Improving Cold Music Recommendation through Hierarchical Audio Alignment

Collaborative and content-based filtering are the majorly used approaches towards implementing recommender systems which can successfully predict the item(s) to be recommended to users based upon their preferences. Each of the methods has its own advantages and disadvantages that are specific to the situation they are used in. The hybrid approach aims to combine both techniques in multiple ways to optimize results and overcome shortcomings. In this paper, we propose a hybrid approach based recommendation engine to improve accuracy. While content based recommendations are made on the similarity of items’ attributes, collaborative recommendations are based on similarity of users. Collaborative filtering is achieved by matrix factorization technique. One of the most effective algorithms for matrix factorization: Singular Value Decomposition (SVD) has been used to perform collaborative filtering and combined with content based model to predict the item ratings per user. The paper aims to achieve the final result with minimum possible errors as per user preferences and recommended items.

This paper presents a prototype for a Web application recommendation system’s content applied to movies’ recommendations. It learns the pattern of user content consumption, predicting what he will consume in future based on similar items to those he has shown interest. It considers similarity with neighbor users, thus creating a user model. Content-based filtering, collaborative filtering, and memory-based on hybrid filtering techniques are used. Content-based filtering allows to extract the fundamental features or attributes of the items and select similar items. Moreover, it proposes predicted classifications for the items of interest not yet classified by the active user. Collaborative filtering allows applying the KNN methodology to identify the similarity between the active user located in the neighborhood and propose predicted classifications for items of interest not yet classified. Hybrid filtering combines the two methodologies to overcome their drawbacks. A weighted approach is applied, allowing a dynamic linear combination of collaborative and content-based filtering. The results obtained were empirically relevant in the experimental evaluation, matching with the results presented in similar studies validated with RMSE metrics.KeywordsInformation systemsRecommender systemsHuman centered computingUser modelsWeb-based interaction

The rating matrix of a personalized recommendation system contains a high percentage of unknown rating scores which lowers the quality of the prediction. Besides, during data streaming into memory, some rating scores are misplaced from its appropriate cell in the rating matrix which also decrease the quality of the prediction. The singular value decomposition algorithm predicts the unknown rating scores based on the relation between the implicit feedback of both users and items, but exploiting neither the user similarity nor item similarity which leads to low accuracy predictions. There are several factorization methods used in improving the prediction performance of the collaborative filtering technique such as baseline, matrix factorization, neighbour-base. However, the prediction performance of the collaborative filtering using factorization methods is still low while baseline and neighbours-base have limitations in terms of over fitting. Therefore, this paper proposes Ensemble Divide and Conquer (EDC) approach for solving 2 main problems which are the data sparsity and the rating scores’ deviation (misplace). The EDC approach is founded by the Singular Value Decomposition (SVD) algorithm which extracts the relationship between the latent feedback of users and the latent feedback of the items. Furthermore, this paper addresses the scale of rating scores as a sub problem which effect on the rank approximation among the users’ features. The latent feedback of the users and items are also SVD factors. The results using the EDC approach are more accurate than collaborative filtering and existing methods of matrix factorization namely SVD, baseline, matrix factorization and neighbours-base. This indicates the significance of the latent feedback of both users and items against the different factorization features in improving the prediction accuracy of the collaborative filtering technique.

Purpose – Project ATHENA aims to develop an application to address information overload, primarily focused on Recommendation Systems (RSs) with the personalization and user experience design of a modern system. Method – Two machine learning (ML) algorithms were used: (1) TF-IDF for Content-based filtering (CBF); (2) Classification with Matrix Factorization- Singular Value Decomposition (SVD) applied with Collaborative filtering (CF) and mean (normalization) for prediction accuracy of the CF. Data sampling in academic Research and Development (R&D) of Philippine Council for Agriculture, Aquatic, and Natural Resources Research and Development (PCAARRD) e-Library and Project SARAI publications plus simulated data used as training sets to generate a recommendation of items that uses the three RS filtering (CF, CBF, and personalized version of item recommendations). Series of Testing and TAM performed and discussed. Results – Findings allow users to engage in online information and quickly evaluate retrieved items produced by the application. Compatibility-testing (CoT) shows the application is compatible with all major browsers and mobile-friendly. Performance-testing (PT) recommended v-parameter specs and TAM evaluations results indicate strongly associated with overall positive feedback, thoroughly enough to address the information-overload problem as the core of the paper. Conclusion – A modular architecture presented addressing the information overload, primarily focused on RSs with the personalization and design of modern systems. Developers utilized Two ML algorithms and prototyped a simplified version of the architecture. Series of testing (CoT and PT) and evaluations with TAM were performed and discussed. Project ATHENA added a UX feature design of a modern system. Recommendations – High-end hardware specs of v-Parameter are recommended, at least 8-cores of vCPU with 16GiB of memory and sufficient driver-type size to run the model and End-to-end jobs execution for continuously maintaining the application. Research Implications – Future Developers must use/integrate a large dataset. Other ML approaches can expand Hybrid-RS better. Implicit data and additional filtering methods may enhance the application in the future.

