Registration and Welcome Coffee
Time: 09:00 - 11:00
Conference Opening Ceremony
Time: 11:00 - 11:15
Session: Invited Plenary Lecture 1
Time: 11:15 - 12:00
Chairman: Andrzej M.J. Skulimowski
Place: "Carmen" Lecture Hall
A Japanese Problem Solving Approach: the KJ Ho Method
Susumu Kunifuji, the V-ce President of JAIST, Japan
A Japanese Problem Solving Approach: the KJ Ho Method
In Japan, by far the most popular creative problem-solving methodology using creative thinking is the KJ Ho method. This method puts unstructured information on a subject matter of interest into order through alternating divergent and convergent thinking steps. In this paper we explain basic procedures associated with the KJ Ho.
Session: S2
Time: 12:00 - 13:30
Chairman: Susumu Kunifuji
Place: "Carmen" Lecture Hall
Idea Planter: A Backchannel Function for Fostering Ideas in a Distributed Brainstorming Support System
Hiroaki Furukawa, Takaya Yuizono, and Susumu Kunifuji
Idea Planter: A Backchannel Function for Fostering Ideas in a Distributed Brainstorming Support System
This paper describes the development and evaluation of a backchannel function for distributed brainstorming support system. It also considers the effect of backchannels on ideas in a distributed environment. We defined 'Ability to see or sensitivity to problems', 'Originality', and 'Elaboration' as factors of backchannel that contribute to idea creation. The backchannel function was implemented as a single button click. The experiment was carried out with two sets of four people in a distributed environment. The comparison of the quantity of ideas showed that there was no statistical difference (Mann-Whitney U test: p > 0.75). Moreover, the ratio of feasibility between created ideas also did not show a statistical difference (Mann-Whitney U test: p > 0.2). On the contrary, the ratio of fluency between created ideas showed a statistical difference (Mann-Whitney U test: p < 0.046). As a result, it was suggested that a backchannel function might improve the outcome of distributed brainstorming sessions. In particular, the fluency of individual ideas might be improved significantly.
Modeling and Empirical Investigation on the Microscopic Social Structure and Global Group Pattern
Zhenpeng Li, Xijin Tang
Modeling and Empirical Investigation on the Microscopic Social Structure and Global Group Pattern
In this paper, we investigate the microscopic social mechanisms through agent based modeling and empirical data analysis with the aim to detect the intrinsic link between local structure balance and global pattern. Both investigations suggest that three types of social influences give rise to the emergence of macroscopic polarization, and the polarization pattern is closely linked with local structure balance.
A Comparative Study on Single and Dual Space Reduction in Multi-label Classification
Eakasit Pacharawongsakda and Thanaruk Theeramunkong
A Comparative Study on Single and Dual Space Reduction in Multi-label Classification
Multi-label classification has been applied to several applications since it can assign multiple class labels to an object. However, its effectiveness might be sacrificed due to high dimensionality problem in both feature space and label space. To address these issues, several dimensionality reduction methods have been proposed to transform the high dimensional spaces to lower-dimensional spaces. This paper aims to provide a comprehensive review on ten dimensionality reduction methods that applied to multi-label classification. These methods can be categorized into two main approaches: single space reduction and dual space reduction. While the former approach aims to reduce the complexity in either feature space or label space, the latter approach transforms both feature and label spaces into two subspaces. Moreover, a comparative study on single space reduction and dual space reduction approaches with five real world datasets are also reported. The experimental results indicated that dual space reduction approach tends to give better performance comparing to the single reduction approach.
Knowcations - Conceptualizing a Meme and Cloud-basedPersonal 2nd Generation Knowledge Management System
Ulrich Schmitt
Knowcations - Conceptualizing a Meme and Cloud-basedPersonal 2nd Generation Knowledge Management System
The first generation of Organizational Knowledge Management (OKM) focused on the capturing, storing, and reusing of existing knowledge. To be classed as second generation, systems need to facilitate the creation of new knowledge and innovation which requires creativity and the awareness that old knowledge becomes obsolete. Recent suggestions also urged to advance Personal Knowledge Management (PKM) as an overdue support tool for knowledge workers in the rising Creative Class and Knowledge Societies. Based on the assumption of creative conversations
between many individuals' PKM devices, the autonomous systems are supposed to enable the emergence of the distributed processes of collective extelligence and intelligence, which in turn feed them. With a PKM prototype system pursuing these qualities, the paper illustrates the interaction between a user and external information-bearing hosts and vehicles. The resulting feedback loop incorporates Boisot's I-Space Model, Dawkins'
Memes, Probst's KM Building Blocks and Pirolli's Sensemaking Model for Intelligence Analysis
Lunch
Time: 13:30 - 15:00
Session: S3
Time: 15:00 - 16:20
Chairman: Władysław Homenda
Place: "Carmen" Lecture Hall
Tree Representation of Image Key Point Descriptors
Patryk Najgebauer, Marcin Gabryel, Marcin Korytkowski, Rafał Scherer
Tree Representation of Image Key Point Descriptors
This paper describes a concept of image comparison method based on descriptors of key points generated by SURF algorithm. Proposed method for speed up comparison process uses tree-based representation
of descriptors. We assume that descriptors tree representation of image key points is more efficient than standard list representation. The number of steps to compare sets of image descriptors will be smaller
than in a case of list-based, all to all methods. The proposed method assumes generation of a tree structure from a set of image descriptors generated by SURF algorithm. The descriptors are stored as leaves in the tree structure and other parent tree nodes are used to group similar descriptors. Each next parent node of the tree forms a wider, more general, group of descriptors. We store average values of the
descriptors in the nodes making it possible to quickly compare sets of descriptors by traversing the tree from the root to a leaf by choosing the smallest deviation between searched descriptor and values of nodes.
Each step in a tree traversing can drastically reduce the final number of descriptors that will be needed to compare. The proposed structure also allows for the comparison of whole trees of descriptors that speed up the process of images comparison. Our method involves generating trees of descriptors for single images or groups of related images accelerating the process of searching for similarities among others. In future the method will be used as a base to develop tools for indexing images by their context.
