Content deleted Content added
link fix |
m Open access bot: doi, arxiv added to citation with #oabot. |
||
Line 54:
=== Hybrid MF ===
In recent years many other matrix factorization models have been developed to exploit the ever increasing amount and variety of available interaction data and use cases. Hybrid matrix factorization algorithms are capable of merging explicit and implicit interactions <ref name="Zhao16">{{cite journal |last1=Zhao |first1=Changwei |title=Hybrid Matrix Factorization for Recommender Systems in Social Networks |last2=Sun |first2=Suhuan |last3=Han |first3=Linqian |last4=Peng |first4=Qinke |journal=Neural Network World |date=2016 |volume=26 |issue=6 |pages=559–569 |doi=10.14311/NNW.2016.26.032|doi-access=free }}</ref> or both content and collaborative data <ref name="Zhou12">{{cite book |last1=Zhou |first1=Tinghui |last2=Shan |first2=Hanhuai |last3=Banerjee |first3=Arindam |last4=Sapiro |first4=Guillermo |title=Kernelized Probabilistic Matrix Factorization: Exploiting Graphs and Side Information |journal=Proceedings of the 2012 SIAM International Conference on Data Mining |date=26 April 2012 |pages=403–414 |doi=10.1137/1.9781611972825.35 |publisher=Society for Industrial and Applied Mathematics|isbn=978-1-61197-232-0 }}</ref><ref name="Adams10">{{cite arxiv |last1=Adams |first1=Ryan Prescott |last2=Dahl |first2=George E. |last3=Murray |first3=Iain |title=Incorporating Side Information in Probabilistic Matrix Factorization with Gaussian Processes 1003.4944|date=25 March 2010 |eprint=1003.4944|class=stat.ML }}</ref><ref name="Fang11">{{cite book |last1=Fang |first1=Yi |title=Proceedings of the 2nd International Workshop on Information Heterogeneity and Fusion in Recommender Systems - Het ''Rec'' '11 |last2=Si |first2=Luo |date=27 October 2011 |pages=65–69 |doi=10.1145/2039320.2039330 |publisher=ACM|chapter=Matrix co-factorization for recommendation with rich side information and implicit feedback |isbn=9781450310277 }}</ref>
===Deep-Learning MF===
In recent years a number of neural and deep-learning techniques have been proposed, some of which generalize traditional [[Matrix factorization (recommender systems)|Matrix factorization]] algorithms via a non-linear neural architecture <ref>{{cite journal |last1=He |first1=Xiangnan |last2=Liao |first2=Lizi |last3=Zhang |first3=Hanwang |last4=Nie |first4=Liqiang |last5=Hu |first5=Xia |last6=Chua |first6=Tat-Seng |title=Neural Collaborative Filtering |journal=Proceedings of the 26th International Conference on World Wide Web |date=2017 |pages=173–182 |doi=10.1145/3038912.3052569 |url=https://dl.acm.org/citation.cfm?id=3052569 |accessdate=16 October 2019 |publisher=International World Wide Web Conferences Steering Committee|isbn=9781450349130 |arxiv=1708.05031 }}</ref>.
While deep learning has been applied to many different scenarios: context-aware, sequence-aware, social tagging etc. its real effectiveness when used in a simple [[Collaborative filtering]] scenario has been put into question. A systematic analysis of publications applying deep learning or neural methods to the top-k recommendation problem, published in top conferences (SIGIR, KDD, WWW, RecSys), has shown that on average less than 40% of articles are reproducible, with as little as 14% in some conferences. Overall the study identifies 18 articles, only 7 of them could be reproduced and 6 of them could be outperformed by much older and simpler properly tuned baselines. The article also highlights a number of potential problems in today's research scholarship and calls for improved scientific practices in that area.<ref>{{cite journal |last1=Ferrari Dacrema |first1=Maurizio |last2=Cremonesi |first2=Paolo |last3=Jannach |first3=Dietmar |title=Are We Really Making Much Progress? A Worrying Analysis of Recent Neural Recommendation Approaches |journal=Proceedings of the 13th ACM Conference on Recommender Systems |date=2019 |pages=101–109 |doi=10.1145/3298689.3347058 |url=https://dl.acm.org/authorize?N684126
|accessdate=16 October 2019 |publisher=ACM|hdl=11311/1108996 |arxiv=1907.06902 }}</ref> Similar issues have been spotted also in sequence-aware recommender systems.<ref>{{cite journal |last1=Ludewig |first1=Malte |last2=Mauro |first2=Noemi |last3=Latifi |first3=Sara |last4=Jannach |first4=Dietmar |title=Performance Comparison of Neural and Non-neural Approaches to Session-based Recommendation |journal=Proceedings of the 13th ACM Conference on Recommender Systems |date=2019 |pages=462–466 |doi=10.1145/3298689.3347041 |url=https://dl.acm.org/citation.cfm?id=3347041 |accessdate=16 October 2019 |publisher=ACM|isbn=9781450362436 }}</ref>
==See also==
| |||