INTERNATIONAL CONFERENCE ON COMPUTER AND KNOWLEDGE ENGINEERING (ICCKE) | Year:2014 | دوره:4 |Papers Count:56

THIS PAPER PRESENTS AN OPTIMUM SWITCHING PATTERN FOR THE SPACE VECTOR MODULATION (SVM) OF MATRIX CONVERTERS BASED ON A GENETIC ALGORITHM. THE POSSIBILITY OF CHOOSING THE SEQUENCE OF ACTIVE DUTY CYCLES, AND THE ABILITY TO ALLOCATE THE DURATION OF THE INACTIVE DUTY CYCLE TO EACH ONE OF THE ZERO DUTY CYCLES IN A MATRIX CONVERTER SVM, PROVIDE SOME DEGREES OF FREEDOM IN DESIGNING OF THE SWITCHING PATTERN. THIS LEADS TO AN OPTIMIZATION PROBLEM. THEREFORE, A SUITABLE GENETIC ALGORITHM FOR SOLVING THE SWITCHING PATTERN OPTIMIZATION PROBLEM IS ADOPTED, AND AN OBJECTIVE FUNCTION TO MINIMIZE THE WEIGHTED TOTAL HARMONIC DISTORTION (WTHD) AND THE LOW-ORDER HARMONICS OF THE OUTPUT VOLTAGE IS PROPOSED. SIMULATIONS IN MATLAB/SIMULINK CONFIRM THE VALIDITY OF ANALYTICAL ACHIEVEMENTS.

IN THE RECENT YEARS THERE HAS BEEN AN INTEREST WITHINTHE PHYSICS COMMUNITY IN THE PROPERTIES OF NETWORKS OF MANYTYPES. GRAPH CLUSTERING IS THE PROCESS OF IDENTIFYING THENETWORK STRUCTURE IN TERMS OF GROUPING THE VERTICES OF A GRAPHINTO CLUSTERS TAKING INTO CONSIDERATION THE EDGE STRUCTURE OF THEGRAPH THAT IN SUCH A WAY THERE SHOULD BE MANY EDGES WITHINEACH CLUSTER AND RELATIVELY FEW BETWEEN THE CLUSTERS. BASED ONHIGH COMPUTATIONAL COST, THE CLASSICAL ALGORITHMS WILL SLOWMUCH SINCE DATA SIZE IN REAL APPLICATION INCREASES RAPIDLY. INSUCH A SITUATION, MODEL BASED GRAPH CLUSTERING ALGORITHMS AREAN EFFICIENT ALTERNATIVE TO CLASSICAL ONES. THE PERFORMANCE OFTHE MODEL BASED GRAPH CLUSTERING ALGORITHMS DEPENDS ON THECORRECT INITIAL PARAMETER SETTING. WE ARE PROPOSED ANEVOLUTIONARY ALGORITHM TO FIND PROPER VALUES FOR THE MODELBASED GRAPH CLUSTERING ALGORITHMS. THE PROPOSED METHOD ISTESTED ON BOTH SIMULATED AND REAL DATA SETS AND GAVE IMPROVINGRESULTS IN COMPARISON WITH RANDOM PARAMETER SETTING.

THE QUESTION CLASSIFICATION SYSTEM IS ONE OF THE IMPORTANT SUBSYSTEMS IN THE QUESTION ANSWERING SYSTEMS (QAS). IN SUCH SYSTEMS THROUGH RETRIEVAL METHODS AND INFORMATION EXTRACTION THE TEXTS ARE RETRIEVED IN ORDER TO GET TO A CORRECT ANSWER. THE CURRENT STUDY IS DESIGNED TO PRESENT THE ARCHITECTURE OF QUESTION CLASSIFICATION (QC) IN PERSIAN BASED ON THE CONDITIONAL RANDOM FIELDS (CRF) MACHINE LEARNING MODEL AND EVALUATE EFFECTS OF VARIOUS FEATURES ON ITS ACCURACY. IN THIS STUDY, SENTENCES WERE CLASSIFIED INTO TWO LEVELS OF COARSE ANDFINE CLASSES BASED ON THE TYPE OF THE ANSWER TO EACH QUESTION.AFTER EXTRACTING FEATURES AND SETTING SLIDING WINDOW ON THE CRF MODEL, CRF QUESTION CLASSIFIER (QC) IS TRAIN. THEN, THE QC PREDICTS LABELS FOR EVERY TOKEN IN QUESTION. NEXT, A MAJORITY VOTING ON THE QUESTION CLASSIFICATION OUTPUT, IS USED TO EXTRACT A UNIQUE LABEL FOR EACH QUESTION. FURTHER, THE EFFECTS OF DIFFERENT FEATURES ON THE ULTIMATE ACCURACY OF THE SYSTEM WERE EVALUATED. FINALLY RESULTS OF THIS QUESTION CLASSIFIER, ILLUSTRATE A SATISFACTORY ACCURACY.

IN THIS PAPER A NEW ELECTROMAGNETISM-LIKE MECHANISM IS PROPOSED FOR COMBINATORIAL OPTIMIZATION OF CAPACITATED VEHICLE ROUTING PROBLEM. ELECTROMAGNETISM-LIKE MECHANISM IS A NEW METAHEURISTIC METHOD AND INSPIRED BY THE ATTRACTION AND REPULSION MECHANISM OF THE ELECTROMAGNETISM THEORY. WE PROPOSE A NEW ELECTROMAGNETISM-LIKE MECHANISM THAT IT INCLUDES A NEW DISTANCE MEASURE BETWEEN SOLUTIONS AND NEW EFFECTIVE PROCESS OF ATTRACTION AND REPULSION. IN ORDER TO ANALYZE THE PROPOSED ALGORITHM A COMPARISON IS DONE WITH EXISTING ALGORITHM FOR THIS PROBLEM. COMPUTATIONAL RESULTS SHOW THAT THE PROPOSED ELECTROMAGNETISM-LIKE MECHANISM ALGORITHM HAS A GOOD PERFORMANCE FOR THE CONSIDERED PROBLEM.

ONE OF THE NOTICEABLE TOPICS IN FUZZY LOGIC CONTROLLERS IS PARAMETER CONTROLLING OF HEURISTIC SEARCH ALGORITHMS. IN THIS PAPER, ONE OF THE PARAMETERS OF GRAVITATIONAL SEARCH ALGORITHM, GSA, IS CONTROLLED USING FUZZY LOGIC CONTROLLER TO ACHIEVE BETTER OPTIMIZATION RESULTS AND TO INCREASE CONVERGENCE RATE. SEVERAL EXPERIMENTS ARE PERFORMED AND RESULTS ARE COMPARED WITH THE RESULTS OF THE ORIGINAL GSA. EXPERIMENTAL RESULTS CONFIRM THE EFFICIENCY OF THE PROPOSED METHOD.

IN THIS PAPER, A ROBUST LOSSLESS DATA HIDING SCHEME ON UNCOMPRESSED VIDEO DATA BASED ON HISTOGRAM DISTRIBUTION CONSTRAINED (HDC) SCHEME IS PROPOSED. THE PROPOSED METHOD IS A REVERSIBLE DATA HIDING ALGORITHM WHICH ENABLES EXACT RECOVERY OF THE ORIGINAL VIDEO WITHOUT ANY DISTORTION AFTER EXTRACTING HIDDEN INFORMATION, IF THE WATERMARKED VIDEO IS NOT AFFECTED BY ANY OTHER PROCESS. THE ALGORITHM IS ROBUST AGAINST H.246/AVC VIDEO COMPRESSION. IN THE PROPOSED METHOD, THE LUMINANCE COMPONENTS OF VIDEO FRAMES ARE USED TO EMBED THE WATERMARK DATA BITS. THE LUMINANCE COMPONENT IS DIVIDED INTO NON-OVERLAPPING BLOCKS AND THE ARITHMETIC DIFFERENCE OF EACH BLOCK IS COMPUTED. IN THE NEXT STEP, THE HIDING DATA ARE EMBEDDED INTO THE BLOCKS BY SHIFTING THE ARITHMETIC DIFFERENCE VALUES. THE PERFORMANCE OF THE PROPOSED ALGORITHM IS EVALUATED WITH RESPECT TO IMPERCEPTIBILITY, ROBUSTNESS AGAINST H.264 VIDEO CODING AND DATA PAYLOAD CAPACITY BY SIMULATIONS.SIMULATION RESULTS DEMONSTRATE A HIGH PERFORMANCE FOR THE PROPOSED METHOD.

IN THIS PAPER A ROBUST INTELLIGENT CONTROLLER IS ADOPTED IN ORDER TO FULFIL MAXIMUM POWER POINT TRACKING (MPPT) ALGORITHM FOR PERMANENT MAGNET SYNCHRONOUS GENERATOR (PMSG) WITH THE APPLICATION OF WIND TURBINES.CONVENTIONAL LINEAR PI CONTROLLER IS SIMPLE AND EASY TO IMPLEMENT BUT IT HAS THE DRAWBACK OF DEPENDENCE ON SYSTEM PARAMETERS ESPECIALLY IN A WIND ENERGY CONVERSION SYSTEM WITH LARGE AMOUNT OF UNCERTAINTIES AS A SEQUENCE OF WIND SPEED VARIATION. DUE TO INHERENT ROBUSTNESS TO UNCERTAINTIES IN FUZZY LOGIC CONTROLLER, IT CAN BE USED EFFECTIVELY IN WIND ENERGY CONVERSION SYSTEM. HOWEVER DESIGNING PROCEDURE AND GAIN TUNING COULD BE A CHALLENGING APPROACH. IN THIS PAPER THE INITIAL PARAMETERS OF FLC ARE OBTAINED FROM THE PARAMETERS OF ITS WELL-TUNED LINEAR COUNTERPART. THIS IS A FLEXIBLE CONTROL SCHEME THAT TAKES ADVANTAGE OF THE BEST FEATURES OF FUZZY CONTROL, ROBUSTNESS TO UNCERTAINTIES, AND OF PI CONTROLLER, SIMPLE TUNING PROCEDURE. THIS INTELLIGENT CONTROLLER IS DESIGNED IN PRESENCE OF MECHANICAL PARAMETERS AS WELL AS MECHANICAL TORQUE DISTURBANCE, WHICH IS THE MOST CONCERNED UNCERTAINTY IN WIND TURBINES. SIMULATION RESULTS CLEARLY SHOWS PERFORMANCE OF PROPOSED ROBUST FUZZY PI CONTROL SYSTEM IN MAXIMUM POWER POINTS TRACKING IN A WIND ENERGY CONVERSION SYSTEM.