Creativity Effects of Idea-Marathon System (IMS): Torrance Tests of Creative Thinking (TTCT) Figural Tests for College Students
Takeo Higuchi, Takaya Yuizono, Kazunori Miyata
Creativity Effects of Idea-Marathon System (IMS): Torrance Tests of Creative Thinking (TTCT) Figural Tests for College Students
Idea-Marathon System (IMS) is a creativity training process, based on the use of notebooks, in which we make a daily habit of writing our ideas immediately by managing to create any new idea regardless of any thinking area. This paper presents an experimental analysis conducted at Ohtsuki City College (OCC) to quantitatively measure creativity effect on college students before and after a 3 month IMS training. TTCT (Torrance Tests of Creative Thinking) Figural Pre and Post tests were used to confirm the creativity effects on students quantitatively. The group with 3 months of IMS training showed significant increases in "Total Score", "Fluency", "Originality" and "Resistance to Premature Closure (RPC)" while students in the control group showed a significant increase in "RPC" only. Support system of IMS "e-Training System (ETS)" was found moderately correlated with "Fluency". Top, Middle and Low analysis showed improvement in Middle and Low through 3 month IMS.
Complementarity and Similarity of Complementary Structures in Spaces of Features and Concepts
Wladyslaw Homenda and Agnieszka Jastrzebska
Complementarity and Similarity of Complementary Structures in Spaces of Features and Concepts
Authors present an approach to real-life phenomena modeling through concepts' descriptions gathered in features vectors. The start point of analysis are imprecise, fuzzied features, which describe objects. Developed model (feature and concept spaces) formalizes units and groups of knowledge granules. The goal of this study is to discuss complementarity and similarity of complementary features structures. Complementarity of features provides auxiliary knowledge, which should be taken into account. In the article two methodologies for calculating similarity between complementary sets of features are presented. First approach is based on authors generalized similarity measure based on the underlying concept space. The second methodology relies on distance measure, computed directly between features vectors. Addressed issues (most importantly complementarity) are authors contribution to the area of knowledge modeling and structuralization.
Detecting Context Free Grammar for GUI Operation
Ryo Hatano and Satoshi Tojo
Detecting Context Free Grammar for GUI Operation
We apply an algorithm of grammar compression to the detection of regularity in our operation of a computer. When we observe only a surface sequence of inputs, we may be able to find a simple grammar as is often the case in usual grammar learning algorithms. In order to learn such regularity, however, we consider the resultant state of each
operation, that is, the output when we regard an operation as an input. In this paper, we aim at finding a hidden grammar rules, considering the each state change in a computer system per an input. As a result, we find the regularity of human behavior in system manipulation, as well as hidden grammar rules.
Coffee break
Time: 16:20 - 16:40
Session: S4
Time: 16:40 - 18:00
Chairman: Tomoko Kajiyama
Place: "Carmen" Lecture Hall
Classification with Rejection: Concepts and Formal Evaluations
Wladyslaw Homenda, Marcin Luckner and Witold Pedrycz
Classification with Rejection: Concepts and Formal Evaluations
Standard classification allocates all processed elements to given classes. Such type of classification assumes that there are only native and no foreign elements, i.e. all processed elements are included in given classes. The quality of standard classification can be measured by two factors: numbers of correctly and incorrectly classified elements, called True Positives and False Positives. Admitting foreign elements in standard classification increases False Positives and, in this way, deteriorates quality of classification. In this context, it is desired to reject foreign elements, i.e. to not assign them to any of given classes. Rejecting foreign elements will reduce the number of False Positives, but can also reject native elements reducing True Positives as side effect. Therefore, it is important to build well designed rejection, which will reject significant part of foreigners and only few natives. In this paper, concepts of evaluations of classification with rejection are presented. Three main models: a classification without rejection, a classification with rejection, and a classification with reclassification are presented. The concepts are illustrated by
flexible ensembles of binary classifiers with theoretical evaluations of each model. The proposed models can be used, in particular, as classifiers working with noised data, where recognized input is not limited to elements of known classes.
A Color Extraction Method from Text for Use in Creating a Book Cover Image. That Reflects Reader Impressions
Takuya Iida, Tomoko Kajiyama, Noritomo Ouchi and Isao Echizen
A Color Extraction Method from Text for Use in Creating a Book Cover Image. That Reflects Reader Impressions
The image on a book cover gives potential buyers not only an impression of the book's contents but also a clue for search and browsing before or after buying the book. We propose using a color extraction method as the first step in automatically creating book cover images that reflect readers' impressions. We constructed a database expressing the relationships between adjectives and colors and extracted colors from text such as sentences in the book and user reviews. In an experiment with 20 participants who were tasked with reading a book, writing a review of the book, and drawing an image of the book cover, we demonstrated that the colors extracted using this method were more
consistent with the colors in the images drawn by the participants than the colors in the actual cover, especially for novels, regardless of the amount of text in the book.
Removing Redundant Features via Clustering: Preliminary Results in Mental Task Separation
Renato Cordeiro de Amorim and Boris Mirkin
Removing Redundant Features via Clustering: Preliminary Results in Mental Task Separation
Recent clustering algorithms have been designed to take into account the degree of relevance of each feature, by automatically calculating their weights. However, as the tendency is to evaluate each feature at a time, these algorithms may have difficulties dealing with features containing similar information. Should this information be relevant, these algorithms would set high weights to all such features instead of removing some due to their redundant nature. In this paper we introduce an unsupervised feature selection method that targets redundant features. Our method clusters similar features together and selects a subset of representative features for each cluster. This selection is based on the maximum information compression index between each feature and its respective cluster centroid. We empirically validate out method by comparing with it with a popular unsupervised feature selection on three EEG data sets. We find that ours selects features that produce better cluster recovery, without the need for an extra user-defined parameter.
On the Application of Fourier Series Density Estimation for Image Classification Based on Feature Description
Piotr Duda, Maciej Jaworski, Lena Pietruczuk, Rafał Scherer, Marcin Korytkowski, Marcin Gabryel
On the Application of Fourier Series Density Estimation for Image Classification Based on Feature Description
This paper presents an image classification algorithm called Density Based Classifier (DBS). The proposed method puts together the image representation based on keypoints and the estimation of the probability density of descriptors with the application of orthonormal series. For each class of images a separate classifier is constructed. The presented procedure ensures that different descriptors affect the final decision in a different degree. The trained classifier determines whether the query image is assigned to the class or not. The obtained experimental results show that proposed method provides good results. The algorithm can be applied to many tasks in the field of image processing.
Welcome Reception
Time: 19:00 - 21:00
Parallel Sessions
Session: S5-A
Time: 09:00 - 10:35
Chairman: Grzegorz Nalepa
Place: "Carmen" Lecture Hall
On Potential Usefulness of Inconsistency in Collaborative Knowledge Engineering
Weronika T. Adrian, Grzegorz J. Nalepa, Antoni Ligęza
On Potential Usefulness of Inconsistency in Collaborative Knowledge Engineering
Inconsistency in knowledge bases traditionally was considered undesired. Systematic eradication of it served to ensure high quality of a system. However, in Collaborative Knowledge Engineering (CKE), where distributed, hybrid knowledge bases are developed and maintained collectively, inconsistency appears to be an intrinsic phenomena. In this paper, we analyze inconsistency in CKE in terms of its origin, level, type and significance. We claim that in some cases inconsistency should be tolerated and show examples where it can be used constructively.
