U radu je dan pregled područja povezanog s procesiranjem prirodnih jezika i njihova međusobnog odnosa, počevši od šire domene kao što je umjetna inteligencija, putem strojnog učenja, računalne ...lingvistike, metoda strojnog prevođenja te posebice onih zasnovanim na dubokom učenju. Opisane su karakteristike, primjene, faze i glavni problemi obrade prirodnih jezika s leksičke, sintaktičke, semantičke, govorne i pragmatičke perspektive. Opisane su faze prepoznavanja i analize prirodnog jezika kao i faza generiranja prirodnih jezika. Postupci pre-editinga i post-editinga uz korištenje kontroliranih prirodnih jezika dani su kao primjeri prakse kojom se povećava točnost i kvaliteta automatskog prevođenja i općenito procesiranja teksta. Poseban je fokus stavljen na strojno prevođenje te metode strojnog prevođenja. Pristupi strojnom prevođenju kao statistički, temeljen na pravilima, hibridni i pristup temeljen na dubokom učenju opisani su i predstavljeni s obzirom na njihove prednosti i nedostatke i prikladnu primjenu u praksi. Na kraju su dani još uvijek neriješeni izazovi kao smjer daljnjih istraživanja vezanih uz obradu prirodnih jezika te značaj razvoja pristupa temeljenog na dubokom učenju.
The paper provides an overview of areas related to the processing of natural languages and their interrelationships, starting from a broader
domain such as artificial intelligence, through machine learning, computational linguistics, machine translation methods and especially those based on deep learning. The characteristics, applications, phases and main problems of natural language processing from the lexical, syntactic, semantic, speech and pragmatic perspective
are described. The phases of natural language recognition and analysis
as well as the natural language generation phase are described. Pre-editing and post-editing procedures using controlled natural languages are given as examples of practices that increase the accuracy
and quality of automatic translation and text processing in general. Special focus is given to machine translation and machine translation methods. Approaches to machine translation as statistical, rule-based, example-based, hybrid and deep learning-based approach are described and discussed with regard to their advantages and disadvantages including appropriate application in practice. In the end, still unresolved challenges are given as a direction of future research related to natural language processing and the importance of further development of a deep learning-based approach.
The paper describes our current research activities and results related to developing knowledge‐based systems to support the creation of entity‐relationship (ER) models. The authors based obtaining ...an ER model in textual form on translation from one language into another, that is, from an English controlled natural language into the formalized language of an ER data model. Our translation method consisted of creating translation rules of sentential form parts into ER model constructs based on the textual and character patterns detected in the business descriptions. To enable the computer analyses necessary for creating translation mechanisms, we created a linguistic corpus that contains lists of the business descriptions and the texts of other business materials. From the corpus, we then created a specific dictionary and linguistic rules to automate the business descriptions' translation into the ER data model language. Before that, however, the corpus was enriched by adding annotations to the words related to ER data model constructs. In this paper, we also present the main issues uncovered during the translation process and offer a possible solution with utility evaluation: applying information‐extraction performance measures to a set of sentences from the corpus.
Today, the amount of data in and around the business system requires new ways of data collection and processing. Discovering sentiments from hotel reviews helps improve hotel services and overall ...online reputation, as potential guests largely consult existing hotel reviews before booking. Therefore, hotel reviews of Croatian hotels (categories three, four, and five stars) in tourist regions of Croatia were studied on the Booking.com platform for the years 2019 and 2021 (before and after the start of the pandemic COVID-19). Hotels on the Adriatic coast were selected in the cities that were mentioned by several sources as the most popular: Rovinj, Pula, Krk, Zadar, Šibenik, Split, Brač, Hvar, Makarska, and Dubrovnik. The reviews were divided into four groups according to the overall rating and further divided into positive and negative in each group. Therefore, the elements that were present in the positive and negative reviews of each of the four groups were identified. Using the text processing method, the most frequent words and expressions (unigrams and bigrams), separately for the 2019 and 2021 tourism seasons, that can be useful for hotel management in managing accommodation services and achieving competitive advantages were identified. In the second part of the work, a machine learning (ML) model was built over all the collected reviews, classifying the reviews into positive or negative. The results of applying three different ML algorithms with precision and recall performance are described in the Results and Discussion section.