Memory-based collaborative filtering (CF) aims at predicting the rating of a certain item for a particular user based on the previous ratings from similar users and/or similar items. Previous studies in finding similar users and items have several drawbacks. First, they are based on user-defined similarity measurements, such as Pearson Correlation Coefficient (PCC) or Vector Space Similarity (VSS), which are, for the most part, not adaptive and optimized for specific applications and data. Second, these similarity measures are restricted to symmetric ones such that the similarity between A and B is the same as that for B and A, although symmetry may not always hold in many real world applications. Third, they typically treat the similarity functions between users and functions between items separately. However, in reality, the similarities between users and between items are inter-related. In this paper, we propose a novel unified model for users and items, known as Similarity Learning based Collaborative Filtering (SLCF) , based on a novel adaptive bidirectional asymmetric similarity measurement. Our proposed model automatically learns asymmetric similarities between users and items at the same time through matrix factorization. Theoretical analysis shows that our model is a novel generalization of singular value decomposition (SVD). We show that, once the similarity relation is learned, it can be used flexibly in many ways for rating prediction. To take full advantage of the model, we propose several strategies to make the best use of the proposed similarity function for rating prediction. The similarity can be used either to improve the memory-based approaches or directly in a model based CF approaches. In addition, we also propose an online version of the rating prediction method to incorporate new users and new items. We evaluate SLCF using three benchmark datasets, including MovieLens, EachMovie and Netflix, through which we show that our methods can outperform many state-of-the-art baselines.

Recommender systems are used to generate meaningful recommendations to users based on their preferences, which will be determined following several approaches. This work targets Arab readers by providing accurate and reliable results that match their needs and desirability. Eventually, it will enhance the reading experience for any Arab readers. The main approach is to filter the recommendations, and this can be achieved either by Content-Based filtering or by Collaborative Filtering. The collaborative filtering techniques presented in this paper compute the similarity matrix between items and users' ratings, and then evaluate the recommendations for users. The techniques cover User-Based and Item-Based Collaborative Filtering, as well as Matrix Factorization through an SVD algorithm. A comparison between these techniques is presented in terms of the fitting and testing time, and accuracy. The KNN-based algorithms showed better performance than the matrix factorization method with respect to fitting and testing time. However, the matrix factorization (SVD) algorithm had the best results in terms of accuracy.

Day to day, the uploading of data into the world wide web and e-commerce directed the development of recommender systems.Recommender system filters the information based on the user's interest.Nowadays, recommender systems are being used in every domain.The advantage of a recommender system is that it makes searching easy.Recommender systems are classified into content-based filtering, collaborative filtering and hybrid approach.In this paper, we analysed the performance of item similarity, matrix factorisation and popular recommender algorithms and evaluated with precision-recall and root mean square error metrics.

When we buy items in online stores, it is common to face recommended items that meet our interest. These recommendation system help users not only to find out related items, but also find new things that may interest users. Recommendation system has been widely studied and various models has been suggested such as, collaborative filtering and content-based filtering. Though collaborative filtering shows good performance for predicting users preference, there are some conditions where collaborative filtering cannot be applied. Sparsity in user data causes problems in comparing users. Systems which are newly starting or companies having small number of users are also hard to apply collaborative filtering. Content-based filtering should be used to support this conditions, but content-based filtering has some drawbacks and weakness which are tendency of recommending similar items, and keeping history of a user makes recommendation simple and not able to follow up users preference changes. To overcome this drawbacks and limitations, we suggest weighted window assisted user history based recommendation system, which captures user’s purchase patterns and applies them to window weight adjustment. The system is capable of following current preference of a user, removing useless recommendation and suggesting items which cannot be simply found by users. To examine the performance under user and data sparsity environment, we applied data from start-up trading company. Through the experiments, we evaluate the operation of the proposed recommendation system.

