Search

Digitisation of archival materials is a lengthy process in which the act of digitisation is usually the shortest and the most straightforward one. It is the processing of the digitised materials that take a lot of time and effort. It is valuable to have digitised materials available online, even for the reason of not having to travel to the physical location of the originals, but the tendency of extracting additional value from the materials increases. To that end, the archival materials can be processed by new techniques such as artificial intelligence (AI) and crowdsourcing. The authors use the example of digitisation of the minutes from the Faculty (of Humanities and Social Sciences) Council meetings dating from 1913 to 1996 and the application of AI to improve the OCR results and NER for semantic enrichment as well the example of digitisation of the food rationing cards, used between 1941 and 1945 in Zagreb, Croatia and the application of AI and crowdsourcing to data extraction, analysis and visualisation.

Cilj ovoga rada bio je istražiti mogućnosti GPT-3 jezičnog modela na platformi GPT-3 Playground te istražiti kako kompanija OpenAI približava implementaciju umjetne inteligencije široj publici. U teorijskom dijelu rada analizirana je ljudska inteligencija i ponašanje u usporedbi s umjetnom inteligencijom i racionalnim ponašanjem. Prikazana je tehnologija transformer modela te način na koji se mijenja pristup računalnoj obradi prirodnih jezika. Prikazan je razvoj GPT modela od prve do treće generacije. U istraživačkom dijelu analizirane su tri različite primjene GPT-3 transformer modela: pojednostavljivanje i sažimanje teksta, pretvaranje Python koda u tekst i chatbot za pitanja i odgovore. Svaka od primjena analizirana je na četiri modela koji su trenirani na različitoj količini podataka. Rezultati pokazuju da je GPT transformer model vrlo napredna tehnologija te da su mogućnosti primjene te tehnologije raznolike. Očituje se da je model Davinci, koji je imao najviše podataka za treniranje i najveći broj parametara najuspješnije obavio sva tri zadatka. Model Curie je približno uspješan kao model Davinci, no naišao je na neke probleme sa razumijevanjem konteksta, ali pokazao se vrlo uspješnim u generiranju prijateljskog govora. Model Babbage je imao više problema sa generiranjem teksta koji je nalik na ljudski govor iz razloga što je ponavljao fraze, te nije koristio gramatičke oblike koji bi tekst napravili prirodnijim. Model Ada je imao velikih problema sa razumijevanjem rečenice ako mu nije bilo dano dovoljno konteksta. The goal of this research paper was to explore the possibilities of the GPT-3 language model on the platform GPT-3 Playground and to see how the company OpenAI brings the implementation of artificial intelligence closer to the public. The theoretical part of the research paper analyses human intelligence and behaviour in comparison to artificial intelligence and rational behaviour. It also shows the transformer technology and the way that it changes the approach to natural language processing. The paper presents the development of the GPT model from the first to the third generation. The research part of the paper analyses three different implementations of the GPT-3 transformer model: text simplification and summarization, translating Python code to natural text, and a Q&A chatbot. Each of the implementations is analysed on four models that were trained on different amounts of data. The results show that the GPT transformer model is a highly advanced technology and that the possibilities of the implementation of this technology are diverse. The research shows that the model Davinci, which had the most training data and the largest number of parameters, was the most successful at completing the three tasks. The model Curie was nearly as successful as Davinci, however it ran into some problems with understanding the context of a text, but it proved to be successful at generating friendly chat. The model Babbage had more problems with generating natural language text because it repeated phrases and it did not use some grammatical forms that would make the text sound more natural. The model Ada had large problems with understanding sentences if it was not given enough context.

Cilj završnog rada je izgradnja online rječnika hrvatsko-poljskih lažnih prijatelja. Rad se sastoji od teorijskog i praktičnog dijela. U teorijskom prikazana je uloga tehnologije u sporazumijevanju te uloga rječnika u komunikaciji i jezičnim tehnologijama. Analizirane su karakteristike online rječnika, kao i karakteristike rječnika lažnih prijatelja. Objašnjen je fenomen lažnih prijatelja kao aspekt šuma u komunikacijskom kanalu te upotreba online rječnika za smanjenje šuma i poboljšanje komunikacije. U praktičnom djelu rada prikazan je način izgradnje online rječnika uz pomoć alata HTML, CSS i PHP. Provedena je analiza online rječničke strukture te su prikupljeni podaci za izradu stranice. Rječnički korpus pohranjen je u SQL bazi, dok se podaci filtriraju i prikazuju na stranici iz tabličnih zapisa. Kao produkt rada dobivena je web-stranica online rječnika hrvatsko poljskih lažnih prijatelja koja može koristiti raznim istraživanjima online rječnika ili međujezične sličnosti. The aim of the final paper is to build an online dictionary of Croatian-Polish false friends. The paper consists of both theoretical and practical work. The theoretical part of the paper shows the role of modern technologies in communication and the role of dictionaries in human language technologies. The phenomenon of false friends is explained as an aspect of noise in the communication channel and the use of online dictionaries to reduce noise and improve communication. In the practical part of the paper, the method used for building an online dictionary with the help of HTML, CSS, and PHP tools is explained. An analysis of the online dictionary structure was carried out and data was collected for the creation of the site. The dictionary corpus is stored in the SQL database, while the data is filtered and displayed on the page from tabular records. A website, Online Dictionary of Croatian-Polish False Friends, was obtained as a product of the work, which can be used for various research purposes of online dictionaries or cross-linguistic similarities.

Analiza informacija nacionalnog identiteta ima važnu ulogu u osmišljavanju, razvoju i strategiji brendiranja neke države. Informacije dobivene mišljenjem stranih turista ukazuju na percepciju nacionalnog brenda te na kulturna i kompetitivna obilježja koja diferenciraju Hrvatsku na nacionalnoj i međunarodnoj razini. U radu će se kroz istraživanje na populaciji stranih državljana ispitati stavovi o kulturnom turizmu i dimenzijama brendiranje Hrvatske kao države. Rad se sastoji od teorijskog i praktičnog dijela. U teorijskom dijelu, u radu se prikazuje uloga tehnologije i pristup informacijama u doba globalizacije. Analizira se pojam kompetitivnog identiteta države na primjerima stranih država, kao i elementi koji prema međunarodnim kriterijima utječu na rangiranje neke države kao što su ekonomski učinak, digitalna potražnja, strategija brenda neke zemlje i online prisutnost. Prikazuje se važnost pristupa informacijama kroz pojam modernog turizma. U praktičnom dijelu provedeno je istraživanje na 175 stranih državljana – stranih turista, ali uglavnom studentske populacije, u Republici Hrvatskoj. Istraživanjem se ispituje percepcija Hrvatske kao države brenda kroz kulturni i kompetitivni identitet. Upitnik sadrži 31 pitanje u kojima se ispituju demografski podaci, preferencije vezane uz putovanje i odabir destinacije te stavovi vezani uz hrvatski nacionalni brend prema Simon Anholtovom (2005) modelu Indeksa nacionalnog brendiranja (eng. Nation Brands Index - NBI). Kroz šest dimenzija ispituje se percepcija stranih turista kroz sljedeće dimenzije: Ljudi, Upravljanje državom, Izvoz, Turizam, Kultura, Ulaganje i imigracije te se ispituju stavovi prema odabiru Hrvatske kao destinacije. Cilj ovog rada jest utvrditi jesu li obilježja modernog turizma primjenjiva na ispitanicima, odnosno odrediti preferencije ispitanika vezane uz putovanje i stavove prema elementima nacionalnog identiteta Hrvatske kao brenda. The analysis of national identity information plays an important role in the design, development, and branding strategy of a country. The information obtained from the opinion of foreign tourists indicates the perception of the national brand and the cultural and competitive features that differentiate Croatia at the national and international levels. The paper will examine attitudes about cultural tourism and branding dimensions of Croatia as a country through a survey of the population of foreign citizens. The work consists of a theoretical and a practical part. In the theoretical part, the paper presents the role of technology and access to information in the age of globalization. The concept of a country's competitive identity is analyzed using the examples of foreign countries, as well as elements that, according to international criteria, influence the ranking of a country, such as economic performance, digital demand, a country's brand strategy, and online presence. The importance of access to information through the concept of modern tourism is shown. In the practical part, research was carried out on 175 foreign citizens - foreign tourists, but mostly student population, in the Republic of Croatia. The research examines the perception of Croatia as a state brand through cultural and competitive identity. The questionnaire contains 31 questions that examine demographic data, preferences related to travel and destination selection, and attitudes related to the Croatian national brand according to Simon Anholt's (2005) Nation Brands Index (NBI) model. Through six dimensions, the perception of foreign tourists is examined through the following dimensions: People, Government, Export, Tourism, Culture, Investment, and immigration, and the attitudes towards choosing Croatia as a destination are examined. This paper aims to determine whether the features of modern tourism apply to the respondents, that is, to determine the respondents' preferences related to travel and attitudes towards the elements of Croatia's national identity as a brand.

