Nr.3

AKTET Journal of Institute Alb-Shkenca Revistë Shkencore e Institutit Alb-Shkenca Vol. IV, 2011

AKTET Journal of Institute Alb-Shkenca Revistë Shkencore e Institutit Alb-Shkenca Vol. IV, Nr 3 © 2011 Institute Alb-Shkenca

ISSN 2073-2244 www.alb-shkenca.org

Editor‐in‐chief Robert Magari, Ph.D., Beckman Coulter Inc., Miami, Florida, USA

Editors Anila Hoda, Dr, University of Agriculture, Tirana, Albania Ardian Harri, Ph.D., Mississippi State University, Mississippi State, Mississippi, USA Eshref Januzaj, Dipl.‐Inform. Univ. (M.Sc.) Technische Universität Braunschweig, Germany Evan Rroço, Prof., Dr., University of Agriculture, Tirana, Albania Kristaq Jorgo, Dr., University of Tirana, Tirana, Albania Lulzim Dragidella, Dr., University of Applied Sciences, Regensburg, Germany and University of Prishtina, Kosovë Margarita Ifti, Ph.D., University of Tirana, Tirana, Albania Musa Rizaj, Dr, University of Prishtina, Prishtina, Kosovë Naim Shabani, Dr., Ludwig ‐ Maximilians ‐ University, Munich, Germany Nebi Caka, Dr., University of Prishtina, Prishtina, Kosovë Nikolla P. Qafoku, Ph.D., Pacific Northwest National Laboratory, Richland, Washington, USA Pandi Zdruli, Ph.D., Mediterranean Agronomic Institute of Bari Valenzano, Bari, Italy Roza Allabashi, Dipl.Ing. Dr., University of Natural Resources and Applied Life Sciences, Vienna, Austria Zeqirja Neziri, Prof., University “St. Cyril and Metodij”, Skopje, Macedonia

Technical editor Roland Lelaj, Millennium Research Group, Toronto, Canada

ii

AKTET, Vol. IV, 3, 2011

CONTENT

DISTRIBUTION OF VOTES IN ALBANIAN PARLIAMENTARY ELECTIONS MARGARITA IFTI, DODË PRENGA, INVA BUZI

1 367

EXPERIMENTAL STUDY OF A COMBINED AC AND DC MAGNETIC FIELD EFFECTS ON ION CURRENT OF CELL CHANNELS ENTELË GAVOÇI, DHURATA KUQI, MARGARITA KUQALI

6 372

AN APPLICATION OF LOGISTIC REGRESION FOR EVALUATING TEACHING PRACTICES SADRI ALIJA, LAZIM KAMBERI, LLUKAN PUKA

1 1 377

INVERTIMI I OPERATORIT TË MBULIMIT ME ANË TË ALGORITMIT ME DY RRJETA NË QCD-NË RRJETORE AFINA XHAKO, ARTAN BORIÇI

1 8 384

HYPERCUBIC SYMMETRY BREAKING EFFECTS ON PI AND RHO MESON MASSES FROM SIMULATIONS WITH MINIMALLY DOUBLED ACTIONS RUDINA ZEQIRLLARI, INVA BUZI, ARTAN BORIÇI

