JSI Brown Bag Seminar

Janez Kranjc, IJS: Cloud data mining: constructing and executing data mining workflows using a service oriented web application

noreply@blogger.com (Tadej) — Tue, 10 Apr 2012 11:39:00 +0000

I will present a freshly launched browser-based platform for construction and execution of data mining workflows with an interface similar to those of Orange, RapidMiner, and Weka. What differentiates my platform from the aforementioned is its cloud computing nature - executions of workflows are handled by the server and may utilize (but are not limited to) web services. The non local nature of the workflows and experiments also makes it easier for sharing workflows as no installation is required (apart from a semi-decent web browser).
The presentation will include a brief description of the development of the platform, a use case for novice users, and an example of an advanced user adding features that suit his own need.

Friday, 13.4.2012, 14:00, Orange room (note the different time!)

Matic Perovšek, IJS: Visual divisive hierarchical clustering using k-means

noreply@blogger.com (Tadej) — Fri, 30 Mar 2012 14:16:00 +0000

In this presentation I will show a browser-based semi-automatic taxonomy construction tool Vdhcuk, similar to Ontogen. The Vdhcuk system incorporates text and data-mining algorithms into an user-friendly ontology construction interface. The main features of the presented system are: browser-based, usefulness for textual and numerical data, usage of unsupervised learning for concept suggestion and visualization.

Thursday, 5.4.2012, 13:00, Orange room

Biljana Mileva Boshkoska, Applications of qualitative option ranking with copulas

noreply@blogger.com (Tadej) — Tue, 13 Mar 2012 09:07:00 +0000

Copulas are joint cumulative functions that have been widely used in finance, hydrology and biology and recently they attract popularity in machine learning. In this presentation I will show how copulas may be used for decision making, in particular for multi-attribute qualitative option ranking. First I will give a short introduction to copulas, then I will explain the process of performing regression using copulas and finally I will demonstrate their applicability for option ranking on two real examples:
1. Ranking of workflows;
2. Ranking of 840 real case EC motors.

Thursday, 15.3.2012 13:00, Orange room

Dragana Miljković, IJS - Constructing signalling network topology for modelling plant-virus interaction

noreply@blogger.com (Tadej) — Wed, 15 Feb 2012 07:57:00 +0000

The presentation will address my work that will be submitted as a journal publication. It addresses the issue of developing a topology for global defence response model in plants. This model still does not exist even though the biologists are motivated to develop it since decades. As the experimental data is still lacking in this research field, we have started to manually build the model topology of plant-virus interaction by knowledge elicitation from biology experts and the literature. To accelerate this time-consuming approach, we employed additionally the automatic method to extract information from biological literature. To sum up, this work results in two major contributions to the scientific community:

1. A workflow that extracts the relations between compounds from biological text in the form of triplets: .
2. An augmented topology of the plant defence response model that is accomplished by combining manual and automatic approach.

Thursday, 16.2.2012 13:00, Orange room

Tomaž Erjavec: Language Resources and Tools for Historical Slovene, 2.2.2012

noreply@blogger.com (Tadej) — Tue, 24 Jan 2012 13:12:00 +0000

Recent years have seen an explosive growth in the number of texts that are available in digital libraries, such as Google books and the dLib.si library of the National and University Library of Slovenia. Most of these books are old - they promote our cultural heritage and are out of copyright, which makes publishing them on the Web much easier. However, digital historical texts bring with them a number of problems. It is difficult to do full-text search on them, as spelling of words has changed over time and there is no support for lemmatisation. Furthermore, such books are typically available only as PDF scans, and automatic OCR is of poor quality, esp. for materials older than 1900.

This talk presents work of the last two years in producing a set of language resources for historical Slovene, and associated tools, aimed at alleviating these problems, as well as enabling computer supported studies of historical Slovene. I will present an annotated reference corpus of 1,000 pages of historical texts, a text collection of a few million words, a computational lexicon, and a tool for text annotation. For the resources and tools each word is first modernised, and then tagged and lemmatised. The modernisation relies on a transcription rules, and has the benefit of making the text easier to read by today's speakers, as well as enabling standard tagging and lemmatisation models to be used on the text. We present the workflow and tools used in developing the resources, and show the results.

This work in progress, to be finished in 2012, is supported by the EU project IMPACT "Improving Access to Text" and the Google Humanities Research Award "Developing computational models for historical Slovene".

