Gábor Szűcs

The goal of this paper is to find a solution for route planning in a transport network where the network type can be arbitrary: a network of bus routes, a network of tram rails, a road network or any other type of a transport network. Furthermore, the costs of network elements are uncertain. The concept is based on the Dempster–Shafer theory and Dijkstra's algorithm which helps with finding the best routes. The paper focuses on conventional studies without considering traffic accidents or other exceptional circumstances. The concept is presented by an undirected graph. In order to model conventional real transport, the influencing factors of traffic congestion have been applied in the abstract model using uncertain probabilities described by probability intervals. On the basis of these intervals, the cost intervals of each road can be calculated. Taking into account the uncertain values of costs, an algorithm has been outlined for determining the best routes from one node to all other nodes comparing cost intervals and using decision rules that can be defined by the end user, and if necessary, node by node. The suggested solution can be applied for both one type of network as well as for a combination of a few of those.

This paper consists of two parts: planning Co‐operative Transport Systems (CTS) with general rules (communication is a new idea in this system) and an investigation into route planning as a part of services provided in CTS. Part one reveals developing new systems for cooperative sensing and predicting flow, infrastructure and environmental conditions surrounding traffic, with a view to improve the safety and efficiency of road transport operations. Part two introduces the enhancement of the performance of the route planning algorithms. The key issues of these parts are communication and co‐operation between the elements of the whole Co‐operative Transport System.

The goal of this paper was to recognize such marks, signs in human-computer communication, which refers to state of the partner. The paper deals with two tasks: speech style classification and emotion recognition of the speakers based on only the written text of the communication; and an answer generation task based on the recognized speech style or emotion. Our work

... He is an Assistant Professor at Department of Telecommunications and Media Informatics of BME. He is the Vice President of the Hungarian Simulation Society (EUROSIM), and Deputy Director of the McLeod Institute of Simulation Sciences Hungarian Centre. ...

The goal of this paper is to work out a concept for route planning in a road network, where the costs of roads are not constant, but changing in a linear way. The solution developed is based on the classical Dijkstra's algorithm, which helps to find the route with minimal cost. The new algorithm takes the varying into account in order to find out the best route. This search refers not only to a moment of the departure but to the whole duration of the travel. A speed-up technique has been developed for preprocessing before run time. This preprocessing phase helps to give back the route with minimal cost for the user quickly in run time query. A numerical example has been presented to show the detailed steps of the algorithm and the speed-up technique.

The paper is concerned with managing image album, where the picture set (containing very similar or different images) in each album is given. The goal has been to select the most representative pictures from the album. Our solution is based on the clustering of the images. The developed clustering procedure takes the large variety of the pictures and different type of image features into account. We have solved the incomplete feature value problem as well. The central pictures of the largest clusters are selected for representing the album.

The aim of the paper was to support the answer to a query with a ranked list of video segments (search sub-task) and to generate possible hyperlinks (in ranked order) to other video segments in the same collection that provide information about the found segments (linking sub-task). Our solution is based on concept enrichment i.e. the set of words is extended with their synonyms or other conceptually connected words. The other contribution is the content mixing using the combination of all transcripts and manual subtitles of the videos.

In this paper we introduce a refinement and diversification process for re-ranking image search results based on social metadata and visual characteristics of the photos. The goal of the developed re-ranking algorithm is to construct a new sequence with maximal value of the harmonic mean of precision and diversity. Our contribution is twofold: estimation of precision using the statistical average and mixing of clustering results in order to get better diversity. In the combined clustering the new label set is the Cartesian product of the two original cluster label sets.

Protection of content of sensitive text documents is important in enterprise intranets. An index structure is needed to support efficient search and retrieval, but it can lead to information leakage; by statistical attacks an adversary can draw probabilistic inference about the contents of document collection. Zerr and others present a confidential index structure and the ranking of retrieved documents for the query, but only for singleterm queries. The solution proposed in the paper generalizes Zerr’s method by using an anonymization parameter and query-dependent anonymized inverse document frequency factors; thereby it provides better ranking and gives possibility of multi-term queries.

The goal of this paper is to find Wardrop equilibrium in transport networks at case of uncertainty situations, where the uncertainty comes from lack of information. We use simulation tool to find the equilibrium, which gives only approximate solution, but this is sufficient for large networks as well. In order to take the uncertainty into account we have developed an interval-based procedure for finding the paths with minimal cost using the Dempster-Shafer theory.