IN THIS PAPER, THE DISTRIBUTED ESTIMATION IN WIRELESS SENSOR NETWORK USING OPTIMAL TASK SCHEDULING IS CONSIDERED.THE MAIN GOAL IS TO MAXIMIZE NETWORK LIFETIME WHERE IT CONSISTS OF A FUSION CENTER AND A SET OF SENSOR NODES. LIFETIME DEFINED AS THE NUMBER OF ROUNDS ACCOMPLISHED BEFORE NETWORK BECOMES NONFUNCTIONAL. IN ORDER TO PROLONG NETWORK LIFETIME, WE DETERMINE THE OPTIMAL NUMBER OF ACTIVE SENSOR NODES AND THE NUMBER OF SAMPLES PROVIDED BY EACH OF THEM USING LINEAR OPTIMIZATION PROBLEM. SENSOR OBSERVATIONS ARE QUANTIZED INTO MESSAGES AND THEN FORWARDED TO A FUSION CENTER WHERE A FINAL ESTIMATION IS PERFORMED BASED ON DEGREE OF CERTAINTY.SIMULATION RESULTS SHOW THAT OUR OPTIMAL PROPOSED ALGORITHM HAS ACHIEVED SIGNIFICANT EFFICIENCY WHEN COMPARED WITH OTHER HEURISTIC METHODS.

NOWADAYS, UTILIZING HEURISTIC ALGORITHMS IS HIGHLY APPRECIATED IN SOLVING OPTIMIZATION PROBLEMS. THE FUNDAMENTAL OF THESE ALGORITHMS ARE INSPIRED BY NATURE. THE GRAVITATIONAL SEARCH ALGORITHM (GSA) IS A NOVEL HEURISTIC SEARCH ALGORITHM WHICH IS INVENTED BY USING LAW OF GRAVITY AND MASS INTERACTIONS. IN THIS PAPER, A NEW OPERATOR IS PRESENTED WHICH IS CALLED "THE BLACK HOLE". THIS OPERATOR IS INSPIRED BY THE CONCEPT OF AN ASTRONOMY PHENOMENON. BY ADDING THE BLACK HOLE OPERATOR, THE EXPLOITATION OF THE GSA IS IMPROVED. THE PROPOSED ALGORITHM IS EVALUATED BY SEVEN STANDARD UNIMODAL BENCHMARKS. THE RESULTS OBTAINED DEMONSTRATE BETTER PERFORMANCE OF THE PROPOSED ALGORITHM IN COMPARISON WITH THOSE OF THE STANDARD GSA AND OTHER VERSION OF GSA WHICH IS EQUIPPED WITH THE DISRUPTION OPERATOR.

IN THIS PAPER, MEMBERSHIP FUNCTION SHAPES AND TYPES AND THE FUZZY RULES OF FUZZY SYSTEMS ARE ADJUSTED BY USING A INTELLIGENT GRAVITATIONAL SEARCH ALGORITHM (IGSA) IN ORDER TO OBTAIN AN OPTIMAL FUZZY SYSTEM. THE ADVANTAGES OF THIS METHOD IN CLASSIFYING VARIOUS DATA SETS ARE ILLUSTRATED. POSSIBLE EXTENSIONS OF THIS TECHNIQUE ARE BRIEFLY DISCUSSED.

REINFORCEMENT LEARNING IS A POPULAR CONTEXT OF MACHINE LEARNING THAT AIMS AT IMPROVING THE BEHAVIOR OF AUTONOMOUS AGENTS THAT LEARN FROM INTERACTIONS WITH THE ENVIRONMENT. HOWEVER, IT IS OFTEN COSTLY, TIME CONSUMING, AND EVEN DANGEROUS. TO DEAL WITH THESE PROBLEMS, REWARD SHAPING HAS BEEN USED AS A POWERFUL METHOD TO ACCELERATE THE LEARNING SPEED OF THE AGENT. THE PRINCIPLE IDEA IS TO INCORPORATE A NUMERICAL FEEDBACK, OTHER THAN ENVIRONMENT REWARD, FOR THE LEARNING AGENT. HOWEVER, FINDING AN EFFICIENT POTENTIAL FUNCTION TO SHAPE THE REWARD IS STILL AN INTERESTING AREA OF RESEARCH. IN THIS PAPER, A NEW ALGORITHM HAS BEEN PROPOSED THAT RECEIVES THE ENVIRONMENT GRAPH, PERFORMS SOME NEW ANALYSIS, AND PROVIDES THE EXTRACTED INFORMATION FOR THE LEARNING AGENT TO ACCELERATE THE SPEED OF LEARNING. THIS INFORMATION INCLUDES SUB GOALS, BAD STATES, AND SUB ENVIRONMENTS WITH DIFFERENT EXPLORATION, OR REWARD, VALUES. TO EVALUATE THIS ALGORITHM AN EXPERIMENTAL STUDY HAS BEEN CONDUCTED ON TWO BENCHMARK ENVIRONMENTS, SIX ROOMS AND MAZE. THE OBTAINED RESULTS DEMONSTRATE THE EFFECTIVENESS OF THE PROPOSED ALGORITHM.

RECENT ADVANCES IN WIRELESS SENSOR NETWORK'S TECHNOLOGY HAVE DEVELOPED USING THIS TECHNOLOGY IN VARIOUS FIELDS. FOR THE IMPORTANCE AND INTENSE APPLICATION OF THESE NETWORKS, MANY EFFORTS AND RESEARCHERS HAVE BEEN DONE TO CONFRONT CHALLENGES. IN THIS PAPER, TO CONQUER ONE OF THE MOST IMPORTANT CHALLENGES OF THESE NETWORKS, THAT IS THE LIMITATION IN ENERGY RESOURCES, AN ADAPTIVE SCHEDULING ALGORITHM BASED IN POTTS MODEL WAS APPLIED. ACCORDING TO THIS MODEL, FOR EVERY ELEMENT OF A SYSTEM, Q DIFFERENT STATES ARE CONSIDERED. EACH ELEMENT CONVERTS ITS STATE TO A NEW STATE OR REMAINS IN THAT STATE DUE TO ITS CURRENT STATE AND ITS ADJACENT NEIGHBOURS AND WITH THE EFFECT OF ENVIRONMENT. IN THIS PROJECT, THIS MODEL WAS USED IN WIRELESS SENSOR NETWORKS SUCH THAT EACH SENSOR NODE IS CONSIDERED AS AN ELEMENT OF SENSOR NETWORK SYSTEM AND THREE ACTIVE, INACTIVE AND STANDBY STATES ARE DEFINED FOR THAT NODE, WHICH IT SELECTS ONE OF THESE THREE STATES ACCORDING TO ITS CURRENT STATE AND NEIGHBOURS AND ENVIRONMENT EFFECT AND ADAPTS ITS ACTIVITY WITH THE ENVIRONMENT IN A WAY THAT THE LEAST ENERGY TO BE CONSUMED. BY COMPARING THIS ALGORITHM AND SIMILAR ALGORITHM (WITH TWO STATES), IT IS OBSERVED THAT IN IDENTICAL CONDITIONS, POTTS MODEL WITH THREE STATES REPRESENTS BETTER RESULTS AND MORE LIFETIMES FOR NETWORK.

ONE OF THE MOST SIGNIFICANT PRACTICAL CHALLENGES FOR FACE RECOGNITION IS A LIKENESS OF FACES WHICH LEADS TO A BIG PROBLEM IN CLASSIFICATION OF DIFFERENT CLASSES. TO TACKLE THIS PROBLEM, WE PRESENT A NOVEL METHOD BASED ON SIMILARITY OF EACH FACE WITH OTHER FACES USING THE PEARSON CORRELATION COEFFICIENTS. BESIDES, ANOTHER PROBLEM IS VARIABILITY IN LIGHTING INTENSITY WHICH ITS PHYSICS ARE DIFFICULT FOR ACCURATE MODEL. IN THIS PAPER, FIRST, DISCRETE WAVELET TRANSFORM (DWT) IS USED FOR FEATURE EXTRACTION. NEXT, WITH RESPECT TO THE CORRELATION MATRIX, TWO ALGORITHMS ARE EMPLOYED, NAMELY, K-MEANS CLUSTERING AND PARTICLE SWARM OPTIMIZATION (PSO) BASED SCATTERING RATIO MATRIX OF CORRELATION FEATURES. THEN FOR EACH CLUSTER, THE PROCESS OF CLASSIFICATION IS CONTINUED BY NORMALIZATION OF THE EACH SUBSET FIRSTLY AND THEN THE DECISION MAKING FOR EACH SUBSET IS PERFORMED BY SUPPORT VECTOR MACHINE (SVM). THE EXPERIMENTS ARE PERFORMED ON THE ORL AND YALE DATABASES AND THE RESULTS SHOW THAT THERE ARE A SIGNIFICANT IMPROVEMENT IN 45 FEATURES BASED WEIGHTED RECOGNITION RATE.

CONSIDERING AN UNDER SUPERVISED 3D SPACE WHERE A GROUP OF MOBILE DEVICES WITH LIMITED SENSING AND COMMUNICATING CAPABILITIES ARE DEPLOYED, THIS PAPER AIMS AT PROPOSING A DECENTRALIZED SELF-DEPLOYMENT ALGORITHM FOR AGENTS TO GET MAXIMUM CONNECTED COVERAGE TOPOLOGY. THE PROBLEM IS MODELED AS MAXIMIZATION WHICH IS SOLVED COMPLETELY DISTRIBUTED. IN FACT EACH AGENT TRIES TO MAXIMIZE ITS SENSING VOLUME WHILE PRESERVING CONNECTIVITY AND HAVING HIGH QUALITY COMMUNICATION LINKS. THEN EACH AGENT KNOWING DISTANCE OF ITS NEIGHBORS DEFINES AN ERROR FUNCTION AND TRIES TO DECREASE THE ERROR BY CHANGING ITS POSITIONS THROUGH GRADIENT DESCENT ALGORITHM.