2 6 392

INTEGRAL PRESENTATION OF A CONVEX FUNCTION LUIGJ GJOKA, QEFSERE GJONBALAJ

3 1 397

REGULAR INVERSE Γ-SEMIGROUP ISLAM BRAJA, EDMOND PISHA

3 6 402

POSSIBILITIES FOR IMPROVING THE CUTTING EFFICIENCY IN BAND SAWS ARBEN BEJTJA, FIRDUS HAVERAJ, PANDELI MARKU

4 1 407

ISSUES OF WOOD PROCESSING INDUSTRY IN ALBANIA ARBEN BEJTJA, ARBEN BOÇARI, ALKETA GREPCKA

4 6 412

ACCELERATING SIMULTANEOUS METHODS FOR THE DETERMINATION OF POLYNOMIAL MULTIPLE ROOTS EGLANTINA XHAJA, FATMIR HOXHA

5 2 418

TOTAL QUALITY MANAGEMENT AND COMPETITIVE ADVANTAGE OF ALBANIAN FIRMS ENRIKO CEKO

5 7 423

THE ASSESSMENT OF APPLE ORCHARDS WITH VISUAL SOIL ASSESSMENT METHOD ROBERT DAMO, PIRRO IÇKA

6 4 430

FUZZY LOGIC–BASED CONTROLLER DESIGN FOR CONTROL OF INDUCTION MACHINE AIDA SPAHIU, PETRIKA MARANGO, GENCI SHARKO

7 0 436


iii

WIRELESS MESHED ACCESS NETWORK SALEM LEPAJA

7 7 443

STUDY OF OPTIMAL CONSTRUCTIVE SOLUTIONS OF THE FURNITURES MADE IN LAMINATED PARTICLE BOARD ARMOND HALEBI, HOLTA ÇOTA, MANDI MARKU, HEKTOR THOMA

8 3 449

THE ASSESSMENT OF THE RAINFALL EVENTS AT THE PLAIN OF KORÇA THROUGH THE SPI PIRRO IÇKA, ROBERT DAMO, LORENC EKONOMI

8 7 453

LIDHJA MES NORMAVE TË INTERESIT DHE ÇMIMIT TË PASURIVE TË PATUNDSHME ESMERALDA URUÇI, NEVILA KIRI

9 4 460

APPROACHES TO TRACKING IN COMPUTER VISION SONILA DOBI, BESNIK DOBI

1 0 2 468

ELIMINATION OF THE CONTRADICTING STATES OF DETERMINED ASYNCHRONOUS SEQUENTIAL MACHINES BY MATRIX METHOD SHKELZEN SHABANI, MIRLIND BRUQI, AGRON PAJAZITI, SHABAN SHABANI

1 0 8 474

SYNTHESIS OF ASYNCHRONOUS SEQUENTIAL AUTOMATS OF STOCHASTIC SYSTEMS BY THE MATRIX METHOD AND MODIFIED METHOD SHABAN SHABANI, AVDYL BUNJAKU, SHKELZEN SHABANI

1 1 4 480

PRODHIMI I KOKSIT TË NAFTËS DHE IMPAKTI MJEDISOR NË SHQIPËRI I. BEQIRAJ, S. DRUSHKU, E.F. MULLA, A. MALJA, H. MANAJ, D. KALLUSHI

1 2 1 487

THE IMPACT OF TIRANA’S GASOLINE CARS ON THE AIR POLLUTION BY UNBURNED HYDROCARBONS EDLIRA F. MULLA, ANGJELIN SHTJEFNI, ANDONAQ LONDO

1 2 9 495

THE IMPORTANCE OF WATER RESOURCES MANAGEMENT AS AN OBLIGATION OF WATER FRAMEWORK DIRECTIVES MIRIAM BOGDANI NDINI, PALLE D. GREVY

1 3 5 501

SOME NEW RESULTS FOR FUZZY WEAKLY CONTRACTIVE MAPPINGS ELIDA HOXHA , ARBEN ISUFATI

1 4 1 507

FIXED POINT THEOREMS FOR FUZZY CONTRACTIVE MAPPINGS ARBEN ISUFATI, ELIDA HOXHA

1 4 7 513

POMOLOGICAL AND CHEMICAL PROPERTIES OF BLUEBERRY (VACCINIUM MYRTILLUS L.) POPULATIONS IN KOSOVA SYLË SYLANAJ

iv


1 5 4 519

AFATET E MBJELLJES DHE NUMRI I BIMËVE TË VARIETETEVE MË PRODHUES TË THJERRËZËS NË SHQIPËRI ADRIAN DOKO, RAMADAN SHEHU, AGIM CANKO, ASLLAN CELAMI, ISUF KAZIU

