1. Key word analysis of discourses in Slovene speechIztok Kosem, Darinka Verdonik, 2012, izvirni znanstveni članek Opis: One of the aspects of speech that remains underresearched is the internal variety of speech, i. e. the differences and similarities between different types of speech. The paper aims to contribute to filling this gap in research by making a comparison between different discourses of Slovene spontaneous speech, focusing on the use of vocabulary. The key word analysis (Scott 1997), conducted on a millionword corpus of spoken Slovene, was used to identify lexical items and groups of lexical items typical of a particular spoken discourse, or common to different types of spoken discourse. The results indicate that the presence or absence of a particular word class in the key word list can be a good indicator of a type of spoken discourse, or discourses. Ključne besede: corpus analysis, media discourse, private discourse, official discourse, spoken language, Slovene, key words Objavljeno: 17.05.2017; Ogledov: 10; Prenosov: 1 Polno besedilo (142,91 KB) 
2. 
3. Gesture recognition from data streams of human motion sensor using accelerated PSO swarm search feature selection algorithmIztok Fister, Iztok Fister, Simon Fong, Justin Liang, Sabah Mohammed, 2015, izvirni znanstveni članek Opis: Human motion sensing technology gains tremendous popularity nowadays with practical applications such as video surveillance for security, hand signing, and smarthome and gaming. These applications capture human motions in realtime from video sensors, the data patterns are nonstationary and ever changing. While the hardware technology of such motion sensing devices as well as their data collection process become relatively mature, the computational challenge lies in the realtime analysis of these live feeds. In this paper we argue that traditional data mining methods run short of accurately analyzing the human activity patterns from the sensor data stream. The shortcoming is due to the algorithmic design which is not adaptive to the dynamic changes in the dynamic gesture motions. The successor of these algorithms which is known as data stream mining is evaluated versus traditional data mining, through a case of gesture recognition over motion data by using Microsoft Kinect sensors. Three different subjects were asked to read three comic strips and to tell the stories in front of the sensor. The data stream contains coordinates of articulation points and various positions of the parts of the human body corresponding to the actions that the user performs. In particular, a novel technique of feature selection using swarm search and accelerated PSO is proposed for enabling fast preprocessing for inducing an improved classification model in realtime. Superior result is shown in the experiment that runs on this empirical data stream. The contribution of this paper is on a comparative study between using traditional and data stream mining algorithms and incorporation of the novel improved feature selection technique with a scenario where different gesture patterns are to be recognized from streaming sensor data. Ključne besede: algorithms, human motion sensors, PSO Objavljeno: 07.04.2017; Ogledov: 122; Prenosov: 3 Polno besedilo (0,00 KB) 
4. Particle swarm optimization for automatic creation of complex graphic charactersMatjaž Perc, Iztok Fister, Iztok Fister, Salahuddin M. Kamal, Karin Ljubič Fister, Andres Iglesias, 2015, izvirni znanstveni članek Opis: Natureinspired algorithms are a very promising tool for solving the hardest problems in computer sciences and mathematics. These algorithms are typically inspired by the fascinating behavior at display in biological systems, such as bee swarms or fish schools. So far, these algorithms have been applied in many practical applications. In this paper, we present a simple particle swarm optimization, which allows automatic creation of complex twodimensional graphic characters. The method involves constructing the base characters, optimizing the modifications of the base characters with the particle swarm optimization algorithm, and finally generating the graphic characters from the solution. We demonstrate the effectiveness of our approach with the creation of simple snowman, but we also outline in detail how more complex characters can be created. Ključne besede: optimizacija roja, kompleksni sistem, kaos, particle swarm optimization, complex system, graphics, chaos Objavljeno: 07.04.2017; Ogledov: 106; Prenosov: 1 Polno besedilo (0,00 KB) 
5. A collection of sport activity files for data analysis and data miningSamo Rauter, Iztok Fister, Iztok Fister, 2015, zaključena znanstvena zbirka podatkov ali korpus Opis: Dataset consists of the data produced by nine cyclists. Data were directly
exported from their Strava or Garmin Connect accounts. Data format of
sport’s activities could be written in GPX or TCX form, which are basically
the XML formats adapted to specific purposes. From each dataset, many
following information can be obtained: GPS location, elevation, duration,
distance, average and maximal heart rate, while some workouts include also
data obtained from power meters. Objavljeno: 07.04.2017; Ogledov: 121; Prenosov: 7 Polno besedilo (53,71 KB) Gradivo ima več datotek! Več...

