Ponedeljkov seminar računalništva in informatike - Arhiv
| 2024 | 2023 | 2022 | 2021 | 2020 | 2019 | 2018 | 2017 | 2016 | 2015 | 2014 | 2013 | 2012 |
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
sobota, 28. november 2020 Challenges for Data Mining in Distributed Sensor Networks
V ponedeljek, 30. novembra 2020, bo ob 16.00 uri prek spletnih orodij na daljavo izvedeno predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 7. december 2020 ob 16.00 na daljavo
-----------------------------------------
PREDAVATELJ: Branko KAVŠEK
-----------------------------------------
Branko Kavšek has a PhD in Computer Science from the University of Ljubljana. He is an assistant professor at the University of Primorska, Faculty of Mathematics, Natural Sciences and Information Technologies, a researcher at the Department of Information Sciences and Technologies and member of the Artificial Intelligence Laboratory at the Jozef Stefan Institute in Ljubljana. His research fields include Artificial Intelligence, Data Mining, Machine Learning. In particular, the sub-fields of Supervised and Unsupervised Learning, Frequent Pattern Discovery and Association Rule Learning, Learning Probabilistic Models and Bayesian Networks Learning, Clustering, and Data Mining applied to Big Data.
---------------------------------------------------------------------------------------------
NASLOV: Challenges for Data Mining in Distributed Sensor Networks
---------------------------------------------------------------------------------------------
POVZETEK:
In this seminar we will present an overview of Distributed Sensor Networks (DSNs) with particular emphasis on Wireless Sensor Networks (WSNs) and the use of Data Mining (DM) to analyze the data collected by WSNs. We will start by giving an overview of the history of sensor networks, stressing the technological issues, distributed data management and distributed hypothesis formation. In the second part we will present the application of simple DM methods on data collected during the ARRS project "Autonomic edge computing for air quality monitoring". The last part of the seminar will be devoted to presenting evolution opportunities and challenges.
---------------------------------------------------------------------------------------------------
Predavanje bo potekalo v angleškem jeziku prek spletnega orodja Zoom.
Do predavanja dostopate tako, da se povežete prek sledeče povezave:
https://us02web.zoom.us/j/297328207
Vabljeni!
četrtek, 29. oktober 2020 Đorđe KLISURA: Analyzing Global Terrorism Database with R
V ponedeljek, 2. novembra 2020, bo ob 16.00 uri prek spletnih orodij na daljavo izvedeno predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 2. november 2020 ob 16.00 na daljavo
-----------------------------------------
PREDAVATELJ: Đorđe KLISURA
-----------------------------------------
Đorđe Klisura comes from Serbia. He is a first-year student of Data
Science program at UP FAMNIT. He graduated from UP FAMNIT and received
Bachelor’s degree in Computer Science.
-------------------------------------------------------------------------
NASLOV: Analyzing Global Terrorism Database with R
-------------------------------------------------------------------------
POVZETEK:
With the increase of terrorist activity around the world, it has become
more important than ever to analyze and understand these activities over
time. Although the data on terrorist activities are detailed and
relevant, the complexity of the data has rendered the understanding and
analysis difficult. We present a visual analytical approach to
effectively identify related entities. We will be working with The
Global Terrorism Database (GTD), that is an open-source database
including information on domestic and international terrorist attacks
around the world from 1970 through 2017, and now includes more than
180,000 cases.
---------------------------------------------------------------------------------------------------
Predavanje bo potekalo v angleškem jeziku prek spletnega orodja Zoom.
Do predavanja dostopate tako, da se povežete prek sledeče povezave:
https://us02web.zoom.us/j/297328207
Vabljeni!
petek, 25. september 2020 Sead JAHIĆ: Handling tweet opinion by using weighted lexicon and lexicon-based approach
V ponedeljek, 28. septembra 2020, bo ob 16.00 uri prek spletnih orodij na daljavo izvedeno
predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE
Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 28. september 2020 ob 16.00 na daljavo
-----------------------------------
PREDAVATELJ: Sead JAHIĆ
-----------------------------------
Sead Jahić is a teaching assistant at University of Primorska, Department of Information Sciences and Technologies.
After receiving High School diploma, he went to University of Tuzla (UNTZ), where in 2012 he graduated as Mathematician (Math Teacher).
His first working experience was teaching assistant at UNTZ, in the field of Applied Mathematics and Computer Science.
The main topic of his research is Sentiment Analysis of languages, where the focus is on developing overall sentiment analysis for West Balkan languages from the same West Slavic group.
The idea is to use different Machine Learning and Deep Learning techniques in sentiment analysis with those languages, and show huge connectivity between them.
-----------------------------------------------------------------------------------------------------------------------
NASLOV: Handling tweet opinion by using weighted lexicon and lexicon-based approach
-----------------------------------------------------------------------------------------------------------------------
POVZETEK:
In this seminar, we will present an introduction to creating a lexicon with weight values of words from our database. Given that the lexicon is made in such a way that the amount / number of positive and negative words affects the final outcome, we come to the conclusion that at some point, especially in neutral sentiment, we have words whose power can change the polarity of neutral to positive / negative. Therefore, when calculating sentiment, we relied on the weight lexicon created for a given amount of data.
---------------------------------------------------------------------------------------------------
Predavanje bo potekalo v angleškem jeziku prek spletnega orodja Zoom.
