Andreas Lommatzsch

The slashdot zoo: mining a social network with negative edges

We analyse the corpus of user relationships of the Slashdot technology news site. The data was collected from the Slashdot Zoo feature where users of the website can tag other users as friends and foes, providing positive and negative endorsements. We adapt social network analysis techniques to the problem of negative edge weights. In particular, we consider signed variants of global network characteristics such as the clustering coefficient, node-level characteristics such as centrality and popularity measures, and link-level characteristics such as distances and similarity measures. We evaluate these measures on the task of identifying unpopular users, as well as on the task of predicting the sign of links and show that the network exhibits multiplicative transitivity which allows algebraic methods based on matrix multiplication to be used. We compare our methods to traditional methods which are only suitable for positively weighted edges.

Learning spectral graph transformations for link prediction

We present a unified framework for learning link prediction and edge weight prediction functions in large networks, based on the transformation of a graphs algebraic spectrum. Our approach generalizes several graph kernels and dimensionality reduction methods and provides a method to estimate their parameters efficiently. We show how the parameters of these prediction functions can be learned by reducing the problem to a one-dimensional regression problem whose runtime only depends on the methods reduced rank and that can be inspected visually. We derive variants that apply to undirected, weighted, unweighted, unipartite and bipartite graphs. We evaluate our method experimentally using examples from social networks, collaborative filtering, trust networks, citation networks, authorship graphs and hyperlink networks.

Benchmarking News Recommendations in a Living Lab

Most user-centric studies of information access systems in literature suffer from unrealistic settings or limited numbers of users who participate in the study. In order to address this issue, the idea of a living lab has been promoted. Living labs allow us to evaluate research hypotheses using a large number of users who satisfy their information need in a real context. In this paper, we introduce a living lab on news recommendation in real time. The living lab has first been organized as News Recommendation Challenge at ACM RecSys’13 and then as campaign-style evaluation lab NEWSREEL at CLEF’14. Within this lab, researchers were asked to provide news article recommendations to millions of users in real time. Different from user studies which have been performed in a laboratory, these users are following their own agenda. Consequently, laboratory bias on their behavior can be neglected. We outline the living lab scenario and the experimental setup of the two benchmarking events. We argue that the living lab can serve as reference point for the implementation of living labs for the evaluation of information access systems.

Pattern recognition and classification for multivariate time series

Nowadays we are faced with fast growing and permanently evolving data, including social networks and sensor data recorded from smart phones or vehicles. Temporally evolving data brings a lot of new challenges to the data mining and machine learning community. This paper is concerned with the recognition of recurring patterns within multivariate time series, which capture the evolution of multiple parameters over a certain period of time. Our approach first separates a time series into segments that can be considered as situations, and then clusters the recognized segments into groups of similar context. The time series segmentation is established in a bottom-up manner according the correlation of the individual signals. Recognized segments are grouped in terms of statistical features using agglomerative hierarchical clustering. The proposed approach is evaluated on the basis of real-life sensor data from different vehicles recorded during car drives. According to our evaluation it is feasible to recognize recurring patterns in time series by means of bottom-up segmentation and hierarchical clustering.

Real-Time News Recommendation Using Context-Aware Ensembles

With the rapidly growing amount of items and news articles on the internet, recommender systems are one of the key technologies to cope with the information overload and to assist users in finding information matching the their individual preferences. News and domain-specific information portals are important knowledge sources on the Web frequently accessed by millions of users. In contrast to product recommender systems, news recommender systems must address additional challenges, e.g. short news article lifecycles, heterogonous user interests, strict time constraints, and context-dependent article relevance. Since news articles have only a short time to live, recommender models have to be continuously adapted, ensuring that the recommendations are always up-to-date, hampering the pre-computations of suggestions. In this paper we present our framework for providing real-time news recommendations. We discuss the implemented algorithms optimized for the news domain and present an approach for estimating the recommender performance. Based on our analysis we implement an agent-based recommender system, aggregation several different recommender strategies. We learn a context-aware delegation strategy, allowing us to select the best recommender algorithm for each request. The evaluation shows that the implemented framework outperforms traditional recommender approaches and allows us to adapt to the specific properties of the considered news portals and recommendation requests.