RETINAL NEOVASCULARIZATION (NV) IS ONE OF THE MOST IMPORTANT CAUSES OF DIABETIC BLINDNESS WHICH EMPHASIZES THE SERIOUS NEED FOR AUTOMATIC SCREENING TOOLS TO DIAGNOSE DIABETES AND TREAT IT AS EARLY AS POSSIBLE. IN THIS WORK, THE PROBLEM OF NV DETECTION IN THE COLOR IMAGES OF RETINA IS CONSIDERED.PREVIOUS RESEARCHES IN THE ANALYSIS OF RETINAL IMAGES EITHER IGNORE SUCH COMPLEX LESION OR IN FEW CASES APPROACH IT SUPERFICIALLY AND POSE MANY LIMITATIONS. OBTAINING A COMPLETE VESSEL MAP AND THEN TRYING TO IDENTIFY NV IS A RECENT METHODOLOGY. UNFORTUNATELY, NV MAINLY AFFECTS SMALL AND HARD TO DETECT VESSELS, HENCE, WE BELIEVE SUCH AN APPROACH IS INHERENTLY LIMITED. THEREFORE, IN THIS PAPER SIX DIFFERENT TEXTURE DESCRIPTIONS ARE PROPOSED AS FEATURES IN THE CLASSIFICATION OF RETINAL IMAGES INTO NV AND NORMAL GROUP. WE STUDY APPLICATION OF THE PROPOSED FEATURES USING A SUPPORT VECTOR MACHINE (SVM) AS A SIMPLE CLASSIFIER TO EMPHASIZE THE IMPORTANCE OF FEATURES AGAINST THE COMPLEXITY OF THE CLASSIFIER.FEATURE EXTRACTION IS PERFORMED ON LOCAL REGIONS OF THE IMAGES (128X128) TO LOCALIZE THE LESION.A DATASET OF 1613 REGIONS IS CONSIDERED THAT CONTAINS 591 NV REGIONS WHICH ARE MANUALLY MARKED BY AN EXPERT. THE RESULTS SHOW THE PROPOSED TEXTURE DESCRIPTORS ARE ABLE TO REVEAL THE NV LESION WITH AN ACCEPTABLE ACCURACY OF AROUND 90%.

SPAMMING IS BECOMING A MAJOR THREAT THAT NEGATIVELY IMPACTS THE USABILITY OF E-MAIL. ALTHOUGH LOTS OF TECHNIQUES HAVE BEEN PROPOSED FOR DETECTING AND BLOCKING SPAM MESSAGES, SPAMMERS STILL SPREAD SPAM E-MAILS FOR DIFFERENT PURPOSES SUCH AS ADVERTISING, PHISHING, ADULT AND OTHER PURPOSES AND THERE IS NOT ANY COMPLETE SOLUTION FOR THIS PROBLEM. IN THIS WORK WE PRESENT A NOVEL SOLUTION TOWARD SPAM FILTERING BY USING A NEW SET OF FEATURES FOR CLASSIFICATION MODELS. THESE FEATURES ARE THE SEQUENTIAL UNIQUE AND CLOSED PATTERNS WHICH ARE EXTRACTED FROM THE CONTENT OF MESSAGES.AFTER APPLYING A TERM SELECTION METHOD, WE SHOW THAT THESE FEATURES HAVE GOOD PERFORMANCE IN CLASSIFYING SPAM MESSAGES FROM LEGITIMATE MESSAGES. THE ACHIEVED RESULTS ON 6 DIFFERENT DATASETS SHOW THE EFFECTIVENESS OF OUR PROPOSED METHOD COMPARED TO CLOSE SIMILAR METHODS. WE OUTPERFORM THE ACCURACY NEAR+2% COMPARED TO RELATED STATE OF ARTS. IN ADDITION OUR METHOD IS RESILIENT AGAINST INJECTING IRRELEVANT AND BOTHERSOME WORDS.

THE RECOGNITION OF FACE IMAGES IS A COMPLICATED PROBLEM. FACE IMAGES ARE OFTEN SUFFERED FROM VARIATIONS IN BRIGHTNESS, HEAD ROTATION, FACIAL EMOTIONS AND SO ON. BESIDES, AMAZING ABILITIES OF HUMAN BRAIN IN FACE RECOGNITION IN THE PRESENCE OF THESE VARIATIONS, CONTRIBUTE TO DESIGN FACE RECOGNITION SYSTEMS BASED ON PROCEDURE OF HUMAN BRAIN. SURVEYING THE RECOGNITION AND PERCEPTUAL SYSTEM OF HUMAN, SHOWS THAT, THIS SYSTEM HAS HIERARCHICAL AND BIDIRECTIONAL STRUCTURE. FURTHERMORE, THE PERFORMANCE OF THE SYSTEM WOULD STRONGLY BE IMPROVED BY APPLYING THE INFORMATION OF UPPER LAYERS OF FACE RECOGNITION SYSTEM IN INTERPRETING AND PROCESSING THE INPUT DATA.IN THIS PAPER, NOVEL BIDIRECTIONAL ARCHITECTURE FOR FACE RECOGNITION INSPIRED BY HUMAN FACE RECOGNITION SYSTEM IS PRESENTED VIA APPLYING INVERSION IN ARTIFICIAL NEURAL NETWORKS (ANN’S). IN THIS APPROACH, STORED DATA IN THE INVERSE NETWORK IS APPLIED IN THE RECOGNITION SYSTEM ITERATIVELY AND THEN THE CORRECTNESS OF FACE RECOGNITION MODEL HAS BEEN CONSEQUENTLY IMPROVED BY 8%. THE PROPOSED MODEL IS ABLE TO PRODUCE 12 VARIOUS FACIAL EXPRESSIONS ON THE OUTPUT, FROM ONLY ONE INPUT EXPRESSION OF EACH PERSON, AFTER TRAINING WITH AUT DATABASE IMAGES.

CONSTRUCTING GOOD TEST COLLECTION IS AN EXPENSIVE AND TIME-CONSUMING PROCESS. TRADITIONALLY, TEST COLLECTIONS CONTAIN BINARY JUDGMENTS. IN RECENT YEARS, HOWEVER, THERE HAS BEEN INCREASINGLY INTEREST IN TEST COLLECTIONS WITH MULTI-LEVELS JUDGMENTS AND OF CERTAIN QUALITIES. SUCH COLLECTIONS ARE EVEN MORE EXPENSIVE TO CONSTRUCT. THEREFORE, ABILITY TO REUSE TEST COLLECTIONS CAN NOT ONLY SAVE CONSTRUCTION COSTS, BUT ALSO BOOSTS OUR CONFIDENCE IN THEIR QUALITY. THIS PAPER PROPOSES A METHOD FOR ASSESSING REUSABILITY OF A TEST COLLECTION WITH MULTI-LEVEL JUDGMENTS. THE PROPOSED METHOD CAN HELP IR RESEARCHERS TO DETERMINE WHETHER AN EXISTING TEST COLLECTION WITH A SET OF MULTI-LEVEL JUDGMENTS IS SUITABLE FOR EVALUATING A NEW IR SYSTEM OR NOT. RESULTS OF OUR EXPERIMENTS (ON MAHAK TEST COLLECTION) SUGGEST THAT THIS METHOD CAN HELP ASSESSING REUSABILITY OF A TEST COLLECTION.

THE UNIVERSITY COURSE TIMETABLING PROBLEM (UCTP) IS A CLASSICAL, OLD AND FAMOUS PROBLEM IN THE FIELD OF OPTIMIZATION PROBLEMS. THE PURPOSE OF UCTP IS TO SCHEDULE A NUMBER OF EVENTS (COURSES) IN PROPER TIMESLOTS AND SUITABLE ROOMS. IN THIS PROBLEM, THERE ARE SOME HARD AND SOFT CONSTRAINTS. A FEASIBLE TIMETABLE MUST SATISFY ALL HARDCONSTRAINTS. IN ADDITION, EACH SOFT CONSTRAINT VIOLATION CAUSES A PENALTY. AS FAR AS UCTP IS AN NP-COMPLETE PROBLEM, IT IS REASONABLE TO USE POPULATION-BASED METAHEURISTIC ALGORITHMS AND EVOLUTIONARY ALGORITHMS (EA). ALTHOUGH VARIOUS METHODS HAVE BEEN PRESENTED THE BEST RESULTS ARE REFERRED TO AS HYBRID EVOLUTIONARY ALGORITHMS (HEA) AND METAHEURISTICS. THE PROPOSED METHOD IS A HYBRID GENETIC ALGORITHM (HGA). IN OUR INNOVATIVE HGA, THE INITIAL POPULATION WHICH COMES FROM HEURISTICS IS STORED INTO RED-BLACK TREE DATA STRUCTURE. AFTER THAT, OUR HGA CREATES NEW OFFSPRINGS FROM PREVIOUS INDIVIDUALS BY ITS OPERATORS. MOREOVER, TO IMPROVE LOCALEXPLOITATION, WE USED HILL CLIMBING. THE RESULTS WERE COMPARED WITH OTHER AVAILABLE ONES USING THE 11 DATASETS OF SOCHAET AL.THE RESULTS WERE PROMISING AND SHOWED THAT THE PROPOSED HGA METHOD IS A GOOD METHOD TO SOLVE UCTP.

REPRESENTATIVE VECTOR IS A KIND OF VECTOR WHICH INCLUDES RELATED WORDS AND THE DEGREE OF THEIR RELATIONSHIPS.IN THIS PAPER THE EFFECT OF USING THIS KIND OF VECTOR ON AUTOMATIC CLASSIFICATION OF PERSIAN DOCUMENTS IS EXAMINED.IN THIS METHOD, PREPROCESSED DOCUMENTS, EXTRA WORDS AS WELL AS WORD STEMS ARE AT FIRST FOUND. NEXT, THROUGH ONE OF THE KNOWN WAYS, SOME FEATURES ARE EXTRACTED FOR EACH CATEGORY.THEN, THE REPRESENTATIVE VECTOR, WHICH IS MADE BASED ON THE ELICITED FEATURES, LEADS TO SOME MORE DETAILED WORDS WHICH ARE BETTER REPRESENTATIVES FOR EACH CATEGORY. FINDINGS OF THE EXPERIMENTS SHOW THAT PRECISION AND RECALL CAN BE INCREASED SIGNIFICANTLY BY EXTRA WORDS OMISSION AND ADDITION OF FEW WORDS IN THE REPRESENTATIVE VECTORS AS WELL AS THE USE OF A FAMOUS CLASSIFICATION MODEL LIKE FUZZY SUPPORT VECTOR MACHINE (FSVM).

