Search results for: Ivica Boticki

Authors: Martina Holenko Dlab, Natasa Hoic-Bozic, Ivica Boticki

Abstract:

This paper presents technologically innovative and scalable mobile learning solution within the SCOLLAm project (“Opening up education through Seamless and COLLAborative mobile learning on tablet computers”). The main research method applied during the development of the SCOLLAm mobile learning system is design-based research. It assumes iterative refinement of the system guided by collaboration between researches and practitioners. Following the identification of requirements, a multiplatform mobile learning system SCOLLAm [in]Form was developed. Several experiments were designed and conducted in the first and second grade of elementary school. SCOLLAm [in]Form system was used to design learning activities for math classes during which students practice calculation. System refinements were based on experience and interaction data gathered during class observations. In addition to implemented improvements, the data were used to outline possible improvements and deficiencies of the system that should be addressed in the next phase of the SCOLLAm [in]Form development.

Keywords: Adaptation, collaborative learning, educational technology, mobile learning, tablet computers.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1229

Authors: Ivica Martinjak, Mario-Osvin Pavcevic

Abstract:

When trying to enumerate all BIBD-s for given parameters, their natural solution space appears to be huge and grows extremely with the number of points of the design. Therefore, constructive enumerations are often carried out by assuming additional constraints on design-s structure, automorphisms being mostly used ones. It remains a hard task to construct designs with trivial automorphism group – those with no additional symmetry – although it is believed that most of the BIBD-s belong to that case. In this paper, very many new designs with parameters 2-(13, 5, 5), 2-(16, 6, 5) and 2-(21, 6, 4) are constructed, assuming an action of an automorphism of order 3. Even more, it was possible to construct millions of such designs with no non-trivial automorphisms.

Keywords: BIBD, incidence matrix, automorphism group, tactical decomposition, deterministic algorithm.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1319

Authors: Thomas Bryan, Veton Kepuska, Ivica Kostanic

Abstract:

A simple adaptive voice activity detector (VAD) is implemented using Gabor and gammatone atomic decomposition of speech for high Gaussian noise environments. Matching pursuit is used for atomic decomposition, and is shown to achieve optimal speech detection capability at high data compression rates for low signal to noise ratios. The most active dictionary elements found by matching pursuit are used for the signal reconstruction so that the algorithm adapts to the individual speakers dominant time-frequency characteristics. Speech has a high peak to average ratio enabling matching pursuit greedy heuristic of highest inner products to isolate high energy speech components in high noise environments. Gabor and gammatone atoms are both investigated with identical logarithmically spaced center frequencies, and similar bandwidths. The algorithm performs equally well for both Gabor and gammatone atoms with no significant statistical differences. The algorithm achieves 70% accuracy at a 0 dB SNR, 90% accuracy at a 5 dB SNR and 98% accuracy at a 20dB SNR using 30d B SNR as a reference for voice activity.

Keywords: Atomic Decomposition, Gabor, Gammatone, Matching Pursuit, Voice Activity Detection.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1793

Authors: Jelena Slugan, Ivica Ružić

Abstract:

We are witnessing the rapid development of technologies that change the world around us. However, curriculums and teaching processes are often slow to adapt to the change; it takes time, money and expertise to implement technology in the classroom. Therefore, the University of Split, Croatia, partnered with local school Marko Marulić High School and created the project "Modern competence in modern high schools" as part of which five different curriculums for STEM areas were developed. One of the curriculums involves combining information technology with physics. The main idea was to teach students how to use different circuits and microcomputers to explore nature and physical phenomena. As a result, using electrical circuits, students are able to recreate in the classroom the phenomena that they observe every day in their environment. So far, high school students had very little opportunity to perform experiments independently, and especially, those physics experiment did not involve ICT. Therefore, this project has a great importance, because the students will finally get a chance to develop themselves in accordance to modern technologies. This paper presents some new methods of teaching physics that will help students to develop experimental skills through the study of deterministic nature of physical laws. Students will learn how to formulate hypotheses, model physical problems using the electronic circuits and evaluate their results. While doing that, they will also acquire useful problem solving skills.

Keywords: ICT in physics, curriculum, laboratory activities, STEM.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 990

Authors: Thomas Bryan, Veton Kepuska, Ivica Kostanic

Abstract:

Speaker recognition is performed in high Additive White Gaussian Noise (AWGN) environments using principals of Computational Auditory Scene Analysis (CASA). CASA methods often classify sounds from images in the time-frequency (T-F) plane using spectrograms or cochleargrams as the image. In this paper atomic decomposition implemented by matching pursuit performs a transform from time series speech signals to the T-F plane. The atomic decomposition creates a sparsely populated T-F vector in “weight space” where each populated T-F position contains an amplitude weight. The weight space vector along with the atomic dictionary represents a denoised, compressed version of the original signal. The arraignment or of the atomic indices in the T-F vector are used for classification. Unsupervised feature learning implemented by a sparse autoencoder learns a single dictionary of basis features from a collection of envelope samples from all speakers. The approach is demonstrated using pairs of speakers from the TIMIT data set. Pairs of speakers are selected randomly from a single district. Each speak has 10 sentences. Two are used for training and 8 for testing. Atomic index probabilities are created for each training sentence and also for each test sentence. Classification is performed by finding the lowest Euclidean distance between then probabilities from the training sentences and the test sentences. Training is done at a 30dB Signal-to-Noise Ratio (SNR). Testing is performed at SNR’s of 0 dB, 5 dB, 10 dB and 30dB. The algorithm has a baseline classification accuracy of ~93% averaged over 10 pairs of speakers from the TIMIT data set. The baseline accuracy is attributable to short sequences of training and test data as well as the overall simplicity of the classification algorithm. The accuracy is not affected by AWGN and produces ~93% accuracy at 0dB SNR.

Keywords: Time-frequency plane, atomic decomposition, envelope sampling, Gabor atoms, matching pursuit, sparse dictionary learning, sparse autoencoder.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1570