Search

Ovaj rad predstavlja prvi prikaz sadržaja i rezultata evaluacije jednog programa za podsticanje veštinasaradnje i timskog rešavanja problema kod mladih. Reč je o treningu koji je osmišljen u okviru projektaPEERSolvers (https://peersolvers.f.bg.ac.rs), a zasnovan na PEER modelu uspešne saradnje, koji je većpredstavljen našoj naučnoj i stručnoj javnosti. Shodno postavkama tog modela, istoimeni (PEER) treningima četiri osnovna segmenta. Prvi segment posvećen je formulisanju i utvrđivanju pravilakonstruktivnog dijaloga i razmene ideje; drugi podrazumeva rad na prepoznavanju i uvažavanju osobinaličnosti odnosno individualnih razlika u timskom radu; treći segment tiče se veština iz oblastiemocionalne inteligencije i podstiče pravilno uočavanje, razumevanje i regulisanje emocija tokomsaradnje; najzad, četvrti deo treninga bavi se umešnošću članova tima da koriste spoljne resurse –prevashodno digitalne izvore podataka – koji su od pomoći u rešavanju problema s kojim se tim suočava.Svakom segmentu posvećen je po jedan tročasovni blok, s tim što prvom segmentu prethodi i kratka uvodnasesija (koja služi uzajamnom upoznavanju i upoznavanju učesnika sa programom rada), dok se poslednjiuliva u završnu rekapitulaciju i evaluaciju treninga. Celokupan trening odvija se u četiri dana isastoji od niza grupnih vežbi i interaktivnih zadataka. Trening je prvi put realizovan u martu tekućegodine, a polaznici su bili učenici drugog razreda iz šest beogradskih srednjih škola (tri gimnazije,tri srednje stručne škole), pri čemu je u svakoj školi treningom obuhvaćena po jedna grupa veličine N= 21. Iskustva trenera potvrđuju da trening može uspešno da se izvede prema predviđenom scenariju;istovremeno, evaluacije učesnika ukazuju na to da je sadržaj treninga za njih relevantan (prosečne ocenerelevantnosti četiriju segmenata iznose 3,39–3,58, na skali od 1 do 4) i da su njime stekli odgovarajućeveštine (prosečne ocene samoopažene kompetentnosti su u rasponu 3,61–3,80). Ipak, uočen je i prostorza

U ovom radu, izložićemo jedan aspekt treninga kreiranog u okviru projekta „Saradnjom do rešenja: PEER model osnaživanja mladih za konstruktivni dijalog i timski rad” koji se realizuje u okviru programa Ideje, Fonda za nauku Republike Srbije. Cilj rada je predstavljanje treninga namenjenog unapređenju veština mladih u korišćenju interneta kao resursa u saradničkom rešavanju problema. S obzirom na to da su digitalne kompetencije prepoznate kao veštine od izuzetne važnosti u budućnosti, naše interesovanje je bilo usmereno na osnaživanje ove veštine kod srednjoškolaca. Trening je imao nekoliko ciljeva: a) podstaći korišćenje digitalnih resursa za rešavanje problema, pre svega u vidu traganja za informacijama; b) upoznati učesnike sa mogućnostima koje pruža internet kao metaizvor informacija; c) razviti kritički odnos učesnika prema dostupnim informacijama, pre svega u pogledu njihove pouzdanosti. Trening je realizovan tokom marta 2023. godine u šest srednjih škola iz Beograda (tri gimnazije i tri srednje stručne škole). Ukupno 126 učenika drugog razreda srednje škole je završilo trening u celosti (55% ženskog pola). Deo treninga usmeren ka digitalnim kompetencijama realizovan je u okviru jednog dana, u trajanju od tri sata. U prvom delu treninga učesnici su, kroz vođene grupne aktivnosti, došli do zajedničkog određenja pojma resursa i njihove uloge u saradničkom rešavanju problema. Uz to, kroz obuku i zajednički rad upoznati su sa adekvatnim strategijama pretraživanja informacija na internetu. U drugom delu, učesnici su obučeni za primenu kriterijuma procene relevatnosti i tačnosti informacija dobijenih putem zajedničke internet pretrage. Potom, učesnike smo upoznali sa popularnom alatkom za internet pretragu – ChatGPT, uz naglašavanje njegove prednosti u odnosu na uobičajene pretraživače, ali i ograničenja koja se tiču praćenja izvora informacija. Završna aktivnost ticala se primene naučenog, kroz timski rad, na društveno i naučno relevantnom problemu. Rezultati evalua