Matrix factorization in Cross-domain Recommendations Framework by Shared Users Latent Factors

The Netflix Prize (NP) competition gave much attention to collaborative filtering (CF) approaches. Matrix factorization (MF) based CF approaches assign low dimensional feature vectors to users and items. We link CF and content-based filtering (CBF) by finding a linear transformation that transforms user or item descriptions so that they are as close as possible to the feature vectors generated by MF for CF. We propose methods for explicit feedback that are able to handle 140,000 features when feature vectors are very sparse. With movie metadata collected for the NP movies we show that the prediction performance of the methods is comparable to that of CF, and can be used to predict user preferences on new movies. We also investigate the value of movie metadata compared to movie ratings in regards of predictive power. We compare our solely CBF approach with a simple baseline rating-based predictor. We show that even 10 ratings of a new movie are more valuable than its metadata for predicting user ratings.

Cooking your own food at home is a good step toward reducing fast food consumption. Fast food increases the risk of dangerous diseases. The diversity of recipe information available on the internet makes it difficult to choose recipes that match user preferences. Mobile technology can help with this by recommending recipes that better suit users' eating habits. This makes the transition to a healthier diet easier. Therefore, in this study, a recommendation system was developed that can recommend recipes based on the preferences of Android users. Two main recommendation methods are used in this study: content-based filtering and collaborative filtering. Using cosine similarity, a content-based recommendation system identifies the proximity between a recipe for food and its related context. The history of user comments on recipes serves as implicit feedback for the collaborative recommendation algorithm. This eliminates the need for explicit evaluations, such as ratings. This recommendation system generates recommendations in the form of the top ten food recipes with an evaluation matrix, referred to as NDCG@k and Hit-Ratio@k. The tests revealed that a content-based filtering technique may produce helpful recommendations, with the highest similarity score of 0.41 for the entry "chocolate cake that you can easily make at home." Meanwhile, in the collaborative filtering method using the Neural Collaborative Filtering (NCF) approach, the system shows consistent performance improvements, with the MAP@10 value increasing from 0.705 to 0.767 and the NDCG@10 from 0.78 to 0.83 after 10 training epochs. Keywords: Recommendation systems; content-based filtering; neural collaborative filtering; cosine similarity; implicit feedback

Item-based recommender systems suggest products based on the similarities between items computed either from past user preferences (collaborative filtering) or from item content features (content-based filtering). Collaborative filtering has been proven to outperform content-based filtering in a variety of scenarios. However, in item cold-start, collaborative filtering cannot be used directly since past user interactions are not available for the newly added items. Hence, content-based filtering is usually the only viable option left.

This article presents the Goal-based Framework for providing personalized similarities between multi users profile preferences in formal e-Learning scenarios. It consists of two main approaches: content-based filtering and collaborative filtering because only traditional content-based filtering is not sufficient to generate the recommendations for new-users / learners. Therefore, the proposed work hybridized multi users collaborative filtering functionalities with personalized content-based profile preferences filtering. The main purpose of this proposed work is to (a) overcome the user-based cold-start profile recommendations and (b) improve the recommendations accuracy for new-users in formal e-learning recommendation systems. The experimental results of proposed Goal-based framework are tackled by using famous `MovieLens' dataset while the evaluation of experimental results have been performed with precision mean and recall mean to test the effectiveness of goal-based recommendation framework. Experimental results (precision mean: 76.284% and recall mean: 82.413%) show that the proposed framework goals performed well for the improvement of user-based cold-start issue as well as for content-based profile recommendations, using multi users personalized collaborative similarities, in formal e-Learning scenarios effectively.

A Personalized Recommender System from Probabilistic Relational Model and Users’ Preferences

The popularity of movies has increased in recent years. There are thousands of films produced each year. These films make it challenging for movie lovers to pick the ideal film to see. We propose a recommendation system that strives to offer guidance in selecting films. Depending on the method employed, recommendation systems can be categorized into three groups: collaborative filtering, content-based filtering, and hybrid filtering. In this work, collaborative filtering, one of the methods frequently used in recommendation systems was used. There are two ways to the Collaborative Filtering approach: User-Based Collaborative Filtering (UBCF) and Item-Based Collaborative Filtering (IBCF). There are two methods for finding similar items or users: Cosine and Pearson similarities. The Cosine similarity approach is one way to determine how similar two items are. Additionally, the Pearson Correlation Coefficient approach, which determines similarities between objects by calculating linear correlations between two sets, is the most widely employed. This study aims to determine which system produces the highest item similarity in IBCF and predicted ratings to actual ratings using 90% training and 10% testing data. The data set taken from MovieLens.org consists of 943 users from 1664 movies with 99392 ratings. The MovieLens data collection will be analyzed with the RStudio and the R package recommenderlab. The results reveal that the IBCF with Cosine similarities shows the number of items recommended n top-rated movies to each user for 10 movies. The IBCF can identify the most recommended films and creates a frequency distribution of items.