Do predavanja dostopate tako, da se povežete prek sledeče povezave:
https://us02web.zoom.us/j/297328207
Vabljeni!
torek, 22. september 2020 Patrik KOCJANČIČ: Primerjava vnosa besedila v virtualnem okolju na različnih postavitvah tipkovnice
Tokrat, izjemoma v četrtek, 24. septembra 2020, bo ob 10.00 uri prek spletnih orodij na daljavo izvedeno
predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE
Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 24. september 2020 ob 10:00 na daljavo
--------------------------------------------
PREDAVATELJ: Patrik KOCJANČIČ
--------------------------------------------
Patrik Kocjančič je študent magistrskega študijskega programa Računalništvo in informatika na UP-FAMNIT. Trenutno se ukvarja z magistrskim delom, v katerem skuša najti optimalno rešitev (postavitev tipkovnice) za vnos besedila v virtualnem okolju.
------------------------------------------------------------------------------------------------------------------------------
NASLOV: Primerjava vnosa besedila v virtualnem okolju na različnih postavitvah tipkovnice
------------------------------------------------------------------------------------------------------------------------------
POVZETEK:
Vnos besedila je precej raziskano področje tako v realnem svetu kot tudi v virtualnem svetu. Raziskovalci preučujejo različne načine vnosa besedila bodisi tip vnosa bodisi vnos na različnih postavitvah tipkovnice. V tem seminarju bomo predstavili področje tehnologije VR, pripravo virtualnega okolja, študijo, ki je potekala v okviru magistrskega dela ter rezultate le-te. Cilj študije je bil preizkusiti različne postavitve tipkovnice in le-te primerjati s standardom QWERTZ. Testirali smo 4 različne postavitve tipkovnice, to so: QWERTZ, DVORAK, OPTI in CIRCLE. Na koncu bo potekala še krajša razprava o rezultatih in kakšna možna izboljšava oz. predlogi pri izvedbi nadaljnjih študij.
---------------------------------------------------------------------------------------------------
Predavatelj bo svoj prispevek predstavil v okviru Raziskovalnega seminarja na magistrskem študijskem programu Računalništvo in informatika.
Predavanje bo potekalo v slovenskem jeziku prek spletnega orodja Zoom.
Do predavanja dostopate tako, da se povežete prek sledeče povezave:
https://us02web.zoom.us/j/297328207
Vabljeni!
sobota, 19. september 2020 Jordan Aiko DEJA: Revisiting AR Piano Projection
V ponedeljek, 21. septembra 2020, bo ob 16.00 uri prek spletnih orodij na daljavo izvedeno
predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE
Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 21. september 2020 ob 16.00 na daljavo
--------------------------------------------
PREDAVATELJ: Jordan Aiko DEJA
--------------------------------------------
Jordan Aiko Deja is a PhD student from UP FAMNIT, working on computational interactions and adaptive visualizations on music learning. He is part of the HICUP Lab Research group led by Dr Matjaž Kljun. Prior to that, he earned his bachelor's and master's degree in Computer Science from De La Salle University in Manila, Philippines.
------------------------------------------------------
NASLOV: Revisiting AR Piano Projection
------------------------------------------------------
POVZETEK:
---------------------------------------------------------------------------------------------------
Predavanje bo potekalo v angleškem jeziku prek spletnega orodja Zoom.
Do predavanja dostopate tako, da se povežete prek sledeče povezave:
https://us02web.zoom.us/j/297328207
Vabljeni!
petek, 11. september 2020 Jamolbek MATTIEV: Distance based Clustering of Class Association Rules to Build a Compact, Accurate and Descriptive Classifier
V ponedeljek, 14. septembra 2020, bo ob 16.00 uri prek spletnih orodij na daljavo izvedeno predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 14. september 2020 ob 16.00 na daljavo
----------------------------------------------
PREDAVATELJ: Jamolbek MATTIEV
----------------------------------------------
Jamolbek Mattiev has a Master degree in Computer Science from National University of Uzbekistan. He was awarded with first degree diploma at “the best Master Dissertation Work of Uzbekistan” competition in his master studies. He is a Teaching Assistant at the University of Primorska, Faculty of Mathematics, Natural Sciences and Information Technologies and he is final year PhD student at the Department of Information Sciences and Technologies.
------------------------------------------------------------------------------------------------------------------------------------
NASLOV: Distance based Clustering of Class Association Rules to Build a Compact, Accurate and Descriptive Classifier
------------------------------------------------------------------------------------------------------------------------------------
POVZETEK:
---------------------------------------------------------------------------------------------------
Predavanje bo potekalo v angleškem jeziku prek spletnega orodja Zoom.
Do predavanja dostopate tako, da se povežete prek sledeče povezave:
https://us02web.zoom.us/j/297328207
Vabljeni!
sreda, 9. september 2020 Sead JAHIĆ: Determining sentiment of tweets using first Bosnian lexicon and (AnA)-affirmative and non-affirmative words
Tokrat, izjemoma v četrtek, 10. septembra 2020, bosta ob 10.00 uri prek spletnih orodij na daljavo izvedeni dve predavanji v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE
Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 10. september 2020
1. predavanje bo ob 10.00:
======================
---------------------------------------
PREDAVATELJ: Nikola MARIN
---------------------------------------
Nikola Marin is a MSc student of Computer Science at UP FAMNIT. He recently became interested in the business application industry and explores business application design and development.