Ovaj rad predstavlja prvi prikaz sadržaja i rezultata evaluacije jednog programa za podsticanje veština saradnje i timskog rešavanja problema kod mladih. Reč je o treningu koji je osmišljen u okviru projekta PEERSolvers (https://peersolvers.f.bg.ac.rs), a zasnovan na PEER modelu uspešne saradnje, koji je već predstavljen našoj naučnoj i stručnoj javnosti. Shodno postavkama tog modela, istoimeni (PEER) trening ima četiri osnovna segmenta. Prvi segment posvećen je formulisanju i utvrđivanju pravila konstruktivnog dijaloga i razmene ideje; drugi podrazumeva rad na prepoznavanju i uvažavanju osobina ličnosti odnosno individualnih razlika u timskom radu; treći segment tiče se veština iz oblasti emocionalne inteligencije i podstiče pravilno uočavanje, razumevanje i regulisanje emocija tokom saradnje; najzad, četvrti deo treninga bavi se umešnošću članova tima da koriste spoljne resurse – prevashodno digitalne izvore podataka – koji su od pomoći u rešavanju problema s kojim se tim suočava. Svakom segmentu posvećen je po jedan tročasovni blok, s tim što prvom segmentu prethodi i kratka uvodna sesija (koja služi uzajamnom upoznavanju i upoznavanju učesnika sa programom rada), dok se poslednji uliva u završnu rekapitulaciju i evaluaciju treninga. Celokupan trening odvija se u četiri dana i sastoji od niza grupnih vežbi i interaktivnih zadataka. Trening je prvi put realizovan u martu tekuće godine, a polaznici su bili učenici drugog razreda iz šest beogradskih srednjih škola (tri gimnazije, tri srednje stručne škole), pri čemu je u svakoj školi treningom obuhvaćena po jedna grupa veličine N = 21. Iskustva trenera potvrđuju da trening može uspešno da se izvede prema predviđenom scenariju; istovremeno, evaluacije učesnika ukazuju na to da je sadržaj treninga za njih relevantan (prosečne ocene relevantnosti četiriju segmenata iznose 3,39–3,58, na skali od 1 do 4) i da su njime stekli odgovarajuće veštine (prosečne ocene samoopažene kompetentnosti su u rasponu 3,61–3,80). Ipak, uo

This study is part of the Project PEERSolvers, which aims at conceptualization and designing an evidence-based intervention to promote young people’s capacities for constructive peer interaction and collaborative problem-solving. Previous research indicates that spontaneous peer interaction is often unproductive (Howe and Mercer, 2010). Thus, scholars (e.g. Dawes & Sams, 2004; Mercer et al., 2004; Wegerif & Scrimshaw, 1997) consider it is important to support young people to constructively participate in a dialogue during collaborative problem-solving. Although much is known about the process of peer dialogue (Howe & Abedin, 2013), the systematization of knowledge about characteristics of dialogue that make collaborative problem solving among peers productive or unproductive is lacking. Therefore, we will set out to determine such features in order to conceptualize the intervention model which will promote constructive dialogue and at the same time prevent the occurrence of undesirable ways of communication in teamwork. Furthermore, research findings suggest that adult scaffolding, especially teachers’ plays important role in building youngsters' capacities for constructive dialogue and that digital media could act as resources for effective collaborative problem-solving (e.g. Hathcock et al., 2015; Ratnasari et al., 2019; Sobko et al., 2020). Thus, we intend to systematically review the relevant body of literature not just to identify productive and unproductive dimensions of peer dialogue, but also to target adult scaffolding techniques and the role of digital media as the potential resources in collaborative problem-solving. (see also HYPOTHESES and STUDY TYPE).

