View
11
Download
0
Category
Preview:
Citation preview
123
LNAI
987
5
8th International Conference, ICCCI 2016Halkidiki, Greece, September 28–30, 2016Proceedings, Part I
ComputationalCollective Intelligence
Ngoc-Thanh NguyenYannis ManolopoulosLazaros IliadisBogdan Trawinski (Eds.)
Fuzzy Cognitive Maps for Long-TermPrognosis of the Evolution of Atmospheric
Pollution, Based on Climate Change Scenarios:The Case of Athens
Vardis-Dimitris Anezakis1, Konstantinos Dermetzis1(&),Lazaros Iliadis1, and Stefanos Spartalis2
1 Lab of Forest-Environmental Informatics and Computational Intelligence,Department of Forestry and Management of the Environmentand Natural Resources, Democritus University of Thrace,
193 Pandazidou Street, 68200 N Orestiada, Greece{danezaki,kdemertz,liliadis}@fmenr.duth.gr
2 Laboratory of Computational Mathematics, Department of Productionand Management Engineering, School of Engineering, Democritus University
of Thrace, V. Sofias 12, Prokat, Building A1, 67100 Xanthi, Greecesspart@pme.duth.gr
Abstract. Air pollution is related to the concentration of harmful substances inthe lower layers of the atmosphere and it is one of the most serious problemsthreatening the modern way of life. Determination of the conditions that causemaximization of the problem and assessment of the catalytic effect of relativehumidity and temperature are important research subjects in the evaluation ofenvironmental risk. This research effort describes an innovative model towardsthe forecasting of both primary and secondary air pollutants in the center ofAthens, by employing Soft Computing Techniques. More specifically, FuzzyCognitive Maps are used to analyze the conditions and to correlate the factorscontributing to air pollution. According to the climate change scenarios till2100, there is going to be a serious fluctuation of the average temperature andrainfall in a global scale. This modeling effort aims in forecasting the evolutionof the air pollutants concentrations in Athens as a consequence of the upcomingclimate change.
Keywords: Fuzzy Cognitive Maps � Air pollutants � Climate change models �Soft Computing Techniques
1 Introduction
Air pollution is the condition in which air is contaminated by foreign substances,radiation or other forms of energy, in quantities or duration that can have harmful orpoisonous effects in the health of living organisms. Moreover, they can upset theecological balance in large or small geographical scale. These pollutants are eitheremitted directly by various human activities, like energy production from solid orliquid fuels, transport, industry and heating and they are known as primary ones
© Springer International Publishing Switzerland 2016N.T. Nguyen et al. (Eds.): ICCCI 2016, Part I, LNAI 9875, pp. 175–186, 2016.DOI: 10.1007/978-3-319-45243-2_16
kdemertz@fmenr.duth.gr
(e.g. CO, NO, NO2, SO2) or they are formed in the atmosphere under proper conditionsand they are known as secondary ones (e.g. O3). The assessment of air pollutionconsequences requires a comprehensive spatiotemporal analysis of the favoring con-ditions and the search for interrelations between pollutants and meteorological plusphotochemical factors affecting it [2–4]. This research effort proposes an innovativeanalytical Soft Computing model towards long term estimation of the pollutants con-centrations. More specifically it carries out a descriptive representation of complexcorrelations between atmospheric pollutants with the method of Fuzzy Cognitive Maps(FCM), based on historical data of the Athens center for the period 2000–2012.Additionally, it proposes a sophisticated method of predicting the values of pollutantsin relation to the fluctuation of temperature and rainfall values, as reflected by thenumber of projections of climate model GFDL_CM2.0 for the period 2020–2099.
1.1 Related Literature - Data
Important research efforts have been carried out on the short term, towards forecastingof the air quality in medium cities or major urban centers like Athens, using statisticalmethods and without taking into serious consideration the effect of the fluctuation oftemperature and precipitation in the 21st century due to climate change [15].
Gordaliza et al. [6] developed coherent storylines about ordinary people livingunder diverse scenarios of low/high CO2. Luiz et al. [12] constructed FCM in order tounderstand the viability of Clean Development Mechanism (CDM) projects in SouthAfrica and how they would influence greenhouse gas (GHG) emissions. Zhang et al.[20] explored the application of fuzzy cognitive maps on getting stakeholders’ per-spectives and they employed graph theory indices on quantifying them. Pathinathanand Ponnivalavan [16] analyzed the hazards of plastic pollution using Induced FuzzyCognitive Maps (IFCMs). IFCMs are a fuzzy-graph modeling approach based onexpert’s opinion. Amer et al. [1] developed three future scenarios using FCM for thenational wind energy sector of a developing country. Mesa-Frias et al. [13] developed anovel method based on FCM to quantify the framing assumptions in the assessment ofhealth impact. Fons et al. [5] proposed a model of an eco-industrial park and used FCMto analyze the impacts of this model in terms of pollution and waste disposal.
