A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf
.
Filters
Music Sequence Prediction with Mixture Hidden Markov Models
[article]
2018
arXiv
pre-print
In this paper, we propose a novel mixture hidden Markov model (HMM) for music play sequence prediction. ...
Nowadays, most music recommendation systems rely on item-based or user-based collaborative filtering or content-based approaches. ...
Bruno Ribeiro for interesting lectures and well-designed projects in Purdue CS 573, Spring 2018 [50] . This work was supported by a grant from the Goodata Foundation (GDF grant b5d-8d-564-f0). ...
arXiv:1809.00842v3
fatcat:h7f44ys6nffoli3qptxuo7l4z4
Interpretable Deep Generative Recommendation Models
2021
Journal of machine learning research
User preference modeling in recommendation system aims to improve customer experience through discovering users' intrinsic preference based on prior user behavior data. ...
The observed-level disentanglement on items is achieved by modeling the intra-user preference diversity in a prototype learning strategy, where different user intentions are captured by item groups (one ...
We are also grateful for the anonymous reviewers and the editor for their helpful comments. ...
dblp:journals/jmlr/LiuJWXWYN21
fatcat:tvc5ue7l5fh5ngpoe4acd34vcm
Analyzing User Preference for Social Image Recommendation
[article]
2016
arXiv
pre-print
In such a scenario, extensive amount of heterogeneous information such as tags, image content, in addition to the user-to-item preferences, is extremely valuable for making effective recommendations. ...
STM jointly considers the problem of image content analysis with the users' preferences on the basis of sparse representation. ...
Finally, the mAPS is reported with the mean of the APS scores for all target users. ...
arXiv:1604.07044v1
fatcat:md4unswkrzendeiffvedq3kjdq
Denoising User-aware Memory Network for Recommendation
[article]
2021
arXiv
pre-print
Meanwhile, the existing methods utilize item sequence for capturing the evolution of user interest. ...
For better user satisfaction and business effectiveness, more and more attention has been paid to the sequence-based recommendation system, which is used to infer the evolution of users' dynamic preferences ...
CTR Model Early methods of CTR prediction construct the interactive data of users and items into a user-item rating matrix, and the user-based or item-based CF method [8] is used for rating prediction ...
arXiv:2107.05474v1
fatcat:nfrvl5epi5fztm4r6qfqcqa6qu
Attentive Collaborative Filtering
2017
Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval - SIGIR '17
We argue that, in multimedia recommendation, there exists item-and component-level implicitness which blurs the underlying users' preferences. e item-level implicitness means that users' preferences on ...
However, the majority of existing collaborative ltering (CF) systems are not well-designed for multimedia recommendation, since they ignore the implicitness in users' interactions with multimedia content ...
ACKNOWLEDGMENTS We would like to thank the anonymous reviewers for their valuable comments. ...
doi:10.1145/3077136.3080797
dblp:conf/sigir/ChenZ0NLC17
fatcat:eruye76yufaphl3prsu6fwuunu
BayesMallows: An R Package for the Bayesian Mallows Model
2020
The R Journal
BayesMallows is an R package for analyzing preference data in the form of rankings with the Mallows rank model, and its finite mixture extension, in a Bayesian framework. ...
These posteriors are fully available to the user, and the package comes with convienient tools for summarizing and visualizing the posterior distributions. ...
Acknowledgments The authors would like to thank Arnoldo Frigessi and Elja Arjas for fruitful discussions.
Bibliography ...
doi:10.32614/rj-2020-026
fatcat:tbfj774wcbdzrb7zqo6e3rffpe
Outfit Generation and Style Extraction via Bidirectional LSTM and Autoencoder
[article]
2018
arXiv
pre-print
In a fashion item prediction task (missing prediction task), the proposed model outperformed a baseline method. ...
This capability allows us to generate fashionable outfits according to various preferences. ...
, the proposed model will be able to recommend items and out ts that re ect the preferred styles of individual users. ...
arXiv:1807.03133v3
fatcat:sjqdsva5mnaejnzytnmcibzfnu
Survey for Trust-aware Recommender Systems: A Deep Learning Perspective
[article]
2020
arXiv
pre-print
A significant remaining challenge for existing recommender systems is that users may not trust the recommender systems for either lack of explanation or inaccurate recommendation results. ...
filter untruthful noises (e.g., spammers and fake information) or enhance attack resistance; explainable recommender systems that provide explanations of recommended items. ...
