Search

The thesis will present different ways of using artificial intelligence in computer games with an emphasis on a tool that uses deep reinforcement learning. We will go deep into the rabbit hole that is game AI and try to understand years of research. The whole history of game AI will not be covered, but it will be enough to figure out what it means. V diplomski nalogi bodo predstavljeni različni načini uporabe umetne inteligence v računalniških igrah s poudarkom na orodju, ki uporablja globoko spodbujevano učenje. Poglobili se bomo v zajčjo luknjo, ki je umetna inteligenca iger in poskušali razumeti leta raziskav. Celotne zgodovine umetne inteligence iger ne bomo zajeli, vendar bo dovolj da ugotovimo kaj pomeni.

U radu su opisane kriptovalute, njihove prednosti i rizici. Objašnjen je pojam blockchaina, način rada kriptovaluta, kripto novčanika i platformi za trgovanje kriptovalutama. Također, objašnjeni su principi podržanog učenja, objašnjena je veza između pojmova i prijašnja istraživanja na temu podržanog učenja kod kriptovaluta. Implementirani modeli simulacije okruženja za trgovinu kriptovalutama, priloženi rezultati, mišljenja i zaključak. This paper deals with the advantages and risks of cryptocurrencies. It also explains blockchain, ways people use cryptocurrencies, crypto wallets and exchanges. This paper also talks about reinforcement learning principles, the connection between cryptocurrencies and reinforcement learning, and previous work done in this field. Following these topics, reinforcement learning models are created to simulate cryptocurrency trading inside an environment.

Prepoznavanje ručno pisanih znamenki jedan je od pionirskih problema računalnog vida. Za rješavanje problema korištene su razne varijacije klasifikatora, metode temeljene na izračunima udaljenosti, slojevite potpuno povezane neuronske mreže, ali prekretnicu predstavljaju Yann LeCun i njegovi suradnici koji su 1998. prikupili MNIST bazu podataka i na njoj istrenirali prvu konvolucijsku neuronsku mrežu naziva LeNet 1. Za konvolucijsku mrežu specifični su konvolucijski sloj u kojem se izdvajaju bitne značajke slike te sloj sažimanja koji doprinosi točnosti i smanjuje računalnu zahtjevnost mreže. LeNet mreže postižu dobre rezultate na neviđenim podacima uz prihvatljivu memorijsku opterećenost. U današnje vrijeme modeliranje konvolucijske mreže jednostavno je zbog široke dostupnosti jakih računala i raznih programskih biblioteka za modele dubokog učenja. Koristeći Python i biblioteku Keras okvira TensorFlow izgradila sam konvolucijsku mrežu koja se u kratkom vremenskom roku istrenira te prepoznaje nove primjere znamenki s velikom pouzdanošću. Handwritten digit recognition is one of the pioneer problems in the field of computer vision. Many methods were used to solve this problem, various versions of linear classifiers, different distance classifiers, fully connected multy-layer networks, but the turning point happened in 1998. when Yann LeCun and his associates collected the MNIST database and used it to train the first ever convolutional neural network named LeNet 1. A convolutional network is made of two special layers: a convolutional layer that extracts the important features of the picture and a pooling layer that improves network's predicition skills and decreases network's computational complexity. LeNet networks achieve good results on unseen data with acceptable memory requirements. These days it is easy to build a convolutional neural network because of the availabilty of powerful computers and many deep learning programing libraries. Using Python and TensorFlow's Keras API I have created a convolutional network that needed a small amount of time to be trained and is quick and reliable in recognising a new digit example.

U diplomskom radu naveden je utjecaj napretka metoda dubokog učenja na zastupljenost njihove upotrebe u područjima računalnog vida i problematike 3D rekonstrukcije. Navedene su prednosti i mane upotrebe dosadašnjih rješenja za 3D rekonstrukciju koja ne koriste metode dubokog učenja. Objašnjena je važnost digitalizacije ljudskog tijela i njegove primjene u područjima virtualne i proširene stvarnosti. Opisuju se rješenja koja koriste konvolucijske neuronske mreže za problematiku 3D rekonstrukcije ljudskog tijela te njihova podjela prema obliku i načinu prikupljanja skupova podataka za treniranje. Navedene su arhitekture slojeva odabranih neuronskih mreža i prikupljenih podataka za treniranje. Također je opisan postupak predobrade 3D modela i slika u svrhu generiranja sintetiziranih podataka za treniranje mreže. Mreže, implementirane u programskom jeziku Python upotrebom biblioteke Tensorflow, eksperimentalno su evaluirane u procesu treniranja, a rezultati analizirani i uspoređeni ovisno o tipu korištenih sintetiziranih podataka. This Master's thesis presents the impact of the progress of deep learning methods on the prevalence of their use in the fields of computer vision and the task of 3D reconstruction. The advantages and disadvantages of using existing solutions for 3D reconstruction without deep learning methods are listed. The importance of digitalization of the human body and its application in the fields of virtual and augmented realities is explained. Solutions using convolutional neural networks for the task of 3D human body reconstruction and their division according to the form and method of dataset collection is described. Layer architectures of selected neural networks and collected datasets for training are listed. Moreover, the method of preprocessing 3D models and images for the purpose of generating synthesized data for training the networks is described. Networks, implemented in the Python programming language using the Tensorflow library, are experimentally evaluated in the process of training, and the results are analyzed and compared according to the use type of synthesized data that is used.

