Michal Kopecky

Software Confederations - An Architecture for Agile Development in the Large

There are two important shifts in software development: from object-oriented monoliths to service orientation and from the rigid specification-development-delivery to agile development when the needs of users are better taken into account as it can be reacted on the changes of the needs even during the development. In the point of intersection of these movements is the use of software confederations. The concept of software confederation uses decomposition of problem and application into smaller parts (services) and at the same time it is expected very tight cooperation with users as it is usual in agile development. Software confederations can be developed incrementally, quite easily modified during the development and even after the delivery. During the development of software confederations legacy systems and third party products may be used what reduces development and learning effort and opens opportunity for development subcontraction.

Monotonization of User Preferences

We consider the problem of user-item recommendation as a multiuser instance ranking learning. A user-item preference is monotonizable if the learning can restrict to monotone models. A preference model is monotone if it is a monotone composition of rankings on domains of explanatory attributes (possibly describing user behavior, item content but also data aggregations). Target preference ordering of users on items is given by a preference indicator (e.g. purchase, rating).

Understanding Transparent and Complicated Users as Instances of Preference Learning for Recommender Systems

In this paper we are concerned with user understanding in content based recommendation. We assume having explicit ratings with time-stamps from each user. We integrate three different movie data sets, trying to avoid features specific for single data and try to be more generic. We use several metrics which were not used so far in the recommender systems domain. Besides classical rating approximation with RMSE and ratio of order agreement we study new metrics for predicting Next-k and at least 1-hit at Next-k. Using these Next-k and 1-hit we try to model display of our recommendation --- we can display k objects and hope to achieve at least one hit. We trace performance of our methods and metrics also as a distribution along each single user. We define transparent and complicated users with respect to number of methods which achieved at least one hit. We provide results of experiments with several combinations of methods, data sets and metrics along these three axes.

Relaxed Objects - Object Model for Context-Aware Applications

It is sometimes necessary to operate on data having different sets of attributes. It can be caused by the fact that the data come from different sources or that we collected different amount of information about different real-world entities. In such cases a unification of the data to be stored into tables causes either data loss or too frequent changes in the table/database layout. Therefore we propose a data model allowing effective handling of such data. The model, originally developed for storage and processing of non-graphic data in a temporal GIS, ensures that all procedures/functions/methods are performed only on objects having all data necessary for their successful completion. Such features enable easier creation of context-aware and context-adaptable applications.

Visual Similarity in Sign Language

In this paper, we present an algorithm for evaluating the similarity of pairs of signs used in sign language. The proposed algorithm is based on the visual similarity on the sign and compares sequences of static positions of the signing person. The algorithm does not rely on any particular way of obtaining necessary geometric form of signs. The usability of the proposed algorithm has already been tested in its pilot implementation in a dictionary between Czech and Czech sign language. Geometric data available for a 3D animated model was used to enhance the dictionary with the possibility of searching for signs in dictionary according to partial information obtained from the user as well as retrieval of visually similar signs for a currently selected one.

Basis Functions as Pivots in Space of Users Preferences

Our starting motivation is a user visiting an e-shop. E-shops usually offer conjunction of sharp filter conditions and one attribute ordering of results. We use a top-k query system where results are ordered by a multi-criterial monotone combination of soft filter conditions.

Objects through Relations

Object databases are usually referring to those data management products that are specifically designed for use with an object programming language and are very closely coupled with one or more object programming languages. Typically, the most of the recent OODBMSs use C++ and/or Smalltalk languages as their database programming language. The application programmer may access database objects directly using the database operations in the programming language, or may perform associative lookups of objects using the query language. Unfortunately, the latter includes many problems concerning both the design of such languages and their implementation. Two significant development lines are recently at disposal: ODMG-93 standard with its language OQL [Ca93] and the work proposed by the ANSI X3H2 group well-known as SQL3 [Me94] that stores objects in special kinds of columns in relations.