Daniela Stojanova: Global and Local Spatial Autocorrelation in Predictive Clustering Trees

noreply@blogger.com (Tadej) — Tue, 22 Feb 2011 13:50:00 +0000

Spatial autocorrelation is the correlation among data values, which is strictly due to the relative location proximity of the objects that the data refer to. This statistical property clearly indicates a violation of the assumption of observation independence - a pre-condition assumed by most of the data mining and statistical models. Inappropriate treatment of data with spatial dependencies could obfuscate important insights when spatial autocorrelation is ignored:.

We propose a data mining method that explicitly considers autocorrelation when building the models.
The proposed approach combines the possibility of capturing both global and local effects (common to top-down model tree learners) and detecting / dealing with positive spatial autocorrelation (common to spatial statistical methods). As a consequence, the discovered models adapt to local properties of the data, providing at the same time spatially smoothed predictions.

Thursday, 24 February 2011, 11:00, MPŠ predavalinica

Brownbag seminar: Primož Škraba: Topological Data Analysis

noreply@blogger.com (Tadej) — Mon, 31 Jan 2011 13:41:00 +0000

Topology is the branch of mathematics which studies spaces by how they are locally connected. Recently, quantatative topology,
specifically the notion of perisistence has found numerous applications in computer science. In this talk, I will introduce three applications of these ideas to data analysis. Specifically, I will discuss: 1.) clustering/unsupervised learning, 2.) study of periodic systems, 3.) robustness of maps.

The notion of stability is crucial if we are to have any hope of working with real data sets. Using these examples, I will define the stability of persistence and show how it helps overcome noise and limited knowledge of the underlying system/space. Finally, I present preliminary results on extending stability using statistical tools.

Thursday, 3 February 2011, 13:00, Orange room

Rok Piltaver: Generating accurate AND understandable hybrid classification trees

noreply@blogger.com (Tadej) — Tue, 11 Jan 2011 06:58:00 +0000

Most algorithms that induce classifiers primarily aim for high accuracy while they take understandability into account as a secondary measure of classifier's quality. The algorithm that will be presented efficiently generates a range of hybrid classification trees ranging from the most understandable to the most accurate ones by combining an easy to understand decision tree with a black-box classifier with a high accuracy. That enables the user to choose how much of accuracy he/she is willing to sacrifice for a higher understandability or vice versa.

Thursday, 13.1.2011, 13:00, Orange room

Mitja Trampuš: Graph pattern mining for article template discovery

noreply@blogger.com (Tadej) — Wed, 22 Dec 2010 12:47:00 +0000

We will have a look at a generalized variant of frequent tree pattern mining where each pattern's instances don't need to match exactly, but rather have to be "similar enough", where similarity is measured with the help of a taxonomy of node labels.
We will then apply the approach to discovering article templates for a set of related articles (e.g. news articles on bombings; or wikipedia biographies of physicists).

Thursday, 23.12.2010, 13:00, Orange room

Andreea Bizau: Determining Adjective Semantic Orientation. The beginnings of TweetMDb

noreply@blogger.com (Tadej) — Tue, 14 Dec 2010 12:42:00 +0000

This Thursday, Andreea Bizau will present her work on sentiment mining, focused on constructing a map of interesting adjectives, related to expressing sentiment orientation in social media.

Abstract:
There is a huge volume of user-generated content expressing sentiments, preferences, opinions. We are interested in how people express opinions in a certain domain (e.g. movies) and how can opinion indicators determined from one opinion source (reviews) be used to analyze another opinion source (twitter)?

Thursday, 16.12.2010, 13:00, Orange room

Laura Langohr: Biomine and beyond - a biomedical data warehouse and related graph mining research

noreply@blogger.com (Tadej) — Tue, 26 Oct 2010 07:59:00 +0000

Information is often modeled as a network of objects or concepts. In the talk I will introduce Biomine, a biomedical network and data warehouse, which consists of about one million of nodes (biological concepts) and about nine million edges (relationships). Further, I will discuss some related graph mining research, especially, finding representative nodes as well as retrieving relevant and non-redundant objects.