Jedinstvenost i prepoznatljivost karakteristike su gotovo svake države na svijetu. Komunikacija nacionalnog brenda predstavlja važan segment u prepoznatljivosti zemlje te utječe na poslovanje, turizam, investicije, itd. Stoga analiza informacija kojima se procjenjuje razina nacionalnog brendiranja predstavlja važan korak u mjerenju napretka ostvarenog nacionalnim brendiranjem. Cilj ovog rada bio je provesti analizu informacija o Hrvatskoj kao nacionalnom brendu primjenom Indeksa nacionalnog brendiranja Republike Hrvatske kroz šest dimenzija: Ljudi, Upravljanje, Izvoz, Turizam, Ulaganja i imigracije, Kultura i baština. Mogući nedostaci kod provedenog istraživanja su dominacija dobi od 19 do 25 godina u odnosu na ostale dobne skupine koje su pristupile istraživanju, premali broj ispitanika koji su sudjelovali u istraživanju ali i neodgovaranje svih ispitanika na sva postavljena pitanja. Upitnik je ispunilo 342 ispitanika, od čega 167 domaćeg stanovništva i 175 stranog. Prevladavala je dob od 19 do 25 godina i kod domaćih ispitanika (60%) i kod stranih (66%). Rezultati istraživanja pokazuju da se percepcija domaćih i stranih ispitanika od prilike podudara što i dokazuju prosječne ocjene koje iznose od domaćih ispitanika 4,85 i od stranih 4,85. Prosječne ocjene domaćih i stranih ispitanika su jednake no ocjene se međusobno razlikuju za svaku dimenziju posebno. Također Hrvatska je sa svojom prosječnom ocjenom od 4,85 zauzela srednju poziciju. Navedeni rezultati ukazuju na to da Hrvatska nije loše percipirana od strane domaćih i stranih ispitanika te da svakako ima veliki potencijal za napredak. Uniqueness and recognizability are the characteristics of almost every country in the world. The communication of the national brand represents an important segment in the recognition of the country and affects business, tourism, investments, etc. Therefore, the analysis of information that is used to assess the level of national branding represents an important step in measuring the progress made by national branding. The aim of this paper was to analyze information about Croatia as a national brand by applying the National Branding Index of the Republic of Croatia through six dimensions: People, Management, Exports, Tourism, Investments and immigration, Culture and heritage. Possible shortcomings of the conducted research are the predominance of the age group from 19 to 25 years old compared to the other age groups that participated in the research, the insufficient number of respondents who participated in the research, but also the failure of all respondents to answer all the questions. The questionnaire was completed by 342 respondents, of which 167 were local and 175 were foreign. The age of 19 to 25 years prevailed for both domestic respondents (60%) and foreign respondents (66%). The results of the research shows that the perception of domestic and foreign respondents are about the same, which is evidenced by the average ratings of 4,85 from domestic respondents and 4.85 from foreign respondents. The average scores of domestic and foreign respondents are equal, but they differ from each other for each dimension separately. Also, Croatia, with its average rating of 4.85 took a middle position. The above results indicate that Croatia is not badly perceived by domestic and foreign respondents and that it certainly has great potential for progress.

Mnoge internetske usluge kao što su Amazon, Deezer i Netflix koriste sustave za preporuku (eng. recommender systems) kako bi korisnicima pružili preporučene personalizirane sadržaje. Ovakvi sustavi mogu postati i sastavni dio aplikacija. Cilj ovog rada je bio predstaviti sustave za preporuku u kontekstu u kojem se koriste na internetu, njihove vrste i primjene te utvrditi stavove mladih ljudi u dobi od 16-30 godina o sustavima za preporuke, ali i saznati koliko su upoznati s ovim sustavima, koriste li određene internetske usluge isključivo zbog preporuka te smatraju li da im je preporučen sadržaj relevantan. U teorijskom dijelu rada prikazani su osnovni principi rada i algoritama sustava za preporuku, predstavljena je povijest razvitka sustava za preporuke, objašnjene su razlike između primijenjenih tipova filtriranja (kontekstno utemeljeni sustavi, kolaborativno filtriranje, filtriranje temeljeno na znanju, demografski i hibridni sustavi) i algoritmi, te predstavljeni načini i metrike evaluacije sustava za preporuku. Navedeni su primjeri korištenja sustava za preporuku na popularnim internetskim uslugama Netflix i Deezer. U praktičnom dijelu prikazani su rezultati provedenog istraživanja kojim su ispitani stavovi i upoznatost mladih sa sustavima za preporuku te njihov utjecaj na korištenje odabranih internetskih usluga. Na kraju su izrađeni modeli kontekstno utemeljenog sustava i kolaborativnog filtriranja u programskom jeziku Python. Numerous internet services such as Amazon, Deezer and Netflix use recommender systems to provide users with recommended personalized content. These systems can become an integral part of application software. The aim of this thesis is to present recommender systems within the framework of the Internet, enumerate recommender types and applications, and to identify opinions about this topic of young people aged 16-30, as well as to find out the extend of familiarity with recommenders, whether they use any Internet service exclusively because of its recommendations and whether they find the recommended content relevant. In the theoretical segment of this paper the basic principles and algorithms of recommender systems are presented, an outline of the historical development of recommender systems is shown, and types of recommender systems and algorithms are differentiated (content-based recommendation, collaborative filtering, knowledge-based, demographic, hybrid systems), and methods and metrics of recommender system evaluation are given. Then, examples of recommender systems used by popular internet services Netflix and Deezer are given. In the research segment of the paper the results of a conducted survey on the attitudes and awareness of young people about recommender systems are presented, and on their impact on the usage of chosen Internet services. Lastly, a content-based and a collaborative filtering-based recommender system were created in the Python programming language.