------------------------------------------------------------------------------------------------------------------------------
NASLOV: Enterprise Resource Planning (ERP) System for Subcontractors of Telekom Slovenije
------------------------------------------------------------------------------------------------------------------------------
POVZETEK:
Telekom Slovenije has many subcontractors supporting their infrastructure and installations. Those subcontractors are small and medium-size businesses who are using just email and a Microsoft Office package for their e-business. Because of that, they experience data loss, misunderstandings, human errors and non-optimal business. Based on previous research where business process analysis, ERP implementation technologies and methodologies were explored, this seminar aims to analyse business processes inside Telekom Slovenije subcontractor companies and to propose a conceptual model of ERP system that could fit and provide optimal business in the long run for such companies.
---------------------------------------------------------------------------------------------------
The seminar presentation will be part of the Research Seminar on the Computer Science Masters Program.
===========================================================================
2. predavanje bo ob 10.30:
======================
---------------------------------------
PREDAVATELJ: Sead JAHIĆ
---------------------------------------
Sead Jahić is a teaching assistant at University of Primorska, Department of Information Sciences and Technologies.
After receiving High School diploma, he went to University of Tuzla (UNTZ), where in 2012 he graduated as Mathematician (Math Teacher).
His first working experience was teaching assistant at UNTZ, in the field of Applied Mathematics and Computer Science.
The main topic of his research is Sentiment Analysis of languages, where the focus is on developing overall sentiment analysis for West Balkan languages from the same West Slavic group.
The idea is to use different Machine Learning and Deep Learning techniques in sentiment analysis with those languages, and show huge connectivity between them.
------------------------------------------------------------------------------------------------------------------------------------------------------------------
NASLOV: Determining sentiment of tweets using first Bosnian lexicon and (AnA)-affirmative and non-affirmative words
------------------------------------------------------------------------------------------------------------------------------------------------------------------
POVZETEK:
The talk will present a simple model for sentiment tagging of Tweeter messages based on AnA (affirmative and non-affirmative) words and tagged sentiment dictionary (Kadunc-Šikonja dictionary).
Lexicon was used in combination with intensifiers, where intensifier (AnA word), that stand next to the word, produce more powerful (more positive or more negative) sentiment value of word - tweet at all.
This also presents our first step in creating the first Bosnian lexicon and proceed the same on Sentiment analysis of tweets written in Bosnian where the major topic will be tweeting about Bosnian National basketball player Džanan Musa.
---------------------------------------------------------------------------------------------------
Predavanji bosta potekali v angleškem jeziku prek spletnega orodja Zoom.
Do predavanj dostopate tako, da se povežete prek sledeče povezave:
https://us02web.zoom.us/j/297328207
Vabljeni!
sreda, 26. avgust 2020 Niki HROVATIN: Naivni Bayes in umetna konvolucijska nevronska mreža za identifikacijo padcev s pomočjo pametnih tal
Tokrat v petek, 14. avgusta 2020, bo ob 10.00 uri prek spletnih orodij na daljavo izvedeno predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 14. avgust 2020 ob 10.00 na daljavo
----------------------------------------------
PREDAVATELJ: Niki HROVATIN
----------------------------------------------
Niki Hrovatin je študent Magistrskega študijskega programa računalništva in informatike na UP-FAMNIT. Trenutno se ukvarja z magistrsko nalogo v kateri skuša razvit sistem za neintruzivno identifikacijo padcev.
----------------------------------------------
NASLOV: Naivni Bayes in umetna konvolucijska nevronska mreža za identifikacijo padcev s pomočjo pametnih tal
----------------------------------------------
POVZETEK:
Da lahko nudimo takojšnjo pomoč, je potrebno v čim krajšem času zaznati padec, ampak v nekaterih primerih se oseba tako hudo poškoduje, da ne utegne klicati na pomoč in padec je prepozno opažen ter oseba doživi dodatne poškodbe zaradi dodatnega časa.
Na seminarju bo predstavljen sistem za identifikacijo padcev osnovan na prototipu tal z vgrajenimi senzorji pritiska. Predstavili bomo proces zbiranja podatkov, podatke in postopek pred obdelave podatkov, uvedena bo metoda Naivnega Bayesa in umetne konvolucijske nevronske mreže ter na koncu bo predstavljena evalvacija modelov in rezultati.
----------------------------------------------
Predavanje bo potekalo v slovenskem jeziku prek spletnega orodja Zoom.
Do predavanja dostopate tako, da se povežete prek sledeče povezave:
https://us02web.zoom.us/j/297328207
Vabljeni
nedelja, 5. julij 2020 Cuauhtli CAMPOS: Pinhole Paper: Making Static Objects Perceptually Dynamic with Rear Projection on Perforated Paper
V ponedeljek, 6. julija 2020, bo ob 17.00 uri prek spletnih orodij na daljavo izvedeno predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 6. julij 2020 ob 17.00 na daljavo
Pozor! PON seminar RIN tokrat, izjemoma, prične ob 17h (ne ob 16h kot je to običajno)
Warning! The MON CS Seminar starts at 17h this time (not at 16h as usual)
----------------------------------------------
PREDAVATELJ: Cuauhtli CAMPOS
----------------------------------------------
Cuauhtli Campos is a PhD student in Computer Science in the field of Human-Computer Interaction at the University of Primorska, Faculty of Mathematics, Natural Sciences and Information Technologies. He is also a member of the HICUP Lab at the Department of Information Sciences and Technologies.