1 5 8 523

STUDIMI I DISA KULTIVARËVE TË GRURIT NË ZONA TË NDRYSHME TË SHQIPËRISË FETAH ELEZI, VLASH TIRANA, BELUL GIXHARI

1 6 3 529

A RELATED FIXED POINT THEOREM FOR FOUR MAPINGS ON FOUR COMPLETE METRIC SPACES LULJETA GJONI, KRISTAQ KIKINA A FIXED POINT THEOREM FOR m MAPINGS ON m COMPLETE FUZZY METRIC SPACES USING IMPLICIT RELATIONS LULJETA GJONI, KRISTAQ KIKINA, JOLLANDA SHARA ON REGULAR TERNARY SEMIHYPERGROUPS KOSTAQ HILA, KRISANTHI NAKA


1 6 9 535 1 7 7 543 1 8 5 551

v

GUIDE FOR AUTHORS AKTET publishes papers presented during the annual meeting of Institute Alb-Shkenca (IASH). The journal publishes only research papers containing original, novel, and relevant research that enhances the existing knowledge. The editorial board consists of an editor-in-chief, a technical editor, and several subject areas editors. Manuscripts can be submitted in English or Albanian, English being the preferred language. Submission of the manuscript implies that essentially the same research is not published before (except in a form of a summary, lecture, or thesis), is not submitted elsewhere for publication, and that the manuscript is approved by all the authors and authorities where the work is carried out. Authors are requested to disclose any conflict of interest and provide information for any financial support used to conduct the research. An accepted manuscript cannot be published elsewhere without the written permission of IASH. The journal publishes only original work that contributes to the advancement of science. Plagiarism is not acceptable and will be reported to the appropriate authorities. Conclusions should be justified by the experimental design and data. The information presented in the paper should have sufficient details to permit others replicate the work. Authors should retain the research data and should be able to provide these data to the editorial board if asked. Please use appropriate language, be concise and grammatically correct. Authors are responsible for the scientific accuracy and the format of the submitted manuscript as well as the content of the correspondence with the editors.

ORGANIZATION OF MANUSCRIPT Manuscript should be limited to 2500 words including title, summary, text, bibliography, tables, ® graphs, etc. Manuscript should be saved in Microsoft Office Word, A4 format, 2.5 cm (1 inch) margins, single column double spaced, and Times New Roman font of size 12. Do not use footnotes or endnotes or any other formatting. Pages should be numbered while the title, names and contact information of the authors should be provided on the first page. Manuscript should contain: 1.

2. 3. 4. 5. 6.

7.

vi

Title: Be concise and clear. Provide authors contact information. Superscripts a, b, c can be used to indicate authors affiliations. Corresponding author should be indicated with an asterisk. Provide title in both Albanian and English. Key words: Provide 4-5 key words that represent the subject of the paper. Summary: Should represent the essence of the work and should be limited to 150 words. Provide summary in both Albanian and English. Introduction: Describe your work, related research, its importance, and clearly state the objective of the study. Material and methods: Provide details on the materials being used and experimental procedures. Provide sufficient details to allow work to be reproducible. Result and discussions: Results should be clear and illustrated in tables and graphs. Use statistical analysis and draw conclusions based on statistical significance. Discuss in details the implication of the results. Conclusions should be directly related to the results obtained. Conclusions should not be speculative, spontaneous, or/and summarizing the already available information. Bibliography: List literature sources alphabetically base on the last name of the first author. Every literature source must be cited in the text by the related number. A limit of 20 literature sources can AKTET, Vol. IV, 3, 2011

be provided. A citation ‘in press’ implies that the reference is accepted for publication. Unaccepted manuscripts cannot be referenced. Do not use references from the internet except for the online publications. Examples of references: 1. Libby WF, Anderson EC, Arnold JR (1949) Age determination by radiocarbon content. Science 109, 227-228. 2. Strunk W, White EB (1979) The elements of style, third ed. Macmillian, New York 8. 9.

