microsoft Excel based on FAHP Chang’s model. FAHP shows the first rank university having the highest total weight vector. Universiti Malaysia Sabah (UMS) got the first rank based on the criteria chosen.

The Feminist Perspectives Series seeks to provide concise, accessible and engaging introductions to key feminist topics and debates. The texts in the series are designed to be used on a wide range of courses exploring feminist issues and are written by experienced teachers who are also well known in their respective fields. Each book in the series includes the most up-to-date statistics, research data, key sources and suggestions for further reading.Feminist Perspectives On Language provides an accessible introduction to this complex area. It redresses the balance of current feminist texts which tend to concentrate on discourse analysis and fail to connect with feminist thought in other disciplines such as sociology and politics. The text is divided into two parts, the first looks at language itself, how we learn language exploring such questions as; Does language free or trap us? Does our language affect how we come to understand the world around us? Is our language sexist? If so, does that reflect male dominance in society? and many more issues. Part Two explores questions of methodology and interpretation examining language in use, communication styles and the analysis of conversation.

The article reveals the peculiarities of the development of the educational movement in the camp of captured Ukrainian soldiers Freistadt (Austria-Hungary) in 1915-1917. It was established that despite all the initial difficulties (due to the opposition of the Black Hundreds) due to the systematic work of members of the Education Department, the Education Section Social and educational circle, the Society «Prosvita» M. Drahomanov during 1915-1917 it became possible to organize the teaching of a fairly wide range of courses and literacy schools (for the illiterate). The decisive role in the organization of educational life was initially played by the Union for the Liberation of Ukraine Educational Department, which included Ukrainian public figures, educators, cooperators with university or high school education, who had developed a national worldview. Through their efforts, the Ukrainian camp organization in Freistadt created appropriate structures from among the prisoners (educational section and the Enlightenment Society), which over time took over the main burden of educational work. Its content differed significantly – if at first the emphasis was on general education courses (so as not to provoke criticism of «politics» from prisoners), then within a year educational institutions paid more and more attention to Ukrainian studies (history of Ukraine, Ukrainian language and history of Ukrainian literature). At the same time, a large number of training hours were devoted to the elimination of illiteracy among prisoners. For many prisoners, education became the meaning of their camp life, and at the same time their national awareness took place, when they – sometimes even unnoticed – began to worry about the problems of Ukraine's self-affirmation in political terms and became adherents of its independence. Thus, on the one hand, educational organizations purposefully prepared prisoners for the future of civilian life in the homeland, and on the other - to educate them conscious and loyal citizens of free Ukraine.

Abstract: Germany is an attractive research destination for PhD students and young scientists from Europe and around the world. With its innovative research achievements, the country ranks among the top in the European Union. The country's higher education institutions offer a wide range of courses for young researchers, some of which can be studied in a foreign language. The main questions concern the regulations on doctoral studies in Germany at the level of the federal state and the constituent states on the basis of decentralised self-government, the procedure for admission to and completion of doctoral studies at the Free University of Berlin, and the principle of internationalisation at the same university, which are also the subject of the article. А study of the legal framework, strategic documents on doctoral degree and education is made in the article. The methods used are content analysis and secondary data analysis. Conclusions and generalizations are made and presented in summary form in tables. The article has a practical-applied character as it presents practices from Germany that may be useful in other countries and universities. Keywords: Germany, Free University Berlin, PhD study, young researchers, regulatory framework, research achievements, decentralization, internationalization

This paper reports on a pilot project carried out between September 1989 and June 1990 at a large college of Further and Higher Education. Using Interactive software based on Repertory Grid Techniques, the authors monitored the progress of a group of students undertaking theflrst year of the Diploma In Nursing (University of London). The primary focus was on how the students were reacting to the course as a whole and to Illuminate Individual difficulties and concerns, which could then be fed back to the course team. It was hoped that the complete process would provide a unique picture of how individual perceptions and responses changed throughout the year. The authors explain how the software was used, the data analysed, and how the feedback to students and tutors was managed. The prime purpose of the activity was to gain insights into how the technique could be most effectively used across a wide range of courses within the college and to determine whether it was feasible for students to use eac...

Continuing education needs of community nurses, midwives and health visitors for supervising and assessing students