Stefan Holland

Preference Mining: A Novel Approach on Mining User Preferences for Personalized Applications

Advanced personalized e-applications require comprehensive knowledge about their user’s likes and dislikes in order to provide individual product recommendations, personal customer advice and custom-tailored product offers. In our approach we model such preferences as strict partial orders with “A is better than B” semantics, which has been proven to be very suitable in various e-applications. In this paper we present novel Preference Mining techniques for detecting strict partial order preferences in user log data. The main advantage of our approach is the semantic expressiveness of the Preference Mining results. Experimental evaluations prove the effectiveness and efficiency of our algorithms. Since the Preference Mining implementation uses sophisticated SQL statements to execute all data-intensive operations on database layer, our algorithms scale well even for large log data sets. With our approach personalized e-applications can gain valuable knowledge about their customers’ preferences, which is essential for a qualified customer service.

Situated Preferences and Preference Repositories for Personalized Database Applications

Advanced personalized web applications require a carefully dealing with their users’ wishes and preferences. Since such preferences do not always hold in general, personalized applications also have to consider the user’s current situation. In this paper we present a novel framework for modeling situations and situated preferences. Our approach consists of a general meta model for situations, which can be applied as foundation for situation models in a wide range of applications. Furthermore, an XML-based preference repository for the storage and management of situated preferences is developed. Long-term and situated preferences can easily be accessed with the preference repository interface. Particularly, preferences best-matching to a given situation can be queried. This approach allows web applications to react flexibly and personalized to the changing situations of their users.

Preference XPATH:A Query Language for E-Commerce

We present a new XML-based search technology that enables users to formulate complex customer or vendor preferences which typically occur within e-commerce applications. Preferences are modeled in a natural way by partial orders. Since our semantics of multi-attribute preferences implements the Pareto-optimality principle Preference XPATH queries avoid both the unwanted “emptyresult”-effect and the flooding-effect with lots of irrelevant query results. If perfect matches are not available best possible alternatives are found instead. We have extended the XML query language XPATH by the capability to formulate preferences as soft selection conditions. As our extensions are fully compatible with the XPATH standard both hard and soft selection conditions become now available to any XML-based e-commerce application. Several e-shopping examples show how easy and elegant it is to transform customer wishes into Preference XPATH queries. Our prototype implementation is smoothly integrated with the XML database system Tamino of Software AG. Moreover we show how Preference XPATH can be used within the XML query language QUILT. It even merges with XML style sheets (XSLT) and the XML pointer language (XPointer). Thus with Preference XPATH powerful personalized search engines and match-making processes for B2C and B2B can be implemented completely inside the XML framework.

The COSIMA prototype for multi-objective bargaining

Multi-objective bargaining is a challenging business model for B2C or B2B Internet applications. Our visual bargaining agent COSIMA builds on cooperative database technology like Preference SQL or Preference XPATH, which can efficiently implement \linebreak[4] multi-dimensional search engines with a Pareto-optimality semantics to find best alternative offers. COSIMA is a smart avatar with dynamically generated speech output. We have implemented several well-known bargaining strategies for a sample e-shop, where the customer can bargain on more attributes than just the price. The life-like appearance of COSIMA is supported by randomizing her bargaining tactics and by expressing visual emotions during bargaining. She can also learn and adapt dynamically to various bargaining preferences of the customers. The COSIMA prototype is carefully engineered by means of component-ware and IT-standards including SQL, XML and Java. This makes it scalable even to very large e-shop databases. Measurements show that this advanced user interface of COSIMA requires only a small overhead of a few seconds, including the bidding process. Thus real-time performance using state-of-the art software components and hardware is feasible. A preliminary user feedback has been very encouraging. COSIMA gave her debut to a large professional audience at the computer fair CeBIT 2002.

Design and implementation of COSIMA-a smart and speaking e-sales assistant

Presents a new cooperative user interface for e-shopping in B2C (business-to-consumer) e-commerce. COSIMA (COmparison Shopping with Interactive Meta-search Agents) is a smart and animated Internet avatar with synchronized dynamic voice output who assists customers through their e-shopping tours and advises them in the spirit of a real salesperson in the old economy. COSIMA benefits from a Preference SQL-based search engine that relies on on the Pareto principle, computing best-matching results to the customers wishes. This enables COSIMA to offer ideal hits or suitable alternatives with a single query, to evaluate these search results in terms of the stated preferences and to generate proper voice output, following the basic principles of sales psychology. As a sample application, we present a meta-search engine for comparison shopping. Since COSIMA constitutes an entire technological framework, it can easily be adjusted to other e-commerce environments by customizing the search engine, the avatar animations and the speech contexts.

COSIMA- your smart, speaking E-salesperson

We present a new cooperative user interface for e-shopping. COSIMA is an intelligent Internet avatar with dynamic voice output that assists customers through their e-shopping tours and advises them like a real salesperson. COSIMA uses a meta search engine based on Preference SQL, computing best matching results to the customers wishes. COSIMA can qualify these results and generates proper voice output. Our presentation shows COSIMA in action for comparison shopping.

Efficient and flexible Web access to art-historical image collections

Modern object-relational database systems are capable of managing multimedia data, e.g. image, video and audio. In this paper we study how such universal database systems can be used for implementing self-tuning and adaptive digital image archives. The presented framework is based on the idea that multimedia objects axe often stored redundantly to support broadest system access for diverse clients from heterogenous environments. There is a large variety of ways to store and deliver multimedia data in various formats. Partitioning the data into formats that are physically stored in the database and those that are converted into any delivery format on-demand poses a nontrivial optimization problem. We present a new optimization algorithm which has been implemented and evaluated with the commercial database system DB2 Universal Database from IBM. The work is part of the HERON project, a project committed to the creative use of computer science and technology to advance the study of arts and humanities.

Gaining Customer Preferences from E-Shopping Log-Files

Gaining customer preferences from e-shopping log-files / Torsten Ehm ... - In: E-finance / H. U. Buhl ... (Hrsg.). - Berlin u.a. : Springer, 2001. - S. 43-62

User preference mining techniques for personalized applications

Advanced personalized e-applications require comprehensive knowledge about their users’ likes and dislikes in order to provide individual product recommendations, personal customer advice, and custom-tailored product offers. In our approach we model such preferences as strict partial orders with “A is better than B” semantics, which has been proven to be very suitable in various e-applications. In this paper we present preference mining techniques for detecting strict partial order preferences in user log data. Real-life e-applications like online shops or financial services usually have large log data sets containing the transactions of their customers. Since the preference miner uses sophisticated SQL operations to execute all data intensive operations on database layer, our algorithms scale well even for such large log data sets. With preference mining personalized e-applications can gain valuable knowledge about their customers’ preferences, which can be applied for personalized product recommendations, individual customer service, or one-to-one marketing.