-----------------------------------------------------------------------------------------------------------------------------------
NASLOV: Pinhole Paper: Making Static Objects Perceptually Dynamic with Rear Projection on Perforated Paper
-----------------------------------------------------------------------------------------------------------------------------------
POVZETEK:
We explore paper augmentation technique where a tablet computer is placed bellow the paper to modify its content appearance and create the illusion of making static object perceptually dynamic. Our approach is based on Deformation Lamps (DL) technique applied to rear projection (instead of commonly used front projection), which removes the need for fix setups or handling and wearing dedicated hardware. The key challenge is to ensure appropriate mixing of light reflecting from the paper (e.g. what is printed on the paper) and the light that is coming though the paper (e.g. tablet screen). This is particularly the case as a relatively weak light source of a tablet screen is being used. We address this through modification of optical properties of paper with perforation and a new method for generating dynamic luminance imaging which successful make static object perceptually dynamic. The outcomes provide a guide to how paper needs to be modified in order to achieve the illusion of making static object perceptually dynamic.
---------------------------------------------------------------------------------------------------
Predavanje bo potekalo v angleškem jeziku prek spletnega orodja Zoom.
Do predavanja dostopate tako, da se povežete prek sledeče povezave:
https://us02web.zoom.us/j/297328207
Vabljeni
nedelja, 31. maj 2020 Nuwan ATTYGALLE: Pandemic era app development
V ponedeljek, 1. junija 2020, bo ob 16.00 uri prek spletnih orodij na daljavo izvedeno predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 1. junij 2020 ob 16.00 na daljavo.
----------------------------------------------
PREDAVATELJ: Nuwan ATTYGALLE
----------------------------------------------
Nuwan Attygalle is a PhD student in Computer Science in the field of Human-Computer interactions and a teaching assistant at the University of Primorska, Faculty of Mathematics, Natural Sciences and Information Technologies. He is also a member of the HICUP Lab at the Department of Information Sciences and Technologies.
--------------------------------------------------------
NASLOV: Pandemic era app development
--------------------------------------------------------
Key to end, control the spread in an epidemic is rapid identification and quarantine of the infected individuals, determination of whom they have had close contact within the previous days and decontamination of locations the infected individual has visited.
This process is labour intensive, susceptible to memory errors, and fraught with privacy concerns. But the main advantage in this era is the availability of smart-phones, mobile devices, and with the right technology we can overcome above difficulties and track down infected individuals in a much productive way.
---------------------------------------------------------------------------------------------------
Predavanje bo potekalo v angleškem jeziku prek spletnega orodja Zoom.
Do predavanja dostopate tako, da se povežete prek sledeče povezave:
https://us02web.zoom.us/j/297328207
Vabljeni!
nedelja, 24. maj 2020 Jamolbek MATTIEV: Clustering class association rules to form a Compact and Meaningful Associative Classifier
V ponedeljek, 25. maja 2020, bo ob 16.00 uri prek spletnih orodij na daljavo izvedeno predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 25. maj 2020 ob 16.00 na daljavo
----------------------------------------------
PREDAVATELJ: Jamolbek MATTIEV
----------------------------------------------
Jamolbek Mattiev has a Master degree in Computer Science from National University of Uzbekistan. He was awarded with first degree diploma at “the best Master Dissertation Work of Uzbekistan” competition in his master studies. He is a Teaching Assistant at the University of Primorska, Faculty of Mathematics, Natural Sciences and Information Technologies and he is the third year PhD student at the Department of Information Sciences and Technologies.
He is doing his PhD in the field of Computer Science (Data Mining). His research fields include Artificial Intelligence, Data Mining, Machine Learning. In particular, the sub-fields of Supervised and Unsupervised Learning, Frequent Pattern Discovery, Association Rule Learning, Classification and Clustering.
---------------------------------------------------------------------------------------------------------------------------------------
NASLOV: Clustering class association rules to form a Compact and Meaningful Associative Classifier
---------------------------------------------------------------------------------------------------------------------------------------
Huge amounts of data are being collected and analyzed nowadays. By using the popular rule-learning algorithms, the number of rules discovered on those big datasets can easily exceed thousands of rules. To produce compact and accurate classifiers, such rules have to be grouped and pruned, so that only a reasonable number of them are presented to the end user for inspection and further analysis. To solve this problem researchers have proposed several associative classification approaches that combine two important data mining techniques, namely, classification and association rule mining.
---------------------------------------------------------------------------------------------------
Predavanje bo potekalo v angleškem jeziku prek spletnega orodja Zoom.
Do predavanja dostopate tako, da se povežete prek sledeče povezave:
https://us02web.zoom.us/j/297328207
Vabljeni!
petek, 15. maj 2020 Cuauhtli CAMPOS: Interactive paper with back print illumination and rear-projection
V ponedeljek, 18. maja 2020, bo ob 16.00 uri prek spletnih orodij na daljavo izvedeno predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 18. maj 2020 ob 16.00 na daljavo
------------------------------------------
PREDAVATELJ: Cuauhtli CAMPOS
------------------------------------------
Cuauhtli Campos is a PhD student in Computer Science in the field of Human-Computer Interaction at the University of Primorska, Faculty of Mathematics, Natural Sciences and Information Technologies.
