nach oben

2018 | Buch

Kapitel lesen Erstes Kapitel lesen

ICT Innovations 2016

Cognitive Functions and Next Generation ICT Systems

herausgegeben von: Georgi Stojanov, Andrea Kulakov

Verlag: Springer International Publishing

Buchreihe : Advances in Intelligent Systems and Computing

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten

Über dieses Buch

The International Conference on ICT Innovations was held in September 2016, in Ohrid, Macedonia, with the main topic “Cognitive Functions and Next Generation ICT Systems”.

We live in the era where technologies are intimately woven into virtually all aspects of daily life and are becoming almost invisible. While these technologies have considerable benefits, they also have a number of shortcomings and unforeseen consequences. For example, on the one hand, bodily sensors that track physical activity, physiological parameters and sleep patterns can help promote healthy habits and can enable early detection of problems. On the other hand, attention spans are becoming shorter and shorter due to constant interruptions by notifications, emails, and instant messages being delivered to cell phones or watches, and similar disturbances. Moreover, the privacy issues involved in storing and manipulation of these data must not be neglected.

The technological convergence of sciences that were considered separate in the past, like information and communication technologies, cognitive sciences, nanotechnologies and biotechnologies, determines not only our society, health and economy, but also our education and culture.

The conference gathered academics, professionals and practitioners involved in developing solutions and systems in the industrial and business arena, especially innovative commercial implementations, to discuss novel applications of these next-generation, emerging technologies in the context of human cognitive functions.

Inhaltsverzeichnis

Frontmatter

Invited Keynote Papers

Frontmatter

Towards Multimodal Affective Stimulation: Interaction Between Visual, Auditory and Haptic Modalities

Abstract

Affective computing is concerned with designing and implementing emotionally intelligent machines. Three major subareas of research in this field are: (1) sensing the emotional state of the user, (2) expressing or displaying emotional states in a robot or an avatar, and (3) manipulating the users emotional state. For example, Picard [1] has designed a system to infer emotional state of the user based on their facial expressions.

Bipin Indurkhya

Cognitive and Emotive Robotics: Artificial Brain Computing Cognitive Actions and Emotive Evaluations, Since 1981

Abstract

This plenary keynote paper marks 35-th anniversary of description of the first artificial brain, a neural network which in addition of computing cognitive actions, computes emotive evaluations of the consequence of those actions. It was designed in 1981, as part of an effort within the Adaptive Networks Group of the COINS Department of University of Massachusetts in Amherst to solve the problems of (1) designing a delayed reinforcement learning mechanism for artificial neural networks and (2) designing a self learning system which will learn without external reinforcement of any kind. This paper describes steps which led toward solution of those problems. The proposed artificial brain, named Crossbar Adaptive Array, can be viewed as a model of cognition-emotion interaction in biological brains, and can be used in building brains for cognitive and emotive robotics.

Stevo Bozinovski

Metaphors of Creativity

Abstract

What is creativity? This question is far more likely to elicit an anecdote, an aphorism or a metaphor than it is a literal definition. Creativity is an elusive phenomenon to study, made all the more vexing by our fundamental inability to pin it down in formal terms.

Tony Veale

Socially Intelligent Robots, the Next Generation of Consumer Robots and the Challenges

Abstract

We are evolving, so as our society, lifestyle and the needs. AI and ICT have been with us for decades, and now penetrating more in our day-to-day life, so as the robots.

Amit Kumar Pandey

Proceeding Papers

Frontmatter

Hand Gesture Recognition Using Deep Convolutional Neural Networks

Abstract

Hand gesture recognition is the process of recognizing meaningful expressions of form and motion by a human involving only the hands. There are plenty of applications where hand gesture recognition can be applied for improving control, accessibility, communication and learning. In the work presented in this paper we conducted experiments with different types of convolutional neural networks, including our own proprietary model. The performance of each model was evaluated on the Marcel dataset providing relevant insight as to how different architectures influence performance. Best results were obtained using the GoogLeNet approach featuring the Inception architecture, followed by our proprietary model and the VGG model.

Gjorgji Strezoski, Dario Stojanovski, Ivica Dimitrovski, Gjorgji Madjarov

Computer-Based Statistical Description of Phonetical Balance for Romanian Utterances

Abstract

Motivated by the advent of security solutions which rely on voice biometrics, we revisit by means of extensive computer-based investigations the concept of phonetical balance for Romanian utterances. We show that the standard distribution of phonems offers only a partial description of the phonetics of the language and that more detailed statistical indicators are needed. To this end, we introduce a simple indicator that measures vowel-consonant (or consonant-vowel) sequences and analyze the distribution of consonant clusters for Romanian words. Our results show that the distribution of consonant clusters is scale-free-like (akin to the distribution of words and phrases in large texts) and that large clusters of vowels or consonants are infrequent. This, in turn, indicates that utterances consisting of words which are statistically unrepresentative with respect to the previous indicators are good candidates for benchmarking the efficency of voice biometrics solutions.