Background: The environment in swimming pools, which contain chlorine, might interact with the airway epithelium, resulting in oxidative stress and/or inflammation during high intensity training periods. Methods: We evaluated pulmonary functional (metacholine challenge test, FEV1 and VC), cellular (eosinophils and neutrophils), inflammatory (FeNo, IL-5, IL-6, IL-8 and TNF-[alpha]), oxidative (8-isoprostanes) and angiogenesis factors (VEGF) in induced sputum and peripheral blood of 41 healthy non-asthmatic elite swimmers (median 16 years) during the period of high intensity training before a national championship. The second paired sampling was performed seven months later after training had been stopped for one month. Results: There was a ten-fold increase (median 82-924 pg/ml P < 0.001) in 8-isoprostanes in induced sputum and five-fold increase (median 82-924 pg/ml P < 0.001) in sera during training in comparison to the period of rest. However, there was no difference in FEV1 (113 vs 116%), VC (119 vs 118%), FeNo (median 34 vs 38 ppb), eosinophils (2.7 vs 2.9% in sputum 180 vs 165 cells/[micro]l in blood), neutrophils, different cytokines or VEGF in induced sputum or sera. The only exception was TNF-[alpha], which was moderately increased in sera (median 23 vs 40 pg/ml P=0.02) during the peak training period. Almost half (18 of 41) of swimmers showed bronchial hyperresponsiveness during the peak training period (PC20 cutoff was 4 mg/ml). There was no correlation between hyperresponsiveness and the markers of oxidative stress or inflammation. Conclusions: High intensity training in healthy, non-asthmatic competitive swimmers results in marked oxidative stress at the airway and systemic levels, but does not lead to airway inflammation. However, we could not confirm that oxidative stress is associated with bronchial hyperresponsiveness (AHR), which is often observed during the peak exercise training period.

The paper offers an answer to the question given by the title, starting from the revision of some specific elements of the appearance of the philosophical counsellor in the business and organisational environment, but also of the coaching as a service for the leaders, and it presents the characteristics of each one. The work makes a comparative analysis between coaching and philosophical counselling to observe the similarities and differences between the two applications that can act successfully in the field of building effective leadership in the business field. The concept of philosophical leadership is presented as a result of the application of philosophy in the business area. It leads implicitly to launching a collaboration proposal between the two practices, which can be materialised by issuing common services for the leadership, but also by taking over some specific tools from philosophy, now implemented in practice in the organisational and business area., Rad pruža odgovor na pitanje postavljeno naslovom, počinjući od revizije nekih karakterističnih sastavnica pojavnosti filozofijskih savjetnika u poslovanju i organizacijskom okruženju, ali i u treniranju kao usluzi za vođe, te se predstavljaju posebnosti svake. Izvodi se komparativna analiza između treniranja i filozofijskog savjetovanja radi motrenja sličnosti i razlike između dviju primjena koje uspješno mogu djelovati na području građenja efikasnog vodstva u području poslovanja. Pojam filozofijskog vodstva predstavljen je kao ishod primjene filozofije u području poslovanja. Implicitno vodi do lansiranja suradnje između dviju praksi, što se može materijalizirati tvorbom zajedničke usluge za vodstvo, ali i preuzimanjem nekih specifičnih alata iz filozofije, sada implementiranih u praksi u području organizacije i poslovanja., Die Arbeit gibt eine Antwort auf die Frage, die der Titel stellt, beginnend mit einer Revision einiger charakteristischer Komponenten des Auftretens philosophischer Berater im geschäftlichen und organisatorischen Umfeld, aber auch beim Coaching als Dienstleistung für Führungskräfte, wobei Besonderheiten jeder einzelnen Komponente wiedergegeben werden. Es wird eine vergleichende Analyse zwischen Coaching und philosophischer Beratung durchgeführt, um die Ähnlichkeiten und Unterschiede zwischen den beiden Anwendungen zu beobachten, die erfolgreich im Sektor des Aufbaus einer effizienten Führung im Geschäftsbereich eingesetzt werden können. Der Begriff der philosophischen Führung wird als Resultat der Anwendung der Philosophie im Geschäftsbereich präsentiert. Dies führt implizit zur Einleitung einer Zusammenarbeit zwischen den beiden Praktiken, die durch die Schaffung eines gemeinsamen Dienstes für die Führungskräfte materialisiert werden kann, indessen auch durch die Übernahme einiger spezifischer Instrumente aus der Philosophie, die gegenwärtig in der Praxis auf dem Gebiet Organisation und Geschäft implementiert werden können., Ce travail offre une réponse à la question posée dans le titre en commençant par réviser certains éléments caractéristiques liés à l’apparition de conseillers philosophiques dans le domaine professionnel et organisationnel, mais également dans la formation des leaders comme service offert pour eux, pour présenter ensuite la particularité de chacun des éléments. Une analyse comparative est réalisée entre la formation et le conseil philosophique afin d’observer les ressemblances et les différences entre ces deux activités qui peuvent être bénéfiques dans le domaine visant à construire un leadership efficace dans le domaine professionnel. Le concept de leadership philosophique est présente comme le résultat de l’activité philosophique dans le domaine professionnel. Elle mène de manière implicite à l’ouverture d’une collaboration entre deux activités, ce qui peut être concrétisé en créant des services communs pour le leadership, mais également en reprenant certains outils de la philosophie, qui sont à présent mis en œuvre dans la pratique dans le domaine de l’organisation et des affaires.