ONE OF THE FEATURES OF THE NEXT GENERATION WIRELESS NETWORKS (NGWNS) IS ITS HETEROGENEOUS COMMUNICATION ENVIRONMENT. HETEROGENEOUS NETWORKS ARE RANGING FROM WIRELESS WAN, LAN, MAN AND PAN. THE MOST IMPORTANT PARAMETERS IN THIS REGARD ARE DIFFERENT DATA RATES AND COVERAGE OF THESE NETWORKS. IN THIS PAPER, WE PROPOSE AN EVOLVABLE FUZZY LOGIC-BASED ALGORITHM FOR MANAGING HANDOVER PROCESS IN HETEROGENEOUS WIRELESS NETWORKS. WE USE FUZZY LOGIC SYSTEM TO DETERMINE THE RIGHT TIME OF INITIATING THE HANDOFF PROCEDURE AND UTILIZE GENETIC ALGORITHM (GA) TO PREDICT THE BEST FORM OF RULES OF THE FUZZY MODULE. WE IMPLEMENT OUR PROPOSED SCHEME ON A SPECIAL SIMULATION MODEL AND DISCUSS ABOUT THE IMPACT OF EVOLUTION ON THE ALGORITHM.

IN THIS PAPER, WE INTRODUCE AUBUE - ADAPTIVE USER-INTERFACE BASED ON USER’S EMOTION IN WHICH A USER’S EMOTIONS ARE DETECTED ACCORDING TO INTERACTIONS OF THE USER WITH KEYBOARD; THEN USER INTERFACE’S COLOR IS ADOPTED REGARDING TO USERS’ EMOTION. AUBUE INCLUDES FOUR ELEMENTS: KEYBOARD INTERPRETATION, EVENT INTERPRETATION, MOOD UPDATE, AND COLOR SELECTION. IN THE KEYBOARD INTERPRETATION ELEMENT, INTERACTIONS OF A USER WITH KEYBOARD ARE ANALYZED WITH RESPECT TO SOME PREDEFINED PARAMETERS. IN THE EVENT INTERPRETATION ELEMENT, KEYBOARD INTERACTIONS ARE INTERPRETED AS ACTIVE EMOTIONS OF THE USER. MOOD UPDATE ELEMENT IS RESPONSIBLE FOR MAPPING ACTIVE EMOTIONS TO A MOOD AND UPDATING CURRENT MOOD. FINALLY, THE COLOR SELECTION ELEMENT SELECTS APPROPRIATE COLORS TO COPE WITH CURRENT MOOD. WE HAVE IMPLEMENTED AUBUE IN A SHAPE GUESS GAME.

IN THIS PAPER, A NEW APPROACH IS PROPOSED FOR TRACKING MANEUVERING TARGET BY IMPROVING THE BASIC SIR ALGORITHM. WE PROPOSED A DYNAMIC MODEL FOR MANEUVERING TARGET. IF CHANGES OF TARGET MODEL ARE LARGER THAN A PREDETERMINED THRESHOLD, THEN THE TARGET MODEL WILL BE UPDATED. PRACTICAL RESULTS SHOW THAT OUR PROPOSED APPROACH CONSIDERABLY IMPROVES ACCURATE TRACKING OF A MANEUVERING OBJECT. WE WILL COMPARE OUR PROPOSED APPROACH WITH THE BASIC SIR ALGORITHM IN TRACKING A SINGLE MANEUVERING AIRCRAFT.RESULTS OF OUR PROPOSED ALGORITHM ARE ALSO ILLUSTRATED IN TRACKING DIFFERENT MANEUVERING TARGETS ON FEW REAL-WORLD CASE STUDY VIDEOS.

SYMMETRIC ENCRYPTION OF DATA USING TIME-VARYING KEYS AT BASE-STATION IS PROPOSED AS AN ATTRACTIVE AND INTERESTING METHOD FOR BROADCASTING IN WIRELESS SENSOR NETWORKS SINCE LAST TIME. IN THIS PAPER, WE HAVE PROPOSED A LIGHT-WEIGHT ENCRYPTION SCHEME USING TIME-VARYING AT COST OF DELAYED-VERIFICATION. MAIN IDEA OF THE PROPOSED MODEL IS THAT INSTEAD OF USING ONE KEY PER EACH PACKET, WE USE ONE KEY PER EACH GIVEN NUMBER OF PACKETS.HOWEVER, A SIGNIFICANT PROBLEM IS THAT INTERFERENCE OR DISCONNECTIONS MAY CAUSE A RECEIVER TO MISS BROADCAST PACKETS AND THE DYNAMIC KEYS CONTAINED THEREIN, RENDERING IT UNABLE TO PARTICIPATE IN SUBSEQUENT BROADCASTS. HENCE, WE IMPROVE OUR PROPOSAL BY A PATCH-LIKE LOSS-RECOVERY SCHEME. AT THE END, WE WILL DISCUSS THE SECURITY AND PERFORMANCE ASPECTS OF OUR METHOD.

THIS PAPER CONSIDERS THE DESIGN OF ROBUST OBSERVERS FOR A CLASS OF CONTINUOUS-TIME NONLINEAR SYSTEMS PRESENTED BY TAKAGI-SUGENO (T-S) MODEL WITH NONLINEAR SUBSYSTEMS AND UNMEASURABLE PREMISE VARIABLES. THE PROPOSED T-S STRUCTURE REDUCES THE NUMBER OF RULES IN THE SUGENO MODEL BY USING LOCAL NONLINEAR RULES. MOREOVER, IT CAN REPRESENT LARGER CLASS OF NONLINEAR SYSTEMS AS COMPARED TO THE MEASURABLE PREMISE VARIABLE CASE. THE PROPOSED OBSERVER GUARANTEES EXPONENTIAL CONVERGENCE OF STATE ESTIMATION ERROR BY LYAPUNOV STABILITY ANALYSIS AND LINEAR MATRIX INEQUALITY (LMI) FORMULATION IN THE PRESENCE OF BOTH PARAMETRIC AND NON-PARAMETRIC UNCERTAINTIES. NUMERICAL EXAMPLES ILLUSTRATE EFFECTIVENESS OF THE PROPOSED METHOD.

IN MOBILE WIRELESS SENSOR NETWORKS, MOBILE ELEMENTS CAN BE EITHER SINK OR SENSOR. IN THESE NETWORKS, ENERGY CONSTRAINT IS VERY IMPORTANT. CLUSTERING ENHANCES NETWORK PERFORMANCE BY CONSERVE MORE ENERGY AS A RESULT OF DATA AGGREGATION AND FEWER DATA TRANSMISSION. MORE ENERGY SAVING IN MOBILE WIRELESS SENSOR NETWORKS RESULTS USING ENERGY AWARE CLUSTERING ALGORITHMS, WHICH CAUSES NETWORK REMAINS LONGER.THE GOAL OF THIS STUDY IS USING LEACH AS A CLUSTERING ALGORITHM, WHICH IS ONE OF THE MOST WELL-KNOWN AND PRACTICAL ENERGY BALANCING ALGORITHMS IN THE NETWORK. WE PERFORM MATHEMATICAL ANALYSIS BASED ON LEACH USING MARKOV MODEL.FURTHERMORE, THE MOST OPTIMAL PATH FOR MOVING SINK IS ACHIEVED BY AN IMPROVEMENT OF ANT COLONY OPTIMIZATION.COMPARING OUR PROPOSED METHOD WITH CIRCULAR SINK METHODOLOGY REVEALED THAT OUR METHOD OUTPERFORMS IN NETWORK LONGEVITY AND ENERGY CONSUMPTION.

ELECTROENCEPHALOGRAM (EEG) IS GENERALLY KNOWN AS A NON-STATIONARY SIGNAL. DIVIDING A SIGNAL INTO THE EPOCHS WITHIN WHICH THE SIGNALS CAN BE CONSIDERED STATIONARY IS THEREFORE VERY IMPORTANT IN MANY APPLICATIONS. NOISE OFTEN INFLUENCES THE PERFORMANCE OF AN AUTOMATIC SIGNAL SEGMENTATION SYSTEM. THEREFORE, IN THIS ARTICLE, A NEW APPROACH FOR SEGMENTATION OF THE EEG SIGNALS BASED ON SINGULAR SPECTRUM ANALYSIS (SSA) AND IMPERIALIST COMPETITIVE ALGORITHM (ICA) IS PROPOSED. AS THE FIRST STEP, SSA IS EMPLOYED TO REDUCE THE EFFECT OF VARIOUS NOISE SOURCES. THEN, FRACTAL DIMENSION (FD) OF THE SIGNAL IS ESTIMATED AND USED AS A FEATURE EXTRACTION FOR AUTOMATIC SEGMENTATION OF THE EEG. IN ORDER TO SELECT TWO ACCEPTABLE PARAMETERS RELATED TO THE FD, ICA THAT IS A MORE POWERFUL EVOLUTIONARY ALGORITHM THAN TRADITIONAL ONES IS APPLIED. BY USING SYNTHETIC AND REAL EEG SIGNALS, THE PROPOSED METHOD IS COMPARED WITH ORIGINAL APPROACH (I.E. WITHOUT USING SSA AND ICA). ALSO, SIMULATION RESULTS SHOW THAT THE SPEED OF SSA IS MUCH BETTER THAN THAT OF THE DISCRETE WAVELET TRANSFORM (DWT) WHICH HAS BEEN ONE OF THE MOST POPULAR PREPROCESSING FILTERS FOR SIGNAL SEGMENTATION. THE SIMULATION RESULTS INDICATE THE PERFORMANCE SUPERIORITY OF THE PROPOSED METHOD.