A. Cocioceanu, T. Ivănoaica, A. I. Nicolin, M. C. Raportaru

Distributed Private Key Generator for ID-Based Public Key Infrastructure

Abstract

We recognize the need of certificateless PKI to reduce the step of obtaining the public key. This leads to ID-Based cryptography where we have PKI with full power to generate private keys for any identity. We solve this problem by implementing distributed key generation to form a group of players which will act as private key generator for ID-Based PKI. The implementation is done on the Android platform, showing the possibilities of running PKI on cheap and widely available hardware.

Pance Ribarski, Ljupcho Antovski

Relation Between Statistical Tests for Pseudo-Random Number Generators and Diaphony as a Measure of Uniform Distribution of Sequences

Abstract

In this paper we investigate the relation between statistical tests for pseudo-random number generators and the diaphony as a measure of uniform distribution of sequences. In order to find some relations many experiments are done. For these experiments we use generators and tests from Diehard battery. For all generated sequences, the diaphony is calculated and the tests from Diehard battery are done. Also, we made experiments using two deterministic sequences: the sequence of Van der Corput and the sequence of equidistant points.

Sashe Gjorgjievski, Verica Bakeva, Vesna Dimitrievska Ristovska

Pattern Recognition of a Digital ECG

Abstract

The process of assisted ECG diagnosing mimics the way a medic would act upon. Such a process inevitably comprises the feature extraction step, when the standard ECG signal components: the QRS complex, the P wave and T wave are detected. Using a pattern recognition algorithm for the purpose is one of the available options. In this article, the pattern recognition approach for the feature extraction routine is explained by analysis of consecutive steps and its effectiveness is discussed in comparison to other means of QRS complex detection.

Marjan Gusev, Aleksandar Ristovski, Ana Guseva

Performance Evaluation of FIR and IIR Filtering of ECG Signals

Abstract

When a wearable ECG sensor transmits signals to a mobile device, the mobile applications needs to be very efficient and save the limited mobile phone resources. This motivates us to find an algorithm implementation that is not computationally intensive, but still very efficient in denoising the ECG signal. The use of a window-based design Finite Impulse Response (FIR) and Infinite Impulse Response (IIR) filters are analysed in this paper. Several filters have been designed and the computational efficiency have been analysed both theoretically and experimentally. The results show that the designed IIR outperforms the FIR filter achieving a better computational efficiency with a minimal distortion of the ECG signal.

Aleksandar Milchevski, Marjan Gusev

Influence of Fuzzy Tolerance Metrics on Classification and Regression Tasks for Fuzzy-Rough Nearest Neighbour Algorithms

Abstract

In this paper, we investigate the influence of the fuzzy tolerance relationship (fuzzy similarity metrics) on two fuzzy and two fuzzy-rough nearest neighbour algorithms for both classification and regression tasks. The fuzzy similarity metric plays a major role in construction of the lower and upper approximations of decision classes, and therefore has high influence on the accuracy of the algorithm. The experimental results evaluated on the four approaches show the difficulty to estimate a single metric that will be good in all cases. Moreover, the choice of similarity metric on some datasets has not influence at all. This require further investigation, not only with similarity metrics, but also for evaluating the algorithms with different T-norms and implicators.

Andreja Naumoski, Georgina Mirceva, Petre Lameski

On the Kalman Filter Approach for Localization of Mobile Robots

Abstract

In this work we analyze robot motion given from the UTIAS Multi-Robot Dataset. The dataset contains recordings of robots wandering in a confined environment with randomly spaced static landmarks. After some preprocessing of the data, an algorithm based on the Extended Kalman Filter is developed to determine the positions of robots at every instant of time using the positions of the landmarks. The algorithm takes into account the asynchronous time steps and the sparse measurement data to develop its estimates. These estimates are then compared with the groundtruth data provided in the same dataset. Furthermore several methods of noise estimation are tested, which improve the error of the estimate for some robots.

Kristijan Petrovski, Stole Jovanovski, Miroslav Mirchev, Lasko Basnarkov

Evaluation of Automatically Generated Conceptual Database Model Based on Business Process Model: Controlled Experiment

Abstract

This paper presents the results of the controlled experiment that we have conducted in order to evaluate an approach to automated design of the initial conceptual database model based on the collaborative business process model. The source business process model is represented by BPMN, while the target conceptual model is represented by UML class diagram. The results of the experiment imply that the approach enables generation of the target conceptual model with a high percentage of completeness (>85%) and precision (>85%), which confirms the results of the initial case-study based evaluation.