This study is part of Project PEERSolvers, which aims at designing an evidence-based intervention to promote young people’s capacities for constructive peer interaction and collaborative problem solving. A major assumption of the Project is that group interaction and collaborative problem solving may be greatly enhanced if participants demonstrate the abilities, competencies and skills encompassed by the construct of emotional intelligence (EI; see e.g., Mayer & Salovey, 1997; Mestre et al., 2016). This assumption is based on prior research which has shown EI to contribute to the quality of social/interpersonal relations (Lopes et al., 2004, 2005; Song et al., 2010) and to play a significant role in teamwork (e.g., Cole et al., 2019; Farh et al., 2012). In the present study we intend to test this assumption and further examine the role of socio-emotional competencies in collaborative problem solving by systematically reviewing the relevant body of literature (see also HYPOTHESES and STUDY TYPE).

This study is part of Project PEERSolvers, which aims at designing an evidence-based intervention to develop young people’s capacities for constructive peer interaction and collaborative problem solving. Informed by several previous studies, the Project builds on the assumption that personality traits play a significant role in team processes and performance (e.g., Bell, 2007; Driskell et al., 2006; Morgeson et al., 2005; Peeters et al., 2006; van Vianen & De Dreu, 2001) and that team work may actually be facilitated by diversity in personality profiles (Hammer & Huszczo, 1996), especially if team members are taught to appreciate and take advantage of this diversity (Clinebell & Stecher, 2003). To establish a firm ground for designing the planned intervention, the present study will systematically review the evidence pertaining to the above assumptions and synthesize available findings on the role of personality differences in collaborative problem solving (see also HYPOTHESES and STUDY TYPE). Bell, S. T. (2007). Deep-Level composition variables as predictors of team performance: A meta-analysis. Journal of Applied Psychology, 92, 595–615. doi: https://doi.org/10.1037/0021-9010.92.3.595 Clinebell, S., & Stecher, M. (2003). Teaching teams to be teams: An exercise using the Myers-Briggs Type Indicator and the Five-Factor personality traits. Journal of Management Education, 27, 362–383. doi: https://doi.org/10.1177/1052562903027003006 Driskell, J. E., Goodwin, G. F., Salas, E., & O’Shea, P.G. (2006). What makes a good team player? Personality and team effectiveness. Group Dynamics Theory Research and Practice, 10, 249–271. doi: https://doi.org/10.1037/1089-2699.10.4.249 Hammer, A. L., & Huszczo, G. E. (1996). Teams. In A. L. Hammer (Eds.), MBTI applications: A decade of research on the Myers-Briggs Type Indicator (pp. 81–103). Palo Alto, CA: Consulting Psychologists Press, Inc. Morgeson, F. P., Reider, M. H., & Campion, M. A. (2005). Selecting individuals in team settings: The importance of social skills, personality characteristics, and teamwork knowledge. Personnel Psychology, 58, 583–611. doi: https://doi.org/10.1111/j.1744-6570.2005.655.x Peeters, M. A. G., Van Tuijl, H. F. J. M., Rutte, C. G., & Reymen, I. M. M. J. (2006). Personality and team performance: A meta-analysis. European Journal of Personality, 20, 377–396. doi: https://doi.org/10.1002/per.588 van Vianen, A. E. M., & De Dreu, C. K. W. (2001). Personality in teams: Its relationship to social cohesion, task cohesion, and team performance. European Journal of Work and Organizational Psychology, 10, 97-120. doi: https://doi.org/10.1080/13594320143000573