Cilj ovog rada jest izrada hrvatsko-englesko-njemačke online terminološke baze koja sadrži informatičku terminologiju preuzetu iz hrvatskih i njemačkih udžbenika nastavnog predmeta „Informatika“, kao i njihove prijevode na engleskom jeziku. Rad se sastoji od teorijskog i praktičnog dijela. U teorijskom dijelu prikazana je važnost ujednačene stručne terminologije te su navedeni najvažniji terminološki alati, resursi i online jezični resursi. Provedena je analiza stručnih pojmova iz hrvatskih i njemačkih udžbenika nastavnog predmeta „Informatika“. Iz analize utvrđeno je kako većina hrvatskih i njemačkih stručnih termina pored sebe nemaju navedenu englesku prijevodnu istovrijednicu, no hrvatski imaju veći udio od njemačkih. Problem korištenja informatičkih obrazovnih sadržaja, a prije svega digitalnih obrazovnih sadržaja jest nedostatak izvora na hrvatskom jeziku. Stoga je nužno osigurati više online resursa kojima bi im se olakšalo proučavanje i razumijevanje predmetnog gradiva, ali i daljnje obrazovanje i korištenje digitalnih tehnologija i digitalnih resursa, kao što su višejezične terminološke baze. Nakon odabira informatičkih termina iz osnovnoškolskih udžbenika za hrvatski jezik, i odgovarajućih prijevoda za engleski i njemački, Izrađena je online terminološka baza za hrvatsko-englesko-njemački jezik. Ova online terminološka baza dostupna je za učenika, nastavnike i za širu populaciju. The aim of this dissertation is to create a Croatian-English-German online term base that contains the informatics terminology taken from Croatian and German elementary school textbooks, as well their English translations. The paper consists of a theoretical and a practical part. The theoretical part presents the importance of a uniformed specific technical terminology and lists the most important terminological tools, resources, and online language resources. An analysis of the technical terms from the Croatian and German informatics textbooks was also conducted. The analysis shows that most Croatian and German informatics terms do not have an English translation equivalent next to them. However, the Croatian textbooks have a larger amount than the German ones. The problem with the use of educational content in the field of informatics, and above that the digital educational content, is the lack of sources in the Croatian language. Therefore, it is necessary to provide more online resources that would allow them easier studies and an understanding of the subject material, but also ease their further education and use of digital technologies and digital resources, such as multilingual term bases. After selecting the specific terms from Croatian primary school textbooks and the appropriate translations for English and German, an online term base for Croatian-English-German language was created. The online term base is available to students, teachers and the general population.

Cilj istraživanja jest primijeniti alat za strojno učenje za predikciju odgovora te provesti evaluaciju, što je učinjeno primjenom ručne evaluacija točnosti. U teorijskom dijelu prikazani su osnovni podaci o strojnom učenju, podatkovnoj znanosti i umjetnoj inteligenciji. Strojno učenje je opisano u kontekstu informacijskih znanosti, stavljeno u poredbu sa srodnim stručnim poljima poput rudarenja podataka i dubokog učenja. Izložena je kratka povijest, prednosti i nedostatci, aktualne primjene te pogledi na budućnost i razvoj strojnog učenja. Provedeno je istraživanje na 282 sudionika, pri čemu su odgovori prikupljeni vlastitom anketom. U ispitivanju su korištena dva modela koja se međusobno razlikuju u veličini podatkovnih skupova za treniranje i testiranje u omjeru 25/75 i 50/50. Modeli su namijenjeni za predviđanje binarne i numeričke vrijednosti te vrijednosti višestruka odabira. Rezultati ovoga istraživanja pokazuju da ne postoji jedinstveni model predviđanja univerzalno primjenjiv na bilo koju bazu podataka a da pritom postiže maksimalnu učinkovitost. Model A pokazao je prednost u testiranju na skupovima binarne te numeričke vrijednosti, a Model B na skupovima odgovora višestrukog odabira. Odabir modela uvelike ovisi o vrsti i veličini podatkovnog skupa, a izbor na posljetku odlučuje prioritetniji tip podatka. The purpose of the research is to apply a machine learning tool for predicting answers and to carry out an evaluation, which was done using manual accuracy evaluations. The theoretical part presents basic information about machine learning, data science and artificial intelligence. Machine learning is described in the context of information sciences, compared to related fields of expertise such as data mining and deep learning. A brief history, advantages and disadvantages, current applications and views on the future and development of machine learning are presented. A survey was conducted on 282 participants, where the answers were collected through a self-administered survey. In the test, two models were used, which differ from each other in the size of the data sets for training and testing in the ratio of 25/75 and 50/50. The models are intended to predict binary and numeric values and multiple selection values. The results of this research show that there is no unique forecasting model universally applicable to any database while achieving maximum efficiency. Model A showed an advantage in testing on sets of binary and numerical values, and Model B on sets of multiple choice answers. The choice of model depends largely on the type and size of the data set, and the choice is ultimately decided by the more prioritized data type.

Cilj rada bio je prikaz izrade i implementacija chatbota u programskom jeziku Python u domeni bankarstva. Digitalni asistent (eng. chatbot) je sustav koji oponaša razgovor čovjeka kroz tekstualnu ili audio interakciju. U industriji sve više se koriste za rješavanje jednostavnih zadataka pretraživanja, kako u poslovnom tako i u potrošačkom okruženju. Implementacija chatbota može smanjiti troškove poslovanja smanjenjem radne snage ili prenamjenom osoblja za druge poslove, može se povećati kvaliteta poslovanja. Chatbot sustave dijelimo na dvije glavne vrste: sustav temeljen na pravilima i sustav temeljen na umjetnoj inteligenciji. Kod sustava temeljenog na pravilima jasno su definirana pravila kod interakcije digitalnog asistenta i čovjeka, odgovori digitalnog asistenta u potpunosti su unaprijed definirani i vraćaju se korisniku na osnovu prepoznatih ključnih riječi njegovog upita. Sustav temeljen na umjetnoj inteligenciji koristi strojno i duboko učenje kako bi čovjek u razgovoru s digitalnim asistentom dobio iskustvo pričanja s pravom osobom, gdje kao prava osoba, digitalni asistent zna prepoznati gramatičke greške te i dalje shvatiti smisao upita. Glavni nedostatak vog sustava je je mogućnost unaprijed definiranog toka komunikacije. The aim of the paper was to present the development and implementation of a chatbot in the Python programming language in the field of banking. A digital assistant (chatbot) is a system that mimics a person's conversation through textual or audio interaction. In industry, they are increasingly used to solve simple search tasks, both in the business and consumer environment. Implementing a chatbot can reduce business costs by reducing the workforce or reassigning staff for other jobs, it can increase the quality of business. We divide chatbot systems into two main types: a rules-based system and an artificial intelligence-based system. In a rule-based system, the rules for digital assistant-human interaction are clearly defined, the digital assistant's responses are fully predefined, and are returned to the user based on the identified keywords of his query. The system based on artificial intelligence uses machine and deep learning to give a person in conversation with a digital assistant the experience of talking to the right person, where as a real person, the digital assistant knows how to recognize grammatical errors and continue to understand the meaning of queries. The main disadvantage of this system is the possibility of a predefined flow of communication.