Tables: Present each table on a separate page. Tables should be numbered and titled. Limit the number of tables to five. Figures and graphs: Present figures and graphs on separate pages after the tables. Altogether provide no more than five figures and graphs.

Authors submitting social sciences contributions may disregard sections 5 and 6 when preparing their manuscripts. Manuscripts that do not conform to the above format will not be considered.

SUBMISSION Manuscripts are electronically submitted through the ALPA system at, http://alpa.maliit.eu/journal/aktet_v4_nr1/. Instruction for use of the ALPA system is found at the address above as well as at, www.alb-shkenca.org Submit the manuscript within the deadline announced by the editorial board. Along with the manuscript submit a message directed to the editorial board, stating the desire to publish the manuscript and indicating the participation in the annual meeting of IASH. Submitted manuscripts will be double-blinded reviewed by at least two experts of the subject area. Editorial board will make the decision on the acceptance of the manuscript based on the recommendations from the reviewers and notify this decision to the authors. Authors should be notified on the status of the manuscript within a 3-4 months period. The entire process will be electronically documented.


vii

AKTET Journal of Institute Alb-Shkenca Revistë Shkencore e Institutit Alb-Shkenca

ISSN 2073-2244 www.alb-shkenca.org Copyright © Institute Alb-Shkenca

DISTRIBUTION OF VOTES IN ALBANIAN PARLIAMENTARY ELECTIONS SHPERNDARJA E VOTAVE NE ZGJEDHJET PARLAMENTARE SHQIPTARE MARGARITA IFTI, DODË PRENGA, INVA BUZI Department of Physics, Faculty of Natural Sciences, University of Tirana, Bul. Zog I, Tirana, Albania AKTET IV, 3: 367 - 371, 2011 PERMBLEDHJE Në këtë punim ne studiojmë sjelljen e numrit të votave të marra nga subjektet zgjedhore në proceset elektorale të 10 viteve të fundit në Shqipëri. Jepet denduria e marrjes së përqindjeve të ndryshme të votave nga këto subjekte për zgjedhjet parlamentare. Në shpërndarjet e votave për zgjedhjet në Shqipëri identifikohen dy regjime. Përqindjet e ulëta kanë një shpërndarje fuqi me eksponent rreth -1.7. Si rregull në këtë regjim gjenden rreth 80% e pikave, ndërsa me to lidhen 20% e votave. Rezulton se në këtë këtë regjim gjenden votat e subjekteve të vogla. Regjimi tjetër që shtrihet në zonën 15-85% të votave karakterizohet nga shpërndarje Gausiane me bisht të gjatë dhe u korrespondon subjekteve të mëdha. Disa qarqe (Shkodër, Kukës, Berat, Dibër, etj,) shfaqin shpërndarje krejtësisht fuqi, ndërsa në disa të tjerë (Tiranë, Durrës, Elbasan, Korçë) kemi prezencën e Gausianit. Fjalë kyçe: proces zgjedhor, shpërndarje, ligj fuqi, Gausian. SUMMARY In this paper we study the behavior of the number of votes cast for different electoral subjects in the Albanian elections of the last 10 years. We report the frequency of obtaining a certain percentage (fraction) of votes versus this fraction for the parliamentary elections. In the distribution of votes in Albania we identify two regimes. In the low percentages we see a power law distribution, with exponent about -1.7. In the power law regime we find over 80% of the points, while they relate to 20% of the votes cast. Votes of the small electoral subjects are found in this regime. The other regime includes percentages from 15% to 85%, and has Gaussian distribution, followed by a long tail. It corresponds to big parties. Some districts (Shkodër, Kukës, Berat, Dibër, etc.) exhibit a distribution that is entirely power law, while in others (Tiranë, Durrës, Elbasan, Korçë) the Gaussian is present. Key words: electoral process, distribution, power law, Gaussian.