Part 4: Modelling: Structures and Implementations; International audience; Coordination scenarios have high demands on concurrency and interaction. However, these are typical sources for flaws in both design and implementation. A modeling approach enables reasoning about distributed algorithms and finding deficiencies right from the beginning. The Peer Model has been introduced as a modeling tool for distribution, concurrency and blackboard-based collaboration and coordination, relying on known foundations like tuple spaces, Petri Nets and Actor Model. A runtime system exists that serves Java developers for prototyping, but still a feasible tool-chain was missing, like for most academic systems.This paper presents a practical new tool-chain for the Peer Model consisting of a graphical modelling tool, building on a drawing program that exports XML. A translator parses the XML and translates it into a newly developed domain specific language that is the basis for code generation. One target is a new, formal automaton-based runtime written in the Go programming language. It allows systematic simulation runs of user models. The demo shows a peer competition scenario, where several players play a game, a global state holds the players’ scores, and in addition each peer maintains a decentralized state. Before taking a move in the game, a peer asserts its current local state to be the same like the global one. If this is the case, it carries out its action and distributes the information about it to all other players for further verification. The scenario captures core coordination mechanisms found in blockchain systems.

In den meisten Peer-to-Peer Netzwerken werden Peers zufällig oder anhand ihrer geografischen Position verbunden. Das kann zu einem Leistungsengpass führen, worunter die Performanz eines solchen Netzwerkes leiden kann. Dieses Problem kann durch die Nutzung von Peer Clustering Algorithmen, welche die Peers anhand bestimmter Charakteristika gruppieren, gelöst werden. Diese Diplomarbeit liefert ein Framework, das systematische Benchmark-Tests von Peer Clustering Algorithmen, deren Auswertung und Vergleich ermöglicht. Die resultierende Applikation basiert auf dem Peer Model, einem koordinationsbasierten Programmiermodell. Außerdem bietet das Framework vielfältige Konfigurationsmöglichkeiten und erlaubt die einfache Austauschbarkeit der verwendeten Peer Clustering Algorithmen.Als weiterer Beitrag werden zwei innovative Algorithmen, Slime Mold und Slime Mold K-Means, die auf dem Lebenszyklus des Dictyostelium discoideum Schleimpilzes basieren, implementiert. Die beiden Algorithmen werden in Kapitel 5 vorgestellt und detailliert beschrieben. Diese werden kompetitiv gemessen, evaluiert und mit den neun folgenden, bekannten konventionellen und schwarm-basierten Algorithmen verglichen: Artificial Bee Colony, Artificial Bee Colony combined with K-Means, Ant-based Clustering, Ant K-Means, Fuzzy C-Means, Genetic K-Means, Hierarchical Clustering, K-Means and Particle Swarm Optimization. Slime Mold und Slime Mold K-Means übertreffen sämtliche andere schwarm-basierten Algorithmen in Bezug auf Ausführungszeit und Qualität der Clustering Lösung. Denn obwohl sie sich meist im Mittelfeld befinden, zeigen sie, im Gegensatz zu anderen Algorithmen, keine extremen, unerwünschten Schwankungen bezüglich der Clustering Effektivität., In most Peer-to-Peer networks, peers are placed randomly or based on their geographical position. This can lead to a performance bottleneck and thus, performance in such networks can be extremely poor. This problem can be solved by using peer clustering algorithms, aiming at grouping peers, which have certain characteristics in common, together as neighbors. This thesis provides a benchmarking framework, which allows the systematic benchmarking, evaluation and comparison of peer clustering algorithms. The resulting application is based on the Peer Model, a coordination-based programming model. Furthermore, the framework provides extensive configuration possibilities and easy exchangeability of the applied peer clustering algorithms.As additional contributions, two innovative algorithms, Slime Mold and Slime Mold K-Means, based on the life-cycle of the Dictyostelium discoideum slime mold are implemented. Those two algorithms are presented and described in detail in Chapter 5. They are competitively benchmarked, evaluated and compared to the following nine well-known conventional and swarm-based algorithms: Artificial Bee Colony, Artificial Bee Colony combined with K-Means, Ant-based Clustering, Ant K-Means, Fuzzy C-Means, Genetic K-Means, Hierarchical Clustering, K-Means and Particle Swarm Optimization. Slime Mold and Slime Mold K-Means outperform all other swarm-inspired algorithms in terms of execution time and quality of the clustering solution. Although they are usually midranged, they never provide unwanted massive increase or decrease in clustering effectiveness results, unlike other peer clustering algorithms.