-------------------------------------------------------------------------------------------------------------------------
NASLOV: Interactive paper with back print illumination and rear-projection
-------------------------------------------------------------------------------------------------------------------------
POVZETEK:
We live in an era where our society remains to be heavily dependent on paper. The simplicity, sensory richness and versatility enabled the paper to maintain its importance in this digital age. However despite these obvious advantages of paper our lives are becoming more and more digital. This introduces a need for new solutions that will integrate digital technology with paper whilst maintaining its unique aforementioned properties. This talk presents the findings of a study of rear projection and back print illumination using paper as screen material.
---------------------------------------------------------------------------------------------------
Predavanje bo potekalo v angleškem jeziku prek spletnega orodja Zoom.
Do predavanja dostopate tako, da se povežete prek sledeče povezave: https://us02web.zoom.us/j/297328207
Vabljeni!
nedelja, 10. maj 2020 Elham MOTAMEDI: User-centric multimedia characteristics: A Survey
V ponedeljek, 11. maja 2020, bo ob 16.00 uri prek spletnih orodij na daljavo izvedeno predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 11. maj 2020 ob 16.00 na daljavo
-------------------------------------------------
PREDAVATELJICA: Elham MOTAMEDI
-------------------------------------------------
Elham Motamedi is a PhD student at UP FAMNIT. She completed her Master's degree of networked systems at the University of Tehran, Iran and her Bachelor's degree of information technology at the Isfahan University of Technology, Iran. Her research interests include multimedia and recommender systems.
--------------------------------------------------------------------------------
NASLOV: User-centric multimedia characteristics: A Survey
--------------------------------------------------------------------------------
The properties of multimedia content are usually represented as content-centered metadata, such as type (e.g. video, audio, text, etc.), duration, resolution, etc. However, in user-centric applications, such as personalized recommender systems, it makes more sense to describe multimedia items with user-centric meta-data, such as genre, emotions, etc. For example, a recommender system would make a better job of explaining recommendations in the form of "You have been recommended this film because you liked comedies" than "You have been recommended this film because you liked 1:30 long films". The problem of such a characterization is that it usually requires humans to annotate with user-centric metadata, which is expensive. Hence, research efforts have been made to develop automatic methods for mapping multimedia content to user-centric item characteristics.
In this presentation, I present some of the state-of-the-art studies in studying automatic approaches for predicting user-centric characteristics of multimedia contents and some raw ideas of future work.
---------------------------------------------------------------------------------------------------
Predavanje bo potekalo v angleškem jeziku prek spletnega orodja Zoom.
Do predavanja dostopate tako, da se povežete prek sledeče povezave:
https://us02web.zoom.us/j/297328207
Vabljeni!
nedelja, 3. maj 2020 Vladimir BATAGELJ: NetsJSON - part 2
On Monday, 4th of May 2020 at 4 PM we kindly invite you to attend the (on-line) MONDAY COMPUTER SCIENCE SEMINAR organized by the Department of Information Sciences and Technologies at UP FAMNIT and UP IAM, this time joint with the MATHEMATICAL RESEARCH SEMINAR organized by the Department of Mathematics at UP FAMNIT and UP IAM.
TIME and PLACE: 4th of May 2020 at 4 PM on-line
-------------------------------------------
PRESENTER: Vladimir BATAGELJ
-------------------------------------------
Vladimir Batagelj, Professor Emeritus
IMFM - Institute of Mathematics, Physics, and Mechanics
Jadranska 19, 1000 Ljubljana, Slovenia and
University of Primorska, Andrej Marušič Institute, Koper and
NRU HSE, Applied Network Research Lab, Moscow
W: http://vladowiki.fmf.uni-lj.si/doku.php?id=vlado
RECENT PUBLICATIONS:
On fractional approach to analysis of linked networks
(https://doi.org/10.1007/s11192-020-03383-y)
Clustering and Generalized ANOVA for Symbolic Data
(http://www.iste.co.uk/book.php?id=1597)
Temporal bibliographic networks
(https://authors.elsevier.com/a/1aRuH6EAijpVq-)
Advances in Network Clustering and Blockmodeling (book)
(https://www.wiley.com/en-si/Advances+in+Network+Clustering+and+Blockmodeling-p-9781119224709)
---------------------------------
TITLE: NetsJSON - part 2
---------------------------------
ABSTRACT:
A network consists of a graph and additional data on (properties of) nodes and/or links.
In some applications, the values of properties can be structured objects such us intervals, compositions, sets of keywords, temporal quantities, etc. Standard network description formats do not support such kinds of networks.
We propose a JSON based format NetsJSON as a general network description option. It is supported in Python by a network analysis library Nets. The NetsJSON descriptions can be used as input data for network visualizations based on D3.js.
REFERENCES:
[1] Batagelj, V.: Python Packages for Networks. Encyclopedia of Social Network Analysis and
Mining. Reda Alhajj, Jon Rokne (Eds.), Springer, 2017
https://link.springer.com/referenceworkentry/10.1007%2F978-1-4614-7163-9_110210-1
[2] Batagelj, V., Maltseva, D.: Temporal bibliographic networks. Journal of Informetrics,
14(2020)1, 101006. https://www.sciencedirect.com/science/article/pii/S1751157719301439
[3] Batagelj, V.: Nets. https://github.com/bavla/Nets
[4] Batagelj, V.: NetsJSON. https://github.com/bavla/netsJSON
---------------------------------------------------------------------------------------------------
The seminar will be held in English through the videoconferencing tool Zoom.