Ovaj rad bavi se problematikom autonomne bespilotne letjelice. Na početku opisujemo vrstu letjelice koju ćemo koristiti u implementaciji I način na koji ista leti. Zatim su ukratko opisane umjetne neuronske mreže i potporno učenje, dva pristupa umjetne inteligencije ključna za razumjevanje dubokog potpornog učenja. Opisujemo PPO, metodu dubokog potpornog učenja, odabranu za učenje željenog ponašanja naše letjelice. U Unity Editor-u izrađujemo model letjelice i pomoću ML- Agents alata pripremamo okolinu za učenje. Prevodimo naš problem u sustav nagrada kojima letjelicu navodimo na željeno ponašanje. Nakon završenog učenja grafički prikazujemo dobivene rezultate. The topic of this essay is the problematics of autonomous unmanned aircrafts. In the beginning, we described a type of aircraft that is used in implementation, along with its manner of flying. What follows is a brief description of artificial neural networks and reinforced learning, two approaches in artificial intelligence that are key to understanding deep reinforced learning. We also described PPO, a method of deep reinforced learning, chosen for the purpose of teaching our aircraft a desired behavior. In the Unity Editor, we made a model of the aircraft, and with the ML-Agents tool we prepared an environment for learning. We translated our problem into a reward system, which is used to lead the aircraft onto desired behavior. After the learning process is done, the obtained results are displayed graphically.

Banke se često nalaze u nedoumici prilikom sklapanja ugovora o kreditu jer ne mogu znati hoće li se klijent držati ugovora, zbog toga banke razvijaju razne strategije kako bi mogle odrediti rizike i jedan od rizika je mogućnost da klijent vrati kredit prije dogovorenog ro­ka.U ovom radu opisan pristup određivanju rizika preuranjenog vraćanja kredita. Opisani su postupci obrade skupa podataka. Upoznavanje s algoritmom XGBoost-a i svim nje­govim značajkama koji ga dovode u prednost naspram ostalih algoritama strojnog učenja. Izgradnja inicijalnog modela za procjenu rizika preuranjenog vraćanja kredita te detaljni opis koraka koje je potrebno izvesti kako bi se parametri XGBoost-a podesili na vrijednosti koji će poboljšati točnost modela. Banks often find themselves in doubt when concluding a loan agreement because they can not know if a client will behave as it was agreed in contracts, so banks develop various strategies to determine risks and one of the risks is if the client pay off their loan earlier. In this paper is described approach to determining the risk of early repayment of the loan. Also, the data set processing procedures are described. Acquainted with the XGBoost al­gorithm and all its features that take precedence over other machine learning algorithms.It is built an initial model for estimating the risk of premature loan repayment and a detailed description of the steps that need to be performed to set the XGBoost parameters to a value that will improve the accuracy of the model.