U današnje vrijeme količina podatka koja se nalazi u poslovnom sustavu i oko njega zahtijevanove načine prikupljanja i obrade podataka. Otkrivanje sentimenta iz hotelskih recenzija pridonosipoboljšanju hotelske usluge ali i ukupnoj online reputaciji budući da se potencijalni gosti prijerezervacije uvelike konzultiraju postojećim recenzijama smještaja. Na tragu toga, napravljeno jeistraživanje nad hotelskim recenzijama hrvatskih hotela (kategorija tri, četiri i pet zvjezdica) uturističkim hrvatskim regijama sa platforme Booking.com, za godinu 2019 i 2021 (prije i poslije COVID19 pandemije). Odabrani su hoteli sa Jadranske obale i to u gradovima koji su na više izvora odabranikao najpopularniji: Rovinj, Pula, Krk, Zadar, Šibenik, Split, Brač, Hvar, Makarska te Dubrovnik Recenzijesu grupirane u četiri grupe po ukupnom ratingu i dodatno podijeljene u svakoj grupi na pozitivnei negativne kako bi se identificirale stavke koje su prisutne u pozitivnim i negativnim recenzijamasvake od četiri grupe. Metodom procesiranja teksta identificirane su najčešće riječi i izrazi (unigramii bigrami) prisutni u spomenutim grupama recenzija, zasebno za 2019. i 2021. turističku sezonu, kojemogu poslužiti hotelskom menadžmentu kod upravljanja uslugama hotelskog smještaja i ostvarivanjakonkurentske prednosti. U drugom dijelu rada, izrađen je model strojnog učenja nad svim prikupljenimrecenzijama koji klasificira recenzije u pozitivne ili negativne. Rezultati primjene tri različita algoritmastrojnog učenja sa performansama preciznosti i odziva opisani su u sekciji rezultati i diskusija.
Automated creation of a conceptual data model based on user requirements expressed in the textual form of a natural language is a challenging research area. The complexity of natural language ...requires deep insight into the semantics buried in words, expressions, and string patterns. For the purpose of natural language processing, we created a corpus of business descriptions and an adherent lexicon containing all the words in the corpus. Thus, it was possible to define rules for the automatic translation of business descriptions into the entity–relationship (ER) data model. However, since the translation rules could not always lead to accurate translations, we created an additional classification process layer—a classifier which assigns to each input sentence some of the defined ER method classes. The classifier represents a formalized knowledge of the four data modelling experts. This rule-based classification process is based on the extraction of ER information from a given sentence. After the detailed description, the classification process itself was evaluated and tested using the standard multiclass performance measures: recall, precision and accuracy. The accuracy in the learning phase was 96.77% and in the testing phase 95.79%.
Knowledge-Based Systems for Data Modelling Šuman, Sabrina; Jakupović, Alen; Kuljanac, Francesca Gržinić
International journal of enterprise information systems,
04/2016, Volume:
12, Issue:
2
Journal Article
Peer reviewed
Data modelling is a complex process that depends on the knowledge and experience of the designers who carry it out. The quality of created models has a significant impact on the quality of successive ...phases of information systems development. This paper, in short, reviews the data modelling process, the entity-relationship method (ERM) and actors in the data modelling process. Further, in more detail it presents systems, methods, and tools for the data modelling process and identifies problems that occur during the development phase of an information system. These problems also represent the authors' motivation for conducting research that aims to develop a knowledge-based system (KBS) in order to support the data modelling process by applying formal language theory (particularly translation) during the process of conceptual modelling. The paper describes the main identified characteristics of the authors' new KB system that are derived from the analysis of existing systems, methods, and tools for the data modelling process. This represents the focus of the research.