Developing distributed systems with multiple stakeholders and evolving requirements is a highly complex task, which can be simplified by the usage of middleware with suitable coordination abstractions. However, in open environments like the Internet, also security and trust among the participants have to be considered. Each participant must be able to protect access to its own data and services in a flexible way. This also applies to space-based middleware, which enables data-driven coordination among autonomous processes using decoupled communication via shared spaces. This thesis therefore aims at integrating space-based coordination with security by creating a novel authorization concept that adapts well-established access control principles to the characteristic properties of space-based middleware. The concept relies on simple yet expressive authorization rules that restrict operations on specific space partitions, thus allowing for fine-grained access control. Permissions may depend on authenticated subject attributes, properties of the accessed content, and additional context information. This approach enables administrators to grant each participant only permissions that are actually necessary for planned interactions. It is presentedby means of access control models for two related middleware technologies that cover different aspects of space-based coordination. XVSM provides configurable sub-spaces with extensible query features, while the Peer Model supports a hierarchical space structure with customizable coordination logic for conditional message routing and service invocations. Using the intrinsic coordination mechanisms of the respective middleware, authorization policies can be configured independently for each distributed space, whereas administrator privileges for dynamic policy modifications are specified in the same way as regular permissions. Security is further increased by the usage of multiple protection layers, so that permissions need to be acquired at different levels. Due to an integrated delegation and trust concept, the approach is suitable for open environments without fixed trust assumptions. To enable their practical application, the conceptualized access control models are integrated into the respective middleware architectures and their prototypical runtime implementations. Reusability is promoted via the specification of patterns for secure coordination, which provide generic solutions for common coordination tasks by combining the required coordination logic with suitable authorization policies for protecting all involved spaces. The feasibility of the approach is demonstrated via a series of case studies that cover different security constraints and application domains.