DIABETIC RETINOPATHY IS ONE OF THE LEADING CAUSES OF BLINDNESS IN THE WORLD. DETECTION OF HARD EXUDATES IS AN IMPORTANT STEP FOR EARLY DIAGNOSIS IN EYE DISEASES SUCH AS MACULAR EDEMA (ME). IF HARD EXUDATES WERE SEGMENTED PRECISELY, LASER TREATMENTS CAN BE APPLIED MORE EFFECTIVE FOR PATIENTS BY SURGEONS. THE POSSIBILITY OF BLINDNESS IS VERY HIGH WHEN THE HARD EXUDATES ARE VERY CLOSE TO MACULA REGION OR OPTIC DISC. THEREFORE, FAST AND ACCURATE SEGMENTATION IS ONE OF THE MOST IMPORTANT FACTORS IN ELIMINATION OF HARD EXUDATES. IN THIS PAPER, A METHOD IS PROPOSED FOR SEGMENTATION OF HARD EXUDATES IN RETINAL COLOR IMAGE BASED ON MORPHOLOGICAL OPERATION. IN THE PROPOSED METHOD, THE RETINAL IMAGES PREPROCESSED AND OPTIC DISC AND THE BLOOD VESSELS IDENTIFIED PRIMARILY AND THEN THEY ELIMINATE FROM THE IMAGE. FINALLY, THE HARD EXUDATES (HES) ARE SEGMENTED BY MIXTURE OF MORPHOLOGICAL OPERATION SUCH AS TOP-HAT, BOTTOM-HAT AND RECONSTRUCTION OPERATIONS. THE PROPOSED METHOD WAS TESTED ON DIARETDB1 DATABASE AND 78.28% OF SENSITIVITY WAS OBTAINED.COMPARING TO OTHER RECENT AUTOMATIC METHOD AVAILABLE IN THE LITERATURE, OUR PROPOSED METHOD CAN OBTAIN ACCEPTABLE EXUDATES DETECTION RESULT IN TERM OF SENSITIVITY.

IN THIS PAPER, A NEW METHOD FOR FARSI FONT RECOGNITION BASED ON COMBINATION OF FEATURES IS PROPOSED. THE FEATURES ARE EXTRACTED AND COMBINED FROM TEXTURES OF SIZE 128×128 USING SRF AND WAVELET TRANSFORM. WAVELET AND SRF ARE NATURALLY DIFFERENT METHODS OF FEATURE EXTRACTION, SO THEIR ERRORS HAVE LOW CORRELATION. IN THIS CONDITION, THE COMBINATION OF THESE FEATURES WHICH ARE BOTH APPLICABLE FOR TEXTURE RECOGNITION WAS EXPECTED TO REDUCE TOTAL ERROR AND THE EXPERIMENTAL RESULTS APPROVED THIS HYPOTHESIS. THE PROPOSED ALGORITHM IS TESTED ON 21000 SAMPLES PROVIDED FROM 10 COMMON FARSI FONTS. IN THE METHOD PRESENTED HERE, THE FONT CHARACTERISTICS ARE EXTRACTED WELL AND THIS IS CLEAR IN THE RESULTS. WE ACHIEVED THE RECOGNITION RATE OF 95.56% USING MLP CLASSIFIER WHICH IS 2.37% AND 11.79% MORE THAN SRF AND WAVELET TRANSFORM RESPECTIVELY.

SHAPE IS ONE OF THE MAIN FEATURES IN CONTENT-BASED IMAGE RETRIEVAL (CBIR). THIS PAPER PROPOSES A NOVEL CBIR TECHNIQUE BASED ON SHAPE FEATURE. THIS TECHNIQUE USES THE DISTANCES BETWEEN THE BOUNDARY POINTS OF A SHAPE AND THE SMALLEST RECTANGLE THAT COVERS IT. THE PROPOSED TECHNIQUE IS A FOURIER BASED TECHNIQUE AND IT IS INVARIANT TO TRANSLATION, SCALING AND ROTATION. THE RETRIEVAL PERFORMANCE BETWEEN SOME COMMONLY USED FOURIER BASED SIGNATURES AND OUR SMALLEST RECTANGLE DISTANCE (SRD) SIGNATURES HAS BEEN TESTED USING MPEG-7 DATABASE. EXPERIMENTAL RESULTS ARE SHOWN THAT THE SRD SIGNATURE HAS BETTER PERFORMANCE COMPARED WITH MANY OF THOSE SIGNATURES.

A SUPER-RESOLUTION RECONSTRUCTION FROM SINGLE IMAGE ALGORITHM DESIGNED FOR LICENSE PLATE RECOGNITION IS PROPOSED IN THIS PAPER. LOW RESOLUTION IMAGES DATABASE IS GENERATED BY DOWN SAMPLING AND ADDING WHITE GAUSSIAN NOISE TO THE SUPER RESOLUTION LICENSE PLATE DATABASE. THE LOW-RESOLUTION IMAGE CAN BE VIEWED AS A DOWN SAMPLED VERSION OF A HIGH-RESOLUTION IMAGE, WHERE ITS PATCHES ARE ASSUMED TO HAVE A SPARSE REPRESENTATION WITH RESPECT TO AN OVER-COMPLETE DICTIONARY OF PROTOTYPE SIGNAL ATOMS. THE PRINCIPLE OF COMPRESSED SENSING ENSURES THAT UNDER MILD CONDITIONS, THE SPARSE REPRESENTATION CAN BE CORRECTLY RECOVERED FROM THE DOWN SAMPLED SIGNAL.THEREFORE, TWO DICTIONARY OF LOW AND HIGH RESOLUTION FROM SAME IMAGES PATCHES ARE TRAINED. FINALLY, SUPER RESOLUTION IMAGES FROM SINGLE LOW RESOLUTION IMAGE ARE RECOVERED, BY SOLVING AN OPTIMIZATION PROBLEM BY GENETIC ALGORITHM.

THIS PAPER PROPOSES A MARKERLESS VIDEO ANALYTIC SYSTEM FOR QUANTIFYING BODY PARTS MOVEMENT WHILE LYING. THESE MOVEMENTS INCLUDE: HAND, LEG, BOTH HAND & LEG AND TURNING TO LEFT OR RIGHT MOVEMENTS. COMBINATION OF PIXEL INTENSITY AND AREA DIFFERENCE OF BOTH SEGMENTED AND THE WHOLE PARTS OF EACH SILHOUETTE COMPARED WITH THE FOLLOWING SILHOUETTES WOULD PROVIDE A USEFUL CUE FOR DETECTION OF DIFFERENT BODY PARTS MOVEMENT WHILE LYING. EXTRACTED FEATURE VECTORS AFTER APPLYING PCA ALGORITHM FOR DIMENSION REDUCTION ARE FINALLY FED TO A MULTICLASS SUPPORT VECTOR MACHINE FOR PRECISE CLASSIFICATION OF PROPOSED MOVEMENTS. UNLIKE MOST OF THE EXISTENT HUMAN ACTION DETECTION SYSTEMS THAT ONLY DEAL WITH HUMAN MOVEMENTS WHILE STANDING, WE HAVE CONSIDERED MOVEMENTS THAT A PERSON DOES WHILE LYING, WHICH HAS A WIDE RANGE OF APPLICATION IN SPORT AND MEDICAL SCIENCE. RELIABLE RECOGNITION RATE OF EXPERIMENTAL RESULTS UNDERLIES SATISFACTORY PERFORMANCE OF OUR SYSTEM.

TODAY, THE CHURN PHENOMENON HAS BEEN CONSIDERED IN MANY APPLICATIONS AS AN IMPORTANT OUTCOME. SOCIAL NETWORKS CAN BE CONSIDERED AS ONE OF THE MOST IMPORTANT APPLICATIONS WITH THE MENTIONED OUTCOME. CHURN IN SOCIAL NETWORKS DEPENDS ON THE USERS’ ACTIVITY IN A COMMUNICATION ENVIRONMENT AND APPEARS IF THIS ACTIVITY IS LESS THAN A REQUIRED EXTENT. SWARM INTELLIGENCE ALGORITHMS (SI), ASSUMED TO BE THE PROPER TOOLS TO MODEL THE COMMUNICATIONS IN A SOCIAL NETWORK. THIS BUNCH OF ALGORITHMS ACCORDING TO THE LOCAL AGENTS’ BEHAVIOR, TRY TO RESULT THE GLOBAL BEHAVIOR. THIS PAPER AIMS TO MEASURING THE USER’S CHURN BY THE MENTIONED METHOD AND INCLUDING THE COMMUNICATION MESSAGES TRANSFERRED BY THE USERS IN THE NETWORK. CONSIDERING THE MEASURED ACTIVITY RATE, A CHURN THRESHOLD IN VARIOUS AREAS OF COMMUNICATION WILL BE OBTAINED. SIMULATION RESULTS REFERRING TO CONFIRMED THE PRESENTED MODEL OF COMMUNICATION. THE MODEL VALIDATION AND OTHER VALUES ARE OBTAINED BY A DISCRETE EVENT SIMULATOR. THE COMMUNICATIONS USED IN THIS SIMULATION RESULT FROM MINING A DATA SET INCLUDING REAL COMMUNICATIONS FOR ONE SPECIES OF THE MENTIONED NETWORKS.

FLEXIBLE JOB SHOP SCHEDULING PROBLEM (FJSP) IS AN IMPORTANT EXTENSION OF THE CLASSICAL JOB SHOP SCHEDULING PROBLEM, WHERE EACH OPERATION COULD BE PROCESSED ON MORE THAN ONE MACHINE AND VICE VERSA. SINCE IT HAS BEEN PROVEN THAT THIS PROBLEM IS STRONGLY NP-HARD, IT IS DIFFICULT TO ACHIEVE AN OPTIMAL SOLUTION WITH TRADITIONAL OPTIMIZATION ALGORITHMS. IN THIS PAPER A NEW APPROACH IS PROPOSED TO SOLVE THE MULTIOBJECTIVE FJSP. THIS NEW APPROACH HAS THREE STEPS. FIRST, AN INITIAL POPULATION OF FEASIBLE SOLUTIONS WITH GOOD DISTRIBUTION IN THE SEARCH SPACE IS CREATED BY USING A PARAMETER CALLED NEIGHBORHOOD. SECOND, THIS POPULATION, BASED ON FITNESS AND NEIGHBORHOOD PARAMETERS, EXPLORES THE SEARCH SPACE UNTIL IT WILL FORM SEVERAL DYNAMIC CLUSTERS AROUND GOOD AREAS, INCLUDING LOCAL OPTIMUMS. FINALLY, IN PARALLEL, A LOCAL SEARCH IS PERFORMED ON THE BEST SOLUTION FOR EACH CLUSTER BY USING TABU SEARCH ALGORITHM AND EVENTUALLY THE OPTIMAL SOLUTION IS OBTAINED AMONG THEM. COMPUTATIONAL RESULTS ON BENCHMARK PROBLEMS SHOW THAT THE OPTIMAL SOLUTIONS ARE OBTAINED MUCH FASTER THAN OTHER APPROACHES.