On Quick Sort Algorithm Performance for Large Data Sets
Marcin Woźniak, Zbigniew Marszałek, Marcin Gabryel, Robert K. Nowicki
On Quick Sort Algorithm Performance for Large Data Sets
Sorting algorithms help to organize data. However sometimes it is not easy to determine the correct order in large data sets, especially if they present special poses of the input series. It often complicates sorting, results in time prolongation or even unable sorting. In such situations, the most commonly used method is to perform sorting process to reshuffled input data or change the algorithm. In this paper, the authors examined quick sort algorithm in two versions for large scale data sets. The algorithms have been examined in performance tests and the results helped to compare them.
A System Using n-grams for Visualizing the Human Tendency to Repeat the Same Patterns and the Difficulty of Divergent Thinking
Taro Tezuka, Shun Yasumasa, Fatemeh Azadi Naghsh
A System Using n-grams for Visualizing the Human Tendency to Repeat the Same Patterns and the Difficulty of Divergent Thinking
One of the factors that inhibits creative thinking is that humans tend to think in the same patterns repetitively and cannot easily come up with a totally new combination of concepts. In other words, humans are not talented at evenly exploring combinatorial space. In order to visualize how strong this tendency is, we implemented a system that asks users to type a long sequence of numbers and then evaluates the frequency of the appearance of the same subsequences, or n-grams. This system can also be used to train oneself to avoid such tendency. We call it "the Creativity Test". The reason for the name is because we believe that efficiency in exploring a wider part of a combinatorial space without being caught in few patterns is important for divergent thinking, which constitutes an integral part of creativity. When we tested the system on a group of participants, we discovered that, for most of them, surprisingly long subsequences appeared repeatedly, making the participants realize how inefficient they were at coming up with new combinations.
The Impact of Changing the Way the Fitness Function Is Implemented in an Evolutionary Algorithm for the Design of Shapes
Andrés Gómez de Silva Garza
The Impact of Changing the Way the Fitness Function Is Implemented in an Evolutionary Algorithm for the Design of Shapes
Evolutionary algorithms (EA's) have been used in many ways for design and other creative tasks. One of the main elements of these algorithms is the fitness function used by the algorithm to evaluate the quality of the potential solutions it proposes. The fitness function guides, constrains, and biases the algorithm's search for an acceptable solution. In this paper we explore the degree to which the fitness function and its implementation affects the search process in an evolutionary algorithm. To do this, the reliability and speed of the algorithm, as well as the quality of the designs produced by it, is measured for different fitness function implementations.
Intelligent Auto-adaptive Web E-content Presentation Mechanism
Wiesław Pietruszkiewicz and Dorota Dżega
Intelligent Auto-adaptive Web E-content Presentation Mechanism
Herein, we present an intelligent auto-adaptive web e-content presentation mechanisms, responsible for the presentation of learning materials and being a technological core of e-learning software. As the e-learning is based on various pieces of software it is possible to efficiently gather data and extract meaningful information about learner's needs. Together with the delivered knowledge about course, we can use them in the reasoning mechanism, deployed to select proper pieces of content - called as the learning pills - according to the learner's requirements. In the first part of this article, we analyse the organisation of learning process and basic pieces of knowledge delivered in it. Later, we introduce the framework for an intelligent auto-adaptive content presentation mechanism. Finally, we discuss its abilities, future research and summarise the presented material.
Session: S5-B
Time: 09:00 - 10:35
Chairman: Jerzy Michnik
Place: "Halka" Conference Room
Intuitionistic Fuzzy Dependent OWA Operator and Its Application
Cuiping Wei , Xijin Tang and Yanzhao Bi
Intuitionistic Fuzzy Dependent OWA Operator and Its Application
In this paper, we propose a novel approach, based on entropy and similarity measure of intuitionistic fuzzy sets, to determine weights of the IFOWA operator. Then we define a new intuitionistic fuzzy dependent OWA (IFDOWA) operator which is applied to handling multi-attribute group decision making problem with intuitionistic fuzzy information. Finally, an example is given to demonstrate the rationality and validity of the proposed approach.
A Fuzzy - Genetic System for ConFLP Problem
Krzysztof Pytel and Tadeusz Nawarycz
A Fuzzy - Genetic System for ConFLP Problem
The article presents the idea of the fuzzy-genetic system, that support making decisions in multiobjective optimization problems. The Genetic Algorithm realizes the process of multiobjective optimization and search Pareto-optimal solutions in a given area of the search space. The Fuzzy Logic Controller (FLC) is used for making decisions, which solution from the Pareto-optimal set will be used. The FLC uses additional fuzzy logic criteria obtained from experts. The article presents the results of solving the Connected Facility Location Problem (ConFLP). The ConFLP is a theoretical model used in telecommunication network design. ConFLP is a NP-hard problem, based on the graph theory. The Genetic Algorithm optimizes three different objective functions: looking for a tree with the minimal edge length that connects each clients' terminals, optimizing the number of network nodes which interlink the terminals and designing their distribution in the given network area. All objective functions are mutually dependent, which additionally makes problem solving very difficult. The experiments show, that the proposed algorithm is an efficient tool for solving the Connected Facility Location Problem. The algorithm can be also used for solving similar optimization problems.
Eval-net: Implementation Evaluation Nets Elements as Petri Net Extension
Michał Niedzwiecki, Krzysztof Rzecki and Krzysztof Cetnarowicz
Eval-net: Implementation Evaluation Nets Elements as Petri Net Extension
Evaluation nets are an easy, readable and functional method for visualizing states and communication between computer systems using diagrams. They are, unfortunately not so popular and there are no tools to help their creation, verification and simulation. However evaluation nets have many common characteristics with Petri nets; and Petri nets have many such tools. In this article an extension to Petri nets called Eval-nets is presented. Eval-nets introduce the most useful elements of evaluation nets. This extension is capable of being used in existing tools and it is not necessary to implement the common elements for Petri and evaluation nets to achieve new functionality. As a result a functional tool for creating, analysing, running, debugging and simulating communication protocols may be build based on Petri nets.