Cilj ovog rada bio je prikazati postupak treniranja, prepoznavanja i zamjene lica primjenom alata za duboko učenje, koje predstavljaju najsuvremeniji pristup u području umjetne inteligencije i primjena neuronskih mreža prilikom učenja i definiranja lica, te ukazati na potencijalne stvarne prijetnje korištenja takve tehnologije, ali i pozitivne strane. Rad se sastoji od teorijskog i praktičnog dijela. U teorijskom dijelu definirani su pojmovi strojno učenje i duboko učenje i ključni elementi i razlike. Prikazana je struktura i način rada neuronske mreže, te osnovni principi rada koji se koriste za duboko učenje. Naglasak rada je na primjenama dubokog učenja te su prikazane tehnike koje koriste razne aplikacije u kojima je implementirano duboko učenje za prepoznavanje lica. U praktičnom dijelu rada provedeno je istraživanje primjenom alata za prepoznavanje lica. Računalni program FaceSwap primjenjuje princip dubokog učenja za stvaranje „dubokih lažnjaka“ (eng. deepfakes) u kojima algoritam uči glavne značajke lica te izvorna lica na slikama ili videozapisima zamjenjuje željenim licem neke druge osobe. Nakon prikaza sustava i analize načina rada, prikazan je praktičan dio zamjene lica u videozapisu te su na kraju prikazani rezultati. Posljednje poglavlje odnosi se na etičke principe i ukazuje na opasnosti i moguću zlouporabu te zaštitu od neprimjerene upotrebe. The aim of this paper was to present the process of training, recognizing and replacing faces using deep learning tools, which represent the most modern approach in the field of artificial intelligence and neural networks in learning and defining faces, and point out potential real threats of using such technology. sides. The paper consists of a theoretical and a practical part. In the theoretical part, the terms machine learning and deep learning and key elements and differences are defined. The structure and mode of operation of the neural network are presented, as well as the basic principles of operation used for deep learning. The emphasis of the paper is on the applications of deep learning and the techniques used by various applications in which deep learning for facial recognition is implemented are presented. In the practical part of the paper, research was conducted using facial recognition tools. The computer program FaceSwap applies the principle of deep learning to create "deep fakes" in which the algorithm learns the main features of the face and replaces the original faces in pictures or videos with the desired face of another person. After showing the system and analyzing the mode, the practical part of the face replacement in the video is shown and finally the results are shown. The last chapter deals with ethical principles and points out the dangers and possible abuse and protection against inappropriate use.

Cilj ovog rada je prikaz različitih pristupa sustavima za strojno prevođenje, te izgradnja vlastitog sustava za strojno prevođenje iz odabrane domene, za hrvatsko-engleski jezični par. Rad je sastavljen od dva osnovna dijela, teorijskog i praktičnog. U teorijskom dijelu prikazane su različite arhitekture sustava za strojno prevođenje: sustava temeljenog na pravilima, statističkog sustava i neuronskog sustava. U drugom, praktičnom dijelu, analiziran je proces izgradnje vlastitog sustava za strojno prevođenje koji uključuje prikupljanje resursa, izgradnju sustava i zatim evaluaciju te vrednovanje kvalitete dobivenog strojnog prijevoda. The aim of this thesis is to present different approaches to machine translation systems, as well as to construct a custom machine translation system in a selected domain, for the Croatian-English language pair. In the theoretical part of the thesis, different types of architectures of machine translation systems are presented: the rule-based system, the statistical system, and the neural system. In the practical part, the process of building a custom machine translation system is analyzed. It includes gathering necessary resources and the construction of the system, followed by evaluation of the quality of the obtained machine translations