Danijela Banjac, Drazen Brdjanin, Goran Banjac, Slavko Maric

Analysis of Protein Interaction Network for Colorectal Cancer

Abstract

In this paper we create and analyze a protein-protein interaction network (PPIN) of colorectal cancer (CRC). First we identify proteins that are related to the CRC (set of seed proteins). Using this set we generate the CRC PPIN with the help of Cytoscape. We analyze this PPIN in a twofold manner. We first extract important topological features for proteins in the network which we use to determine CRC essential proteins. Next we perform a modular analysis by discovering CRC significant functional terms through the process of GO enrichment within densely connected subgroups (clusters) of the PPIN. The modular analysis results in a mapping from the CRC significant terms to CRC significant proteins. Finally, we combine the topological and modular evidence for the proteins in the CRC PPIN, exclude the initial seed proteins and obtain a list of proteins that could be taken as possible bio-markers for CRC.

Zlate Ristovski, Kire Trivodaliev, Slobodan Kalajdziski

Using Sentiment Analysis of Twitter Data for Determining Popularity of City Locations

Abstract

The paper considers mining and analyzing data generated by Twitter social network, regarding content classification, language determination and sentiment analysis of tweets. Analyzes are based on geospatial tweets collected in timespan of four months within region Vračar in Belgrade, Serbia. All of collected data is first being preprocessed, filtered and classified by given criteria, by using “Twitter search engine” (TSE) application, that has been upgraded in order to detect tweet language and execute sentiment analysis of the tweets written in English. This type of analysis can be used for determining popularity of city locations of interest and public spaces in general.

Nikola Dinkić, Nikola Džaković, Jugoslav Joković, Leonid Stoimenov, Aleksandra Đukić

Internet Addiction: Evaluating the Psychometric Properties of the IAT in Macedonia

Abstract

In this digital age there is a growing importance in global research of Internet addiction. The purpose of this research study is to establish a valid instrument for measuring the addictive use of Internet. We investigated the psychometric properties of the established Young’s Internet Addiction Test (IAT) on a Macedonian population. Since its development, there has been a number of culture-specific validation studies of the IAT, but never in cultures from south-east Europe. In a sample of 322 undergraduate participants, exploratory factor analysis (EFA) determined the presence of a three factor structure of the IAT with 17 items. The three factors, “Withdrawal and social problems”, “Time management” and “Failure and neglect” explained 38.43% of the total variance with good internal consistency for the IAT (Cronbach’s a = .89) and good reliability for the factors (.65–.84). Hence, this version of the IAT is a valid instrument with sound psychometric properties for measuring Internet addiction in a sample of participants from south-east Europe and may be used for further research.

Martin Mihajlov, Aleksandar Stojmenski

Health Care Domain Mobile Reminder for Taking Prescribed Medications

Abstract

Nowadays, majority of people as a main problem for their poor health and bad psycho-physical condition states the lack of time. Commitments, dynamic and stressful way of life lead to people being negligent of themselves. Thus, their health condition is damaged. The big problem are acute patients, but also the patients who have a chronic disease and who should take prescribed medications regularly, and who, for some reason, are prevented or have forgotten to take the prescribed dosage of the medicaments. Due to the inadequate taking of the therapy, the time for patient’s recovery is significantly prolonged or the existing problem is not relieved (with chronic patients). PersonalMedicationReminder is an Android application that downloads the prescriptions from the server of a health care institution or allows the user to insert the over-the-counter drugs (without prescription therapy). The application allows patients to create reminders and receive notifications which would inform them about the time for the next receiving of the therapy. The application downloads the prescriptions from the electronic medical record of the patient from the medical information system MEDIS.NET which is used in health institutions for primary health care in Republic of Serbia [1, 2]. The problems that occurred during the application development are also shown in the paper. In the conclusion are stated the directions of the further research and improvement of the mobile application.

Eleonora Milić, Dragan Janković, Aleksandar Milenković

Enhancing Text-Based Relatedness Measures with Semantic Web Data

Abstract

Entity relatedness measures quantify the amount of association between two entities, such as people, places or events, and are fundamental part of many Natural Language Processing and Information Retrieval applications. Calculating entity relatedness requires access to entity specific information, so a very common practice is to use Wikipedia or its Semantic Web representations as source of knowledge. This paper explores which of the different semantic relationships that associate two entities in DBpedia are good indicators of their relatedness and could be used to enhance some of the standard text-based relatedness measures. The ultimate goal is learning a well performing relatedness calculation method that does not require vast amount of preprocessing, but is applicable in cases when entities lack either textual context or semantic relationships. The KORE entity relatedness dataset was used for learning a convenient and well performing method for measuring relatedness and its evaluation.