INTRODUCTION There is a class of systems, traditionally not studied by physics, in which details or history might not influence the global behaviour, e.g. individuals in a community may have opinions that take on two or few values about an issue, such as agree/not agree, seller/buyer, believer/atheist, Linux/Windows, left/right, etc. in all cases evolution drives the system toward a final state, in which one can identify the dominant opinions. In these cases the evolution of the system can be described very well using the techniques and tools of statistical physics

[20]. Statistical physicists who study opinion dynamics aim to identify states of the opinion of population, and to know the elementary processes that determine transitions between them, in order to understand the mechanisms and nature of the interaction that produces a certain ordering. This resembles a return to the origins of statistical physics in the 19th century, when Maxwell and Boltzmann were inspired by social statistics, and founded statistical physics. In 1952 R.B. Potts [15] proposed a model where each agent (spin) can take on one of the q possible values of opinion (orientation) (1, 2, ...,

Ifti et al. q), and where the interactions with the neighbours favour parallel orientation. For q=2 this model gives the well-known Ising model [10] of ferromagnets. The analogies between the two models: Potts and Ising inspired an elegant simplification, known as the voter model [3, 9]. In this model each node (agent) has two possibilities of “orientation” related to a given issue. Each time step a node i, picked at random, takes on the opinion of their neighbour j, and the process goes on forever. This model has analytical solution in any dimension d. In the real life there are individuals who do not change their opinion. Taking this into consideration, the voter model has been changed by introducing to it the presence of “zealots”, i.e. spins that never change their orientation [12]. For d≤2 the zealot node influences the whole system, and their opinion is adopted by the whole community [13]. For higher dimensions the situation is more complex. If there are many zealots in the system, consensus is never reached, and the distribution of magnetisations is Gaussian with width of the order √(1/z), where z is the number of zealots [14]. Another modification that makes the model more realistic is that is which agents can be in one of three states: left (A), right (B), and centre (C), but only the centrists can interact with other species [19]. Depending on the initial conditions, this model produces finite probabilities of finding the system in one of the possible states, or in a mixture of those. There are several other models, that we are describing here briefly. Majority rule. In a community of N agents, whose opinion can have one of two values, the opinion of a node will be determined by the opinion of the majority of a group of his r neighbours [6]. The system converges toward consensus as logN. This model has been modified further [7, 1]. Sznajd model. In this model [17, 18] a pair of nodes with the same value of spin (individuals who agree with eachother) persuade their neighbours, or, if they can not agree among them, (have different orientations of spin) then nothing is changed in the configuration of opinions. Models with 368

continuous values of opinion consider a continuous spectrum of opinions s [0, 1]. The most widely discussed are models Deffuant and Hegselmann-Krause. In the Deffuant model [4] rules are such that, if a pair of agents (i, j) have opinions that differ by more than a threshold value, nothing changes in the system, otherwise their opinions will change in a way that their difference is reduced by an amount that depends on their difference from an “opinion of convergence”. The Hegselmann-Krause model [8] is similar to this, except that in this case the interaction happen when the neighbour’s opinion is within a surrounding area of the opinion of the central node. What about the election results in the real world? The first empirical study was done in the case of the Federal elections of 1998 in Brazil [2], where was reported a power law distribution of the fraction of votes obtained by candidates according to Zipf law . This result was reconfirmed in the elections of 2002 [11]. This results is reproduced if the fraction of votes is treated as a product of n independent “persuasive” sub-processes with the electors, each of which succeeds with probability pi. The same distribution was later obtained as a result of dynamics of the modified Sznajd model in a scale-free network. Later studies have shown that the 1/ν law is not universal [1]. Exponents that are different from -1 have been found in the case of German, French, Polish, and Italian elections [5]. These countries use proportional electoral systems with open lists. If the list of Q candidates has obtained N votes, the quantity gives the average number of votes per candidate, and the ratio characterizes the performance of the candidate [5]. The mechanism that produces this distribution is of the “word of mouth” type: the dynamics starts with the candidate, who has a well-defined opinion (they vote for themselves), and then spreads this opinion to other voters. They are persuaded with probability r1) ose zvogëlohet (O.R.