Komunikacija zdravstvenih djelatnika i pacijenta danas se može smatrati temeljnom i jednom od najvažnijih vještina u kliničkoj praksi kod uspješnog postavljanja dijagnoze i planiranja te provedbe liječenja. Komunikacijske kompeticije po kojima zdravstveni djelatnici razvijaju odnos i komunikaciju s korisnicima zdravstvenih usluga trebaju se temeljiti na istraživačkim i empirijskim konceptima. Učinkovita komunikacija između zdravstvenih djelatnika i pacijenata stvara veće mogućnosti za uspješnu suradnju tijekom postupaka liječenja i rehabilitacije. Cilj doktorskog rada je jasno definirati profesionalne komunikacijske kompetencije fizioterapeuta potrebne za uspješnu komunikaciju i utvrditi utjecaj asertivnog modela komuniciranja na učinkovitost suradnje s pacijentima i unutar zdravstvenog tima. U radu postoje i tri specifična istraživačka cilja: (1) analizirati pravne i javno dostupne dokumente koji se odnose na klinički rad fizioterapeuta i formalno obrazovanje fizioterapeuta u Hrvatskoj s aspekta komunikacijskih kompetencija; (2) procijeniti komunikološki profil fizioterapeuta te istražiti različite aspekte komunikacijskih kompetencija – empatičnost i asertivnost u komunikaciji te tolerantnost na neslaganje u komunikaciji na radnom mjestu; (3) istražiti kod korisnika fizioterapije učinkovitost komunikacije s fizioterapeutom, ponašanje fizioterapeuta tijekom komunikacije, strah od komunikacije te korelaciju ishoda rehabilitacije i zadovoljstva komunikacijom s fizioterapeutom. U istraživanju su korištene četiri metode prikupljanja podataka: (1) analiza pravne dokumentacije o fizioterapeutskoj profesiji, (2) anketno istraživanje o komunikacijskim kompetencijama fizioterapeuta, (3) anketno istraživanje o učinkovitosti komunikacije korisnika fizioterapije s fizioterapeutima, (4) analiza kolegija u formalnom obrazovanju fizioterapeuta. Anketno istraživanje provedeno je u razdoblju od 1. svibnja do 31. prosinca 2018. godine, u 74 zdravstvene ustanove na području cijele Hrvatske. Testni skup čine 502 fizioterapeuta zaposlena u zdravstvenim ustanovama u javnom, državnom ili privatnom vlasništvu, te 2902 pacijenta ispitana u istim zdravstvenim ustanovama. Znanstveni doprinos ovog doktorskog rada može se razmatrati na teorijskoj i empirijskoj razini. Na teorijskoj razini znanstveni doprinos sastoji se u definiranju nužnih komunikacijskih kompetencija potrebnih za učinkovit profesionalni rad fizioterapeuta te u definiranju utjecaja asertivnog modela na učinkovitost komunikacije u kliničkoj praksi fizioterapeuta. Na empirijskoj razini znanstveni doprinos se sastoji u analizi i primjeni rezultata provedenog istraživanja o komunikacijskim kompetencijama fizioterapeuta te u izradi sugestija za dodatnom edukacijom na razini formalnog i neformalnog obrazovanja. INTRODUCTION: Clinical medicine and the work of healthcare workers, as well as the results of treatment and medical rehabilitation, are all dependant on the effectiveness of communication throughout this process, as well as on the quality of the relationships between the physician and other healthcare workers and patients and their families, and within the healthcare team itself. Factors that are of importance for effective communication within clinical practice are the atmosphere in which communication unfolds, the manner in which interlocutors relate to each other, the words that are used in communication and the quality of communication on a verbal and non-verbal level. The communication competencies on the basis of which healthcare workers develop their relationships and communication with users of healthcare services should be rooted in research and empirical concepts. Effective communication between healthcare workers and patients increases the possibility of achieving successful cooperation during the treatment and rehabilitation process. A review of published literature has shown that, as of yet, no research related the communication competencies of physiotherapists has been performed in Croatia while, at an international level, scientific papers have been published only in relation to certain thematic areas concerning communication competencies, whether in physiotherapy or clinical practice in general. Woodward et al. (2012) have identified five main categories in the assessment of the quality of communication in physiotherapy: non-verbal communication, manner of the physiotherapist, the language that the physiotherapist uses, hospital organisation and the content of therapy procedures and communication tools. Amoudi et al. (2017) have studied the effect of positive communication skills on reducing pain in orthopaedic patients during physiotherapy in Palestinian rehabilitation centres. The authors state that the communication competencies of physiotherapists play a key role during the physiotherapy programme when it comes to alleviating pain and depression and increasing the motivation of patients. The results of their research showed that 50% of respondents agreed that positive communication skills were of great help in the alleviation of pain, while around 30% of respondents stated that such skills were of moderate help. This means that, to a greater or slightly lesser extent, 80% of patients consider communication during rehabilitation to be of importance for the positive outcome of rehabilitation. Włoszczak-Szubzda and Jarosz (2013) stress that effective communication skills are necessary in order to correctly gather information in healthcare consultations, as well as for adapting to the needs and thoughts of the patients. ˝During physiotherapy and discharge itself, effective communication is necessary for the patient to gain a better understanding of their further progress, lifestyle and behaviour˝ (Włoszczak-Szubzda , Jarosz, 2013; p. 194). Hargreaves (1982) and Ambady et al. (2002) stress that an effective understanding and use of non-verbal communication in a therapeutic environment is necessary in order to achieve a successful rehabilitation outcome. Patients that have confidence in their physiotherapists are more likely to adhere to therapeutic instructions, as opposed to patients who received identical verbal instructions, but sensed the disinterest and absence of their physiotherapists. Doubts in the sincerity and credibility of the sender of the message lead to unsatisfactory interactions. From the research described above, it can be concluded that the communication competencies of physiotherapists are of great importance for both the successful and effective implementation of physiotherapy and the satisfaction of patients. RESEARCH OBJECTIVES AND HYPOTHESES: The objective of this doctoral dissertation is to clearly define the communication competencies that physiotherapists need to possess in order to communicate successfully, as well as to establish the influence of the assertive communication model on the effectiveness of cooperation with patients and within the healthcare team itself. The dissertation also has three specific research objectives: analysing legal and publicly available documents related to the clinical work of physiotherapists and the formal education of physiotherapists in Croatia; assessing the communication profile of physiotherapists and studying the different aspects of communication competencies: empathy and assertiveness in communication, and tolerance to disagreements at the workplace; researching the effectiveness of communication of users of physiotherapy with their physiotherapists, the manner of physiotherapists during communication, fear of communication and the correlation between the outcome of rehabilitation and the patient’s satisfaction with their communication with their physiotherapist. Research hypotheses: H1: The professional communication competencies of physiotherapists are represented neither in Croatian legal documentation, nor Croatian education documentation. H2: The use of an assertive communication style by physiotherapists significantly increases the effectiveness of cooperation with patients. H3: Physiotherapists with a greater tolerance to disagreements achieve more effective communication within the healthcare team. H4: Developed communication skills in physiotherapists have a positive effect on the outcome of medical rehabilitation. H5: Patients who evaluate worse their communication with a physiotherapist during medical rehab express greater fear of physical therapist, are older (over 58 years of age) and are the lowest in education. This dissertation aims to contribute to the development of communication skills in physiotherapists, which leads to the development of one’s own professional work, both in physiotherapy and clinical practice in general. The outcomes of this research may be of assistance in the evaluation of professional communication competencies related to physiotherapy as well as other healthcare professions, which benefits both patients and physiotherapists within the therapeutic process. The expected scientific contribution of this doctoral dissertation can be elaborated on a theoretical and empirical level: On a theoretical level: the scientific contribution relates to the definition of the communication competencies that are necessary for the effective professional work of physiotherapists, and to the definition of the influence of the assertive model on the effectiveness of communication in the clinical practice of physiotherapists. On an empirical level: the scientific contribution relates to the analysis and application of the results of the conducted research on the communication competencies of physiotherapists, and in the drafting of suggestions for additional training at the level of the formal and informal education of physiotherapists. RESEARCH METHODS: The subject of the doctoral dissertation is gaining insight into the condition of the objective parameters of the communication competencies of physiotherapists within the Croatian healthcare system. The research used four methods of data collection: 1. Analysis of legal documentation on the physiotherapy profession 2. Questionnaire on the communication competencies of physiotherapists 3. Questionnaire on physiotherapy patients’ views on the effectiveness of communication with their physiotherapists 4. Analysis of courses that are part of the formal education of physiotherapists Analysis of legal documentation was performed with the objective of researching the extent to which the public availability and validity of legal documents that regulate the profession of physiotherapy in Croatia is connected with the definition of the professional competencies of physiotherapists. This analysis encompassed the Act on Physiotherapy, Standards in Physiotherapy Practice and the Code of Physiotherapy Ethics and Deontology. The analysis of the communication competencies of physiotherapists was performed by means of a questionnaire that assessed physiotherapists’ knowledge of and attitude towards communication with users of physiotherapy, as well as towards communication within the healthcare team itself. The objective of this questionnaire was to determine the knowledge and skills that are necessary in order to make communication in the clinical work of physiotherapists as effective as possible (empathy in communication, assertiveness and tolerance to disagreements at the workplace). The analysis of the effectiveness of the communication of patients with their physiotherapists was performed by means of a questionnaire that examined patients’ views of and satisfaction with the communication competencies of their physiotherapists. The objective of this questionnaire, which was filled in by users of physiotherapy, was to examine the effectiveness of their communication with their physiotherapist, their opinions concerning the manner of their physiotherapist during communication, fear of communicating with their physiotherapist and the correlation between the outcome of rehabilitation and patients’ satisfaction with their communication with their physiotherapist. The analysis of courses that are part of the formal education of physiotherapists encompassed the syllabi of the undergraduate and graduate university and professional study programmes in physiotherapy performed at higher education institutions in Croatia, in which course content involving the communication competencies of physiotherapists was the subject of research. RESULTS: In the first part of the research, an analysis of the publicly available legal documentation that regulates the profession of physiotherapy in Croatia was performed: the Physiotherapy Act, Standards in Physiotherapy Practice and the Code of Physiotherapy Ethics and Deontology. The Act on Physiotherapy regulates the content and manner of performing physiotherapy activities, educational standards, prerequisites for performing the profession, obligations, quality control and professional supervision of the work of physiotherapists in the Republic of Croatia. Pursuant to the Physiotherapy Act, in 2018 the Croatian Chamber of Physiotherapists published the Code of Physiotherapy Ethics and Deontology, which defines the principles and rules of conduct for physiotherapists for the purpose of upholding the dignity and reputation of the profession and activity of physiotherapy. The principles of this Code define the importance and role of the physiotherapist in healthcare and in society in general, and they are aimed at providing guidance on how to pursue the profession of physical activity in a responsible and professional manner. Pursuant to the Physiotherapy Act, in 2009 the Croatian Chamber of Physiotherapists issued a document titled ˝Standards in Physiotherapy Practice˝, which were developed in accordance with the recommendations of the World Confederation for Physical Therapy (WCPT) and the European Region of the World Confederation for Physical Therapy (ER-WCPT). Standards in the practice of physiotherapy represent principles for the implementation of physical therapy and requirements that are demanded from physiotherapists in order to provide a high standard of physiotherapy services to the community. The standards represent the professional obligation of the physiotherapy profession in the promotion of the health of individuals and the community at large, all for the purpose of increasing the effectiveness of physiotherapy services, and they encompass: administration and principles in the organisation of physiotherapy practice, communication with patients and within the healthcare team, responsibility towards the community, cultural competencies, documentation, education and training, ethical code, informed consent, legality, management of patient care, personal professional development, quality assurance, research and assisting staff in physiotherapy. According to the analysis performed, it is evident that professional communication competencies have not been defined in Croatian legal documentation. In order to establish the professional communication competencies of physiotherapists, two questionnaires were implemented: (1) Questionnaire on the communication competencies of physiotherapists and (2) Questionnaire on physiotherapy patients’ views on the effectiveness of communication with their physiotherapists. The questionnaires were implemented during a period of 8 months, between 01 May and 31 December 2018, on a sample of 2 test groups in 74 private and public healthcare institutions in Croatia that perform physiotherapy, and the research involved 502 physiotherapists and 2902 patients. The analysis of the results obtained from the questionnaire on the communication competencies of physiotherapists shows that an assertive style of communication adopted by physiotherapists significantly increases the effectiveness of their cooperation with patients. However, it was established that physiotherapists that are older or somewhat older both in terms of age (the group between the ages of 35-49 and the group above the age of 50) and in terms of length of service (group with 15-29 years of service and group with more than 30 years of service) have lower indicators of assertiveness in communication than their younger colleagues (group with less than 15 years of service). Empathy also ranked highly as a desirable trait in working and communicating with patients. Furthermore, in this part of the research, an assessment of attitudes towards disagreements in communication at the workplace was also performed, in which it was observed that physiotherapists perceived disagreements, i.e. conflicts as ˝undesirable˝, in the sense that they were inclined towards avoiding open conflict at the workplace. They were more inclined towards switching conversation to another topic, or avoiding communication with conflictive persons. The research of physiotherapy patients’ views on the effectiveness of communication with their physiotherapists was based on a questionnaire that assessed patient’s views of and satisfaction with the communication competencies of their physiotherapists. This questionnaire implemented among patient tested their own assessment of their health status, assessment of fear of communicating with their physiotherapist, assessment of the effectiveness of their communication with their physiotherapist, assessment of physiotherapy patients’ views on interpersonal communication with their physiotherapist, assessment of the behaviour of physiotherapists in communication, assessment of the manner of speech of physiotherapists. The results show that the majority of respondents (85.5%) reported having a moderate fear of their physiotherapist, a smaller part of the respondents reported a high degree of fear (14.5%), while none of the respondents reported not having a fear of physiotherapists. Furthermore, patients’ views on the effectiveness of their communication with their physiotherapist was assessed. The effectiveness of communication was tested according to the following parameters: (1) thoroughness of the physiotherapist’s questions on the patient’s health status and symptoms; (2) attentiveness of the physiotherapist in communication with the patient; (3) elaborations of the medical problems of the patient or the procedures planned for the patient; (4) the physiotherapist’s responses to the questions asked by the patient; (5) length of conversation and time dedicated by the physiotherapist to talking with the patient; and (7) the physiotherapist’s level of intimacy with the patient. The analysis of the data concerning effectiveness of communication with the physiotherapist showed that patients valued the traits of patience and attentiveness the most in conversation with their physiotherapist. The analysis of the results pertaining to physiotherapy patients’ views on their interpersonal communication with their physiotherapist also shows what patients find important during physiotherapy: they want to be heard, to be able to openly express their problems without feeling criticised, for the physiotherapist to express a sincere interest in their health concerns and to show initiative in explaining the further course of treatment – through advice, support and instructions. As for the behaviour of physiotherapists towards patients in communication, users of physiotherapy highly value having enough time to consult with the physiotherapist during physiotherapy, being attentively listened to by the physiotherapist while they were talking about their health concerns; having their physiotherapists help them feel well enough to perform their everyday life activities; receiving explanations on the purpose of the planned tests and procedures from their physiotherapist at the beginning of therapy; receiving help from their physiotherapist in understanding the advice that they received during therapy; being warned of the consequences of not performing the recommended therapeutic exercises as recommended and instructed. The research results also show that the respondents, as an error in communication, most commonly stressed that physiotherapists talked too fast in conversation with them. The average values obtained also show that they used incomprehensible speech, that they were confused in communication and neglected them in communication. The analysis of courses that are part of the formal education of physiotherapists established that, within the syllabi of courses that are part of the formal education of physiotherapists, there exist courses where communication is taught. The research encompassed syllabi at 2 polytechnics and 5 universities in Croatia, on the undergraduate and graduate level. On the territory of Croatia, the formal education of physiotherapists is performed through university and professional study programmes. The analysis set apart the following 5 compulsory courses on the undergraduate level with content and objectives related to the communication competencies of future physiotherapists: 1. Ethics of Healthcare, 2. Physiotherapy Assessment, 3. Clinical Propaedeutics, 4. Communication Skills, 5. Foreign Language; together with 3 elective courses: 1. Palliative Care, 2. Psychology of Sports, 3. Gerontology. At the level of graduate study programmes of physiotherapy, 3 compulsory courses related to improving the communication competencies of future physiotherapists were found: 1. Management in Healthcare, 2. Working with the Disabled, 3. The Skill of Medical Teaching; together with 3 elective courses: 1. Teamwork, 2. Working with a Group, 3. Business Communication. Educational programmes have recognised the need for teaching communication skills as soon as at the undergraduate level, and the course ˝Communication Skills˝ was introduced at all 7 institutions for training physiotherapists, with content and objectives encompassing the knowledge that future physiotherapists will need in their work with patients, their families and work colleagues. However, the results of this analysis did not prove conclusive as to whether the students are familiarised with this content only informatively, or whether they are trained to resolve concrete communication problems. CONCLUSION Further research and guidelines within the profession should take into consideration the educational framework of communication competencies and include it in the legal framework of the profession, for example including it in the ethical code of physiotherapy, standards in physiotherapy and similar. The legal and educational framework within the profession should certainly be in harmony, so that results in the field of communication competencies may both gain importance and be systematically implemented and monitored in practice. On the basis of the questionnaires, both those given to patients and those given to physiotherapists, important new knowledge on the work processes of physiotherapists concerning communication and the necessary competencies was obtained. Above all, one needs to spend time in rehabilitation in order for rehabilitation to be as effective as possible, and this also applies to communication as an integral part of rehabilitation. Elderly patients demand extra attention not only in the rehabilitation process, but also when it comes to the manner of communication itself. Likewise, patients with a lower level of education need extra care for their fears to be assuaged, as do elderly patients. Within all systems (forms) of treatment, it is necessary to implement training on the manner, methods and procedures of communication, as well as to insist on good and effective communication within the healthcare team itself.