This seminar is the follow-up of the seminar given on the 6th of April 2020.
You can connect to the seminar using the following link:
You are very welcome!
nedelja, 19. april 2020 Sidra ASLAM: Personal Information Management on Distributed Ledger Technology
V ponedeljek, 20. aprila 2020, bo ob 16.00 uri prek spletnih orodij na daljavo izvedeno predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 20. april 2020 ob 16.00 na daljavo
-----------------------------------------
PREDAVATELJICA: Sidra ASLAM
-----------------------------------------
Sidra Aslam is a PhD student at the University of Primorska. She did her Master in Computer Science from COMSATS University, Pakistan. She followed a complementary Master in 2019 at the University of Saarland, Germany. Her research interests include information security and privacy, blockchain, and semantic web technologies.
-----------------------------------------------------------------------------------------------------------
NASLOV: Personal Information Management on Distributed Ledger Technology
-----------------------------------------------------------------------------------------------------------
POVZETEK:
In wood supply chain environments, stakeholders need full traceability of individual products, and at the same time, they need protection from disclosure of their global activity. Existing solutions rely on centralized system to manage privacy-sensitive data and suffer from single point of failure, lack of trust, and performance issues. Distributed ledger technology ensures data security, however, user-sensitive information remains publicly available, which leads to privacy issues. In our research, we explore how to protect sensitive information over distributed ledger technology while providing product traceability through the supply chain.
In this presentation, we will illustrate our research problem with a use case scenario inspired from a wood supply chain. Then, we will give a brief overview of the state of the art, before describing open issues and discussing possible solutions that we are currently exploring.
---------------------------------------------------------------------------------------------------
Predavanje bo potekalo v angleškem jeziku prek spletnega orodja Zoom.
Do predavanja dostopate tako, da se povežete prek sledeče povezave:
https://us02web.zoom.us/j/297328207
Vabljeni!
petek, 27. marec 2020 András BÓTA: Network-based framework for estimating the risk factors of epidemic spreading
V ponedeljek, 30. marca 2020, bo ob 16.00 uri prek spletnih orodij na daljavo izvedeno predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 30. marec 2020 ob 16.00 na daljavo
------------------------------------------
PREDAVATELJ: András BÓTA
------------------------------------------
András Bóta finished his postgraduate studies in Computer Science at the University of Szeged under the supervision of Dr. Miklós Krész and Dr. András Pluhár in 2015, then worked as a Research Associate at the University of New South Wales with Dr. Lauren Gardner for two years. Following that, he worked as a postdoctoral fellow at Umeå University. His main research interests lie in the fields of complex network analysis and artificial intelligence.
-------------------------------------------------------------------------------------------------------------------------
NASLOV: Network-based framework for estimating the risk factors of epidemic spreading
-------------------------------------------------------------------------------------------------------------------------
POVZETEK:
Network-based adaptations of traditional compartmental infection models such as SIR or SEIR can be used to model the spreading of diseases between cities, countries or other geographical regions. One of the common challenges arising in these applications is the lack of available transmission probabilities between these geographical units. The task of inverse infection is the systematic estimation of these values. Several methods have been proposed recently for solving this task. One of them is the Generalized Inverse Infection Model (GIIM) [1]. GIIM offers a large amount of modeling flexibility and allows transmission probabilities to be defined as a function of known attributes, or risk factors in an epidemic context. In this presentation we will see how GIIM works in two specific real-life outbreaks.
Both examples are embedded in a geographical and temporal setting. The first one considers the 2015-2016 Zika virus outbreak in the Americas, where the countries and overseas territories of the continent form the nodes of the network and air travel routes define the links [2]. The second application models the 2009 H1N1 outbreak between the municipalities of Sweden, with links between the municipalities indicating frequent travel routes.
Our first goal in both of these studies is to discover the relationship between the transmission risk between geographical units and a variety of travel, environmental, meteorological and socioeconomic risk factors. Our second goal is to estimate the risk of exportation and importation of the diseases for the territories involved in these studies. We will show that the GIIM model is able to identify the most critical risk factors in both scenarios, and in the influenza study, it is able make predictions about future outbreaks with good accuracy.
REFERENCE:
1. A. Bóta, L. M. Gardner: A generalized framework for the estimation of edge infection probabilities. arXiv:1706.07532 (2017)
2. L. M. Gardner, A. Bóta, N. D. Grubaugh, K. Gangavarapu, M. U. G. Kramer: Inferring the risk factors behind the geographical spread and transmission of Zika in the Americas.PLoS Neglected Tropical Diseases. http://journals.plos.org/plosntds/article?id=10.1371/journal.pntd.0006194
---------------------------------------------------------------------------------------------------
Predavanje bo potekalo v angleškem jeziku prek spletnega orodja Zoom.
Do predavanja dostopate tako, da se povežete prek sledeče povezave: https://zoom.us/j/297328207
Vabljeni!
sobota, 14. marec 2020 Jordan Aiko DEJA: Using EEG for Emotion Experience Design in TV Commercials
V ponedeljek, 16. marca 2020, bo ob 16.00 uri prek spletnih orodij na daljavo izvedeno
predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE
Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 16. marec 2020 ob 16.00 na daljavo
Predavanje bo potekalo v angleškem jeziku prek spletnega orodja Zoom.