The motivation for this research was the development of a rational system, capableof analyzing effectively the actual conditions during incidents of high air pollution inurban areas and also to model the long term evolution of emissions due to climatechange. FCM were developed based exclusively on measurable factors resulting fromthe correlation analysis of the actual data and not on the opinion of experts as usually.The above perspectives add validity and reliability in the overall inference process.Moreover, the development and use of a fuzzy system to forecast future values of airpollutants based on climate change scenarios is an interesting innovation that signifi-cantly improves the quality and value of the proposed model. The design, developmentand testing of this system are described herein. The topography of Athens prevents thediffusion of pollutants. The pollution data come from the “Patissia” area in the center ofAthens. Relativity data analysis with FCM was performed for the air pollution mea-suring station of “Patissia” for the period 2000 to 2012, in order to obtain a symbolic
176 V.-D. Anezakis et al.
kdemertz@fmenr.duth.gr
representation of existing complex correlations between the atmospheric pollutants.The above measuring station (who is distinguished for its consistency and reliability asa few missing values were observed) is storing hourly values of CO (in mg/m3), (NO,NO2, O3 and SO2 in lg/m
3). During a day the station of Patissia is full of traffic and thisis the reason of high values of air pollution concentrations. Additionally, recordsrelated to six meteorological factors namely: air temperature (Temp), relative humidityRH), air pressure(PR), solar radiation (SR), wind speed (WS) and wind direction(WD) were obtained from the station of “Thiseion” 9 km far from the sea.
During data pre-processing all the records with missing values for one or moreparameters were removed from the dataset. Outliers are very important as they arealways considered for the activation of the civil protection mechanisms. For this rea-son, they were not removed from the datasets in order to obtain representative trainingsamples offering potential generalization in future forecasting models. Finally in orderto tackle the problem of features with different range, in which the higher values mostaffect the cost function with respect to the characteristics of the smaller ones, withoutbeing more important, a normalization process was performed in the interval [−1, +1].
2 Theoretical Framework and Methodology
2.1 Correlation Analysis
In order to test the level of linear relationship between meteorological parameters andair pollutants, the typical relativity analysis was performed, using the parametric cor-relation coefficient of Pearson (r). The Pearson linear correlation coefficient betweentwo parameters X and Y is defined based on a sample of n pairs of observationsxi; yið Þ i ¼ 1; 2; . . .; n; and it is denoted as r X; Yð Þ or more briefly as r. The variables�x and �y are the averages of xi, yið Þ. The r is the covariance (CovX,Y) of the twovariables divided by the product of their standard deviations (sx,sy). It is given by thefollowing Eq. 1:
r ¼ sxysxsy
¼Pv
i¼1 xi � �xð Þ yi � �yð ÞffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiPvi¼1 xi � �xð Þ2
q ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiPvi¼1 yi � �yð Þ2
q ¼Pv
i¼1 xiyi � v�x�yffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiPvi¼1 x
2i � v�x2
p ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiPvi¼1 y
2i � v�y2
pð1Þ
The correlation coefficient is a pure number in the interval [−1, 1]. More specifi-cally, when 0\r� 1, then X, Y are linearly positively correlated and when �1\r\0,then X, Y are negatively correlated. When r = 0 or close to zero there is no correlationbetween them.
2.2 Fuzzy Cognitive Maps
FCM are fuzzy-graph structures. In the model of a fuzzy cognitive map, the nodes arelinked together by edges and each edge connecting two nodes describes the change inthe activation value. The direction of the edge implies which node affects the other.
Fuzzy Cognitive Maps for Long-Term Prognosis 177
kdemertz@fmenr.duth.gr
The sign of the causality relationship is positive if there is a direct influence, negative ifthere is an inverse influence relation and zero if the two nodes are uncorrelated. Thecausal relationships are described by the use of fuzzy linguistics and they are fuzzifiedby using membership functions taking values in the closed interval [−1, 1] [14, 17, 18].
Unlike the majority of complex dynamic systems, characterized by nonlinearity andhigh uncertainty, the fuzzy cognitive maps use advanced learning techniques in order tochoose appropriate weights for the causal connections between the examined variables.This is done in order to reflect the examined problem with absolute realism.
Combining the theoretical background of fuzzy logic, FCM cover the comparisonand characterization purpose of the reference sets, towards modeling and solvingcomplex problems for which there is no structured mathematical model.