EXTREMAL OPTIMIZATION (EO) ALGORITHM IS A KIND OF EVOLUTIONARY OPTIMIZATION METHOD WHICH HAS BEEN APPLIED SUCCESSFULLY IN DIFFERENT FIELDS. IN THIS PAPER A NEW FRAMEWORK IS PROPOSED FOR APPLYING EXTREMAL OPTIMIZATION IN IMAGE SEGMENTATION. OVERSEGMENTED IMAGES ARE THE INITIAL TO EO WHICH WORKS ON TWO LEVELS: SEGMENTS AND PIXELS. A NEW ENERGY FUNCTION IS DEFINED FOR SEGMENTS AND THE ENERGY FUNCTION IN MARKOV RANDOM FIELDS (MRF) IS USED FOR PIXELS. APPLYING EO IN SEGMENT LEVEL ACCELERATES THE SPEED OF THE ALGORITHM. THE RESULTS SHOW THAT BY DEFINING A SUITABLE ENERGY FUNCTION, EO CAN BE SUCCEED IN MERGING SIMILAR SEGMENTS AND PROVIDE A VISUALLY GOOD SEGMENTATION.

SIMPLE STRUCTURE, QUICK, AND ROBUST RESPONSE OF DIRECT TORQUE CONTROL (DTC) MAKE IT ATTRACTIVE IN AC DRIVES. HOWEVER, IN SENSORLESS APPLICATIONS, PRECISE ESTIMATION OF ROTOR SPEED IS A CRUCIAL ISSUE AS WELL AS THE TORQUE AND FLUX ESTIMATION. IN THIS PAPER AN ADAPTIVE FLUX OBSERVER IS INTRODUCED WITH STATOR CURRENT AND FLUX VECTOR COMPONENTS AS STATE VARIABLES AND ROTOR SPEED AS AN UNKNOWN PARAMETER WHICH ARE ESTIMATED SIMULTANEOUSLY USING AN ADAPTIVE SCHEME. ALSO TORQUE, FLUX, AND CURRENT PULSATIONS DURING STEADY STATE ARE CONSIDERED AS THE DISADVANTAGES OF THE CLASSICAL DTC METHOD. A COMBINED DTCSVM STRATEGY IS PRESENTED USING FUZZY LOGIC CONTROL (FLC). FUZZY LOGIC CONTROLLERS ARE DESIGNED SUCH THAT OBTAIN A FAST RESPONSE FOR DYNAMIC PERFORMANCE, AND HIGH ACCURACY IN STEADY STATE.

ONE OF THE MOST CRITICAL ISSUES IN WIRELESS SENSOR NETWORKS IS THE LIMITED ENERGY AVAILABILITY OF THE NETWORK NODES.THIS PAPER IS INVESTIGATING THE ADVANTAGES OF USING CONTROLLED SINK MOBILITY IN CLUSTERED WIRELESS SENSOR NETWORKS (WSNS) WHICH INCREASES NETWORK LIFETIME. IN A CLUSTERED SENSOR NETWORK ALL CLUSTER HEADS (CHS) HAVE TO TRANSMIT THEIR BUFFERED DATA TO THE SINK DURING A SPECIFIED INTERVAL, CALLED DATA REPORTING TIME (TDR). IN THIS PAPER, WE PROPOSE A SCHEME THAT PRESCRIBES THE SINK PATH FOR COLLECTING ALL CHS DATA INTDR TIME SPAN WHILE MAXIMIZING NETWORK LIFE TIME USING THE MATHEMATICAL MODEL MILP (MIXED INTEGER LINEAR PROGRAMMING). THE PROPOSED SCHEME IS COMPARED WITH OTHER RELATED SCHEMES BY MEANS OF VARIOUS SIMULATION SCENARIOS. SIMULATION RESULTS SHOW THAT THE PROPOSED SCHEME SIGNIFICANTLY OUTPERFORMS OTHER SCHEMES.

THE GROWTH OF DIFFERENT TYPES OF APPLICATIONS IN THE INTERNET ARISES THE NEED FOR SUPPORTING QUALITY OF SERVICE (QOS) FOR THEM. DIFFERENTIATING SERVICES AIMS TO DIFFERENTIATE AMONG APPLICATIONS IN ORDER TO PROVIDE THEIR REQUIREMENTS BASED ON THE LEVEL OF QOS THEY NEED AND THEIR CHARACTERISTICS.IN THIS PAPER WE INTRODUCE A DYNAMIC QOS-AWARE QUEUING ALGORITHM TO DIFFERENTIATE SERVICES IN IP BASED NETWORKS, WHICH CONSISTS OF A PACKET-BY-PACKET DYNAMIC CLASSIFICATION ALGORITHM AND A WEIGHTED ROUND ROBIN SCHEDULING ALGORITHM.THE PROPOSED DYNAMIC QOS-AWARE QUEUING PROVIDES THE QOS REQUIRED FOR DELAY SENSITIVE APPLICATIONS AND IMPROVES THE PERFORMANCE OF THE SYSTEM, MEANING SATISFYING MORE USERS. PLUS, THE ALGORITHM IS CAPABLE OF PROVIDING A SATISFACTORY FAIRNESS AMONG FLOWS. THE MAIN ADVANTAGE OF THE ALGORITHM IS ITS ABILITY TO DYNAMICALLY ADAPT ITS PARAMETERS TO THE CHANGES IN THE NETWORK.

IRIS SEGMENTATION IS THE PRINCIPAL TASK OF IRIS BASED BIOMETRIC IDENTIFICATION SYSTEMS. ALWAYS THE LOW CONTRAST BETWEEN PUPIL AND IRIS ALWAYS AFFECTS ON ACCURACY OF DETECTING BOUNDARY BETWEEN THEM. IN ORDER TO INCREASING THE ACCURACY OF BOUNDARY DETECTING, WE PROPOSE A NEW TECHNIQUE USING A DIFFERENCE FUNCTION AND A FACTOR MATRIX. WE ALSO ENHANCE THE TECHNIQUE TO DETECT THE PUPIL AND IRIS AS ELLIPSE INSTEAD OF CIRCLE.EXPERIMENTS SHOW THAT THE PROPOSED TECHNIQUE CAN SEGMENT THE IRIS REGION AND PUPIL REGION PRECISELY. BASED ON OUR RESULT, 99.34% OF EYES HAVE BEEN SEGMENTED ACCURATELY IN 1.24S AVERAGELY.

A NEW STABLE ADAPTIVE INTERVAL TYPE-2 FUZZY PROPORTIONAL INTEGRAL SLIDING MODE CONTROLLER (AI2FPISMC) IS INTRODUCED HERE TO CONTROL A CLASS OF NONLINEAR SYSTEMS. THE PROPOSED METHOD IS BASED ON INTERVAL TYPE-2 FUZZY LOGIC SYSTEM (IT2FLS) WHOSE ANTECEDENT AND CONSEQUENT MEMBERSHIP FUNCTIONS ARE INTERVAL TYPE-2 FUZZY SETS. IT2FLS IS UTILIZED TO APPROXIMATE UNKNOWN NONLINEAR FUNCTIONS. TO ACHIEVE HIGH PERFORMANCE, OPTIMIZING MEMBERSHIP FUNCTIONS (MFS) OF INTERVAL TYPE-2 FUZZY SETS (IT2FS) IS REQUIRED. GENETIC ALGORITHM (GA) IS A PARALLEL SEARCH OPTIMIZATION METHOD; THAT HERE CONTRIBUTES TO OPTIMIZE THE MFS. IN ORDER TO COPE WITH THE CHATTERING OF SLIDING MODE CONTROLLER, PI CONTROL LAW IS PROPOSED AND LYAPUNOV ANALYSIS IS UTILIZED TO PROVE ASYMPTOTIC STABILITY OF THE PROPOSED APPROACH. THE ADAPTATION LAWS ARE DERIVED USING LYAPUNOV APPROACH. TWO NONLINEAR SYSTEM SIMULATION EXAMPLES ARE PRESENTED TO VERIFY THE EFFECTIVENESS OF THE PROPOSED METHOD, AND THEIR RESULTS CONFIRM THE OPTIMIZATION MERITS.

THE HUMAN MOTION ANALYSIS IS AN ATTRACTIVE TOPIC IN BIOMETRIC RESEARCH. COMMON BIOMETRICS IS USUALLY TIME-CONSUMING, LIMITED AND COLLABORATIVE. THESE DRAWBACKS POSE MAJOR CHALLENGES TO RECOGNITION PROCESS. RECENT RESEARCHES INDICATE PEOPLE HAVE CONSIDERABLE ABILITY TO RECOGNIZE OTHERS BY THEIR NATURAL WALKING. THEREFORE, GAIT RECOGNITION HAS OBTAINED GREAT TENDENCY IN BIOMETRIC SYSTEMS. GAIT ANALYSIS IS INCONSPICUOUS, NEEDS NO CONTACT, CANNOT BE HIDDEN AND IS EVALUATED AT DISTANCE. THIS PAPER PRESENTS A BAG OF WORD METHOD FOR GAIT RECOGNITION BASED ON DYNAMIC TEXTURES. DYNAMIC TEXTURES COMBINE APPEARANCE AND MOTION INFORMATION. SINCE HUMAN WALKING HAS STATISTICAL VARIATIONS IN BOTH SPATIAL AND TEMPORAL SPACE, IT CAN BE DESCRIBED WITH DYNAMIC TEXTURE FEATURES. TO OBTAIN THESE FEATURES, WE EXTRACT SPATIOTEMPORAL INTEREST POINTS AND DESCRIBE THEM BY A DYNAMIC TEXTURE DESCRIPTOR. TO GET MORE SUITABLE RESULTS, WE EXTEND LBP-TOP AS A ROTATION INVARIANT DYNAMIC TEXTURE DESCRIPTOR. AFTERWARDS, HIERARCHICAL K-MEANS ALGORITHM IS EMPLOYED TO MAP FEATURES INTO VISUAL WORDS. AT RESULT, HUMAN WALKING REPRESENT AS A HISTOGRAM OF VIDEO-WORDS OCCURRENCES. WE EVALUATE THE PERFORMANCE OF OUR METHOD ON TWO DATASET: THE KTH DATASET AND IXMAS MULTIVIEW DATASET.