Do predavalja dostopate tako, da se povežete prek sledeče povezave: https://zoom.us/j/297328207
--------------------------------------------
PREDAVATELJ: Jordan Aiko DEJA
--------------------------------------------
Jordan Aiko Deja is a UX Practitioner with an HCI and AI background. He has led several teams and projects that have been used for various digital, innovative, and even AI-empowered products and initiatives used both locally and internationally. He has done work fusing HCI and AI that are currently being used in marketing, digital production, therapy, and many others. He is currently a computer science PhD student at UP FAMNIT.
-----------------------------------------------------------------------------------------------
NASLOV: Using EEG for Emotion Experience Design in TV Commercials
-----------------------------------------------------------------------------------------------
POVZETEK:
TV Commercials are audio video segments played by paying advertisers in between the consumers' favorite shows. They are usually aired for marketing purposes especially towards selling specific products or services. In the process of creating TV Commercials, marketers, digital agencies undergo a rigorous process to validate various elements in the said audio video production. This process is usually tedious and time consuming. This report aims to present results of preliminary study that uses Electroencephalogram (EEG) as input in helping production team verify their marketing content. Target consumers for a specific commercial product underwent one-on-one semi-structured interviews and from viewing sessions where their EEG data were collected using the 5-channel EMOTIV Insight headset. To triangulate the emotion modeling phase, external observers referred to as coders annotated their emotions with the aid of emotion recognition via facial recognition. Initial Analysis of the interviews along with the early version of the model revealed several interesting points on visual and audio elements that triggered viewer valence. The collection of EEG data was also able to uncover a few insights that aided in the design of a better viewing experience for the target consumers. We intend to expand the findings in the future work to build a model from a bigger data and user base along with a wider array of emotions to choose from. Also, we intend to employ more advanced experiment designs to determine if we can directly correlate viewer experiences into actual product and commercial conversion.
---------------------------------------------------------------------------------------------------
Vabljeni!
petek, 6. marec 2020 Aleš MAVER: Big data and the human genome
V ponedeljek, 9. marca 2020, bo ob 16.00 uri v prostorih Fakultete za matematiko, naravoslovje in informacijske tehnologije Univerze na Primorskem, Glagoljaška 8, Koper predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 9. marec 2020 ob 16.00 v FAMNIT-VP2
---------------------------------------
PREDAVATELJ: dr. Aleš MAVER
Dr. Aleš Maver has been working in the field of human genetics since 2002, and in his work he focuses on research of complex diseases, bioinformatics and diagnostics of rare genetic disorders. After graduating from medical school in 2011, he continued with his PhD work on rare genetic variations in familial multiple sclerosis using sequence analysis of the complete set of genes in the human genome (exome sequencing). Since then, he's been dealing with the use of big genomic data to diagnose patients with rare diseases. To that end, in 2013, he participated in the establishment of the Centre for Mendelian Genomics, which now functions as a center that offers state-of-the-art technology and bioinformatics to diagnose genetic diseases and is involved in global genome-wide sequencing data initiatives. Today his main interests include identifying new genes for genetic diseases, improving the interpretation of genetic variants at the genomic level and improving the exchange of genomic data in Slovenia and Europe.
---------------------------------------
NASLOV: Big data and the human genome
---------------------------------------
POVZETEK:
New genomic technologies have, in the recent years, provided unimaginable opportunities for human genome investigations, and thus significant advances in disease diagnosis and the discovery of new genes and disease mechanisms. With new approaches, especially through next-generation sequencing, sequencing of the entire human genome has become significantly cheaper, faster and more accessible. At the same time, we require novel bioinformatic approaches and significantly greater computational capacity to process the large amounts of data generated using novel genetic technologies. In my talk, I would like to present how we are facing the new challenges of big data in the time of genomic medicine, how we use big data of genomic sequencing in the discovery of new genes for human hereditary diseases, and how the analysis of such data has become the basis of modern genetic diagnostics in Slovenia today.
---------------------------------------
Predavanje bo potekalo v angleškem jeziku.
Vabljeni!
petek, 28. februar 2020 Julie DUCASSE: Interactive web documentary: audience reception and user engagement. A case study: iOtok.
V ponedeljek, 2. marca 2020, bo ob 16.00 uri v prostorih Fakultete za matematiko, naravoslovje in informacijske tehnologije Univerze na Primorskem, Glagoljaška 8, Koper predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 2. marec 2020 ob 16.00 v FAMNIT-VP2
---------------------------------------
PREDAVATELJICA: Julie DUCASSE
Julie Ducasse holds a PhD in Human-Computer Interaction. She studied in France at The University of Toulouse. Since September 2018, she is a post-doctoral researcher at UP FAMNIT and works for the HICUP Lab.
---------------------------------------
NASLOV: Interactive web documentary: audience reception and user engagement. A case study: iOtok.
---------------------------------------
POVZETEK:
iOtok is an interactive web documentary directed by Miha Čelar, a Slovene documentary filmmaker. The documentary was distributed over 13 weeks in Slovenia and Croatia. Data were collected to better understand how people interacted and engaged with the documentary: Was interactivity efficient in engaging users? Were the social media successful in promoting the web-doc? In this seminar, we will present the key findings of the analysis of 20 000 sessions in terms of audience reception and user engagement.
---------------------------------------
Predavanje bo potekalo v angleškem jeziku.