Modeling Behavioral Biases Using Fuzzy and Balanced Fuzzy Connectives
Agnieszka Jastrzebska and Wojciech Lesinski
Modeling Behavioral Biases Using Fuzzy and Balanced Fuzzy Connectives
Authors present consumer decision making model built with respect to Lewin's field theory and Maslow's needs theory. A two-phase procedure for obtaining the decision is introduced. Consumer's opinions regarding arguments speaking for or against the decision are gathered in premises (general opinions) and priorities (particular attitudes) vectors. Premises and priorities are aggregated with balanced norms. The authors set focus on capturing complex aspects of the decision making process, namely imprecise, gradual information and behavioral biases. Various balanced norms, which allow to include biases and compute decisions, are investigated and described. Presented model joins psychological theories of motivation and decision making with balanced norms. The aim of this study is to show that well-known operators have powerful modeling capabilities and may be applied to describe complex aspects of human behavior.
Functional Dependency Parsing of Nonconfigurational Languages
Petr Homola
Functional Dependency Parsing of Nonconfigurational Languages
The paper presents a dependency-based linguistic formalism which defines two syntactic layers, surface and deep, and the formal relationship between them. This relationship forms the basis of a rule-based grammar description that can be straightforwardly implemented to be used in natural language processing.
Coffee break
Time: 10:35 - 10:50
Parallel Sessions
Session: S6-A
Time: 10:50 - 12:10
Chairman: Paweł Rotter
Place: "Carmen" Lecture Hall
Building Internal Scene Representation in Cognitive Agents
Marek Jaszuk and Janusz A. Starzyk
Building Internal Scene Representation in Cognitive Agents
Navigating in realistic environments requires continuous observation of a robots surroundings, and creating internal representation of the perceived scene. This incorporates a sequence of cognitive processes, including attention focus, recognition of objects, and building internal scene representation. The paper describes selected elements of a cognitive system, which implement mechanisms of scene observation based on visual saccades, followed by creating the scene representation based on a distance matrix. Such internal representation is a foundation for scene comparison, necessary for recognizing known places, or changes in the environment.
Feature Selection Using Cooperative Game Theory and Relief Algorithm
Shounak Gore and Venu Govindaraju
Feature Selection Using Cooperative Game Theory and Relief Algorithm
With the advancements in various data-mining and social network related approaches, data-sets with a very high feature - dimensionality are often used. Various information theoretic approaches have been tried to select the most relevant set of features and hence bring down the size of the data. Most of the times these approaches try to find a way to rank the features, so as to select or remove a fixed number of features. These principles usually assume some probability distribution for the data. These approaches also fail to capture the individual contribution of every feature in a given set of features. In this paper we propose an approach which uses the Relief algorithm and cooperative game theory to solve the problems mentioned above. The approach was tested on NIPS 2003 and UCI datasets using different classifiers and the results were comparable to the state of the art methods.
Similarity of Exclusions in the Concept Space
Agnieszka Jastrzebska
Similarity of Exclusions in the Concept Space
The study is devoted to a developed model of formal knowledge representation. The author presents concept and feature spaces. Concepts correspond to existing entities and are described by their features. Valuation mapping matches evaluated features with the underlying concept space. The author is interested in relations structuring knowledge. The goal of this article is to investigate exclusion and similarity in the spaces of features and concepts. Three types of exclusions: weak, strict, and multiple qualitative are defined and discussed. Similarity of features evaluation vectors satisfying relations of weak or strict exclusion is thoroughly analyzed. Research on similarity is presented from two distinct points of view. First approach relies on direct features comparison. Second methodology uses dedicated similarity relation rooted in the underlying concept space.
Session: S6-B
Time: 10:50 - 12:10
Chairman: Lidia Dutkiewicz
Place: "Halka" Conference Room
Knowledge Extraction from the Behaviour of Players in a Web Browser Game
Joao Alves, Jose Neves, Sascha Lange, and Martin Riedmiller
Knowledge Extraction from the Behaviour of Players in a Web Browser Game
Analysis of player behaviour is a technique with growing popularity in the traditional computer games segment and has been proven to aid the developers creating better games. There is now interest in trying to replicate this attainment in a less conventional genre of games, normally called browser games. Browser games are computer games that have as defining characteristic the fact that they are played directly on the web browser. Due to the increased ease of internet access and the growth of the smart phone market, this game genre has a promising future. One of the advantages of browser games in the area of game mining is that player behaviour is relatively easy to record. In this paper we describe a study where we aim to extract knowledge from the behaviour of players in a browser game, during a short period of time.
Modeling and Recognition of Video Events with Fuzzy Semantic Petri Nets
Piotr Szwed
Modeling and Recognition of Video Events with Fuzzy Semantic Petri Nets
This paper addresses the problem of modeling and automated recognition of complex behavior patterns in video sequences. We introduce a new concept of Fuzzy Semantic Petri Nets (FSPN) and discuss their application to recognition of video events. FSPN are Petri nets coupled with an underlying fuzzy ontology. The ontology stores assertions (facts) concerning classification of objects and detected relations. Fuzzy predicates querying the ontology content are used as guards of transitions in FSPN. Tokens carry information on objects participating in a scenario and are equipped with weights indicating likelihood of their assignment to places. In turn, the places correspond to scenario steps. The Petri net structure is obtained by translating a Linear Temporal Logic formula specifying a scenario in a human-readable form. We describe a prototype detection system consisting of an FSPN interpreter, the fuzzy ontology and a set of predicate evaluators. Initial tests yielding promising results are reported.
Dynamic data discovery
Michal R. Przybylek
Dynamic data discovery
This paper presents a new approach to mine dynamic data. We propose generalised tree languages together with their finite models and show how they can represent systematic series of sequential and parallel actions organised into a process. Then we develop an evolutionary heuristic based on skeletal algorithms to learn tree automata. We present two major applications of our techniques: one in process mining and another in discovering a mathematical theory.
