Lars Schmidt-Thieme

Lars Schmidt-Thieme

Auto-generated profile, based on publicly available information

Lars Schmidt-Thieme

Information Systems and Machine Learning Lab (ISMLL), University of Hildesheim, Hildesheim, Germany

Research Focus & Metrics

Academic Career Length

27Years (2000-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 April 2010
Verified
Conference Paper
Open access
PDF file
SRSteffen Rendle
Steffen Rendle
CFChristoph Freudenthaler
Christoph Freudenthaler
LSLars Schmidt-Thieme
Lars Schmidt-Thieme
Recommender systems are an important component of many websites. Two of the most popular approaches are based on matrix factorization (MF) and Markov chains (MC). MF methods learn the general taste of a user by factorizing the matrix over observed user-item preferences. On the other hand, MC methods model sequential behavior by learning a transition graph over items that is used to predict the next action based on the recent actions of a user. In this paper, we present a method bringing both approaches together. Our method is based on personalized transition graphs over underlying Markov chains. That means for each user an own transition matrix is learned - thus in total the method uses a transition cube. As the observations for estimating the transitions are usually very limited, our method factorizes the transition cube with a pairwise interaction model which is a special case of the Tucker Decomposition. We show that our factorized personalized MC (FPMC) model subsumes both a common Markov chain and the normal matrix factorization model. For learning the model parameters, we introduce an adaption of the Bayesian Personalized Ranking (BPR) framework for sequential basket data. Empirically, we show that our FPMC model outperforms both the common matrix factorization and the unpersonalized MC model both learned with and without factorization.

4 February 2010
Verified
Conference Paper
Open access
PDF file
SRSteffen Rendle
Steffen Rendle
LSLars Schmidt-Thieme
Lars Schmidt-Thieme

24 July 2011
Verified
Conference Paper
Open access
PDF file
SRSteffen Rendle
Steffen Rendle
ZGZeno Gantner
Zeno Gantner
CFChristoph Freudenthaler
Christoph Freudenthaler
LSLars Schmidt-Thieme
Lars Schmidt-Thieme

24 August 2014
Verified
Conference Paper
Open access
PDF file
JGJosif Grabocka
Josif Grabocka
NSNicolas Schilling
Nicolas Schilling
MWMartin Wistuba
Martin Wistuba
LSLars Schmidt-Thieme
Lars Schmidt-Thieme

9 May 2012 in arXiv
Verified
Preprint
Open access
PDF file
SRSteffen Rendle
Steffen Rendle
CFChristoph Freudenthaler
Christoph Freudenthaler
ZGZeno Gantner
Zeno Gantner
LSLars Schmidt-Thieme
Lars Schmidt-Thieme
Item recommendation is the task of predicting a personalized ranking on a set of items (e.g. websites, movies, products). In this paper, we investigate the most common scenario with implicit feedback (e.g. clicks, purchases). There are many methods for item recommendation from implicit feedback like matrix factorization (MF) or adaptive knearest-neighbor (kNN). Even though these methods are designed for the item prediction task of personalized ranking, none of them is directly optimized for ranking. In this paper we present a generic optimization criterion BPR-Opt for personalized ranking that is the maximum posterior estimator derived from a Bayesian analysis of the problem. We also provide a generic learning algorithm for optimizing models with respect to BPR-Opt. The learning method is based on stochastic gradient descent with bootstrap sampling. We show how to apply our method to two state-of-the-art recommender models: matrix factorization and adaptive kNN. Our experiments indicate that for the task of personalized ranking our optimization method outperforms the standard learning techniques for MF and kNN. The results show the importance of optimizing models for the right criterion.

1 July 2010
Verified
Conference Paper
Closed access
NTNguyen Thai-Nghe
Nguyen Thai-Nghe
ZGZeno Gantner
Zeno Gantner
LSLars Schmidt-Thieme
Lars Schmidt-Thieme
Class imbalance is one of the challenging problems for machine learning algorithms. When learning from highly imbalanced data, most classifiers are overwhelmed by the majority class examples, so the false negative rate is always high. Although researchers have introduced many methods to deal with this problem, including resampling techniques and cost-sensitive learning (CSL), most of them focus on either of these techniques. This study presents two empirical methods that deal with class imbalance using both resampling and CSL. The first method combines and compares several sampling techniques with CSL using support vector machines (SVM). The second method proposes using CSL by optimizing the cost ratio (cost matrix) locally. Our experimental results on 18 imbalanced datasets from the UCI repository show that the first method can reduce the misclassification costs, and the second method can improve the classifier performance.