Aufgrund der F��higkeit verschiedene Ressourcen zu teilen, sind P2P Netzwerke seit Jahrzehnten im Fokus der Forschung. Vor allem die Umsetzung von Suchalgorithmen, um verteilte Ressourcen in unstrukturierten P2P Netzwerken zu finden, ist von spezieller Wichtigkeit, da keine zentrale Sicht auf das Netzwerk existiert. Diese Arbeit geht die Notwendigkeit zur Evaluierung und dem Vergleichen von Suchalgorithmen f��r unstrukturierte P2P Netzwerke, unter der Verwendung von Standardmetriken, an. Sie liefert zwei Umsetzungen eines Frameworks f��r systematische Benchmark-Tests von Suchalgorithmen in unstrukturierten P2P Netzwerken und erm��glicht deren Vergleich. Das Framework wird dabei einmal mit dem Actor Model und einmal mit dem auf koordinationsbasiertem Programmiermodel Peer Model umgesetzt. Beide Umsetzungen unterst��tzen dabei die leichte Austauschbarkeit von Suchalgorithmen. Das zweite Ziel dieser Arbeit ist das Erstellen eines voll verteilten Suchalgorithmus f��r unstrukturierte P2P Netzwerke, basierend auf dem kollektiven Futtersuchverhalten von Borkenk��fern. Des Weiteren wird ein bereits bestehender Algorithmus basierend auf dem Physarum Polycephalum Schleimpilz an die Bed��rfnisse von voll verteilter Suche in unstrukturierten P2P Netzwerken angepasst. Um diese Ziele zu erreichen wird folgende methodische Vorgehensweise angewendet. Nach ausf��hrlicher Literaturrecherche werden die Frameworks zun��chst entworfen und anschlie��end umgesetzt. Als n��chstes werden die beiden neuen Suchalgorithmen entwickelt und in die Frameworks eingebunden. Abschlie��end werden beide Algorithmen kompetitiv gemessen, evaluiert und mit vier bekannten Suchalgorithmen verglichen: Gnutella Flooding, k-Walker, AntNet f��r P2P und SMP2P. Bark Beetle und der angepasste Physarum Polycephalum Algorithmus zeigen beide insgesamt sehr gute Ergebnisse hinsichtlich der Skalierbarkeit in Bezug auf Netzwerkgr����e und -last. Hinsichtlich absoluter Zeit zeigen beide Algorithmen sehr vielversprechende Ergebnisse, wobei lediglich k-Walker etwas bessere Ergebnisse bei Netzwerken mit vielen Replikaten erzielt. In puncto Erfolgsrate zeigt Bark Beetle beinahe so gute Ergebnisse wie Gnutella Flooding, ohne dessen wesentliche M��ngel aufzuweisen. Obwohl die Erfolgsrate f��r die Physarum Polycephalum Anpassung sehr niedrig f��r Netzwerke mit wenig Replikaten ist, wird diese signifikant besser f��r Netzwerke mit vielen Replikaten., Through its ability to share various resources, P2P networks have been in the focus of research for the last decades. Especially the utilization of search algorithms to retrieve the distributed resources in unstructured P2P networks is of major importance, since no global view of the network exists. This thesis addresses the need to evaluate and compare search algorithms for unstructured P2P networks with each other by using standard metrics. It provides two frameworks for systematic benchmarking and comparison of search algorithms in unstructured P2P networks. The first framework is implemented based on the Actor Model and the second one based on the Peer Model (a coordination based programming model). Both frameworks support easy exchangeability of search algorithms. The second goal of this thesis is to create a fully distributed search algorithm for unstructured P2P networks based on the collective feeding of bark beetles. Additionally, an already existing algorithm from a different domain based on the Physarum Polycephalum slime mold is adapted to fully distributed search for unstructured P2P networks. To achieve these goals the following methodological approach is applied. After an extensive literature research, the frameworks are first designed and afterwards implemented. As the next step, the two new search algorithms are developed and implemented into the frameworks. As the final step, both algorithms are benchmarked, evaluated and compared to four existing search algorithms: Gnutella Flooding, k-Walker, AntNet for P2P and SMP2P. Overall, Bark Beetle and the adapted Physarum Polycephalum algorithm show very good scalability regarding growing network size and load. In terms of absolute time, both algorithms show very promising results with only k-Walker having slightly better results for networks with high replication. In terms of success rate, Bark Beetle shows an almost equal good success rate as Gnutella Flooding, without having the same major drawbacks. Although the success rate for the Physarum Polycephalum adaption is very low for networks with small replication, it is significantly better for networks with high replication.

WOS: 000466168300011, Autism spectrum disorder (ASD) is a neurobiological disorder characterized by significant difficulties in social interaction and communication, restricted range of behavior and limited areas of interest and stereotypic behaviors. Literature involves a rich variety of studies utilizing an array of effective methods to teach different skills to individuals with ASD. One of these methods has been video modeling which involves watching the video recording of model performing a target behavior. This method has received considerably empirical support. The present study intended to examine the effectiveness of teaching skills to individuals with ASD how to follow signs in finding the restroom in a public setting by utilizing video modeling in which the model is a peer with an ASD. Participants of the study were three male students whose age ranged between 10 and 13. Results of the study showed that video modeling was effective in acquiring, maintaining and generalizing targeted skills. Findings, limitations of the study and its implications for future research as well as practitioners were discussed.

