Markus Franke

Scheduling of electrical household appliances with price signals

Due to the increasing competition in liberalized electricity markets, a succesful customer retention as well as a cost efficient allocation of electric energy become more and more important. Therefore, new, innovative strategies are sought, which promise on the one hand a long-term customer retention and assure, on the other hand, a more cost-efficient provision of electric energy.

On the Analysis of Irregular Stock Market Trading Behavior

In this paper, we analyze the trading behavior of users in an experimental stock market with a special emphasis on irregularities within the set of regular trading operations. To this end the market is represented as a graph of traders that are connected by their transactions. Our analysis is executed from two perspectives: On a micro scale view fraudulent transactions between traders are introduced and described in terms of the patterns they typically produce in the market’s graph representation. On a macro scale, we use a spectral clustering method based on the eigensystem of complex Hermitian adjacency matrices to characterize the trading behavior of the traders and thus characterize the market. Thereby, we can show the gap between the formal definition of the market and the actual behavior within the market where deviations from the allowed trading behavior can be made visible. These questions are for instance relevant with respect to the forecast efficiency of experimental stock markets since manipulations tend to decrease the precision of the market’s results. To demonstrate this we show some results of the analysis of a political stock market that was set up for the 2006 state parliament elections in Baden-Wuerttemberg, Germany.

An update algorithm for restricted random walk clustering for dynamic data sets

In this article, we present a randomized dynamic cluster algorithm for large data sets. It is based on the restricted random walk cluster algorithm by Schöll and Schöll-Paschinger that has given good results in past studies. We discuss different approaches for the clustering of dynamic data sets. In contrast to most of these methods, dynamic restricted random walk clustering is also efficient for a small percentage of changes in the data set and has the additional advantage that the updates asymptotically produce the same clusters as a reclustering with the static variant; there is thus no need for any reclustering ever. In addition, the method has a relatively low computational complexity which enables it to cluster large data sets.

Recommender Services in Scientific Digital Libraries

In this article we give a survey of the current practice and state-of-the-art of recommender services in scientific digital libraries. With the notable exception of amazon.com and CiteSeer which do not qualify as proper scientific libraries our survey revealed that in scientific libraries recommender services are still not in wide use — despite the considerable benefits they offer for students and scientists. This fact can at least partially be explained by mechanism design problems which exist for the basic types of recommender systems and decreased funding for scientific libraries. Next, we present the principles of four recommender services developed at the Universitat Karlsruhe (TH), namely the explicit review and rating service of the library of the Universitat Karlsruhe (TH), the implicit basic “Others also searched …” service (BibTip) of the library of the Universitat Karlsruhe (TH), the prototypes of its small sample and its adaptive variant. A discussion of the current industry trend towards social spaces and societies and its potential for scientific digital libraries concludes this contribution.

Future power markets

In this article we discuss the impact of the increasingly higher dynamics and transparency on future electricity markets due to a higher degree of automation along the energy supply chain. As part of the SESAM (Self Organization and Spontaneity in Liberalized and Harmonized Markets) project, we analyze the consequences of the customers’ higher willingness to change their supplier. In addition, we evaluate the major changes in load profiles caused by the use of load management systems in combination with dynamic retail electricity prices. We demonstrate that not only the consumer load but also the application of decentralized combined heat and power plants can be rescheduled. Finally, we outline further research activities.

Clustering of Large Document Sets with Restricted Random Walks on Usage Histories

Due to their time complexity, conventional clustering methods often cannot cope with large data sets like bibliographic data in a scientific library. We will present a method for clustering library documents according to usage histories that is based on the exploration of object sets using restricted random walks.

USING RESTRICTED RANDOM WALKS FOR LIBRARY RECOMMENDATIONS AND KNOWLEDGE SPACE EXPLORATION

Implicit recommender systems provide a valuable aid to customers browsing through library corpora. We present a method to realize such a recommender especially for, but not limited to, libraries. The method is cluster-based, scales well for large collections, and produces recommendations of good quality. The approach is based on using session histories of visitors of the librarys online catalog in order to generate a hierarchy of nondisjunctive clusters. Depending on the users needs, the clusters at different levels of the hierarchy can be employed as recommendations. Using the prototype of a user interface we show that, if, for instance, the user is willing to sacrifice some precision in order to gain a higher number of documents during a specific session, he or she can do so easily by adjusting the cluster level via a slider.

Automated Indexing with Restricted Random Walks on Large Document Sets

We propose a method based on restricted random walk clustering as a (semi-)automated complement for the tedious, error-prone and expensive task of manual indexing in a scientific library. The first stage of our method is to cluster a set of (partially) indexed documents using restricted random walks on usage histories in order to find groups of similar documents. In the second stage, we derive possible keywords for documents without indexing information from the frequencies of keywords assigned to other documents in their respective cluster.

Building Recommendations from Random Walks on Library OPAC Usage Data

In this contribution we describe a new way of building a recommender service based on OPAC web-usage histories. The service is based on a clustering approach with restricted random walks. This algorithm has some properties of single linkage clustering and suffers from the same deficiency, namely bridging. By introducing the idea of a walk context (see Franke and Thede (2005) and Franke and Geyer-Schulz (2004)) the bridging effect can be considerably reduced and small clusters suitable as recommendations are produced. The resulting clustering algorithm scales well for the large data sets in library networks. It complements behavior-based recommender services by supporting the exploration of the revealed semantic net of a library network’s documents and it offers the user the choice of the trade-off between precision and recall. The architecture of the behavior-based system is described in Geyer-Schulz et al. (2003).

On the Analysis of Asymmetric Directed Communication Structures in Electronic Election Markets

In this article we introduce a new general method of representing trading structures as complex adjacency matrices and transforming these into Hermitian adjacency matrices which are linear self-adjoint operators in a Hilbert space. The main advantages of the method are that no information is lost, no arbitrary decision on metrics is involved, and that all eigenvalues are real and, therefore, easily interpretable. The analysis of the resulting eigensystem helps in the detection of substructures and general patterns. While this approach is general, we apply the method in the context of analyzing market structure and behaviour based on the eigensystem of market transaction data and we demonstrate the method by analyzing the results of a political stock exchange for the 2002 federal elections in Germany.