Program događaja

srijeda, 28.5.2014 15:00 - 19:00, Galija, Hotel Admiral, Opatija Pozvani rad  P. Marcel (Université François Rabelais Tours, Blois, France) Log-driven user-centric OLAP  While OLAP (On-Line Analytical Processing) is now a mature, efficient technology, very little attention has been paid to the effectiveness of the analysis and the user-friendliness of a technology often considered tedious of use. This work presents various contributions to developing user-centric OLAP, focusing on the use of former queries to enhance subsequent analyses. We first introduce a logical model to describe and manipulate queries and logs, including a language parametrized by binary relations over analytical sessions. In particular, these relations can be specialization relations or based on similarity measures tailored for OLAP queries and sessions. We then describe how logs can be mined for knowledge representing the user behavior. This knowledge includes simple preferences, navigational habits and discoveries made during former explorations. We show how it can be used in various query personalization or query recommendation approaches, that vary in terms of formulation effort, proactiveness, prescriptiveness and expressive power. Finally, we remark that the models and approaches introduced can be seen as a starting point for assessing the effectiveness of analytical sessions, and bridging the gap between OLAP and exploratory search. Radovi  1. K. Royer (EDF, Moret, France), L. Bellatreche, S. Jean (LIAS, Futuroscope, France) Combining domain and business ontologies in a modular construction method: EDF study case  Ontologies are seen by companies such as EDF as an answer to the data heterogeneity problem. However the design and management of consensual and complete ontologies in new and evolving domains remain important challenges that must be tackled for the widespread adoption of this notion. In this paper we present the approach we have developed to overcome these issues in the context of a data warehouse project with EDF concerning the electric mobility development. We first propose an incremental and modular design of ontologies. Then we suggest to formalize not only the domain knowledge as an ontology but also the business processes followed to exploit this knowledge. As a result, this method allows engineers to use the ontology as an interface for manipulating heterogeneous data as well as for reusing existing business processes. Finally we report the results and benefits of our approach for the project conducted with EDF. 2. V. Jovanovic (Georgia Southern University, Statesboro, United States), D. Subotic (University of Rijeka, Department of Informatics, Rijeka, Croatia), S. Mrdalj (Eastern Michigan University, Ypsilanti, United States) Data Modeling Styles in Data Warehousing  The paper presents a coordinated set of data modeling styles relevant for data warehouse design in the context of relational databases. The scope of models covers: a) entity relationship models of existing relational DB sources, b) logical data vault model for integrated enterprise data warehouses as a system of records, c) dimensional fact model for attributes and measures analysis leading to query prototyping and dimensional models, and d) dimensional models for presentation data marts. There is no claim of sufficiency, uniqueness, and/or universality of the model set presented by selected styles. The principal contributions of the paper are: definition of data modeling styles as distinct and useful modeling mechanisms, and substantial coordination of selected styles (achieved among other means by using the same baseline standardized notation and an industrial strength case tool). The two of the styles/models are relatively less known (Dimensional Fact Model and Data Vault) and thus require more basic illustrations, and as the point of the paper is practical presentation of a set of visual modeling styles in a coordinated manner, that is showing related model representations and visualizing some transformation steps, appropriate diagrams are used instead of the text as much as possible. 3. A. Mihanović, H. Gabelica, Ž. Krstić (Poslovna inteligencija d.o.o., Zagreb, Croatia) Big Data and Sentiment Analysis using KNIME: Online Reviews vs. Social Media   Text analytics and sentiment analysis can help an organization derive potentially valuable business insights from text-based content such as word documents, email and postings on social media streams like Facebook, Twitter and LinkedIn. The system described here analyses opinions about gadgets collected from two different sources and in two different forms; online reviews and Twitter posts. Sentiment analysis can be applied to online reviews in easier and more detailed way than to the Twitter posts. Namely, online reviews are written in clear and grammatically more accurate form, while in Twitter posts, internet slang, sarcasm and allegory are often used. System explains differences in sentiment analysis of online reviews and social media and describes analysis possibilities for both data sources. 