WOS:000406884300003 Bu araştırmanın amacı, bilgisayar destekli video öğretiminin (BDVÖ) bankamatikten para çekme becerisinin zihinsel yetersizliği olan bireylere öğretilmesinde etkili olup olmadığını incelemektir. Bu amaç doğrultusunda bankamatikten para çekme becerisi, zihinsel yetersizliği olan üç bireye hazırlanan BDVÖ ile öğretilmiştir. Araştırmaya hafif düzeyde zihinsel yetersizliği olan, 16 yaşlarında üç birey katılmıştır. Araştırmada tek denekli araştırma yöntemlerinden katılımcılar arası yoklama evreli çoklu yoklama modeli kullanılmıştır. Zihinsel yetersizliği olan bireylerin bankamatikten para çekme becerisinin BDVÖ aracılığıyla kazanımları ve hedeflenmeyen bilgi kazanım düzeyleri incelenmiştir. Araştırmanın bulguları, BDVÖ'nün hafif derecede zihinsel yetersizliği olan bireylerin bankamatikten para çekme becerisini kazanmalarında etkili olduğunu göstermiştir. Ayrıca, öğretimin tamamlanmasının ardından öğrenciler kazanılmış beceriyi sergilemeye devam etmiş ve bu beceriyi farklı bankamatiklere, bankamatik kartlarına, farklı miktarda para ve farklı şifrelere genellemişlerdir. BDVÖ aracılığı ile sunulan hedeflenmeyen bilgi kazanım düzeyine ilişkin bulgular, araştırmaya katılan bireylerin hedeflenmeyen bilgi kazanımını edindiklerini göstermiştir. Araştırmanın sosyal geçerlilik bulguları, zihinsel yetersizliği olan bireylerin bilgisayar destekli video öğretimi ile bankamatikten para çekme becerisinin öğretimine ilişkin görüşlerinin olumlu olduğunu göstermiştir. The purpose of this research is to examine whether computer aided video instruction (CAVI) is effective in teaching of money withdrawal skill from ATM cash machine to individualas with intellectual disabilities. For this purpose, the ability to withdraw money from the ATM was taught by CAVI, which was prepared for three individuals with intellectual disabilities. Three individuals, aged 16, participated in the study with mild intellectual disabilities. In the research, a single subject research method with multiple probe design with probe conditions across subjects was used. Individuals with intellectual disabilities were assessed for their ability to withdraw money from the ATM by means of CAVI and the degree of untargeted knowledge acquisition. The findings of the study show that CAVI is effective in gaining the skill to withdraw money from the ATM for individuals with mild intellectual disabilitiy. In addition, after completion of the instruction, the students continued to exhibit their acquired skill and this skill was generalized to different ATMs, debit cards, different amounts of money and different passwords. Findings related to the level of untargeted knowledge acquisition provided by CAVI have showed that individuals participating in the research received untargeted information. The social validity findings of the study showed that the opinions of individuals about teaching cash withdrawal skills from ATMs through CAVI were positive.