A model and methodology for simulating traffic in conurbation are described. The aim of the simulation was the optimization of traffic with special regard to the air pollution caused. For the determination of the solution aimed at intelligent demons monitoring and modifying the model are used. Furthermore knowledge based mobile entities – based on Knowledge Attributed Petri Nets – for

This paper deals with multi criteria decision analysis in an engineering area (info-communication networks) by combining two effective mathematical modelling tools: analytic hierarchy process and Bayesian network. An extension of Bayesian network is introduced, this vector Bayesian network is able to calculate the effects of the node faults in info-communication networks on different errors occurred at end users. We have

The paper deals with classification in privacy-preserving data mining. An algorithm, the Random Response Forest, is introduced constructing many binary decision trees, as an extension of Random Forest for privacy-preserving problems. Random Response Forest uses the Random Response idea among the anonymization methods, which instead of generalization keeps the original data, but mixes them. An anonymity metric is defined for undistinguishability of two mixed sets of data. This metric, the binary anonymity, is investigated and taken into consideration for optimal coding of the binary variables. The accuracy of Random Response Forest is presented at the end of the paper.

ABSTRACT The aim of this paper was to develop an image browsing system, where only visual information is available for users and computers as well. Large image set is structured in albums and the idea was to select the most representative image from each album in the level of hierarchy, and then the next upper level in the hierarchy consists of these representative images. Selection of the most representative image is based on semantic features, which helps users in browsing process. Quantitative and qualitative analysis have been completed, where the goodness of the representative image selection is tested and comparison with other systems is accomplished respectively. Finally it can be concluded that the system developed possesses semantic and structured features.

The goal of this paper was to recognize such marks, signs in human-computer communication, which refers to state of the partner. The paper deals with two tasks: speech style classification and emotion recognition of the speakers based on only the written text of the communication; and an answer generation task based on the recognized speech style or emotion. Our work

This paper deals with vehicular traffic management by communication technologies from Traffic Control Center point of view in road networks. The global goal is to manage the urban traffic by road traffic operations, controlling and interventional possibilities in order to minimize the traffic delays and stops and to improve traffic safety on the roads. This paper focuses on transient transport,

The aim of this paper is to find a solution for route planning in a road network, where the costs of roads are uncertain. Our concept is based on the Dempster-Shafer theory and Dijkstra's algorithm, which helps to find the best routes. We model the influencing factors of the traffic congestion on the roads using uncertain probabilities described by probability

The goal of this paper is to find a solution for route planning in a transport network where the network type can be arbitrary: a network of bus routes, a network of tram rails, a road network or any other type of a transport network. Furthermore, the costs of network elements are uncertain. The concept is based on the Dempster–Shafer theory and Dijkstra's algorithm which helps with finding the best routes. The paper focuses on conventional studies without considering traffic accidents or other exceptional circumstances. The concept is presented by an undirected graph. In order to model conventional real transport, the influencing factors of traffic congestion have been applied in the abstract model using uncertain probabilities described by probability intervals. On the basis of these intervals, the cost intervals of each road can be calculated. Taking into account the uncertain values of costs, an algorithm has been outlined for determining the best routes from one node to all...

This paper consists of two parts: planning Co‐operative Transport Systems (CTS) with general rules (communication is a new idea in this system) and an investigation into route planning as a part of services provided in CTS. Part one reveals developing new systems for cooperative sensing and predicting flow, infrastructure and environmental conditions surrounding traffic, with a view to improve the safety and efficiency of road transport operations. Part two introduces the enhancement of the performance of the route planning algorithms. The key issues of these parts are communication and co‐operation between the elements of the whole Co‐operative Transport System.

The aim of this paper is to find solution for route planning in road network for a user, and to find the equilibrium in the path optimization problem, where the roads have uncertain attributes. The concept is based on the Dempster-Shafer theory and Dijkstra's algorithm, which help to model the uncertainty and to find the best route, respectively. Based on uncertain influencing factors an interval of travel time (so called cost interval) of each road can be calculated. An algorithm has been outlined for determining the best route comparing the intervals and using decision rules depending on the user's attitude. Priorities can be defined among the rules, and the constructed rule based mechanism for users' demands is great contribution of this paper. The first task is discussed in more general in this paper, i.e. instead of travel time a general cost is investigated for any kind of network. At the solution of the second task, where the goal is to find equilibrium in transport network at case of uncertain situation, the result of the first task is used. Simulation tool has been used to find the equilibrium, which gives only approximate solution, but this is sufficient and appropriate solution for large networks. Furthermore this is built in a decision support system, which is another contribution of this work. At the end of the paper the implementation of the theoretical concept is presented with a test bed of a town presenting effects of different uncertain influencing factors for the roads.