Ana Gjorgjevikj, Riste Stojanov, Dimitar Trajanov

Power Consumption Analysis of Application Layer Protocols for the Internet of Things

Abstract

In this paper, we present power consumption analysis of application layer protocols CoAP, MQTT and XMPP for the Internet of Things. With this modern concept of the future will be connected all devices which can be connected. Sensors, home appliances, vehicles, mobile devices are just some of the physical objects that will be affected. Here especially may be mentioned the sensory devices. These are devices used to detect events and changes in the environment by generating appropriate output. Many of them are with limited performances, memory and battery. They are often placed in inaccessible areas. Therefore, the power consumption is very important, and which of the protocols used for the Internet of Things provide greater energy savings. According to the test results, MQTT and CoAP provide major energy savings, unlike XMPP which consumes more power. For all protocols, the most energy is spent in a state RX, while the least is spent in a state LPM.

Aleksandar Velinov, Aleksandra Mileva

Improving Medical Cases Retrieval Using an Online Fact Database

Abstract

This paper presents an approach for retrieval of medical cases using a novel query expansion method. The approach relies purely on the text data in the medical cases. The cases are indexed with Terrier IR search engine based on their text content including the caption of the figure contained within them. Furthermore, in the retrieval phase there is an input consisted of a long text query in a narrative form. The input query is expanded by using on-line fact databases, such as Freebase, with the aim that this will add more terms relevant to the concepts mentioned in the text. The goal is to provide a way of query expansion, so that the query is more defined, which should provide more narrowed and precise results in the retrieval. The retrieval is done with the BM25 weighting model. Our approach shows that expanding the input text query in this fashion can provide a boost in the retrieval performance.

Ivan Kitanovski, Katarina Trojacanec, Ivica Dimitrovski, Suzana Loshkovska

Improving Scalability of Web Applications by Utilizing Asynchronous I/O

Abstract

The focus of the paper is the use of asynchronous I/O calls in web applications to improve their scalability, by increasing the number of requests per second that it can process and decreasing the average response time of the system. Popular development frameworks have always included only blocking I/O APIs in their base, making asynchronous I/O methods hard to implement and maintain. Significant effort has been made in recent years to enrich these frameworks with better syntax for asynchronous API to improve developers’ experience and encourage its use. Such improvement in .NET’s syntax is put to the test in this paper and the results are presented and evaluated.

Gjorgji Rankovski, Ivan Chorbev

Relevance Re-ranking Through Proximity Based Term Frequency Model

Abstract

In this internet era, people rely on the most significant tool called search engine for retrieving attractive information from the web. Also, the rapid growth in the usage of the web increases the volume of data on the web, due to which most of the documents retrieved by the search engine is overwhelmed with inappropriate and redundant information called outliers. This not only increases the result space, but also roots in wasting the user’s time and effort that makes them to surf uninteresting data. Consequently, a method is essential for the web user community to remove uninteresting information and to present the interesting data in an organized manner based on their request. Web content outlier mining is promising research area that serves these features to the web users. In this research work, proximity based term frequency model has been developed for retrieving the appropriate information and for refining the quality of the results offered by the search engine. Experimental results indicate that proximity based term frequency model improves the performance in terms of relevancy re-ranking of the retrieved documents.

S. Sathya Bama, M. S. Irfan Ahmed, A. Saravanan

Information System for Biosensors Data Exchange in Healthcare

Abstract

This paper presents a novel cloud information system (BIOHIS - Biosensors Healthcare Information System) whose main goal is to enable vital data exchange obtained from biosensors. BIOHIS corresponds to the requirements of the existing protocols for MRMI (medical response to major incidents). This system aims to ease and improve the data exchange between the different institutions involved in the MRMI protocols. BIOHIS is one step closer to the interoperable data flow among the various medical system interfaces, ensuring structured data capture. The system is intended to use cloud leverages as infinite storage, computing capacities and full time authorized access.

Monika Simjanoska, Bojana Koteska, Magdalena Kostoska, Ana Madevska Bogdanova, Nevena Ackovska, Vladimir Trajkovikj

An Automatic Tracking System for Natural Hazard Events with Satellite Remote Sensing

Abstract

The atmosphere satellite data for atmosphere parameters are the most important source of information for monitoring of areas without or with very rare environment research facilities. With growing dynamic of Climate change, the detailed observation, research and risk management is with a vital importance for nations in regions as the South-East Europe. Due to insufficient ground based research infrastructure and qualified personal, the satellites are main source of reliable data of atmosphere process. The presented paper describes the basic available functionalities of a system for automatic atmosphere events location and transport prediction based on available open data form NASA satellites.

Assen Tchorbadjieff

Backmatter

Titel: ICT Innovations 2016
herausgegeben von: Georgi Stojanov
Andrea Kulakov
Verlag: Springer International Publishing
Electronic ISBN: 978-3-319-68855-8
Print ISBN: 978-3-319-68854-1
DOI: https://doi.org/10.1007/978-3-319-68855-8