U ovom radu je kronološki prikazan razvoj modela i tehnika korištenih za obavljanje zadataka računalnog vida, detekcije objekata na slikama i klasifikacije. Upoznali smo se sa osnovnom idejom strojnog i dubokog učenja kao i sa razlikama među njima. Poseban naglasak je stavljen na važnost nastanka i razvoja umjetnih neuronskih mreža, alata koji stoje iza najvećih postignuća umjetne inteligencije modernog doba. U konačnici je prikazan cjelokupan proces prikupljanja podataka (slika) te izrade i primjene detektora voća na slikama. Detektor je treniran koristeći TensorFlow GPU inačicu poznate Googleove biblioteke strojnog učenja. This work presents the chronological development of models and techniques used to perform computer vision tasks, object detection at the images and classifications. We met with the basic concept of machine and deep learning as well as the differences between them. Special emphasis has been put on the importance of the emergence and development of artificial neural networks, the tools which stays behind the greatest achievements of the modern-day intelligence. At the end, there is described entire process of data (pictures) collecting and the creation and application of the fruit detector in the pictures. Detector is trained using the TensorFlow GPU version of the famous Google Machine Learning library.

Cilj ovog diplomskog rada je koristeći predtrenirane konvolucijske neuronske mreže procijeniti spol osobe na temelju ortopantomograma. Rad je podijeljen u tri dijela. U prvom dijelu se govori o povijesti neuronskih mreža, njihovim arhitekturama, tipovima neuronskih mreža. Takod̄er je objašnjeno na koji način rade i umjetne i konvolucijske neuronske mreže. U drugom dijelu je pojašnjen pristup rješenju, koji je podijeljen u nekoliko djelova. Za svaki dio je objašnjeno zašto je potreban, te je prikazan dio koda koji obavlja tu zadaću. Posljednji dio rada prikazuje dobivene rezultate za svaku kori- štenu neuronsku mrežu. Postignuti rezultati za svaku mrežu se med̄usobno uspored̄uju, a i uspored̄uju se ortopantomogrami koji su krivo procjenjeni. Theme of this paper is by using the pretrained convolutional neural networks to determine the sex of the person using the given X-ray image of the jaw. Paper is divided into three parts. First part explains what are neural networks, talks about history of neural networks, their architectures and types. It is also explained how both artificial and convolutional neural networks work. Theme of the second part of the paper is the solution of the problem. The solution is devided in several parts. For every part it is explained why why it is necessary and there is part of code which solves the issue. In the last part of the paper results are shown and compared.

Konvolucijske neuronske mreže danas su često korištene za klasifikacijsku ulaznih podataka u pojedine moguće razrede za koje je model prethodno istreniran. Neovisno o testnom podatku, model će uvijek s određenom sigurnošću predvidjeti pripradnost ulaza u jedan od klasifikacijskih razreda. Ključna problematika ovog pristupa je nemogućnost otkrivanja ulaznih podataka za koje će model najvjerojatnije pogriješiti. Upravo zbog toga potrebno je otkriti izvandistribucijske podatke dovedene na ulaz modela, a to činimo analizom njegovog izlaza. Kod takvih podataka nikad ne možemo sigurno tvrditi jesu li dijelom distribucije na kojoj je model treniran ili su vandistribucijski, no vjerojatnost detekcije izvandistribucijskih podataka moguće je odrediti na više načina, npr. korištenjem AUROC i AUPR površina. Convolutional neural networks are nowadays often used for classifying input data into specific possible classes for which the model was previously trained. Whatever test data we provide to the model, it will always predict the result corresponding class with specific certainty. The key problem with this approach is an inability of determining the data for which the model will most likely make a wrong prediction. Specifically because of that, it is a necessity to determine out-of-distribution data provided as input to the model, and the way to do that is by analyzing its output. Despite the fact that we cannot claim if the test data belongs to the distribution on which the model was trained or it is out-of-distribution, the possibility of detecting out-of-distribution data can be determined with various methods, including using AUROC and AUPR areas.