Coffee break
Time: 12:10 - 12:25
Parallel Sessions
Session: S7-A
Time: 12:25 - 13:30
Chairman: Katherine G. August
Place: "Carmen" Lecture Hall
Symptom -Treatment Relation Extraction from Web-Documents for Construct Know-How Map
Chaveevan Pechsiri, Onuma Moolwat, and Uraiwan Janviriyasopak
Symptom -Treatment Relation Extraction from Web-Documents for Construct Know-How Map
This paper aims to extract the relation between the disease symptoms and the treatments (called the Symptom-Treatment relation), from hospital-web-board documents to construct the Problem-Solving map which benefits for inexpert-people to solve their health problems in preliminary. Both symptoms and treatments expressed on documents are based on several EDUs (Elementary Discourse Units). Our research contains three problems: first is how to identify a symptom-concept EDU and a treatment-concept EDU. Second is how to determine a symptom-concept-EDU boundary and a treatment-concept-EDU boundary. Third is how to determine the Symptom-Treatment relation from documents. Therefore, we apply a word co-occurrence having a symptom/treatment concept to identify a disease-symptom-concept/treatment-concept EDU, respectively, and also their boundaries. We propose using Naïve Bayes to determine the Symptom-Treatment relation from documents with two feature groups, a symptom-concept-EDU group and a treatment-concept-EDU group. Finally, the result of extraction shows successfully the precision and recall of 84% and 72%, respectively.
A Fuzzy Knowledge-Editing Framework for Encoding Guidelines into Clinical DSSs
Aniello Minutolo, Massimo Esposito, Giuseppe De Pietro
A Fuzzy Knowledge-Editing Framework for Encoding Guidelines into Clinical DSSs
Clinical guidelines have been more and more promoted as a means to foster effective and efficient medical practices and improve health outcomes, especially when implemented in clinical Decision Support Systems (DSSs). In this context, Fuzzy Logic has been proposed as the most suitable approach for profitably tackling uncertainty and vagueness in both clinical recommendations and signs triggering them. In this respect, since the task of building and maintaining a fuzzy knowledge base can be very complex and must be carried out carefully, this paper proposes AFEF (A Fuzzy knowledge Editing Framework), an editing and visualization framework for encoding fuzzy linguistic guidelines into clinical DSSs with the aim of providing intuitive solutions specifically devised to: i) define block of rules pertaining the positive evidence of the same abnormal situation; ii) compose ELSE rules for modeling the negative evidence associated to a block of rules; iii) customize the rules inside a block of rules through a common configuration for the inference; iv) simulate an actual DSS for testing the fuzzy rules inserted; v) automatically encode into a machine executable language the fuzzy clinical knowledge that could be functional in the context of clinical DSSs.
Combination of Interpretable Fuzzy Models and Probabilistic Inference in Medical DSSs
Marco Pota, Massimo Esposito, Giuseppe De Pietro
Combination of Interpretable Fuzzy Models and Probabilistic Inference in Medical DSSs
Fuzzy logic have gained increasing importance in Decision Support Systems (DSSs), in particular in medical field, since it allows to build a transparent and interpretable knowledge base. However, in order to obtain a general description of a system, probabilistic approaches undoubtedly offer the most significant information. Moreover, a good system should be useful also to classify data items which are lacking of some input features. In this work, an approach is proposed to construct an interpretable fuzzy system, which furnishes probabilistic information as a result. The resulting fuzzy sets can be interpreted as the terms of the involved linguistic variables, while the resulting weighted rules model probabilistic information. Rules are presented in two forms: the first is a set of one-dimensional models, which can be used if only one input feature is known; the second is a multi-dimensional combination of them, which can be used if more input features are known. As a proof of concept, the method has been applied for the detection of Multiple Sclerosis Lesions from brain images. The results show that this method is able to construct, for each one of the variables influencing the classification, an interpretable fuzzy partition, and very simple if-then rules. Moreover, a multi-dimensional rule base is presented, by means of which improved results are obtained, also with respect to naive Bayes classifier.
Session: S7-B
Time: 12:25 - 13:30
Chairman: Adrian Horzyk
Place: "Halka" Conference Room
How Does Human-Like Knowledge Come into Being in Artificial Associative Systems?
Adrian Horzyk
How Does Human-Like Knowledge Come into Being in Artificial Associative Systems?
Knowledge is fundamental for intelligence and allows us to consider tasks and solve various problems. Knowledge is formed individually in brain during one's whole life. It can be verified, changed, specified and expanded. The knowledge is available through associations that can be automatically triggered in a context of previous thoughts, associations and changing happenings in surroundings. The associations are instantaneously triggered all the time in various brain parts. One association usually triggers another. The way of association can change accordingly to the surroundings, needs, emotions, and the current knowledge in time. The changes in knowledge influence individual processes of association and reasoning. All conclusions are knowledge dependent. This paper reveals and models some associative processes that take place in neural associative systems and enables them to form knowledge in an associative way. Such associative systems allow us also to exploit the knowledge in the similar way people do using associations, various contexts and previous states of neurons of the biological associative systems.
Integrated Analysis System to Improve Performance of Manned Assembly Line
Won K. Hwam, Yongho Chung, Sang C. Park
Integrated Analysis System to Improve Performance of Manned Assembly Line
Presented in this paper is a framework for an integrated analysis system for performance improvement of a manned assembly line. In a manufacturing system, productivity is a key for competitiveness, for product output, and the
performance of assembly line operations is one of the decisive factors of productivity. However, existing approaches to the manufacturing systems are limited to matters of the plant layout or the robot tasks design in the automated factory and to the manual work focused on the ergonomic based analysis. Consequently, a modern approach for line performance improvement has been researched as individual elements that contribute the line performance, but it has not been researched a framework to synchronize the elements as a productivity improvement
activity in terms of an entire line. In other words, there is not a clear solution to integrate analysis results of micro and macro elements, which represent line capabilities in different levels. As a solution of this problem, this study proposes a framework for a system that focuses on the analysis result integration to improve
performance of a manned assembly line, and shows a software tool that is implemented based on the proposed framework.
ALMM Approach for Optimization of the Supply Routes for Multi-location Companies Problem
Edyta Kucharska, Lidia Dutkiewicz, Krzysztof Rączka, Katarzyna Grobler-Debska
ALMM Approach for Optimization of the Supply Routes for Multi-location Companies Problem
Algebraic-logical meta model (ALMM) is a mathematical model of multistage decision process. This formal schema is knowledge representation of a problem and allows to optimize difficult problems on the basis of simulation. It is specially designed for decision problems for which it is impossible to establish all values and parameters a priori. One of such problems is the supply routes for multi-location companies problem. The aim of the paper is to present substitution tasks method (an optimization method based on ALMM) applied to mentioned problem. A formal algebraic-logical model of this problem, an algorithm based on substitution tasks method and results of computer experiments are presented.
