In e-business, recommender systems have been instrumental in guiding users through their online experiences However, these systems are often limited by the lack of labels data and data sparsity Increasingly, data-mining techniques are utilized to address this issue In most research, recommendations to be made are achieved via supervised learning that typically employs the k-nearest neighbor learner However, supervised learning relies on labeled data, which may not be available at the time of learning Data sparsity, which refers to situations where the number of items that have been recommended represents only a small subset of all available items, further affects model performance One suggested solution is to apply cluster analysis as a preprocessing step and thus guide the learning process from natural grouping, typically using similar customer profiles, to improve predictive accuracy In this paper, we study the benefits of applying cluster analysis as a preprocessing step prior to constructing classification models Our HCC-Learn framework combines content-based analysis in the preprocessing stage and collaborative filtering in the final prediction stage Our results show the value of our HCC-Learn framework applied to real-world data sets, especially when combining soft clustering and ensembles based on feature subspaces

Beyond k-NN: Combining Cluster Analysis and Classification for Recommender Systems.

HCC-Learn Framework for Hybrid Learning in Recommender Systems.

Recommendation systems, which are employed to mitigate the information overload e-commerce users face, have succeeded in aiding customers during their online shopping experience. However, to be able to make accurate recommendations, these systems require information about the items for sale and about users’ individual preferences. Making recommendations to new customers, who have no prior data in the system, is therefore challenging. This scenario, called the “cold-start problem,” hinders the accuracy of recommendations made to a new user. In this paper, we introduce the popular users personalized predictions (PUPP-DA) framework to address cold starts. Soft clustering and active learning are used to accurately recommend items to new users in this framework. Additionally, we employ deep learning algorithms to improve the overall predictive accuracy. Experimental evaluation shows that the PUPP-DA framework results in high performance and accurate predictions. Further, focusing on frequent, or so-called popular, users during our active-learning stage clearly benefits the learning process.

Active Learning and Deep Learning for the Cold-Start Problem in Recommendation System: A Comparative Study

The contribution of technological enhancement has played a vital role in strengthening the country’s economy. The Higher Education lnstitutions (HEI) are providing placement assistance to the students by employing various measures as a part of their curriculum. A few of the attempts that HEIs have adopted to enhance soft skills as well as hard skills of the students are, course structure modification, student-mentoring, industry-academia collaboration etc. In spite of all the noble efforts, many unemployed engineers are witnessed in the society. Therefore, this research aims to develop an effective and accurate framework that would be highly beneficial for both students as well as teachers in HEIs. A Semi-Supervised-Sequence-Prediction (SSSP) machine learning framework has been designed to predict students’ academic achievement towards the completion of a four-year degree course in an outcome-based education system. A time series analysis of students’ data has been conducted to analyze the progression of final cumulative GPA and employability prospect in sequential manner. The results show the effectiveness of the proposed framework in predicting employability of students along with the areas where students need improvement in order to be employable. This research provides an inclusive roadmap, empowering the use of machine learning for employability and skill development in outcome-based education.

Sustainable Development of Employability and Skill Development in Outcome-based Education: A Machine Learning Approach

/pdf/towards-personalized-recommendation-systems-domain-driven-3hz43wqyip.pdf

Beyond k-NN: Combining Cluster Analysis and Classification for Recommender Systems.

Citations

HCC-Learn Framework for Hybrid Learning in Recommender Systems.

Active Learning and Deep Learning for the Cold-Start Problem in Recommendation System: A Comparative Study

Sustainable Development of Employability and Skill Development in Outcome-based Education: A Machine Learning Approach

Towards Personalized Recommendation Systems: Domain-Driven Machine Learning Techniques and Frameworks

Related Papers (5)

Recommender system using fuzzy c-means clustering and genetic algorithm based weighted similarity measure

Clustering Method of Weighted Preference Using K-means Algorithm and Bayesian Network for Recommender System

Design of an Unsupervised Machine Learning-Based Movie Recommender System

Probabilistic Unsupervised Machine Learning Approach for a Similar Image Recommender System for E-Commerce

Clustering Algorithms: a Review