Protinapad je pomemben del rokometne igre, pri katerem se uporabljajo vsi elementi, tako tehnični kot taktični. S protinapadi se igra sodobna hitra igra rokometa in zato število zadetkov na tekmah narašča. Učinkovitost je v protinapadu večja kot v igri na postavljeno obrambo. Protinapad se lahko izvede takoj, ko ekipa pridobi žogo v svojo posest in prenese žogo na nasprotnikovo polovico. Poznamo več učinkovitih oblik izvedbe protinapada, zato je potrebna dobra analiza lastnih igralcev in temu primerna izbira protinapada. Protinapad treniramo postopoma z osnovnimi elementi igre v individulani obliki in šele nato v paru in skupini. Za dobro nadaljne delo je potrebno obvladovanje igre na takem nivoju, da lahko izvajamo gibanja v visoki hitrosti. Taktično lahko protinapad izvedemo na več načinov, predvsem pa tako, da otežimo delo nasprotni ekipi, ki se še ni postavila v obrambo. Za dobro izpeljavo pa morajo biti vsi igralci seznanjeni z njihovo nalogo v napadu in s postavljanjem v prehodu v napad ter s kombinatoriko na nasprotnikovi polovici. V nalogi bom predstavil posamezen protinapad in metodiko postopnega treniranja protinapada. Tako se lahko starejši in mlajši igralci postopoma in lažje naučijo ta del. A counterattack is an important part of handball game, where both, the technical and the tactical elements are used. By counterattack modern and quick handball is played and therefore the number of goals at matches is increasing. The effectiveness in the counterattack is better than in the game played against the set defensive formation. The counter attack is executed right after the team gets the ball in its possession and passes it to the opponent´s half of the court. We know many effective ways of executing the counterattack, thus a thorough analysis of the players is needed to choose the appropriate counterattack. We train the counterattack gradually through the basics elements of the game, first individually and then in a pai or in a group. To achieve quality future work, it is necessary to master the game at the level where the movements are done at high speed. The counterattack can be tactically executed in many ways, mainly to make the work of the opponent team, who has not form the defence yet, more difficult. To execute it well, all the players need to know their tasks in the attack, their positions in switching to attack and the combinations at the opponent´s half. In the thesis I will present the individual counterattack and the methodology of gradually training it. The young and the more mature players learn its accurate execution step by step.

Cilj ovog rada je opisati primjenu dubokih neuronskih mreža u klasifikaciji odsječaka teksta različite duljine s obzirom na jednu pripadajuću kategoriju. U prvom dijelu objašnjavaju se motivacija za korištenje modela u vektorskom prostoru te osnovni koncepti u pozadini Word2vec modela. Predstavljena je i detaljno analizirana korištena arhitektura konvolucijske neuronske mreže s naglaskom na inicijalizaciju težina, aktivacijsku funkciju, metodu optimizacije, funkciju pogreške te regularizaciju. U drugom dijelu opisani su korišteni skupovi podataka za treniranje modela neuronske mreže te su navedeni eksperimentalni rezultati dobiveni koristeći TensorFlow biblioteku u okviru programskog jezika Python. The aim of this paper is to describe the use of deep neural networks in the single-label classification of various length texts. In the first part, the motivation for vector space models and the basic concepts behind the Word2vec model are explained. Then the used convolutional neural network architecture is analyzed in detail, with emphasis on weights initialization, activation function, optimization method, error function and regularization. In the second part, the datasets used to train the neural network model are described and experimental results are reported. The results were obtained using the TensorFlow library and the Python programming language.