LIFE TIME AND ENERGY EFFICIENCY ARE IMPORTANT FACTORS IN THE DESIGN OF WIRELESS SENSOR NETWORK. A CRITICAL ISSUE DURING DATA COLLECTION IS THE FORMATION OF ENERGY HOLES NEAR THE SINK. SENSORS NEAR THE SINK HAVE TO PARTICIPATE IN RELAYING DATA ON BEHALF OF OTHER SENSORS AND THUS THEIR ENERGY WILL BE DEPLETED VERY QUICKLY. THE MOBILE SINK MOVEMENT YIELDS THE SIGNIFICANT PERFORMANCE GAINED THROUGH DECREASING THE AMOUNT OF ENERGY CONSUMPTION. IN THIS STUDY, WE PROPOSE A GRID BASED APPROACH FOR DATA DISSEMINATING TO THE MOBILE SINK IN WIRELESS SENSOR NETWORK. MAPPING THE POSITION OF NODES AND SINK TO VIRTUAL GRID INFRASTRUCTURE CONSTRUCT OUR MAIN CONTRIBUTION. SO THAT THE DATA WHICH IS BUFFERED BY NODES COULD BE DISSEMINATED TO THE MOBILE SINK TO THE OPTIMAL MANNER.FINALLY, COMPREHENSIVE SIMULATION REVEALS THE SUPERIORITY OF PROPOSED ALGORITHM IN VARIOUS SCENARIOS.

TRUST IS A CONCEPT TAKEN FROM SOCIAL SCIENCES AND IS CONSIDERED AS A SOFT SECURITY APPROACH THAT IS EFFECTIVE IN REDUCING RISK. IN THIS PAPER, FOR ESTIMATING THE TRUST BETWEEN UNKNOWN NODES, A GROUP-BASED TRUST PROPAGATION METHOD HAS BEEN PROPOSED. MOST OF THE CONVENTIONAL TRUST PROPAGATION METHODS ARE NOT APPLICABLE FOR TRUST EVALUATION OF TODAY'S LARGE TRUST GRAPHS. OUR TRUST PROPAGATION METHOD IS SCALABLE BECAUSE OF USING GROUPING METHOD. FOR BETTER TRUST ESTIMATION INSIDE GROUP THE CONFIDENCE OF TRUST VALUE HAVE BEEN CONSIDERED. WE ALSO CONSIDER TWO FACTORS INTERMEDIATE GROUP CONFIDENCE (IGC) AND GROUP CONFIDENCE (GC) FOR CONFIDENCE OF TRUST BETWEEN TWO GROUPS. TO EVALUATE THIS METHOD A REAL LARGE DATA SET OF ADVOGATO.ORG IS USED. EVALUATION OF ACCURACY IS BASED ON CORRELATION AND MEAN ABSOLUTE ERROR (MAE). COMPARING THE PROPOSED METHOD WITH THE ITERATIVE MULTIPLICATION METHOD (IMS) RESULTS SUGGESTS THAT THE CORRELATION AND ABSOLUTE MEAN ERROR HAVE BEEN IMPROVED. IN ADDITION, DUE TO THE USE OF GROUPBASED METHOD THE SPEED OF THE PROPOSED METHOD HAS BEEN IMPROVED.

THE FOUR-LEG VOLTAGE SOURCE INVERTER (VSI) IS BECOMING INCREASINGLY POPULAR IN POWER APPLICATIONS DUE TO ITS SUPERIOR PERFORMANCE CHARACTERISTICS SUCH AS RELATIVELY LOW DC-BUS VOLTAGE AND CAPABILITY TO HANDLE UNBALANCED OR NONLINEAR LOAD CURRENTS. WITH UNBALANCED LOAD CONDITION, TO BALANCE SUCH A SYSTEM, A NEUTRAL CONNECTION IS NEEDED TO CONTROL EACH PHASE VOLTAGE INDEPENDENTLY. AVERAGE LARGE SIGNAL MODEL IN ABC COORDINATES IS OBTAINED AND STATE SPACE EQUATION IS PRESENTED. THIS PAPER PRESENTS NEW DECOUPLED CONTROL STRATEGY BASED ON STATE FEEDBACK WITH SIMPLE PROPORTIONAL CONTROLLER THUS EACH PHASE IS CONTROLLED INDEPENDENTLY IN ORDER TO TRACK REFERENCES PROPERLY. SIMULATION RESULTS WITH MATLAB/SIMULINK SHOW ACCEPTABLE PERFORMANCE IN UNBALANCED LOAD CONDITIONS.

IN THIS PAPER, A NEW UNSUPERVISED SENTENCE COMPRESSION METHOD IS PROPOSED. SENTENCES ARE TAGGED WITH PART OF SPEECH TAGS AND SEMANTIC ROLE LABELS. THE PROPOSED METHOD RELIES ON THE SEMANTIC ROLES OF SENTENCES’ PARTS.MOREOVER, IN THE PROCESS OF COMPRESSION, OTHER SENTENCES IN THE CONTEXT ARE TAKEN INTO ACCOUNT. THE APPROACH IS APPLIED IN THE CONTEXT OF MULTI-DOCUMENT SUMMARIZATION. EXPERIMENTS SHOWED BETTER RESULTS THAN OTHER STATE OF THE ART APPROACHES.

IN RECENT YEARS, WITH GROWTH OF ONLINE LEARNING TECHNOLOGY, A HUGE AMOUNT OF E-LEARNING RESOURCES HAVE BEEN GENERATED IN VARIOUS MEDIA FORMATS. THIS GROWTH HAS CAUSED DIFFICULTY OF LOCATING APPROPRIATE LEARNING RESOURCES TO LEARNERS. A PERSONALIZED RECOMMENDATION IS AN ENABLING MECHANISM TO OVERCOME INFORMATION OVERLOAD OCCURRED IN THE NEW LEARNING ENVIRONMENTS AND DELIVER SUITABLE LEARNER RESOURCES TO LEARNERS. SINCE USERS EXPRESS THEIR OPINIONS BASED ON SOME SPECIFIC ATTRIBUTES OF ITEMS, THIS PAPER CONSIDERS CONTEXTUAL INFORMATION INCLUDING ATTRIBUTES OF LEARNING RESOURCES AND RATING OF LEARNER SIMULTANEOUSLY TO ADDRESS SOME PROBLEM SUCH AS SPARSITY AND COLD START PROBLEM AND ALSO IMPROVE THE QUALITY ON RECOMMENDATIONS. LEARNING TREE (LT) IS INTRODUCED THAT CAN MODEL THE INTEREST OF LEARNERS BASED ON ATTRIBUTES OF LEARNING RESOURCES IN MULTIDIMENSIONAL SPACE USING LEARNER HISTORICAL ACCESSED RESOURCES. THEN, USING A NEW SIMILARITY MEASURE BETWEEN LEARNERS, RECOMMENDATIONS ARE GENERATED. THE EXPERIMENTAL RESULTS SHOW THAT OUR PROPOSED METHOD OUTPERFORMS CURRENT ALGORITHMS AND ALLEVIATES PROBLEMS SUCH AS COLD-START AND SPARSITY.

DESIGN PATTERNS PROVIDE EXPERIENCE REUSABILITY AND INCREASE QUALITY OF OBJECT ORIENTED DESIGNS. KNOWING WHICH DESIGN PATTERNS ARE IMPLEMENTED IN A SOFTWARE IS IMPORTANT IN COMPREHENDING, MAINTAINING AND REFACTORING ITS DESIGN.HOWEVER, DESPITE THE INTEREST IN USING DESIGN PATTERNS, TRADITIONALLY, THEIR USAGE IS NOT EXPLICITLY DOCUMENTED.THEREFORE, A METHOD IS REQUIRED TO REVEAL THIS INFORMATION FROM SOME ARTIFACTS OF THE SYSTEMS (E.G. SOURCE CODES, MODELS, AND EXECUTABLES). IN THIS PAPER, AN APPROACH IS PROPOSED WHICH USES THE SEMANTIC WEB TECHNOLOGIES FOR AUTOMATICALLY DETECTING DESIGN PATTERNS FROM JAVA SOURCE CODE. IT IS BASED ON THE SEMANTIC DATA MODEL AS THE INTERNAL REPRESENTATION, AND ON SPARQL QUERY EXECUTION AS THE ANALYSIS MECHANISM.EXPERIMENTAL EVALUATIONS DEMONSTRATE THAT THIS APPROACH IS BOTH FEASIBLE AND EFFECTIVE, AND IT REDUCES THE COMPLEXITY OF DETECTING DESIGN PATTERNS TO CREATING A SET OF SPARQL QUERIES.

THIS PAPER PROPOSES A NEW FUZZY-BASED DIRECT POWER CONTROL (FDPC) FOR DOUBLY FED INDUCTION GENERATOR (DFIG) - BASED WIND ENERGY CONVERSION SYSTEMS. IN THE PROPOSED METHOD, THE REQUIRED ROTOR VOLTAGES ARE DIRECTLY CALCULATED BASED ON FUZZY CONTROLLER, STATOR VOLTAGE, ROTOR SPEED AND SOME MACHINE PARAMETERS. THE SWITCHING TABLE AND HYSTERESIS COMPARATOR USED IN CONVENTIONAL DPC ARE REPLACED BY PWM MODULATOR. ALSO THE CONVERTER SWITCHING FREQUENCY IS CONSTANT. FURTHERMORE, THE EFFECT OF MACHINE PARAMETERS MISMATCHES IS FOUND NEGLIGIBLE. SIMULATIONS ARE DONE TO VERIFY THE EFFECTIVENESS OF PROPOSED METHOD UNDER STEADY STATE AND TRANSIENT CONDITIONS AND ALSO MACHINE PARAMETERS MISMATCHES.

IN THIS PAPER, THE ISSUE OF AUTONOMOUS PARALLEL PARKING OF A VEHICLE IN A LIMITED SPACE HAS BEEN CONSIDERED BASED ON A PARKED VEHICLE IN FRONT. A RADIAL BASIS FUNCTION NETWORK (RBFN) HAS BEEN DEVELOPED TO ONLINE PATH GENERATION ACCORDING TO THE MEASURED DISTANCE FROM THE PARKED VEHICLE. THESE MEASUREMENTS ARE OBTAINED BY EMPLOYING TWO SONAR SENSORS MOUNTED IN THE FRONT LEFT CORNER OF THE VEHICLE. FIFTH ORDER POLYNOMIAL REFERENCE PATHS FOR THREE DIFFERENT INITIAL POSITIONS HAVE BEEN USED TO GENERATE TRAINING DATA FOR THE NETWORK. IN ORDER TO USE FEEDBACK LINEARIZATION CONTROLLER FOR TRACKING THE DESIRED PARKING PATH, TWO TIMING LAWS HAVE BEEN PROPOSED FOR FORWARD AND BACKWARD MANEUVERS AND THE CONTROL LAWS HAVE BEEN DESIGNED IN TIMING LAWS-DOMAIN. THESE TIMING LAWS HAVE BEEN CONSIDERED SUCH THAT ZERO VELOCITY IS OBTAINED FOR THE VEHICLE AT THE INITIAL AND GOAL POINTS OF THE PARKING MANEUVER, AS WELL AS, AT THE POINT THAT DIRECTION OF MOTION MUST CHANGE, WITHOUT OCCURRENCE SINGULARITY IN THE CONTROL LAWS. FURTHERMORE, DUE TO CHANGE OF THE DESIRED PATH AND DIRECTION OF MOTION BASED ON THE VEHICLE’S CONFIGURATION, THE VEHICLE CAN BE PARKED IN LIMITED AREA. SIMULATION RESULTS SHOW THE EFFECTIVENESS OF THE PROPOSED APPROACH TO USE THE NONLINEAR CONTROLLER IN ORDER TO INTELLIGENTLY PERFORM THE PARALLEL PARKING IN A LIMITED SPACE WITHOUT KNOWING THE PARKING SPACE WIDTH AND ONLY BASED ON DISTANCE MEASUREMENTS FROM THE PARKED VEHICLE.