4. T. Jagušt (FER, Zagreb, Croatia), J. Stojanovski (IRB, Zagreb, Croatia), M. Baranović (FER, Zagreb, Croatia) Implementing the additional knowledge in the Croatian Scientific Bibliography  The Croatian Scientific Bibliography - CROSBI is a bibliography web site that stores information about scientific papers published in Croatia in the last 15 years. At the moment, CROSBI consists of over 400,000 records. It is daily accessed and updated by a large number of researchers from whole Croatia. As the appearance and purpose of database (and hence its structure) changed over the years, some inconsistencies emerged, and a part of existing data could not be used to its fullest potential, e.g. linking papers from the same journal or conference, or analysis of scientific activity in relation to the place of article publication. In order to clean and sanitize unstructured free-text data a set of custom ETL procedures was performed. The information about proceedings and conferences was extracted using different text similarity approaches. In order to improve data quality, book records from CROSBI were linked to the corresponding Google Books records, and using Semantic Web technologies, a number of links between CROSBI and online bibliographic databases (e.g. PubMed, DBLP) was created. The resulting data set can be used to further improve the quality of CROSBI by creating a sort of recommender system, or to improve and open up new possibilities in the analysis of existing data. 5. M. Karan, D. Pintar, Z. Skočir, M. Vranić (Faculty of Electrical Engineering and Computing, Zagreb, Croatia), A. Alajković, J. Milojević, M. Pleša (Konzum d.d., Zagreb, Croatia) The Impact of Training Data Tailoring on Demand Forecasting Models in Retail  Demand forecasting plays a very important role in retail business. Retail information systems commonly store large amounts of data which are subsequently used by sophisticated data mining tools for building forecasting models. Quality of these models is usually measured through their predictive accuracy as their most important property, followed by other measures which consider average underestimate and overestimate costs etc. Even though the choice of data mining algorithm is usually paramount, train set cleansing and preparation has a significant influence on final model performance. This article discusses and analyses the impact of training set preparation and tailoring on a final forecasting model performance used in a real world example from the retail industry. 6. T. Eterovic, S. Mrdović, D. Đonko, Ž. Jurić (Elektrotehnički fakultet Sarajevo, Sarajevo, Bosnia and Herzegovina) Data mining meets Network analysis: Traffic prediction models  Most research on network traffic prediction has been done on small datasets based on statistical methodologies. This research analyzes an internet traffic dataset spanning multiple months using the data mining process. Each data mining phase was carefully fitted to the network analysis domain and systematized in context of data mining. The second part of the paper evaluates various univariate seasonal time series prediction models, including ANN, ARIMA, Holt Winters etc., as a data mining phase on the given dataset. The experiments have shown that in most cases ANNs are superior to other algorithms for this application. 7. S. Dupor (sIT Solutions HR d.o.o., Rijeka, Croatia), V. Jovanović (Department of Computer Sciences, Georgia Southern University, Statesboro, Georgia, United States) An approach to conceptual modelling of ETL processes  In this paper, we propose a new conceptual model based on the visualization of data flow showing transformations of records accompanied with attribute transformations. A data warehouse development relies on the development of Extract, Transform, and Load (ETL) processes responsible for extracting data from transactional systems, data quality checks, transformation and conversion and loading data to target tables or cubes in a data warehouse. During the planning and design phases for data warehouse, the ETL conceptual model should be developed not only to show an overview of the whole process, but also to map sources, targets and necessary transformations, verify transformation logic, ensure that requirements are met and the needed structures and content exist. Once created, the conceptual model can be used for development of an ETL process, maintenance/optimization and redesign of a process due to new business requirements and database schema changes, all of which makes it a valuable addendum to data warehouse documentation. 8. V. Damjanovic, W. Behrendt (Salzburg Research, Salzburg, Austria) UNDERSTANDER Business Intelligence Seeker - User Agent  UNDERSTANDER is an exploratory R&D project in the field of Business Intelligence (BI). We develop a knowledge-based model of BI, which can be parametrized by the end-user. Our knowledge model combines three concepts such as (a) Business Intelligence Model (BIM) from a Canadian project for BI technologies, (b) taxonomy of Competitive Intelligence (CI), and (c) Conceptual Dependency (CD) Theory to deal with verb-oriented organization of knowledge. We investigate whether knowledge-based models can be made adequate for the problem of gathering BI from web resources. The knowledge model can be used to “prime” a software agent and the agent can then search autonomously for relevant information on the Web. For the interaction between agents, we consider “mechanism design”, which is a framework for designing interaction between self-interested agents in order to achieve specific outcomes. BI seekers, multi-user agents, are developed in WADE (Workflows and Agents Development Environment), which is an extension of JADE, a popular Open Source framework that now brings workflow mechanisms to agents. Finally, we present the design of multi-agent scenario of UNDERSTANDER with an application from the field of Industrial Internet. 