Namen diplomskega dela je analizirati vprašanje, ali sta nasilje in agresivnost povezana s treniranjem borilnih veščin, in sicer z judom in mešanimi borilnimi veščinami (MMA). V delu smo predstavili tradicionalno borilno veščino JUDO in mešane borilne veščine (MMA) vse od nastanka, začetkov v Sloveniji, športnih pravil, do tehnike in tekmovanj. V nadaljevanju diplomskega dela smo skušali povezati koncept agresivnosti in nasilja z omenjenima borilnima veščinama. Sprva smo definirali nasilje, iskali vzroke zanj in navedli nekaj vrst nasilja. Preverjali smo, ali je agresivnost v borilnih športih dobra in kako jo obvladati. Namen treniranja teh dveh borilnih veščin vsekakor ni spodbujanje agresivnosti in nasilja, temveč krepiti telo in duha ter ne nazadnje nadzor čustev in gradnja samozavesti. Nekaj besed smo namenili poškodbam v borilnih športih in nasprotnikom borilnih veščin ter predstavili različne poglede avtorjev o prednostih borilnih veščin in pozitivnih vplivih borilnih veščin na agresivno vedenje. V empiričnem delu smo s pomočjo vprašalnikov, razdeljenih med udeležence vadbe juda, MMA in študentov Fakultete za varnostne vede, preverjali tri hipoteze. Ugotovili smo, da so udeleženci vadbe borilnih veščin večinoma neagresivne osebe, ki niso naklonjene nasilju v družbi. Za omenjena športa je potrebna popolna zbranost, volja in poslušnost, kar pa vsekakor ni vrlina agresivnežev in nasilnežev. Vedno bodo obstajali posamezniki, pri katerih bosta izstopala agresija in nasilje. Borilne veščine tega ne spodbujajo, temveč pomagajo posamezniku, da agresivnost premaga tako, da se jo nauči kontrolirati in nadzorovati. The whole purpose of this thesis is to confirm or disprove the thesis that violence and aggression are associated with martial arts. In our case, judo and mixed martial arts (MMA). In this thesis will be closely presented the traditional martial art Judo and mixed martial arts (MMA) since its creation, beginnings in Slovenia, sporting rules, techniques and competitions. In the continuation of the thesis we have tried to connect the concept of aggression and violence with this two martial arts. Initially, we define violence, searching for causes and indicated some types of violence. We examined whether the aggression in martial arts is good and how to tackle it. The purpose of the training of these two martial arts is definitely not encourage aggressiveness and violence, but to strengthen the mind and body, and, finally, control of emotions and building self-confidence. A few words are devoted to injuries in martial arts and martial arts opponents, presented some different views of some authors of the advantages and positive effects of martial arts to aggressive behavior. In the empirical part, we are using questionnaires, distributed to participants in sports of Judo, MMA and students of Faculty of Criminal Justice and security, to shed light on all three hypotheses which we have set ourselves. We found that the participants of martial arts are mostly non-aggressive people who are not inclined to violence in society. These two sports require full concentration, will and obedience, which is obviously not a virtue of aggressive people and bullies. Always there will be individuals for which will stand out aggression and violence. Martial arts do not encourage, but help individuals to overcome aggression, so that they can learn to control and monitor it.

Cilj ovog rada bio je napraviti web-aplikaciju koja bi pomogla nogometnih trenerima pri planiranju i evidentiranju svakodnevnih sportskih procesa za svoj tim. Korištenjem ove aplikacije, nogometni treneri će sigurno uštedjeti vrijeme pri planiranju treninga i taktika za utakmice te time dobiti mogućnost dizanja kvalitete svoga rada. Aplikacija nastala kao rezultat ovog rada prilagođena je svim internetskim preglednicima te ju je zbog prilagodljivog dizajna moguće ugodno koristiti preko preglednika na pametnim telefonima. Aplikacija je napravljena koristeći programski jezik C# i .NET Core Framework 1.1 te SQL bazu podataka. The goal of this thesis was to make a web-application which would help football coaches to plan and track everyday sport processes of their team. Using this application, football coaches will surely save their time while planning trainings and tactics for matches and raise the quality of their work. Application created as result of this thesis works on every web browser and due to its responsive design it is can be comfortably accessed and used on smartphones. This application is created using C# programming language, .NET Core Framework 1.1 and SQL database.

Moždani udar na prvom je mjestu uzroka invalidnosti u svijetu. Svake godine skoro 15 milijuna ljudi doživi moždani udar, od čega 6 milijuna umre. Oni koji prežive moždani udar, suočavaju se s posljedicama moždanog udara koji se manifestira kroz invalidnost, hendikep i smanjenu funkciju organizma u cjelini. Kroz proces rehabilitacije, bolesnicima se želi omogućiti ponovnu integraciju u društvo. Nositelj oporavka bolesnika je rehabilitacijska medicina, a jedan od glavnih zadataka i najveća prepreka bolesnicima jest oporavak hoda. Pored konvencionalnih tehnika, koje provode fizioterapeuti, u zadnjih deset godina sve je prisutnija uporaba robota u rehabilitaciji hoda. Robotima se pokušava poboljšati efikasnost terapije i napraviti bolje rezultate u konačnom oporavku bolesnika. Roboti se u grubo mogu podjeliti na dvije skupine, egzoskeletne i end effector robote. Obje vrste robota pokazale su pozitivne rezultate u praktičnom radu s bolesnicima. Pojedine veće studije su dokazale korisnost korištenja robotske tehnologije uz konvencionalnu tehniku. Dokazano je da bolesnici koji prolaze kroz robotsku terapiju, uz konvnencionalnu imaju više rezultate na testovima procjene oporavka od bolesnika koji su prolazili samo konvencionalnu terapiju. S ekonomskog aspekta, implementacija robotike u rehabilitacijsku medicinu je izuzetno skup proces s visokim kratkoročnim troškovima. No ako se uzmu u obzir primjeri iz prakse, treba razmišljati dugoročno i sagledati uštedu koja može doći nakon dužeg vremenskog perioda. Najveći dio toga proizlazi iz rasterećenja fizioterapeutskog osoblja, jer je vježba oporavka hoda najzahtjevnija za fizioterapeute u rehabilitacijskoj medicini., Stroke is in the first place cause of disability in the world. Every year nearly 15 million people suffer from stroke, of which 6 million die. Those who survive stroke, are faced with the consequences of it, which is manifested through disability, handicap and reduced function of the organism as a whole. Through the process of rehabilitation, patients is allowed an oportunity to re-integrate into society. Main protagonist of recovery is rehabilitation medicine, and one of the main tasks and the biggest obstacle to patients recovery is walk. In addition to conventional techniques, carried out by physiotherapists, in the last ten years use of robots in rehabilitation walk is very common. Robots are there to improve the efficiency of the therapy and make better results in the final recovery of patients. Robots can be roughly divided into two groups, exoskeletal and end effector robots. Both types of robots have shown positive results in practical work with patients. Some major studies have proven the usefulness of using robotic technology with the conventional technique. It has been shown that patients undergoing robotic therapy, alongside with conventional therapy, have higher scores on functional tests than patients who have passed only conventional therapy. From an economic point of view, the implementation of robotics in rehabilitation medicine is very expensive process with high short-term costs. But if we take into account the examples from practise, we need to think long term and consider savings that may occur after a long period of time. Most of that comes from relieving physiotherapy staff, because walk recovery is the most challenging exercise for physical therapists in rehabilitation medicine.