2.3 GFDL_CM 2.0
Flato et al. (2013) [21] finds robust relationship between the ability of theGFDL_CM2.0 model to represent interannual variability of near-surface air tempera-ture and the amplitude of future warming. Thus, the GFDL_CM2.0 climate changemodel has been chosen as the most suitable for this research area. Also this modelgenerally provides a smooth gradual temperature rise up to 2.9° C plus reduced rainfallof −245.96 mm up to 2099. The GFDL-CM2.0 is a coupled Atmospheric - Oceangeneral circulation model, developed at NOAA’s Geophysical Fluid Dynamics Labo-ratory (GFDL). It is divided into four modules: the atmosphere model (AM2P13), theocean (OM3P4), the dry (LM2) and the ice cover (SIS). The horizontal resolution ofthe atmospheric and land model is 2° latitude � 2.5° longitude whereas for the oceanicit is 1° � 1°. The atmospheric model has twenty-four vertical planes whereas the landmodel has eighteen for the heat storage and the oceanic fifty. The atmospheric modeluses a B-grid dynamical core, a k-profile planetary boundary layer scheme [11] and asimple local parameterization of the vertical momentum transport by cumulus con-vection. The land-cover-type distribution is a combination of a potential natural veg-etation of type one and a historical land use distribution dataset. GFDL-CM2.0 usesexplicit fresh water fluxes to simulate the exchange of water across the air-sea interface,rather than virtual salt fluxes [7]. Subgrid-scale parameterizations of ocean modelOM3.0 include K-profile parameterization (KPP) vertical mixing, neutral physics [8, 9]and a spatially dependent anisotropic viscosity [10]. Air-sea fluxes are computed on theocean model time step, which is 1 h in OM3.0. SIS (sea ice model) is a dynamicalmodel [19] with three-vertical layer thermodynamics (two ice, one snow), and ascheme for prognosing five different ice thickness categories and open water at eachgrid point. GFDL-CM2.0 model make use of the Flexible Modeling System(FMS) coupler for calculating and passing fluxes between its atmosphere, land andocean components.
178 V.-D. Anezakis et al.
kdemertz@fmenr.duth.gr
3 Description of the Proposed Model
The air pollution evolution modeling system comprises of the following four distinctalgorithmic stages: Modeling, Grid, Scenarios and Forecasting. In the first stage all ofthe associated parameters are added and named and then they are interconnected bysynapses to create the causal positive or negative correlations. The fuzzification of thecorrelations, i.e. the description of each interface in verbal common terms wasaccomplished by selecting six Linguistics namely: Three positive scales (low positive(+), middle positive (++), high positive (+++)). Three negative scales (low negative (–),middle negative (−−), high negative (−−−)) corresponding to fuzzy weights (Table 1)(Fig. 1).
Table 1. Effect and value of six linguistics which corresponding to fuzzy weights
Effect Value
high positive (+++) 1middle positive (++) 0.5low positive (+) 0.25low negative (–) −0.25middle negative (−−) −0.5high negative (−−−) −1
Fig. 1. Flowchart of proposed model
Fuzzy Cognitive Maps for Long-Term Prognosis 179
kdemertz@fmenr.duth.gr
The description of the algorithmic steps is done in the next paragraph:
Step 1 (Modeling): Application for the calculation of the degree of correlationbetween the variables under consideration: Carbon monoxide (CO), nitrogen monoxide(NO), nitrogen dioxide (NO2), ozone (O3), sulfur dioxide (SO2), temperature air(AirTemp), humidity (RH), atmospheric pressure (PR), solar radiation (SR), sunshine(SUN), wind speed (WS), wind direction (WD) and rainfall (RF).
Step 2: Partitioning of the variables with negative correlation from the ones withpositive correlation, with the use of the assigned Linguistics over the initial crispvalues. Three successive and overlapping triangular membership functions wereemployed in order to classify the correlations to the corresponding fuzzy sets (Lin-guistics) “Low”, “Medium” and “High”. The following Table 2 presents clearly thefuzzification of the correlation results (assignment of the corresponding Linguistics).
Table 2. Fuzzification of the correlation analysis with proper Linguistics
CO NO NO2 O3 SO2 AirTemp RH PR SR SUN WS WD RF
CO 1 +++ ++ −−−
++ – + + – + −−−
+ −−−
NO +++ 1 ++ −−−
++ −− ++ + − + −− + −−−
NO2 +++ +++ 1 −−−
++ + − − + + −− + −−−
O3 −−−
−−−
−−−
1 −− ++ −−−
− + + +++ − −−−
SO2 +++ +++ ++ −−−
1 −−− + ++ − − −− + −−−
AirTemp − − ++ +++ −− 1 −−−
− +++ +++ + + −−
RH ++ +++ − −−−
+ −−− 1 + −−−
−−− −− − +++
PR + + − − +++ −−− + 1 − − − −− −−−
SR − − + + − ++ −−−
− 1 +++ ++ + −−−
SUN + + ++ ++ − ++ −−−
− +++ 1 ++ + −−−
WS −−−
−−−
−− +++ −− + −− − ++ ++ 1 − +
WD ++ ++ +++ −− + + − −−−
++ ++ −− 1 +
RF −−−
−−−
−−−
−−−
−−−
−− +++ −−−
−−−
−−− + + 1
180 V.-D. Anezakis et al.
kdemertz@fmenr.duth.gr
Step 3 (Grid): It involves the design of the FCM following the input and the inter-connection of all correlated variables, based on the Linguistics that emerged after thefuzzyfication of the crisp numerical values.