This paper presents an overview of terms, concepts, trends and technologies that are relevant to today's business. It describes the basics of data and information integration and flow in a company ...through a central ERP system with concepts of CRM and SCM. The emergence of big data as a tributary of a huge number of often unstructured data from different sources can become a central problem or opportunity for advancement and achievement of competitive advantages of a company. Ignorance of key figures and/or the non-acceptance of new business conditions, new technologies and possible deployment solutions are the main reasons for non-productivity and poor business performance. To demonstrate the dynamics of appearance and popularity of terms, concepts, trends and technologies this paper offers a tabular overview of the frequencies based on the data from 3 global databases. Meta analysis shows the expected future development of analytical trends and technologies. This paper is intended for those who lead, run and participate in projects of implementation of large software systems, dealing with quality management of business, or want to understand the complexity of this area and the future directions of development.
This paper presents an overview of terms, concepts, trends and technologies that are relevant to today's business. It describes the basics of data and information integration and flow in a company ...through a central ERP system with concepts of CRM and SCM. The emergence of big data as a tributary of a huge number of often unstructured data from different sources can become a central problem or opportunity for advancement and achievement of competitive advantages of a company. Ignorance of key figures and/or the non-acceptance of new business conditions, new technologies and possible deployment solutions are the main reasons for non-productivity and poor business performance. To demonstrate the dynamics of appearance and popularity of terms, concepts, trends and technologies this paper offers a tabular overview of the frequencies based on the data from 3 global databases. Meta analysis shows the expected future development of analytical trends and technologies. This paper is intended for those who lead, run and participate in projects of implementation of large software systems, dealing with quality management of business, or want to understand the complexity of this area and the future directions of development. Keywords: ERP; BI; DSS; SCM; CRM; big data.
TQM - A Way To Differentiation Suman, Sabrina; Pavletic, Dusko
Engineering Review,
01/2008, Volume:
28, Issue:
2
Journal Article
Peer reviewed
Quality management includes systematic usage of different methods, guidelines, techniques and tools in order to, through achievement of high quality products and processes, satisfy users' demands and ...to achieve competitive advantages and business success. Raising the quality conscience in all business processes is a main goal of TQM, and it assumes users orientation, continuous improvement and innovations, teamwork, process approach et cetera. The company in which these aspects of business are achieved may prosper and create products that are distinguished by specific quality and uniqueness. The emphasis is that the business achievement of these conditions must come from company leadership. Companies that successfully and continuously implement TQM's principles differentiate on the market; create high-grade and recognizable products, satisfied and loyal clients and motivated employees.
Glavna je tema ovog rada ispitivanje uloge informacijsko-komunikacijske tehnologije (ICT) u današnjem poslovanju, a posebno u građevinskim tvrtkama. Točnije, radi se o istraživanju usklađenosti nove ...specifične strategije ICT-a s poslovnom strategijom, a to predstavlja značajan problem i ozbiljan menadžerski izazov. Informacijsko-komunikacijska tehnologija postala je temeljno sredstvo ne samo za podupiranje operativnog posla, već se koristi i za analizu kvalitete i donošenje odluka, postizanje performansi, poboljšanje odnosa s kupcima i dobavljačima i na kraju za povećanje profita. S ciljem utvrđivanja stanja i obilježja upravljanja ICT-om u jednom određenom poslovnom području izrađen je anketni upitnik i distribuiran među najvećim hrvatskim građevinskim tvrtkama. Pitanja su postavljena na odgovarajući način kako bismo otkrili softverske pakete ICT-a koje tvrtke uglavnom koriste u postupku provedbe ICT rješenja, menadžerskom ponašanju i stavovima prema ICT menadžmentu te glavne probleme i točke zadovoljstva s gledišta poslodavaca.