Explanation on Visual Data Compared with textual data, visual contents often contain more information that can be exploited for recommendation explanations. ...
arXiv:2004.03774v2
fatcat:q7mehir7hbbzpemw3q5fkby5ty
Empirical and Experimental Perspectives on Big Data in Recommendation Systems: A Comprehensive Survey
[article]
2024
arXiv
pre-print
Algorithmic Methods, and Mathematical Modeling Methods, with further subdivisions into sub-categories and techniques. ...
Covering a wide range of algorithms, this taxonomy first categorizes algorithms into four main analysis types: User and Item Similarity-Based Methods, Hybrid and Combined Approaches, Deep Learning and ...
for new users or items with limited data through content-based filtering or hybrid models. ...
arXiv:2402.03368v1
fatcat:alwsy4cpljhyvj5rdvygzhda44
MMALFM
2019
ACM Transactions on Information Systems
Therefore, it is expected that the proposed method can model a user's preferences on an item more accurately for each user-item pair. ...
To this end, our model could alleviate the data sparsity problem and gain good interpretability for recommendation. ...
Different from them, in this work, item visual features are used together with text reviews to model aspect-aware users' preferences and items' properties for rating prediction. ...
doi:10.1145/3291060
fatcat:4352lpe7ybawdc5bwlj5p2ss7m
Choosing the Right Model: A Comprehensive Analysis of Outfit Recommendation Systems
2021
International Journal of Computer Applications
A presentation of these with examples for representative algorithms of each category, and analysis of their predictive performance and their ability to address the challenges is carried out. ...
The need for presenting Outfit recommendation models and the importance of their accuracy is presented. Finally, a comprehensive survey of 4 types of Fashion Recommendation Systems: 1.) ...
The storage of User-Item data is not done in the form of a matrix; it's done in the form of a linked list, where the head is the value of the user, and the remaining nodes make up for the items. ...
doi:10.5120/ijca2021921413
fatcat:ze6npdmhzba7zky5jva57cvyse
Collaborative Ensemble Learning: Combining Collaborative and Content-Based Information Filtering via Hierarchical Bayes
[article]
2012
arXiv
pre-print
For the Reuters-21578 text data set, we simulate user ratings under the assumption that each user is interested in only one category. ...
For both data sets, collaborative ensemble achieved excellent performance in terms of recommendation accuracy. ...
We will later use () to stand for the SVM preference model for user i, with () containing all SVM model parameters D:i,j and bi. ...
arXiv:1212.2508v1
fatcat:am7u3rru4vc7pormkjaoxp6yxe
Preference Modeling by Exploiting Latent Components of Ratings
[article]
2017
arXiv
pre-print
Finally, all latent factor models are combined linearly to estimate predictive ratings for users. ...
Uncovering the latent components of user ratings is thus of significant importance for learning user interests. ...
-Bayesian User Community Model (BUCM): Relied on both item selection and rating emission, the BUCM generate communities for users that experience the same items and individual user is modeled as mixture ...
arXiv:1710.07072v1
fatcat:msmpngmxbjd5dfn6aeh6ll5flu
Telepath: Understanding Users from a Human Vision Perspective in Large-Scale Recommender Systems
[article]
2017
arXiv
pre-print
From a human vision perspective, there're two key factors that affect users' behaviors: items' attractiveness and their matching degree with users' interests. ...
Its CNN subnetwork simulates the human vision system to extract key visual signals of items' attractiveness and generate corresponding activations. ...
In this case, there is an urgent need for an efficient way to model user behaviors. ...
arXiv:1709.00300v2
fatcat:j2enntxjhje5jlpqtqjr3d5cdi
Telepath: Understanding Users from a Human Vision Perspective in Large-Scale Recommender Systems
2018
PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE
From a human vision perspective, we found two key factors that affect users' behaviors: items' attractiveness and their matching degrees with users' interests. ...
For one of the major item recommendation blocks on the JD app, click-through rate (CTR), gross merchandise value (GMV) and orders have been increased 1.59%, 8.16% and 8.71% respectively by Telepath. ...
These methods map each item to a vector through the way that is assigning each item an initialized dense vector and then correcting these vectors with large scale training data by a certain optimization ...
doi:10.1609/aaai.v32i1.11243
fatcat:hd6v7ouxtfhgjmuikvwdw7ct4q
« Previous
Showing results 1 — 15 out of 21,412 results