The purpose of this study was to examine the influence of adopting an external, internal, and neutral focus of attention while executing an underhand and overhead shot put. Using a counterbalanced within-participant design, thirty highly skilled athletes (height 183±6 cm, body mass 78±7 kg, age 22.2±2.4 years) performed five underhand and five overhead shot puts. The results indicated that the shot put distances were greater (p, Cilj ovog istraživanja bio je utvrđivanje utjecaja usvajanja vanjskog, unutarnjeg ili neutralnog fokusa pažnje tijekom izvođenja bacanja kugle pothvatom naprijed te preko glave. Korištenjem eksperimentalnog nacrta s uravnoteženim ponovljenim mjerenjima unutar ispitanika 30 vrhunski obučenih sportaša (tjelesne visine 183±6 cm, tjelesne težine 78±7 kg u dobi od 22,2±2,4 godine) izvelo je pet izbačaja kugle pothvatom prema naprijed i pet izbačaja preko glave. Rezultati su pokazali je udaljenost bacanja bila veća (p

Cilj našeg istraživanja bio je pronaći razlike između trenera različitih sportova, u dimenzijama zadovoljstva životom i općeg životnog iskustva, za skup nezavisnih varijabli, sociodemografskih te vezanih uz bivšu sportsku karijeru te trenerski posao. Ispitali smo ukupno 132 trenera različitih sportova u Hrvatskoj, tijekom seminara za trenere Hrvatske olimpijske akademije. Za mjerenje zadovoljstva životom, koristili smo Proširenu skalu zadovoljstva životom i skalu Općeg životnog iskustva. Najveći broj statistički značajnih razlika među trenerima pronađen je u području rodnih razlika, što smo protumačili potencijalnim rodnim ulogama i stereotipima. Statistički značajne razlike pronađene su u dimenziji Životno postignuće (veći rezultati kod neoženjenih/ neudanih); u dimenziji Fitness (veći rezultati kod trenera koji su to postali odmah nakon sportske karijere); u dimenziji Zadovoljstvo i optimizam (veći rezultati kod trenera koji su ranije osvajali medalje na državnom prvenstvu). Treneri koji su se prestali baviti sportom zbog kritičnih događaja u životu, imaju veće prosječne rezultate u dimenziji Negativne emocije i stres. Pronašli smo mali broj statistički značajnih razlika među trenerima u zadovoljstvu životom i općem životnom iskustvu u odnosu na skup nezavisnih varijabli., The goal of our study was find the differences between the coaches in different sports, in the dimensions of life satisfaction and general life experience, in relation to the set of independent variables, sociomeographic and linked with earlier sport carierre and in variables directly linked with a coaching.We have examined a total of 132 coaches of various sports in Croatia during the seminar for coaches in Croatian Olympic Academy. For measuring satisfaction with life, we used Extended SatisfactionWith Life Scale and the Questionnaire about General Life Experience. The greatest number of statistically significant differences was found among coaches in the area of gender differences, that we have interpreted by the potential gender roles and stereotypes. Statistically significant differences were found in the dimension of Lifetime Achievement (higher scores for unmarried); in the dimension of fitness (higher scores for coaches who became a coach soon after his sports career); in the dimension of satisfaction and optimism (higher scores for coaches who have previously won medals at the state championship). Coaches who have stopped play sports because of the critical events in life, have higher average scores in the dimension of negative emotions and stress.We have found a small number of statistically significant differences among the coaches in the life satisfaction and general life experience, in relation to a set of independent variables.