Um ��berleben zu k��nnen, ben��tigt jeder Mensch Lebensmittel und daher ist eine angemessene Versorgung der gesamten Menschheit mit Essen wichtig. Offensichtlich ist das nicht der Fall - Nahrungsmittelabf��lle und Mangelern��hrung sind die Folgen. In letzter Zeit versuchen immer mehr food sharing Initiativen das Problem zu l��sen, indem sie ��bersch��ssige Lebensmittel weiterverteilen, sind dabei aber nur m����ig erfolgreich. Deshalb ist das Ziel dieser Arbeit einen smarten food sharing Prozess zu beschreiben. Sie besch��ftigt sich mit vielen Aspekten von food sharing und bietet Vorschl��ge f��r Verbesserungen. Ganz besonders wird das Koordinationsproblem das food sharing innewohnt beleuchtet und mit dem Peer Model gel��st. Zu Beginn gibt die Arbeit Einblicke in die globale Lebensmittelversorgung und sch��tzt das m��gliche Potential von food sharing ab. Danach analysiert sie bestehende food sharing L��sungen und Forschung ��ber food sharing und gibt Verbesserungsvorschl��ge. Anschlie��end werden mehrere Informations- und Kommunikationstechnologien beschrieben, die in einem verbesserten food sharing Prozess verwendet werden k��nnen. Dabei vergleicht sie g��ngige Koordinationsframeworks und kommt zu dem Ergebnis, dass f��r food sharing das Peer Model am besten geeignet ist. Deshalb wird das Koordinationsproblem von food sharing mit dem Peer Model modelliert, f��r seine Java Laufzeitumgebung implementiert und eine Simulation zeigt, dass das Model funktioniert. Abschlie��end beschreibt die Arbeit einen smarten food sharing Prozess und Storyboards veranschaulichen wie Benutzer damit interagieren. Die Weltbev��lkerung mit Essen zu versorgen wird auch zuk��nftig herausfordernd, vor allem wenn sie wie erwartet steigt. Mit food sharing k��nnen heute bereits einige Probleme in der Lebensmittelversorgung verbessert werden, aber aktuelle L��sungen sind nicht effizient genug. Ein smarter food sharing Prozess, wie in dieser Arbeit beschrieben, sollte daher so bald als m��glich eingesetzt werden., Everybody needs food to survive and therefore an appropriate supply of all people with foodstuff is of enormous importance. Obvious this is still not ensured - food waste and undernourishment are direct consequences. Lately, emerging food sharing initiatives try to solve these problem by redistribution of surplus food products, but they are not effective enough. So, the aim of this work is to describe an improved food sharing process. Therefore, the thesis deals with numerous aspects of food sharing and provides several prospects to improve current food sharing solutions. Especially the coordination problem that is inherent to food sharing is considered in detail and solved with a coordination framework called Peer Model. The thesis starts with literature review to get insights about food waste as well as undernourishment and evaluates the possible potential of food sharing. Then it analyses current food sharing solutions and research about food sharing and names improvement opportunities. Afterward it describes several information and communication technologies that can be used for an enhanced food sharing process. At that it compares popular coordination frameworks and comes to the result that the Peer Model is suited best to deal with the coordination problem of food sharing. Thus, the thesis models the coordination parts of food sharing with the Peer Model, implements them for its Java runtime environment and a simulation shows that the model works. Finally, the work describes an advanced smart food sharing process and storyboards illustrate how users interact with it. Supplying the world's population with food will be a challenging task in the future, particularly when the expected growth in population occurs. Today's food sharing solutions can improve some problems of the food supply chain, but current solutions are not efficient enough. Therefore, a smart food sharing process like the one described in this thesis should be implemented as soon as possible.

Over the last few years consumer engagement has highly attracted the interest of marketing academics and practitioners. The present paper reports the preliminary findings of an ongoing research project on the development of a model for consumer engagement in eWOM via social media. It is suggested that the consumer engagement process in social media moves through four sequential stages; participation, engrossment, emotion sharing and relationship building (the PEER model). The validity of the model in the context of twitter was empirically tested through the use of an online version of the LIWC software. Overall 10.801 tweets for two highly advertised mega-brand events (organized by Apple and Nintendo in the USA) were collected and text analyzed providing strong support to the model.

13th Australasian Conference on Information Security and Privacy, ACISP 2008; Wollongong, NSW; Australia; 7 July 2008 through 9 July 2008, In the pre-specified peer model for key agreement, it is assumed that a party knows the identifier of its intended communicating peer when it commences a protocol run. On the other hand, a party in the post-specified peer model for key agreement does not know the identifier of its communicating peer at the outset, but learns the identifier during the protocol run. In this paper we compare the security assurances provided by the Canetti-Krawczyk security definitions for key agreement in the pre- and post-specified peer models. We give examples of protocols that are secure in one model but insecure in the other. We also enhance the Canetti-Krawczyk security models and definitions to encompass a class of protocols that are executable and secure in both the pre- and post-specified peer models. © 2008 Springer-Verlag Berlin Heidelberg.