6. 
7. Treelike isometric subgraphs of hypercubesBoštjan Brešar, Wilfried Imrich, Sandi Klavžar, 2003, izvirni znanstveni članek Opis: Treelike isometric subgraphs of hypercubes, or treelike partial cubes as we call them, are a generalization of median graphs. Just as median graphs they capture numerous properties of trees, but may contain larger classes of graphs that may be easier to recognize than the class of median graphs. We investigate the structure of treelike partial cubes, characterize them, and provide examples of similarities with trees and median graphs. For instance, we show that the cube graph of treelike partial cube is dismantlable. This in particular implies that every treelike partial cube ▫$G$▫ contains a cube that is invariant under every automorphism of ▫$G$▫. We also show that weak retractions preserve treelike partial cubes, which in turn implies that every contraction of a treelike partial cube fixes a cube. The paper ends with several Fruchttype results and a list of open problems. Ključne besede: mathematics, graph theory, Isometric embeddings, partial cubes, expansion procedures, trees, median graphs, graph automorphisms, automorphism groups, dismantlable graphs Objavljeno: 31.03.2017; Ogledov: 81; Prenosov: 1 Polno besedilo (135,80 KB) 
8. nary transit functions in graphsManoj Changat, Joseph Mathews, Iztok Peterin, Prasanth G. NarasimhaShenoi, 2010, izvirni znanstveni članek Opis: ▫$n$▫ary transit functions are introduced as a generalization of binary (2ary) transit functions. We show that they can be associated with convexities in natural way and discuss the Steiner convexity as a natural ▫$n$▫ary generalization of geodesicaly convexity. Furthermore, we generalize the betweenness axioms to ▫$n$▫ary transit functions and discuss the connectivity conditions for underlying hypergraph. Also ▫$n$▫ary all paths transit function is considered. Ključne besede: mathematics, graph theory, narity, transit function, betweenness, Steiner convexity Objavljeno: 31.03.2017; Ogledov: 44; Prenosov: 0 Polno besedilo (143,68 KB) 
9. Median and quasimedian direct products of graphsBoštjan Brešar, Pranava Jha, Sandi Klavžar, Blaž Zmazek, 2005, izvirni znanstveni članek Opis: Median graphs are characterized among direct products of graphs on at least three vertices. Beside some trivial cases, it is shown that one component of ▫$G \times P_3$▫ is median if and only if ▫$G$▫ is a tree in that the distance between any two vertices of degree at least 3 is even. In addition, some partial results considering median graphs of the form ▫$G \times K_2$▫ are proved, and it is shown that the only nonbipartite quasimedian direct product is ▫$K_3 \times K_3$▫. Ključne besede: mathematics, graph theory, median graph, direct product, quasimedian graph, isometric embeddings, convexity Objavljeno: 31.03.2017; Ogledov: 46; Prenosov: 1 Polno besedilo (174,14 KB) 
10. Edgeconnectivity of strong products of graphsBoštjan Brešar, Simon Špacapan, 2007, izvirni znanstveni članek Opis: The strong product ▫$G_1 \boxtimes G_2$▫ of graphs ▫$G_1$▫ and ▫$G_2$▫ is the graph with ▫$V(G_1) \times V(G_2)$▫ as the vertex set, and two distinct vertices ▫$(x_1,x_2)$▫ and ▫$(y_1,y_2)$▫ are adjacent whenever for each ▫$i\in \{1,2\}$▫ either ▫$x_i=y_i$▫ or ▫$x_iy_i \in E(G_i)$▫. In this note we show that for two connected graphs ▫$G_1$▫ and ▫$G_2$▫ the edgeconnectivity ▫$\lambda(G_1 \boxtimes G_2)$▫ equals ▫$\min\{\delta(G_1\boxtimes G_2), \lambda(G_1)(V(G_2)+2E(G_2)), \lambda(G_2)(V(G_1)+2E(G_1))\}$▫. In addition, we fully describe the structure of possible minimum edge cut sets in strong products of graphs. Ključne besede: mathematics, graph theory, connectivity, strong product, graph product, separating set Objavljeno: 31.03.2017; Ogledov: 38; Prenosov: 2 Polno besedilo (173,26 KB) 