Cilj ovoga rada je izrada online informatičkog osnovnoškolskog rječnika koji sadrži informatičke pojmova iz udžbenika od 5. do 8. razreda osnovne škole i njihove prijevode ekvivalente na engleskom jeziku. Rad se sastoji od teorijskog i praktičnog dijela. U teorijskom dijelu prikazana je primjena informacijskih i komunikacijskih tehnologija u obrazovanju te su navedeni ciljevi predmeta Informatike u školama, kao i domene koje se obrađuju na nastavi. Uz udžbenik, opisani su i drugi obrazovni materijali. Provedena je analiza informatičkih pojmova u udžbenicima informatike i utvrđeno je da se u udžbenicima pored većine novih hrvatskih naziva ne nalazi engleska prijevodna istovrijednica. Učenicima je dostupan određen dodatan digitalni obrazovni sadržaja na hrvatskom jeziku, ali veći je izbor na engleskom jeziku. Praktičnim pregledom ponuđenih mrežnih i tiskanih rječnika, leksikona i enciklopedija informatičkog nazivlja koji mogu predstavljati dodatan obrazovni materijal u učenju utvrđena je prevlast rječnika s polaznim engleskim nazivljem, a strojni prijevodi nisu precizni. Poznavanje i hrvatske inačice i engleske istovrijednice pružilo šire mogućnosti korištenja programskih alata i digitalnog sadržaja. Kao programsko rješenje izrađen je online osnovnoškolski informatički rječnik. Rječnik sadrži hrvatske nazive iz udžbenika te njihove precizne engleske prijevodne istovrijednice te tako predstavlja dodatan digitalni obrazovni sadržaj. The aim of this paper is to create an online informatics elementary school dictionary that contains informatics terms from textbooks (5th to 8th grade of elementary school) and their translations equivalent in English. The paper consists of a theoretical and a practical part. The theoretical part shows the application of information and communication technologies in education and lists the goals of the subject of Informatics in schools, as well as the domains that are covered in teaching. In addition to the textbook, other educational materials are described. An analysis of IT terms in IT textbooks was conducted and it was established that in addition to most of the new Croatian terms, there is no English translation equivalent in the textbooks. Students have access to some additional digital educational content in Croatian, but there is a greater choice in English. A practical review of the existing online and printed dictionaries, lexicons and encyclopedias of computer terminology, that can represent additional educational material in learning, has determined the predominance of dictionaries with initial English terminology. In addition, it was found that machine translations are not accurate. Knowledge of both the Croatian version and the English equivalent would provide wider possibilities for the use of software tools and digital content. An online elementary school computer dictionary was created as a software solution,. The dictionary contains Croatian terms from textbooks and their precise English translation equivalents, thus representing additional digital educational content.