Lunch
Time: 13:45 - 15:00
Session: Invited Plenary Lecture 2
Time: 15:00 - 15:30
Chairman: Maciej Nowak
Place: "Carmen" Lecture Hall
Creativity in Online Communication. Empirical Considerations from a Connectivist Background
Thomas Köhler, Technical University of Dresden, Germany
Creativity in Online Communication. Empirical Considerations from a Connectivist Background
Computer-mediated communication (CMC) influences the user`s self. Different predictions were made how users' self and creativity may arise in an 'asocial' setting during the use of modern information and communication technologies. Some of these discourses even observe a new potential for creative
self-constructions. The conceptual frame of this paper is delivered by the combination of different social scientific approaches coming from Social Psychology, Sociology and Communications Studies (cp. Köhler, 2003). Based upon a series of experiments, this paper investigates how the social self develops in CMC. Overall, the self is moderated by the characteristics of the channel, by psycho-social and finally by socialization variables. The influence of the channel is marked by immediate effects of certain cues, the psycho-social variable is the individual versus joint usage of CMC and the sociological dimension is the familiarity with CMC. All those influences have a specific meaning for the Creativity development in online communication. From a connectivist background it is discussed why and to what extent CMC has a prototype character for creative co-constructions in further forms of mediated interpersonal communication.
Parallel Sessions
Session: S8-A
Time: 15:30 - 16:30
Chairman: Maciej Nowak
Place: "Carmen" Lecture Hall
On the Role of Computers in Creativity-support Systems
Bipin Indurkhya
On the Role of Computers in Creativity-support Systems
We report here on our experiences with designing computer-based creativity-support systems over several years. In particular, we present the design of three different systems incorporating different mechanisms of creativity. One of them uses an idea proposed by Rodari to stimulate imagination of the children in writing a picture-based story. The second one is aimed to model creativity in legal reasoning, and the third one uses low-level perceptual similarities to stimulate creation of novel conceptual associations in unrelated pictures. We discuss lessons learnt from these approaches, and address their implications for the question of how far creativity can be tamed by algorithmic approaches.
Learners' Attitudes Toward Knowledge Sharing in the Inter-cultural and High-contextual Cooperative Learning
Pimnapa Atsawintarangkun, Takaya Yuizono
Learners' Attitudes Toward Knowledge Sharing in the Inter-cultural and High-contextual Cooperative Learning
In this paper, we investigated how the cultural differences affect to a cooperative learning based on Jigsaw technique. The experiment measured attitudes of learners from Thai, Japan and China who have different cultures but share the similar style of high-context communication and showed a comparison between learners' feelings in the intra-cultural learning and the inter-cultural learning. The results revealed that the cultural differences and the style of high-context communication caused learners facing more difficult to share their own knowledge in the inter-cultural group than that in the intra-cultural group. Though a quiz score in the experiment showed a fair learning outcome, learners reported a low level of achievement feeling to share knowledge in the inter-cultural group. This work also analyzed the effects of cultural background and cultural dimensions on learning outcome. The results showed that cultural background can enhance inter-cultural competences in cooperative learning. Besides the attitudes of high difficulty and low achievement, the differences in culture and the high-context style also impacted on three feelings including annoyance, interest and understanding. Learners felt more annoyed but less understandable when learning with partners from different cultures. Nevertheless, the results revealed that learners showed more interest in inter-cultural cooperative learning.
Exploring Associations between the Work Environment and Creative Design Processes
Mobina Nouri and Neil Maiden
Exploring Associations between the Work Environment and Creative Design Processes
Creative thinking is a critical activity in design work, and it can be influenced by the climate of a space that designers work in, either individually or in groups. Designers experience different emotions during creative design
processes, and these emotions can influence their levels of both creativity and productivity, so modifying the environments in which design work is done can impact on creative design outcomes. In this paper we report some first empirical research that investigates associations between environment, emotion and creative design work undertaken using different creativity and design techniques.
Session: S8-B
Time: 15:30 - 16:30
Chairman: Antoni Ligęza
Place: "Halka" Conference Room
Recognition of Agreement and Contradiction between Sentences in Support-Sentence Retrieval
Hai-Minh Nguyen and Kiyoaki Shirai
Recognition of Agreement and Contradiction between Sentences in Support-Sentence Retrieval
Automatically classifying semantic relations between sentences is important for text understanding, specically in helping users collect various viewpoints on a given topic (statement). This paper considers two main semantic categories which are agreement and contradiction in the scope of an application, namely support-sentence retrieval. We present here new sentence classification algorithms based on rules and bootstrapping method. Our initial seed data for training the bootstrapping-based classifiers is automatically built. Our best configuration of bootstrapping-based classifiers yields 5.9% higher result than the word overlap baseline in the agreement category. For the contradiction category, applying bootstrapping learning increases the P@10 by 12.1% when compared to the rule-based approach. These results are promising due to the fact that the whole process requires no human interaction.
Understanding Context-Aware Business Applications in the Future Internet Environment
Emilian Pascalau and Grzegorz J. Nalepa
Understanding Context-Aware Business Applications in the Future Internet Environment
The obvious move towards a Future Internet environment that is strongly distributed, mobile, cloud-based, semantically rich has raised and emphasized the need for a different type of applications. The focus of this new type of applications can no longer be on the software itself but directly on the relevant needs and goals of end-users. We argue that because these applications are strongly end-user oriented, context and context-awareness play an important role in their design and development. Hence in this paper we discuss context-aware business applications towards a better understanding of their specific features. We emphasize two major directions related to context oriented applications, their importance, and the order in which they should be followed and interlinked.
Formal Encoding and Verification of Temporal Constraints in Clinical Practice Guidelines
Marco Iannaccone, Massimo Esposito
Formal Encoding and Verification of Temporal Constraints in Clinical Practice Guidelines
In the last decades, clinical practice guidelines have been formulated in order to help decision making about treating specific diseases and promote standards of care quality. Despite the efforts involved to provide solutions for both specifying and verifying temporal constraints in computerized guidelines, none of them is concerned with directly embedding the theoretic semantics of a formal language as the basis of a guideline formalism in order to easily and directly support the temporal perspective. In such a direction, this paper proposes a formal approach which has been seamlessly embedded into a standards-based verifiable guideline model, named GLM-CDS (GuideLine Model for Clinical Decision Support). Such an approach integrates the theoretic semantics of ontology and rule languages to specify and automatically verify a variety of temporal constraints. Such constraints are formulated according to some time patterns, i.e. task duration, periodicity, deadline, scheduling and time lags, and encoded as axioms/formulae verifiable at run-time during the guideline enactment, in order to detect violations or errors occurred with respect to the temporal perspective. As an example of application of the proposed approach, some temporal constraints have been implemented and integrated in GLM-CDS, according to the time patterns identified.