The goal of our study was to fi nd the latent dimensions of the satisfaction with life and general life experience at former athletes (coaches), as well as to fi nd correlations between those dimensions with sociodemographic and variables connected with coaching. We have examined a total of 132 coaches of various sports in Croatia. For measuring satisfaction with life and general life experience, we used Extended Satisfaction With Life Scale (ESWL) and General Life Experience Scale (GLES). We have found three latent dimensions of the satisfaction with life: business and social life, fi tness and life achievement. Four latent dimensions of the general life experience were found: satisfaction and optimism, negative emotions and stress, life quality and daily fatigue. No signifi cant correlations were found between sociodemographic variables two items that describe the coach as someone who could have to be elite former athlete or purposefully educated for coaching. Age and work experience are negatively correlated with business and social life and daily fatigue, but positive with general life achievement. Years in top sport and end of professional career are negatively correlated with business and social life, but positive with general life achievement and negative emotions and stress. We can cautiously say that period of engagement in top sport have more negative than positive implications on subjective well-being.

Upravljanje s človeškimi viri zajema širok in učinkovit sistem ravnanja z ljudmi pri delu, nove stile vodenja in razumevanja psihologije človeka. Zaposleni so pri delu uspešni le, če vedo, katere so zahteve organizacije, oz. kaj se od njih pričakuje. V organizacijah, kjer managerji poskušajo doseči lastne cilje na račun zaposlenih, se v ljudeh razvija zgolj sposobnost, da ustrežejo drugim. Zaposlenim je potrebno omogočiti več, da razvijajo željo po proizvajanju, doseganju in odlikovanju. Le v tem primeru je mogoče ustvariti dobičke, doseči produktivnost, rast in pozitivno spremembo. Zaradi navedenega sem v nalogi namenila poudarek novim stilov vodenja kjer je pomembno znanje motivacije, komunikacije, letnih razgovorov, coachinga, kariernega razvoja, nagrajevanja, ugotavljanja veščin in kompetenc zaposlenih. Zadovoljstvo zaposlenih pri delu se odraža ravno preko naštetih znanj, ki jih uporabljajo managerji pri dobrem upravljanju s človeškimi viri. V organizacijah je zato priporočljivo letno merjenje zadovoljstva oz. klime kjer se lahko preko anket in analiz odkrijejo morebitne pomanjkljivosti, ki lahko služijo nenehnim izboljšavam. Z diplomskim delom smo želeli bolje spoznati odnose med vodji in zaposlenimi v podjetju CM Celje, d.d. in ugotoviti določene probleme, ki jim v podjetju posvečajo premalo pozornosti. Podali smo predloge za učinkovitejše upravljanje s človeškimi viri, kot je vpeljava letnih razgovorov na vseh ravneh (oddelkih, enotah) družbe. Human Resource Management is consisted from wide and effective handeling with people, new of ways leadership styles and understanding psichology of a human. Employees are succesfull and happy only if they now which are requirements of a organization, and what is expected from them. In those organizations where managers are trying to achieve their own goals, employees are developing only ability to please others. We have to be able to give more to our employees . They need to develop a desire to produce, reach and take a credit for successfull things. Only in those cases it is possible to make profits, reach productivity, growth and positive changes. Because of all those things we dedicated special attention to new styles of leadership, where it is important to have knowledge from motivation, communication, regular annual talks, coaching, career developement, rewarding, finding skills and competencies of employees. Satisfaction with work shows through many skills and knowledge which are used by managers in human resource management. In organizations it is highly recommended to make annual measurements of satisfaction or organizational climate, where through polls and analysis drawbacks can be discovered, which are useful in continuous improvements. We asked ourself two basic questions, to know better relations between leaders and employees in choosen company, and to warn at problems, which company doesn't pay enough attention. We have given a few proposals for more efficient human resource management, like annual interviews on every level (departments, units) of a company.

U ovom radu opisuje se jedan od najatraktivnijih elemenata tehnike u suvremenoj košarci, a to je skok-šut. Utvrđen je način izvođenja skok-šuta u košarkaškoj igri s gledišta strukturalne analize. Također, selektirani su metodski postupci učenja i treniranja skok - šuta za sve uzrasne kategorije.