Završni rad na temu strojnog učenja prepoznavanja rukopisne numeracije detaljno prikazuje podjelu algoritama zaduženih za strojno učenje na način da vrši uvid u same pristupe strojnom učenju i načine na koje se ti pristupi mogu koristiti kada dođe do problematike prepoznavanja ručno napisanih brojčanih znamenaka. Svi navedeni pristupi nadalje su razgranati kako bi se postigao jasan pregled najčešćih oblika klasifikatora zaduženih za istraživanja na ovu temu. Ovaj rad daje uvid u postotak točnosti i pogreške dobivenih na osnovu istraživanja vršenih za svaki pojedini algoritam koji se koristi u obliku sustava za klasifikaciju korištenih ulaznih podataka. Svi rezultati dobiveni testiranjem navedenih algoritama tablično su prikazani i detaljno objašnjeni. Također, vršen je i prikaz eventualnih problema koji se događaju pri stvaranju predikcija o kojoj se ručno napisanoj znamenci uistinu radi, kao i koja su eventualna rješenja za iste. U završnome dijelu rada prikazano je softwaresko rješenje ta temu ove problematike u obliku koda napisanog u programskom jeziku Python. The final work on theme of handwritten digit recognition using machine learning shows in detail the division of machine learning algorithms by insight into the machine learning approaches and ways those approaches can be used in problems related to handwritten numbering. All of the above approaches are further explained in order to achieve a clear overview of the most common forms of classifiers in charge of research on this topic. This paper provides insight into the percentage of accuracy and error obtained based on the research performed for each individual algorithm it uses in the form of a system for classifying the input data used. All results obtained by testing this algorithm are tabulated and published in detail. Also, a presentation of possible problems that occur when creating a prediction about which handwritten digit is truly used, as well as possible solutions for the same. The final part of the paper presents a software solution on the topic in the form of code written in the Python programming language.

Informacijska obilježja vizualnog identiteta imaju izrazito jaku ulogu u brendiranju nacionalnog identiteta i u nacionalnoj i međunarodnoj komunikaciji. Poruka koja se odašilje na određeno tržište ili prema publici, sadrži određene vrste informacija kojima se djeluje na publiku. Upravo iz navedenog razloga, vizualni identitet, danas sve češće u digitalnom obliku, ima izrazito veliku ulogu u odašiljanju poruke kroz neverbalnu komunikaciju. Svaka država želi stvoriti snažan, atraktivan i drugačiji vizualni identitet koji će ju razlikovati od ostalih država. Cilj ovoga rada je analizirati informacijska obilježja u brendiranju nacionalnog vizualnog identiteta. U teorijskom dijelu prikazana je uloga strateške komunikacije i koraci u procesu brendiranja. U praktičnom dijelu rada prikazani su rezultati provedenog istraživanja o informacijskim obilježjima nacionalnog brenda provedenog na stavovima prema vizualnom identitetu Hrvatske kako bi se utvrdila informacijska obilježja po kojima je Hrvatska prepoznatljiva. The information features of visual identity play an extremely strong role in the branding of national identity and in national and international communication. A message that is sent to a specific market or to an audience, contains certain types of information that affect the audience. Precisely for this reason, visual identity, increasingly in digital form today, plays an extremely important role in sending a message through non-verbal communication. Each state wants to create a strong, attractive and different visual identity that will set it apart from other states. The aim of this paper is to analyze the information features in visual identity branding. The theoretical part presents the role of strategic communication and steps in the branding process. The practical part of the paper presents the results of a survey on the information characteristics of the national brand conducted on and attitudes towards the visual identity of Croatia in order to determine the information characteristics by which Croatia is recognizable.

U suvremenom načinu življenja i poslovanja do izražaja dolazi informacijska i komunikacijska tehnologija, a osobito primjena interaktivne online tehnologije gdje korisnici imaju mogućnost trenutnog dobivanja informacije. U suvremenom dobu financijske se institucije sve više fokusiraju na stvaranje personaliziranog odnosa sa svojim korisnicima. U te svrhe investiraju mnogo u razvojna rješenja kako bi korisnicima plaćanja i ostale zadatke učinila puno jednostavnijima. Poduzeća, koja se bave financijskom tehnologijom, već jedno vrijeme rade na umjetnoj inteligenciji, a posebice na agentima za razgovore koji čine suvremena konverzacijska sučelja između banke i klijenta. In the modern way of living and doing business, information and communication technology is coming to the fore, and especially the application of interactive online technology where users have the opportunity to receive information instantly. In the modern age, financial institutions are increasingly focused on creating a personalized relationship with their customers. To that end, they invest heavily in development solutions to make payments and other tasks much easier for users. Financial technology companies have been working on artificial intelligence for some time, and especially on chatbots or chat agents that make up modern conversational interfaces powered by artificial intelligence.