9. S. Maržić, P. Krneta, M. Pavlić (Ris d.o.o., Kastav, Croatia) Spend analysis systems  This paper introduces the fields of spend analysis, spend management and supplier management. It describes the need for a spend analysis system, its’ benefits and savings that such system can help to achieve. We present a flowchart of the spend analysis process in a form of a dataflow diagram. We present the main issues in the development of a spend analysis software and provide an overview of existing spend analysis systems. 10. D. Matić (Infodom d.o.o., Zagreb, Croatia), D. Kalpić (Fakultet elektrotehnike i računarstva, Zagreb, Croatia), A. Čičak (Infodom d.o.o., Zagreb, Croatia) Poslovna inteligencija u sustavu informacijskog posrednika za elektronički račun  Elektroničko poslovanje je izvršavanje svih elemenata poslovnih aktivnosti ekonomskog procesa elektroničkim putem, odnosno stvaranje dodatnih vrijednosti koristeći se ICT tehnologijama. Informacijski posrednik je fizička ili pravna osoba koja u ime drugih obavlja otpremu, prijem i prijenos elektroničkih dokumenata. U radu će biti prikazana upotreba poslovne inteligencije unutar sustava informacijskog posrednika za elektronički račun. Sustav služi za razmjenu prvenstveno elektroničkih računa, ali i za razmjenu ostalih dokumenata u elektroničkom obliku. Poslovna inteligencija je dio sustava informacijskog posrednika koji om+ogućava praćenje i upravljanje poslovnim procesima informacijskog posrednika. Primatelji elektroničkog računa mogu pratiti iznos troškova iz primljenih e-Računa te imaju uvid u strukturu troškova. Izdavatelji elektroničkog računa imaju uvid u uspješnost odvijanja poslovnih procesa isporuke e-Računa. Potporne službe informacijskog posrednika kroz sustav poslovne inteligencije nadziru odvijanje svih poslovnih procesa unutar sustava. 11. I. Budišćak, V. Mornar, D. Mlinarić (Sveučilište u Zagrebu, Fakultet elektrotehnike i računarstva, Zagreb, Croatia) Analiza podataka u svrhu predlaganja studijskih programa u Nacionalnom informacijskom sustavu prijava na visoka učilišta  Pri odabiru studijskog programa u Nacionalnom informacijskom sustavu prijava na visoka učilišta neki pristupnici suočavaju se s problemom odabira studija. Odabir studija uglavnom se temelji na osobnim interesima pristupnika no može ovisiti i o geografskoj udaljenosti ili o nekim drugim faktorima. Drugi problem koji se javlja je pravo upisa na željeni studijski program. Jedan dio pristupnika nije uspio upisati željeni studijski program iako postoje srodni ili isti studijski programi na kojima je pristupnik mogao ostvariti pravo upisa a nalaze se npr. u drugom gradu. Ovaj problem dolazi do izražaja zbog činjenice da pristupnici imaju uvid u broj ostvarenih bodova na studijskim programima koje su odabrali na svojoj listi prioriteta a koja je ograničena na 10 studijskih programa. Ostvareni bodovi računaju se u određenim vremenskim intervalima zbog čega pristupnici ne mogu trenutno vidjeti svoj mogući plasman na studijskim programima koji nisu na njihovoj listi prioriteta. U sklopu rada izradit će se skladište podataka te provesti analiza postojećih prijava pristupnika, ocjena u srednjoj školi, ocjena na ispitima državne mature, prijava pristupnika iz iste škole odnosno mjesta kako bi se pristupnicima predložili studijski programi koji bi odgovarali njihovim interesima a na kojima bi mogli ostvariti pravo upisa. 12. K. Matejčić, Z. Mršo (IN2 d.o.o., Zagreb, Croatia) Aplikacija za uvoz podataka iz vanjskih izvora u skladište podataka   Aplikacija za uvoz podataka iz vanjskih izvora u DWH omogućuje korisniku da samostalno unosi vanjske podatke u BI sustav. Korisnici BI sustava često imaju potrebu za podacima koji nisu obuhvaćeni redovitom ekstrakcijom podataka iz standardnih izvora (transakcijske baze podataka, ERP sustavi i sl.).. Ti podaci nisu obuhvaćeni ekstrakcijom iz raznih razloga: dolaze sporadično, dolaze svaki put u drugačijem obliku, potječu iz izvora koji nisu povezani s centralnim informacijskim sustavom poduzeća ... Pomoću aplikacije za uvoz podataka korisnik ima mogućnost da sam mapira podatke iz excela (obične ili pivot tablice) na vrijednosti mjera ili dimenzija u skladištu podataka. Aplikacija se brine za konzistentnost skladišta podataka, tj. ne dozvoljava unos na nepoznate dimenzijske vrijednosti, te traži od korisnika da dimenzijske vrijednosti koje nisu u sustavu mapira na neku dimenzijsku vrijednost. Podržana je verifikacija podataka, u smislu da uneseni podaci neće biti vidljivi korisnicima (iako su unijeti u sustav) dok verifikator ne označi podatke vidljivima. Aplikacija je bazirana na Oracle tehnologiji i integrirana je se OBIEE izvještajnim alatom.