Alternative similarity functions for graph kernels

Given a bipartite graph of collaborative ratings, the task of recommendation and rating prediction can be modeled with graph kernels. We interpret these graph kernels as the inverted squared Euclidean distance in a space defined by the underlying graph and show that this inverted squared Euclidean similarity function can be replaced by other similarity functions. We evaluate several such similarity functions in the context of collaborative item recommendation and rating prediction, using the exponential diffusion kernel, the von Neumann kernel, and the random forest kernel as a basis. We find that the performance of graph kernels for these tasks can be increased by using these alternative similarity functions.

Towards reproducibility in recommender-systems research

Numerous recommendation approaches are in use today. However, comparing their effectiveness is a challenging task because evaluation results are rarely reproducible. In this article, we examine the challenge of reproducibility in recommender-system research. We conduct experiments using Plista’s news recommender system, and Docear’s research-paper recommender system. The experiments show that there are large discrepancies in the effectiveness of identical recommendation approaches in only slightly different scenarios, as well as large discrepancies for slightly different approaches in identical scenarios. For example, in one news-recommendation scenario, the performance of a content-based filtering approach was twice as high as the second-best approach, while in another scenario the same content-based filtering approach was the worst performing approach. We found several determinants that may contribute to the large discrepancies observed in recommendation effectiveness. Determinants we examined include user characteristics (gender and age), datasets, weighting schemes, the time at which recommendations were shown, and user-model size. Some of the determinants have interdependencies. For instance, the optimal size of an algorithms’ user model depended on users’ age. Since minor variations in approaches and scenarios can lead to significant changes in a recommendation approach’s performance, ensuring reproducibility of experimental results is difficult. We discuss these findings and conclude that to ensure reproducibility, the recommender-system community needs to (1) survey other research fields and learn from them, (2) find a common understanding of reproducibility, (3) identify and understand the determinants that affect reproducibility, (4) conduct more comprehensive experiments, (5) modernize publication practices, (6) foster the development and use of recommendation frameworks, and (7) establish best-practice guidelines for recommender-systems research.

Stream-Based Recommendations: Online and Offline Evaluation as a Service

Providing high-quality news recommendations is a challenging task because the set of potentially relevant news items changes continuously, the relevance of news highly depends on the context, and there are tight time constraints for computing recommendations. The CLEF NewsREEL challenge is a campaign-style evaluation lab allowing participants to evaluate and optimize news recommender algorithms online and offline. In this paper, we discuss the objectives and challenges of the NewsREEL lab. We motivate the metrics used for benchmarking the recommender algorithms and explain the challenge dataset. In addition, we introduce the evaluation framework that we have developed. The framework makes possible the reproducible evaluation of recommender algorithms for stream data, taking into account recommender precision as well as the technical complexity of the recommender algorithms.

Overview of NewsREEL’16: Multi-dimensional Evaluation of Real-Time Stream-Recommendation Algorithms

Successful news recommendation requires facing the challenges of dynamic item sets, contextual item relevance, and of fulfilling non-functional requirements, such as response time. The CLEF NewsREEL challenge is a campaign-style evaluation lab allowing participants to tackle news recommendation and to optimize and evaluate their recommender algorithms both online and offline. In this paper, we summarize the objectives and challenges of NewsREEL 2016. We cover two contrasting perspectives on the challenge: that of the operator (the business providing recommendations) and that of the challenge participant (the researchers developing recommender algorithms). In the intersection of these perspectives, new insights can be gained on how to effectively evaluate real-time stream recommendation algorithms.

Optimizing and Evaluating Stream-Based News Recommendation Algorithms

Recommender algorithms are powerful tools helping users to find interesting items in the overwhelming amount available data. Classic recommender algorithms are trained based on a huge set of user-item interactions collected in the past. Since the learning of models is computationally expensive, it is difficult to integrate new knowledge into the recommender models. With the growing importance of social networks, the huge amount of data generated by the real-time web e.g. news portals, micro-blogging services, and the ubiquity of personalized web portals stream-based recommender systems get in the focus of research. In this paper we develop algorithms tailored to the requirements of a web-based news recommendation scenario. The algorithms address the specific challenges of news recommendations, such as a context-dependent relevance of news items and the short item lifecycle forcing the recommender algorithms to continuously adapt to the set of news articles. In addition, the scenario is characterized by a huge amount of messages that must be processed per second and by tight time constraints resulting from the fact that news recommendations should be embedded into webpages without a delay. For evaluating and optimizing the recommender algorithms we implement an evaluation framework, allowing us analyzing and comparing different recommender algorithms in different contexts. We discuss the strength and weaknesses both according to recommendation precision and technical complexity. We show how the evaluation framework enables us finding the optimal recommender algorithm for a specific scenarios and contexts.