Evacuation Assist from a Sequence of Disasters by Robot with Disaster Mind
Taizo Miyachi, Gulbanu Buribayeva, Saiko Iga and Takashi Furuhata
Evacuation Assist from a Sequence of Disasters by Robot with Disaster Mind
Great East Japan Earthquake in 2011 caused 118,549 people lost their lives. Especially, 70% of residents around the shore area could not evacuate immediately from the devastated Tsunami by the earthquake. Additionally, the devastated Tsunami by the earthquake caused Fukushima nuclear power plants explosions and nuclear spread influenced that 315,196 people evacuated from their home town. We propose a disaster robot as assistant robot "ACROS" including "disaster mind" to help people to avoid people's frequent psychological problems ("Normalcy Bias" and "Catastrophe Forgetting") and guide them to a safer place. ACROS could provide the latest information to the residents such as real-time Big Data emergency support, a map of radioactive contaminant, etc. We also discuss refugees support simulation by showing the warning pictures on a screen in a prototype of ACROS.
Coffee break
Time: 16:30 - 16:45
Parallel Sessions
Session: S9-A
Time: 16:45 - 18:00
Chairman: Rafał Scherer
Place: "Carmen" Lecture Hall
Optical Music Recognition as the Case of Imbalanced Pattern Recognition: a Study of Single Classifiers
Agnieszka Jastrzebska and Wojciech Lesinski
Optical Music Recognition as the Case of Imbalanced Pattern Recognition: a Study of Single Classifiers
The article is focused on a particular aspect of classification, namely the imbalance of recognized classes. The paper contains a comparative study of results of musical symbols classification using known algorithms: k-Nearest Neighbors, k-means, Mahalanobis minimal distance and decision trees. Authors aim at addressing the problem of imbalanced pattern recognition. Firstly, we theoretically analyze difficulties entailed in classification of music notation symbols. Secondly, in the enclosed case study we investigate the fitness of named single classifiers on real data. Conducted experiments are based on own implementations of named algorithms with all necessary image processing tasks. Results are highly satisfying.
Mining Music Social Networks from an Independent Artist Perspective
Ewa Łukasik
Mining Music Social Networks from an Independent Artist Perspective
The goal of this paper is to discuss various aspects of composing music in the computer era and mining music social networks from the independent artist perspective. Amateur pop composers search for music similar to their composition in order to be sure that certain aspects of their music are common to others, or, on the contrary, to be sure that they did not commit plagiarism. Others are interested in surprising effects of automated composing and put some demands on the produced music, e.g. required emotional character. As an illustration of the support that mining social networks may bring to independent artists some projects related to this subject, accomplished in the Institute of Computing Science, Poznan University of Technology, are presented.
Selflocalization and Navigation in Dynamic Search Hierarchy for Video Retrieval Interface
Tomoko Kajiyama and Shin’ichi Satoh
Selflocalization and Navigation in Dynamic Search Hierarchy for Video Retrieval Interface
We have improved previously proposed graphical search interface,'Revolving Cube Show,' for multi-faceted metadata. This interface can treat discrete, continuous, and hierarchical attributes, enabling users to search flexibly and intuitively by using simple operations to combine attributes. We added two functions, one for displaying a search hierarchy as a guide tree and one for moving to a specific position in the search hierarchy, to solve problems identified through user testing. We created a video retrieval application using the improved interface for the iPad and tested it using data for 10,352 Japanese TV programs. The improved interface enabled users to easily understand the current position in the overall hierarchy and to quickly change specific attribute values.
Polish Artificial Intelligence Society (PAIS) Annual Assembly
Time: 17:00 - 19:00
Place: AGH University of S&T, B1 bldg, room 4
Session: Poster Session
Time: 18:00 - 19:00
Chairman: Thanaruk Theeramunkong
Place: "Carmen" Lecture Hall
Triple Heap Sort Algorithm for Large Data Sets
Marcin Woźniak, Zbigniew Marszałek, Marcin Gabryel, Robert K. Nowicki
Triple Heap Sort Algorithm for Large Data Sets
Sorting algorithms are important procedures to facilitate the order of data. Classic versions of algorithms often can not efficiently determine the correct order in large scale data sets. In this article, authors describe and examine an extended heap sort algorithm performance for large data sets. Extension of the heap structure was subject to performance tests, that showed validity. With the extension, algorithm is able to sort incoming strings faster, regardless to the arrangement of incoming data.
E-Unification of Feature Structures
Petr Homola
E-Unification of Feature Structures
Unlike structural unification, E-unification of feature structures has, to the best of our knowledge, never been used in natural language processing (NLP). We formalize the concept of E-unification for features structures, present a universal E-unification procedure and discuss its computational tractability for arbitrary as well as linguistically motivated E-theories. A number of examples illustrate the usefulness of E-unification in the domain of NLP.
Machine Understanding for Interactive Storytelling
Wim De Mulder, Quynh Do Thi Ngoc, Paul van den Broek, and Marie-Francine Moens
Machine Understanding for Interactive Storytelling
This paper describes our research-in-progress which integrates several domains, in particular natural language processing and the development of virtual immersive environments. Our research aims at "bringing a given text to life" via an immersive environment where the user can freely explore the surroundings and increase his understanding of the given text. We describe some important challenges in achieving this goal and outline our current research results. Our work is practically oriented, aiming at fulfilling some societal needs related to education on which we also report.
The Use of Evaluation Nets in Complex Negotiations Modelling
Michał Niedzwiecki, Krzysztof Rzecki and Krzysztof Cetnarowicz
The Use of Evaluation Nets in Complex Negotiations Modelling
In this work Complex Negotiations are presented by use of evaluation nets. Complex Negotiations is a protocol for negotiating the terms and conditions of a contract to provide a particular service. It includes negotiation of terms prior to concluding a contract as well as their renegotiation while the contract is being executed. Evaluation nets constitute a sort of Petri net extension and they are well-suited for modelling communication protocols. Their application facilitates graphical visualisation, analysis and verification of the validity of Complex Negotiations performed by particular participants.
Range Reverse Nearest Neighbor Queries
Reuben Pereira, Abhinav Agshikar, Gaurav Agarwal, Pranav Keni
Range Reverse Nearest Neighbor Queries
Reverse nearest neighbor (RNN) queries have a broad application base such as decision support, profile-based marketing, resource allocation, data mining, etc. Previous work on RNN, visible nearest neighbor and visible RNN has been done considering only point queries. Such point queries are highly unlikely in the real world. In this paper we introduce a novel variant of RNN queries - Range Reverse Nearest Neighbor queries which consider queries over a region rather than a point.