IN THIS PAPER, A NEW ALGORITHM WHICH IS THE RESULT OF THE COMBINATION OF CELLULAR LEARNING AUTOMATA AND FROG LEAP ALGORITHM (SFLA) IS PROPOSED FOR OPTIMIZATION IN CONTINUOUS, STATIC ENVIRONMENTS.AT THE PROPOSED ALGORITHM, EACH MEMEPLEX OF FROGS IS PLACED IN A CELL OF CELLULAR LEARNING AUTOMATA. LEARNING AUTOMATA IN EACH CELL ACTS AS THE BRAIN OF MEMEPLEX, AND WILL DETERMINE THE STRATEGY OF MOTION AND SEARCH.THE PROPOSED ALGORITHM ALONG WITH THE STANDARD SFLA AND TWO GLOBAL AND LOCAL VERSIONS OF PARTICLE SWARM OPTIMIZATION ALGORITHM HAVE BEEN TESTED IN 30-DIMENSIONAL SPACE ON FIVE STANDARD MERIT FUNCTIONS. EXPERIMENTAL RESULTS SHOW THAT THE PROPOSED ALGORITHM HAS A VERY GOOD PERFORMANCE.

IN THIS PAPER, A NEW TARGET TRACKING METHOD IS PROPOSED IN WIRELESS VISUAL SENSOR NETWORKS (WVSN) USING TARGET MOVEMENT PREDICTION. TWO IMPORTANT CRITERIA IN THIS REGARD ARE ACCURACY OF TRACKING AND EFFICIENT ENERGY CONSUMPTION. BECAUSE THERE IS A DIRECT RELATIONSHIP BETWEEN THE AMOUNT OF COVERAGE AND ACCURACY TRACKING, AN EVOLUTIONARY ALGORITHM (EA) IS USED AS PRE-PROCESSING TO INCREASE THE COVERAGE PERCENTAGE. THEN, A NEURAL NETWORKBASED APPROACH USING HISTORY OF TARGET MOVEMENTS ARE PRESENTED TO PREDICT THE PATH OF THE TARGET. COMPUTER SIMULATIONS SHOWED IMPROVEMENT IN TRACKING ACCURACY, CONFIGURATION OF WVSN COST AND ENERGY CONSERVATION. THE IMPORTANT ADVANTAGE OF THIS APPROACH IS THE CAPABILITY OF TRACKING IN AN ENVIRONMENT, WHICH IS NOT COVERED COMPLETELY.

CITATION RECOMMENDATION SYSTEMS CAN HELP A RESEARCHER FIND WORKS THAT ARE RELEVANT TO HIS FIELD OF INTEREST. CURRENTLY, MOST APPROACHES IN CITATION RECOMMENDATION ARE BASED ON A CLOSED-WORLD VIEW WHICH IS LIMITED TO USING A SINGLE DATA SOURCE FOR RECOMMENDATION. SUCH A LIMITATION DECREASES QUALITY OF THE RECOMMENDATIONS SINCE NO SINGLE DATA SOURCE CONTAINS ALL REQUIRED INFORMATION ABOUT DIFFERENT ASPECTS OF THE LITERATURE. THIS PAPER PROPOSES A CITATION RECOMMENDATION APPROACH BASED ON THE OPEN-WORLD VIEW PROVIDED BY THE EMERGING WEB OF DATA. IT USES MULTIPLE LINKED DATA SOURCES TO CREATE A RICH BACKGROUND DATA LAYER, AND A COMBINATION OF CONTENT-BASED AND MULTI-CRITERIA COLLABORATIVE FILTERING AS THE RECOMMENDATION ALGORITHM. EXPERIMENTS DEMONSTRATE THAT THE PROPOSED APPROACH IS SOUND AND PROMISING.

IN THIS PAPER THREE NEW METHODS BASED ON SMOOTHED NONLINEAR ENERGY OPERATOR (SNEO), FRACTAL DIMENSION (FD) AND STANDARD DEVIATION TO DETECT THE SPIKES FOR NOISY NEURONAL DATA ARE PROPOSED. IN MANY CASES, ESPECIALLY WHEN THERE ARE SEVERAL NOISE SOURCES, THESE METHODS MAY NOT BE ACCEPTABLE AS SPIKE DETECTORS. TO OVERCOME THIS PROBLEM, WE USE SAVITZKY-GOLAY FILTER AND DISCRETE WAVELET TRANSFORM (DWT) AS PRE-PROCESSING STEPS. RESULTS SHOW THAT WHEN THERE IS TOO MUCH NOISE IN THE SIGNAL, THE PROPOSED METHOD USING THE STANDARD DEVIATION AND DWT CAN DETECT THE SPIKES BETTER THAN THE OTHER METHODS. THE AVERAGE DETECTION RATE AND FALSE DETECTION OF SPIKES FOR THE PROPOSED METHOD BASED ON STANDARD DEVIATION AND DWT ARE RESPECTIVELY 100% AND 43% FOR SEMI-REAL SIGNALS WITH SNR=-5 DB.

WITH THE GROWING AMOUNT OF PUBLISHED RDF DATASETS ON SIMILAR DOMAINS, DATA CONFLICT BETWEEN SIMILAR ENTITIES (SAME-AS) IS BECOMING A COMMON PROBLEM FOR WEB OF DATA APPLICATIONS. IN THIS PAPER WE PROPOSE AN ALGORITHM TO DETECT CONFLICT OF SAME PROPERTIES VALUES OF SIMILAR ENTITIES AND SELECT THE MOST ACCURATE VALUE. THE PROPOSED ALGORITHM CONTAINS TWO MAJOR STEPS. THE FIRST STEP FILTERS OUT LOW RANKED DATASETS USING A LINK ANALYSIS TECHNIQUE. THE SECOND STEP CALCULATES AND EVALUATES THE FOCUS LEVEL OF A DATASET IN A SPECIFIC DOMAIN. FINALLY, THE VALUE OF THE TOP RANKED DATASET IS CONSIDERED. THE PROPOSED ALGORITHM IS IMPLEMENTED BY JAVA PROGRAMMING LANGUAGE AND IS EVALUATED BY GEOGRAPHICAL DATASETS CONTAINING "COUNTRY" ENTITIES.

IN THIS ARTICLE, THE PROBLEM OF SELECTION AND CONFIGURATION OF VISUAL SENSORS IN VISUAL SENSOR NETWORKS (VSNS) IS STUDIED FOR COVERAGE OF THE EXISTING TARGETS IN THE THREE-DIMENSIONAL NETWORK AREA. IT IS ASSUMED THAT VISUAL SENSORS ARE CAPABLE OF PAN-TILT-ZOOM (PTZ) ABILITY AND ABLE TO CHOOSE THE COVERED AREA AND DESIRABLE TARGET THROUGH CONFIGURATION OF THE ANGLE OF VIEW IN TWO HORIZONTAL AND VERTICAL DIMENSIONS AND ALTERATION OF THE ENLARGEMENT SCALE. IN ORDER TO SOLVE THE PROBLEM OF SELECTION AND CONFIGURATION, TWO METHODS ACCORDING TO BINARY INTEGER PROGRAMMING (BIP) AND GREEDY SEARCH ARE PRESENTED TO ENSURE TO FIND THE BEST SOLUTION AND LEAST CALCULATION COMPLEXITY RESPECTIVELY. APPLYING THE PROPOSED SOLUTIONS, THE PROBLEMS OF FULL ANGLE COVERAGE HAVE BEEN ASSESSED USING CIRCULAR TARGET MODEL. THE PERCEIVED RESULTS PROPERLY CONFIRM THE EFFICIENCY OF THE PROPOSED METHOD IN TERMS OF EXECUTION SPEED AND DECREASING THE COMPUTATION COMPLEXITY.

DIGITAL WATERMARKING HAS BEEN PROPOSED AS A WAY TO CLAIM OWNERSHIP. IN THIS PAPER, A NEW APPROACH IN DIGITAL IMAGE WATERMARKING BASED ON DISCRETE WAVELET TRANSFORM (DWT) AND SINGULAR VALUE DECOMPOSITION (SVD) IS PRESENTED. WE USE COMPOSITIONAL PATTERN PRODUCING NETWORKS (CPPNS) TO MAKE A VERY COMPACT REPRESENTATION OF WATERMARK. USING NEURO EVOLUTION OF AUGMENTING TOPOLOGIES (NEAT) WILL EVOLVE THE CPPN STRUCTURE TO PRODUCE A SUITABLE WATERMARK IMAGE. IN THE EMBEDDING PHASE, AT FIRST WE PERFORM DECOMPOSING OF THE HOST IMAGE WITH 2D-DWT TRANSFORM AT 5-LEVEL, THEN THE SVD IS APPLIED TO LH3-LH5 SUB-BANDS OF TRANSFORMED IMAGE, AND EMBED THE WATERMARK BY MODIFYING THE SINGULAR VALUES. IN WATERMARK EXTRACTION PHASE, THE EMBEDDED COEFFICIENTS OF CPPN NEAT ARE EXTRACTED FROM THE WATERMARKED IMAGE. THEN THE WATERMARK IMAGE IS RENDERED BY CPPN. THE EXPERIMENTS INDICATE THAT THE WATERMARK IS ROBUST AGAINST THE DIFFERENT ATTACKS, SUCH AS AVERAGE FILTER, JPEG COMPRESSION AND ETC.