Vabljeni!
petek, 21. februar 2020 Peter ROGELJ: Parenchima: Magnetic resonance imaging biomarkers for chronic kidney disease
V ponedeljek, 24. februarja 2020, bo ob 16.00 uri v prostorih Fakultete za matematiko, naravoslovje in informacijske tehnologije Univerze na Primorskem, Glagoljaška 8, Koper predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 24. februar 2020 ob 16.00 v FAMNIT-VP2
---------------------------------------
PREDAVATELJ: Peter ROGELJ
Peter Rogelj is an assistant professor at the Faculty of Mathematics, Natural Sciences and Information Technologies (FAMNIT) at the University of Primorska in Koper, Slovenia. His research includes medical image processing with a special interest in medical image registration. His goal here is to achieve the speed and accuracy of image processing algorithms that are required for use in medical practice. The most important strategies for achieving this goal are the application of prior knowledge, the use of artificial intelligence and the rigorous evaluation of algorithms.
---------------------------------------
NASLOV: Parenchima: Magnetic resonance imaging biomarkers for chronic kidney disease
---------------------------------------
POVZETEK:
In this seminar, we will present the COST action Parenchima and the activities related to the forthcoming meeting of its workgroup 2(WG2) that will take place in Koper on 3-4 March. The WG2 focuses on establishing a database of renal magnetic resonance imaging (MRI) data and on software algorithms for data processing. First, we will present a data storage solution provided by Quibim and problems that needed to be solved to enable the population of the database. Second, the software evaluation initiative will be presented focusing on the evaluation of renal dynamic MRI registration algorithms. Further information about the Parenchima action can be found on its homepage: https://renalmri.org/.
---------------------------------------
Predavanje bo potekalo v angleškem jeziku.
Vabljeni!
sobota, 15. februar 2020 Domen ŠOBERL: Decentralized computation of homology in wireless sensor networks using spanning trees
V ponedeljek, 17. februarja 2020, bo ob 16.00 v prostorih Fakultete za matematiko, naravoslovje in informacijske tehnologije Univerze na Primorskem (Glagoljaška 8, Koper) predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 17. februar 2020 ob 16.00 v FAMNIT-VP2
-----------------------------------------------
PREDAVATELJ: Domen ŠOBERL
Domen Šoberl is a teaching assistant at UP FAMNIT and is currently finishing his PhD thesis at the Faculty of Computer and Information Science in Ljubljana. Research interests: artificial intelligence, agent learning and planning, computational topology.
-------------------------------------------------------------------------------------------------------------
NASLOV: Decentralized computation of homology in wireless sensor networks using spanning trees
POVZETEK:
Wireless Sensor Networks (WSN) are computer networks of small inexpensive sensor nodes with a limited computational power and often without positioning capabilities. They are deployed over an area of interest to perform some measuring tasks and communicate the results. When deployed arbitrarily, the information on signal coverage is crucial to asses the quality of deployment. A homological criteria has been proposed to guarantee a full coverage using only the connectivity data.
In this seminar I will present our decentralized algorithm to verify homology-based WSN coverage criteria. Individual nodes examine the homology of their immediate neighborhood. They organize themselves in a spanning tree formation and merge their local information up the tree. During the process, a simplicial complex known as Vietoris-Rips complex is constructed, which encodes certain information on the shape of the WSN coverage. The algorithm works in arbitrary network topology and its communication demands are suitable for low-bandwidth wireless sensor networks.
Predavanje bo potekalo v angleškem jeziku.
Vabljeni!
petek, 31. januar 2020 Maheshya WEERASINGHE: Investigating the impact of different techniques towards project users themselves in a new environment using their cognitive, emotional and actionable responses
V ponedeljek, 3. februarja 2020, bo ob 16.00 uri v prostorih UP FAMNIT (Glagoljaška 8, Koper)
predavanje v okviru PONEDELJKOVEGA SEMINARJA RAČUNALNIŠTVA IN INFORMATIKE
Oddelkov za Informacijske znanosti in tehnologije UP FAMNIT in UP IAM.
ČAS/PROSTOR: 3. februar 2020 ob 16.00 v FAMNIT-VP2
-------------------------------------------------------
PREDAVATELJ: Maheshya WEERASINGHE
Maheshya Weerasinghe, Second-year Computer Science PhD student, FAMNIT University of Primorska and University of St. Andrews, UK. Research interests: Human-Computer Interaction, Extended Realities, Learning, and Knowledge Retention.
-------------------------------------------------------
NASLOV: Investigating the impact of different techniques towards project users themselves in a new environment using their cognitive, emotional and actionable responses
-------------------------------------------------------
POVZETEK:
This study is mainly focused on investigating how the different techniques which architects would use to convey their ideas of architectural designs (new building) enable future occupants/users to project themselves in their new building.
To do that, we compare three different presentation techniques:
- Floor plans (paper-based) with 3D rendered still images of the building
- Standard desktop/mouse setup showing the 3D designed environment of the building
- HMD display showing the immersive virtual environment of the building
The reason behind this evaluation is to compare current practices/conventional method (Floor plans with 3D rendered still images) vs methods that could be widespread in the near future (Desktop and HMD). These innovative techniques may have some benefits, notably by triggering a higher sense of presence. But we need to be sure that they enable future occupants/users to understand their new building and that they also enable them to project themselves in their new space.
-------------------------------------------------------
Predavanje bo potekalo v angleškem jeziku.
Vabljeni!