Thursday, 11.11.2010, 13:00 Orange room

Nada Lavrač: Advances in data mining for biomedical research (7.10.2010)

noreply@blogger.com (Tadej) — Wed, 06 Oct 2010 13:14:00 +0000

This Thursday, Nada Lavrač will present her keynote talk from the Medinfo 2010 Conference with the title Advances in data mining for biomedical research.

Thursday 7.10.2010, Orange room, 13:00

Blaž Fortuna: best of KDD 2010

noreply@blogger.com (Tadej) — Thu, 16 Sep 2010 07:26:00 +0000

Today on 16.9.2010, we are opening a new season of Brown Bag seminars with Blaž Fortuna, who will give a talk on the highlights of the ACM SIGKDD conference on knowledge discovery and data mining. Everyone welcome.

Thursday, 16.9.2010, 13:00, Orange Room

Janez Kranjc: Machine learning as a Orange4WS web application

noreply@blogger.com (Tadej) — Tue, 31 Aug 2010 14:52:00 +0000

Janez Kranjc bo v četrtek 2.9. ob 13h v oranžni sobi predstavil še nepoimenovano implementacijo okolja Orange4WS za strojno učenje kot spletne aplikacije, ki je predmet njegovega diplomskega dela. V predstavitvi bo govoril o idejah, ki so privedle do izdelave te aplikacije (podobne aplikacije, ki niso "web-based"), o tehnologijah, ki jih je uporabil za njeno izdelavo (JavaScript in PHP) in predstavil njeno zgradbo. Demonstriral bo primere uporabe aplikacije in razložil načine, kako lahko morebiten uporabnik razširi aplikacijo s svojimi dodatki. Na koncu bo razkril še njene manjše pomanjkljivosti in vas povabil da jo skupaj poimenujete.

Blaž Fortuna: Realtime optimization of a news publishing website using machine learning

noreply@blogger.com (Tadej) — Wed, 23 Jun 2010 09:53:00 +0000

In this week's brown bag Blaz Fortuna will present work done during internship, in collaboration with Carolina Fortuna, on real-time optimization of a news-publishing the web site using machine learning.

Thursday, 24.6.2010, Orange room, 13.00

Tadej Štajner: report from WWW2010

noreply@blogger.com (Tadej) — Thu, 10 Jun 2010 07:12:00 +0000

The World Wide Web Conference is a yearly international conference on the topic of the future direction of the World Wide Web. It began in 1994 at CERN and is organized by the International World Wide Web Conferences Steering Committee (IW3C2). The Conference aims to provide the world a premier forum for discussion and debate about the evolution of the Web, the standardization of its associated technologies, and the impact of those technologies on society and culture. The conference brings together researchers, developers, users and commercial ventures—indeed all those who are passionate about the Web and what it has to offer. WWW2010 will focus on “openness” in web technologies, standards and practices, and will showcase the best of the region’s technology and culture.

Thursday (Today), 13:00, Orange Room

Claudiu Mihaila: Making the Invisible Visible: Finding Romanian Zero Pronouns

noreply@blogger.com (Tadej) — Mon, 31 May 2010 15:55:00 +0000

Anaphora resolution is still a challenging research field in NLP, lacking in an algorithm that correctly resolves anaphoric pronouns. In addition, anaphoric zero pronouns pose an even greater challenge, since this category is not lexically realised. Thus, their resolution is conditioned by their prior identification stage. We present a new study on the distribution and identification of zero pronouns in Romanian. A Romanian corpus that includes legal, encyclopaedic, literary, and news texts has been created and manually annotated for zero pronouns. Using a morphological parser for Romanian and machine learning methods, experiments have been performed on the created corpus for the identification of verbs which have a zero pronoun in the subject position. The evaluation results highlight that zero pronouns appear frequently in Romanian, and their distribution depends largely on the genre. Additionally, a search scope for the antecedent has been determined, increasing the chances of correct resolution. Furthermore, more than 70% of the zero pronouns have been accurately identified by various machine learning algorithms. The strong similarity between our results and those obtained for other Romance languages support our conclusions.

Thursday, June 3rd, 13:00, Orange room.