Danas se sve više koristi tehnologija kako bi se poboljšao način poslovanja raznih poduzeća. Stoga je važno unaprijediti i iskoristiti takvu tehnologiju u cilju unapređenja poslovanja. Iako se chatbot, tj. agent za razgovor može implementirati u bilo koje poduzeće koje se oslanja na komunikaciju s korisnicima, u radu će se fokusirati na ugostiteljske oblike poslovanja. U teorijskom dijelu rada ukratko su prikazati elementi poslovnog informacijskog sustava i pojam agenta za razgovore. U praktičnom dijelu rada prikazan je proces izrade, integracije i implementacije agenta za razgovor sa sposobnošću strojnog učenja u cilju poboljšanja usluge poduzeća. Today, technology is increasingly being used to improve the way businesses operate. Therefore, it is important to improve and leverage such technology in order to improve the business. Although a chatbot can be implemented in any business that relies on customer communication, the paper will focus on the catering business. The theoretical part of the paper briefly outlines the elements of a business information system and the principles of machine learning. The practical part of the paper describes the process of designing, integrating and implementing a chatbot with machine learning capability to improve enterprise service.

U radu se obrađuje uloga vizualnog identiteta u procesu prijenosa informacija putem medija. Sadržaj rada obuhvaća analizu uloge vizualnog identiteta u komunikaciji s promatračima i potrošačima. Posebno su obrađeni sastavni elementi vizualnog identiteta - logo, boja i tipografija. Kroz primjere, pokazuje se utjecaj kvalitetnog dizajna, ali i zastarjelog dizajna, utjecaj boje na percepciju i na tisak te utjecaj tipografije na sveopći dojam vizualnog identiteta. U projektnom radu izrađen je logo Odsjeka za informacijske i komunikacijske znanosti sastavljen od navedenih elemenata koji se korist za izradu dizajna vizualnog identiteta za Odsjek za informacijske i komunikacijske znanosti. Logo je integriran u predloške posjetnice, memoranduma i koverte tako da sastavni elementi prenose informacije o Odsjeku u cjelini. This thesis deals with the process of transferring information through visual identity media. The content of the work encompasses the visual identity and its role in communicating with observers and consumers, the analysis of elements of visual identity, which include logo, color, and typography as three basic components. Through examples, the influence of highquality design, but also obsolete design, the influence of color on perception and printing, and the influence of typography on the overall impression of visual identity are shown. In the end, these elements are used for the visual identity design that would belong to the Department of Information and Communication Science, and the way in which these elements transmit information about the department when they are merged is considered.

Industrijske revolucije su drastično promijenile ljudski život. Razvojem tehnologije tijekom prošlih stoljeća došlo je do napretka ne samo u industriji nego u općenito u svakom ljudskom djelovanju. Izumi kao što su parni stroj, električna energija i internet koji je povezao cijeli svijet, donijeli su nezamislive promjene. Živimo u dobu Industrije 4.0 koja je najnaprednija industrijska revolucija do sad. Automatizirani procesi s primjenom robotike stvorili su pametne tvornice i pametne procese u proizvodnji. Proizvodi iz tvornica izlaze brže, kvalitetniji su te se prilagođavaju kupcima. Poslovni modeli također su usmjereni prema kupcima. Internet stvari kojima se povezuju korisnici i uređaji spojeni na internet, upravljaju velikim podatcima te ih mogu spremiti u računalni oblak kojemu se može pristupiti bilo gdje, gdje postoji pristup internetu. Takav oblik tehnologije koji je dostupan bilo gdje je revolucionaran te treba popratiti taj korak u bilo kojem aspektu korištenja tehnologija. Istraživanje je provedeno u jednoj hrvatskoj pivovari proučavajući digitalnu transformaciju u poslovanju. Industrial revolutions have drastically changed human life. With the development of technology over the past centuries, there has been progress not only in industry but in general in every human activity. Inventions such as the steam engine, electricity, and the internet that connected the whole world have made unthinkable changes. We are living in the age of Industry 4.0 which is the most advanced industrial revolution so far. Automated robotic applications have created smart factories and smart manufacturing processes. The products from the factories come out faster, are better quality and adapt to the customers. Business models are also customer oriented. The Internet of Things that connects users and devices connected to the Internet, manages big data and can store it in a cloud-based computer that can be accessed anywhere with internet access. This type of technology that is available anywhere is revolutionary and should be accompanied by this step in any aspect of technology use. The research was conducted at a Croatian brewery studying digital transformation of business model.

Strojno učenje postaje sve češći alat u većini znanstvenih disciplina. Ovaj tekst pokušava objasniti strojno učenje i njegove osnovne koncepte kroz klasifikaciju sustava strojnog učenja i kroz objašnjenja najkorištenijih algoritama. Također prikazuje kako stvoriti sustav strojnog učenja za klasifikaciju korištenjem klasifikatora K najbližih susjeda, logističke regresije te linearnog i RBF stroja potpornih vektora. Također, objašnjeno je kako interpretirati rezultate najkorištenijih metoda za evaluaciju sustava strojnog učenja. Machine learning is becoming a more prevelant tool in most science fields. This text tries to explain machine learning and its basic concepts through classification of machine learning systems and through explanations of most used algorithms. It is also presents how to create a machine learning system for classification using K- nearest neighbors classifier, logistic regression, and linear and RBF kernel support vector machines. Also, it is shown how to interpret results of most used methods for evaluating machine learning systems.

Online advertising developed in 1994 as a means of financing website content production, but a constantly decreasing number of ad impressions and a significant rise in ad blocker users have incurred the loss of 30% of total industry revenue in 2015. However, studies reveal that two-thirds of current ad blocker users would to be willing to turn their blockers off should overall user experience be improved. As marketers continue debating whether the ad block revolution will finish online advertising, this paper presents an alternate viewpoint. It suggests that the rise of the ad block has been beneficial to the industry, because exposing the weaknesses of the current advertising model and the reasons for blocker popularization prompts the damaged model to change. It gathers best practices in dvertising and recommendations for creating ads which do not need to be blocked.

With the turn towards the digital age, a growing number of institutions in the GLAM (Galleries, Archives, Museums and Libraries) sector started to identify a need for digitising their different collections placing them online with goals to preserve and exhibit them in the digital environment. After the initial efforts to develop policies, methodologies and best practices in transferring the collections into the online environment, researchers and practitioners have started to investigate possibilities of communicating those digitised collections with the public and seizing the opportunities that arise from digitisation. One of the approaches that cultural heritage institutions started to explore in order to involve the general public in their activities on the Web is crowdsourcing - taking a job traditionally performed by a designated agent (usually an employee) and outsourcing it to an undefined community in the form of an open call. In the heritage sector this means inviting members of the public, (“the crowd”), to tag and classify, transcribe, organize, and otherwise add value to digital cultural heritage collection content. In this paper we provide an overview of approaches in using the collective intelligence in the cultural heritage domain. Key terms, concepts and corresponding case studies are discussed, providing the framework for crowdsourcing projects within the heritage sector.

Teaching controlled vocabularies and classifications at an introductory level can be challenging in terms of clarifying theoretical notions and motivating the students. In this light, approaches where students are presented with real life data that illustrates key concepts can be very useful. This case study investigated the difference between subject description of a journal article when using uncontrolled vocabulary (free keywords) and controlled vocabulary (UDC notions) as compared to author keywords, title keywords and UDC notions added by the subject specialist. The study was carried out before and after students attended an undergraduate introductory course to the field of controlled vocabularies and classification systems with a special accent on using the Universal Decimal Classification (UDC). The results confirmed two original hypotheses that UDC driven (post-course) tags will show a greater consistency and have higher agreement with the terms added by the subject specialist than free keyword (pre-course) tags. Since the study was carried out in an educational setting one of the goals of the study was to investigate the possible use of the study to illustrate the basic concept differences between uncontrolled and controlled vocabulary in the light of folksonomies and the use of UDC as a pre-coordinated term list.