On a Constrained Regression Problem and its Convex Optimisation Formulation
Michał Przyłuski
On a Constrained Regression Problem and its Convex Optimisation Formulation
We shall consider a constrained regression problem. A shape constraint is applied to a third-degree polynomial regression. We show that some of those problems might be formulated as a problem of finding a proper metric projection. They belong to a class of convex optimisation problems for which efficient solvers exist. A numerical example for a related problem arising in econometrics is given.
Strategic Planning Optimisation Using Tabu Search Algorithm
Wojciech Chmiel, Piotr Kadłuczka, Joanna Kwiecień, Bogusław Filipowicz and Przemysław Pukocz
Strategic Planning Optimisation Using Tabu Search Algorithm
In this paper we introduce a method for optimisation of strategic tasks using an approximation algorithm. We propose the mathematical model and algorithm for solving this problem near optimality. An overall motivation for the development of the proposed algorithm is discussed with detailed efficiency analysis. From the end users' perspective the proposed method has practical application because it can be used in several practical use cases.
Technological Evolution and Social Implications of Recommendation Systems until 2025
Andrzej M.J. Skulimowski, Przemyslaw Pukocz
Technological Evolution and Social Implications of Recommendation Systems until 2025
.
A Logical Foundation For Troubleshooting Agents
Reza Basseda, Paul Fodor, and Steven Greenspan
A Logical Foundation For Troubleshooting Agents
Intelligent software agents interacting with users arise in different applications. One of the applications of such agents, which are called virtual experts, is troubleshooting systems. In this project, we are trying to use different available textual resources to automatically construct a troubleshooting virtual expert. In our solution, we extract the information about the structure of the system from textual document, then generate a conversation with the user in order to identify the problem and recommend appropriate remedies. To illustrate the approach, we have built a knowledge base for a simple use case. A special parser generates troubleshooting conversations that guides the user solve configuration problems.
An Intelligent System with Temporal Extension for Reasoning About Legal Knowledge
Maria Mach-Król, Krzysztof Michalik
An Intelligent System with Temporal Extension for Reasoning About Legal Knowledge
Any enterprise has to adjust its strategy to environmental aspect, one of elements of which is legal environment. This environment is partly of temporal nature, so time has to be taken into account by an intelligent tool. In the paper we present first results of research aimed at building an intelligent system capable of reasoning about legal knowledge with its temporal aspects.
Conference Dinner
Time: 20:00 - 22:00
Departure to Wieliczka by bus, wait in front of the Radisson Blu Hotel, Straszewskiego 17. Time of departure: 8:45
Time: 08:45 - 08:45
Sightseeing in the historic part of the Salt Mine in Wieliczka
Time: 09:30 - 11:00
Session: S10-W
Time: 11:00 - 12:30
Chairman: Tadeusz Trzaskalik
Place: "Carmen" Lecture Hall
The Role of Creativity in the Development of Future Intelligent Decision Technologies: Highlights of the AI-Focused Foresight Project SCETIST
Andrzej M.J. Skulimowski
The Role of Creativity in the Development of Future Intelligent Decision Technologies: Highlights of the AI-Focused Foresight Project SCETIST
.
Evoking Emotion through Stories in Creative Dementia Care
Clare Thompson, Neil Maiden, Mobina Nouri and Konstantinos Zachos
Evoking Emotion through Stories in Creative Dementia Care
This paper reports research to refine the design of a mobile creativity support app to improve person-centred care for older people with dementia. One barrier to previous app use during creative thinking appeared to be the negative activation emotions associated with problem avoidance and prevention exhibited by care staff when resolving challenging behaviours. Therefore we investigated the redesign of the app's content so that care staff were more likely to positive activation associated with creative thinking through storytelling through a first formative evaluation.
An Interactive Procedure for Multiple Criteria Decision Tree
Maciej Nowak
An Interactive Procedure for Multiple Criteria Decision Tree
A lot of real-world decision problems are dynamic, which means that not a single, but a series of choices must be made. Additionally, in serious problems, multiple criteria and uncertainty have to be considered. In the paper an interactive algorithm for multiple criteria decision tree is proposed. Various types of criteria are taken into account, including expected value, conditional expected value and probability of success. The procedure consists of two steps. First, non-dominated strategies are identified. Next, the final solution is selected using interactive technique. An example is presented to show the applicability of the procedure.
Coffee break
Time: 12:30 - 12:45
Session: S11-W
Time: 12:45 - 13:30
Chairman: Thomas Koehler
Place: Lecture Hall in the Wieliczka Salt Mine
Communities of Practice for Developers: HelpMe tool
D. Assimakopoulos, M. Tzagarakis, J. Garofalakis
Communities of Practice for Developers: HelpMe tool
The term Web 2.0 focuses on: user, software development and content, which results from many users that share experiences and interests. Due to the fact that many people gather on web sites looking for a solution to their problem, communities of practice (CoPs) were developed. CoPs have become important places for people who seek and share experience. In CoPs area, Argumentative Collaboration is developing between users, so that users can help each other. This paper refers to CoPs and especially to the field that refers to computer programmers. We introduce HelpMe tool that receives questions about a subject that is tagged by the init (or start) user. The start user of a query process is the user that brings the initial question to the community. HelpMe tool automatically selects a group of people according to rules and metrics, in order to supply feedback to the community group who deals with the specific subject (according to label tagging). We introduce two new metrics ULQI (user label query importance) and ULCI (user label communication importance) that are responsible for selecting the appropriate group of people in the community and computing the reputation scores based on the received ratings. HelpMe tool visualizes conversations through graphs, text clouds and statistics.
Structural Modeling Approach to Management of Innovation Projects
Jerzy Michnik
Structural Modeling Approach to Management of Innovation Projects
In the global and highly competitive economy innovativeness becomes the key factor of competitive advantage. Selecting the proper direction of projects development is the vital point of the innovation management. Such a decision is characterized by many conflicting objectives that additionally influence each other. To overcome the problem of interrelations between criteria the structural approach has been proposed. Two such methods, ANP and WINGS, have been applied to the choice of innovations' development. Their procedures and results have been also compared.
Final Discussion, Conference Closing Ceremony, Announcement of the next KICSS Conference
Time: 13:30 - 14:30
Lunch
Time: 14:30 - 15:45
A visit to the Geological and Ancient Mining Museum in the Salt Mine
Time: 16:00 - 16:45
Transfer to the ground level by mine elevator, return to Kraków by bus
Time: 17:00 - 17:45
An opportunity to attend a classical music concert in the Kraków Philharmonic
Time: 18:00 - 19:30