Boris Vrdoljak (Croatia), Mirta Baranović (Croatia), Matteo Golfarelli (Italy), Roberto Sandri (Croatia)

Alberto Abello Gamazo (Spain), Marko Banek (Croatia), Mirta Baranović (Croatia), Ladjel Bellatreche (France), Ljiljana Brkić (Croatia), Alfredo Cuzzocrea (Italy), Jamel Feki (Tunisia), Dragan Gamberger (Croatia), Matteo Golfarelli (Italy), Igor Mekterović (Croatia), Damir Pintar (Croatia), Zoran Skočir (Croatia), A Min Tjoa (Austria), Mladen Varga (Croatia), Boris Vrdoljak (Croatia), Robert Wrembel (Poland)

Boris Vrdoljak
Fakultet elektrotehnike i računarstva
Unska 3
10000 Zagreb, Hrvatska

Tel.: +385 1 6129 532
Fax: +385 1 6129 915
E-mail: boris.vrdoljak@fer.hr 

Opatija, često nazivana i biser Jadrana, jedna je od najpoznatijih destinacija u Hrvatskoj s turističkom tradicijom koja ove godine slavi 170 godina. Smještena uz rub Mediterana, na obali Kvarnerskog zaljeva i padinama planine Učke koje se blago spuštaju prema moru, Opatija svojim klimatskim karakteristikama, prekrasnom arhitekturom, kvalitetnim hotelima, prekrasnim njegovanim parkovima i šetnicama pruža mogućnosti za ugodan boravak tijekom cijele godine.

Prvi je čari Opatije otkrio riječki patricij Iginio Scarpa, sagradivši ovdje 1844. godine svoj ljetnikovac koji je nazvao Villa Angiolina po svojoj preminuloj supruzi, čime je otvoren put razvoju turizma. To je bio tek početak.

Hotel je otvoren 27. ožujka 1884. godine. Nazvan je Hotel Quarnero, a gosti su na raspolaganju imali 60 soba. U to vrijeme nastaju i brojne luksuzne vile, ljetnikovci, hoteli, pansioni, sanatoriji, paviljoni, kupališta, šetališta i parkovi. U Opatiju dolaze najpoznatiji liječnici Monarhije i otvaraju sanatorije, grade se šetališta i kupališta. Opatija je tako uz Nicu, Karlove Vare, Cannes i Biarritz postala jedno od najznačajnijih europskih mondenih lječilišta devetnaestog i prve polovice dvadesetog stoljeća.

Ovdje su boravili kraljevi i carevi, književnici, filozofi, pisci, pjesnici i glazbenici. Spomenimo cara Franju Josipa, cara Vilima II, rumunjsku kraljicu Elizabetu, poznatu kao pjesnikinju koja je svoje pjesme objavljivala pod pseudonimom Carmen Sylva, zatim caricu Sissi, pisce A. P. Čehova i Jamesa Joycea, plesačicu Isadoru Duncan, skladatelje Gustava Mahlera i Giacoma Puccinija. Vidjeti i biti viđen - za to je trebalo doći u Opatiju.

Nakon II. svjetskog rata, turizam u Opatiji orijentira se na ljetnu sezonu i na razvoj kongresne ponude. Kroz svo to vrijeme, Opatija je svoju važnost destinacije koja pruža nešto posebno zadržala do danas.

Osvojit će vas mnogočime - prvorazrednim kongresnim sadržajima, odličnim smještajem, wellness centrima za odmor i opuštanje, te ljubaznim i uslužnim osobljem koje će izaći ususret svim vašim željama. Ne zaboravimo ni gastronomiju kroz koju ćete upoznati naše autohtone namirnice, jela i prvoklasna vina.

Upoznajte Opatiju, i neka Opatija upozna vas!

Detaljnije informacije se mogu potražiti na www.opatija.hr i www.opatija-tourism.hr.