E8 Brownbag on Videolectures

noreply@blogger.com (Tadej) — Wed, 26 May 2010 08:38:00 +0000

In case you had wanted to attend a seminar but weren't able to, some of the presentations are available on videolectures.net:

Anže Vavpetič: An implementation of the Pathfinder algorithm for sparse networks and its application on text networks

noreply@blogger.com (Tadej) — Wed, 26 May 2010 08:36:00 +0000

The Pathfinder algorithm is extensively used for pruning weighted networks. It is particularly useful in the analysis of co-citation networks. The present paper reviews two versions of the algorithm: a version with improved time and space complexities, called the Binary Pathfinder, and a version optimized for sparse networks, called Sparse Pathfinder. These two algorithms were implemented and tested on text networks. The obtained results show that the algorithm optimized for sparse networks works notably faster on such data.

Thursday, May 27th, 14:00, Orange room

Mitar Milutinović: Building a Community Wireless Mesh Network

noreply@blogger.com (Tadej) — Sun, 16 May 2010 08:44:00 +0000

Building a community wireless mesh network along with all hardware and software issues requires also an approach for building a community around the network. I will present how we are tackling this combination in wlan ljubljana open wireless network, how we are using open source technology to solve not just networking problems, but also social ones, where are its limits and what we can learn from it. I will also present a data mining perspective on our problems and why our network is a great (open) testbed for various experiments and ideas.

May 20th, 13:00, Orange room, IJS

Mario Karlovčec: Hand geometry based biometric identity verification system

noreply@blogger.com (Tadej) — Wed, 21 Apr 2010 07:32:00 +0000

Verification system that uses features of the hand geometry for verifying identity of a person is presented. Hand geometry is acquired using standard document scanner. After the image of hand is processed and features are extracted, different methods are used for verification, i.e. identification. Vertices represent points of a hand, while edges represent dimensions of parts of a hand. Method based on concept of minimal spanning tree, as well as some other methods, like: nearest neighbor, euclidean distance, etc., are applied for verification.

Thursday, April 22nd, Orange room, 14:00 (not 13:00!)

Erik Štrumbelj: Sports betting and forecasting

noreply@blogger.com (Tadej) — Fri, 09 Apr 2010 12:38:00 +0000

Sports betting attracts millions of people on a daily basis. Most are casual bettors, some are making a profit, and a few are researchers studying the field. In this presentation we will provide a brief overview of the most common types of sports betting markets, their properties, and the role that each of the aforementioned groups of people plays. Research results (including some of our own work on using alternative sources of information in sports forecasting) and practical examples from soccer betting and horse racing will be discussed, with emphasis on predicting the outcome of a sports event.

Thursday, April 15th, 12:00, (Location TBD)

[ed.: changed from 13:00 to 12:00]

Dragana Miljković: Brainstorming Brown bag seminar - modeling approaches in Systems Biology

noreply@blogger.com (Tadej) — Tue, 06 Apr 2010 13:09:00 +0000

The presentation will address my current thesis work regarding modelling immune response in plants to virus attacks. So far, in collaboration with National Institute Biology (NIB) we have developed manually qualitative model by using Petri Nets formalism. The data are gathered from databases, such as KEGG and ArabidopsisReactome, and extensive literature search and implemented manually in the model. Currently I am investigating how and which machine learning techniques I might employ within the thesis. Therefore, you are fully encouraged to give your feedback by expressing your ideas and remarks. Especially colleagues that have experience with ecology modeling and bioinformatics are welcome to contribute with their suggestions.

Thursday, April 8th 2010, 13:00, Orange Room

Anže Marn: SQL Server 2008 Data mining

noreply@blogger.com (Tadej) — Fri, 26 Mar 2010 15:24:00 +0000

In the following brown bag seminar, Anže Marn will present the capabilities of the Microsoft SQL Server 2008 database management system with regard to data mining tasks.

Thursday, April 1st 2010, 15:00, Orange Room
[update 1.4.2010] Rescheduled! Stay tuned.

Igor Trajkovski: Computer generated news sites

noreply@blogger.com (Tadej) — Tue, 23 Mar 2010 14:44:00 +0000

In this talk, we introduce a general framework for building News Engines (aggregators). Our system, TIME.mk/TIMES.si, is a complete news engine for retrieving, indexing, clustering, classifying, ranking and delivering news information extracted from the Web. Currently TIME.mk/TIMES.si collect news from 50 Macedonian (30 Slovenian) news sources (TV stations, newspapers, news agencies, etc. and every day we get more requests for inclusion), groups similar articles together and displays them in order of their calculated importance.

Thursday, March 24th, 13:00, Orange Room