The algorithm simulating the interactions between two nodes of the FCM wasimplemented by performing a repetitive calculation of the new link value corre-sponding to each node. This value depends on the weight of the node from which anedge begins and also on the weight of the edge joining the two nodes. The transferfunction estimates the new value of each node and the weight of each connection. Thenegative type of influence is depicted with an orange color and the positive with a bluecolor. The degree of influence depends on the thickness of each line. The higher theinfluence the thicker the line, as you can see in the Fig. 2 (Table 3).
The degree of influence between some variables depicted in the Fig. 2.
3.1 Scenarios and Forecasting
In the third stage, the changes in the values of temperature and precipitation due toclimate change (CC) for the period 2020 till 2099 are fuzzified in order to obtain thecorresponding linguistics. The whole process is based on the GFDL_CM2.0 CC model.This phase also includes extended testing of various scenarios.
Step 4 (Scenarios): Partitioning of the scenarios variables, based on the changes intemperature and precipitation, according to the global climate model GFDL_CM2.0.
Fig. 2. A FCM between air temperature, rainfall and air pollutants. (Color figure online)
Table 3. The degree of influence
CO NO NO2 O3 SO2 AirTemp RF
AirTemp − − ++ +++ −− 1 −−
RF −−−
−−−
−−−
−−−
−−−
−− 1
Fuzzy Cognitive Maps for Long-Term Prognosis 181
kdemertz@fmenr.duth.gr
Finally, the obtained crisp numerical values are fuzzified, with the use of four triangularfuzzy membership functions (FMF) and eight semi-triangular fuzzy membershipfunctions (S-FMF). Two of FMF and four of S-FMF are related to temperature changesin the closed interval [0, + 2.86] °C, which is the highest temperature increase for thearea under study in the specified time interval. The first S-FMF, the FMF and the nextS-FMF refer to the interval [0, 1.43 (median)] °C. These FMF correspond to the fuzzysets: low negative (–), middle negative (−−), high negative (−−−), whereas the lin-guistic high negative (−−−) contains values close to the smallest estimated change. Thenext S-FMF, the FMF and the last S-FMF refer to the interval [1.43 (median), 2.86(the highest fluctuation based on the model)] °C. These FMF correspond to the lowpositive (+), middle positive (++), high positive (+++), with the high positive (+++)being close to the maximum temperature change. In the same way, two FMF and fourS-FMF were developed for the precipitation with crisp values in the interval [−108.48,−219.98] mm. Due to the fact that precipitation appears reduced in the future values,the first S-FMF, the FMF and the last S-FMF cover the precipitation reduction in theinterval [−164.23, −219.98] mm which corresponds to high precipitation reduction.The next S-FMF, the FMF and the last S-FMF were used for the smaller changes[−108.48, −164.23 (median)] mm which correspond to the fuzzy sets low positive (+),middle positive (++), high positive (+++), with high positive (+++) declaring thelowest rainfall reduction (−108,48 mm) (Table 4).
Step 5 (Scenarios): It includes extended testing of various scenarios based on thepotential changes in the temperature and precipitation and moreover its influence in theair quality of Athens. The fuzzy Linguistics produced by the use of climate changescenarios are defuzzified in order to obtain the forecast of the potential future crispvalues of the air pollutants. In this way we perform a projection in the distant future forthe problem of environmental degradation due to air pollution.
Step 6 (Forecasting): For the defuzzification the centroid function was used whichestimates the center of gravity of the fuzzy set distribution.
Table 4. FMF and S-FMF boundaries (Temperature, rainfall)
Fuzzy sets corresponding totemperature and precipitationchanges
FMF and S-FMF boundariesin the closed interval[0, + 2.86] °C
FMF and S-FMFboundaries in theclosed interval[−108.48, −219.98] mm
−−− (S-FMF) 0 0.572 −220 −197.7−− (FMF) 0.143 0.715 1.287 −214.4 −192.1 −169.8− (S-FMF) 0.858 1.43 −186.5 −164.2+ (S-FMF) 1.43 2.002 −164.2 −141.9++ (FMF) 1.573 2.145 2.717 −158.6 −136.3 −114+++ (S-FMF) 2.288 2.86 −130.7 −108.5
182 V.-D. Anezakis et al.
kdemertz@fmenr.duth.gr
x ¼Rx � l xð ÞdxRl xð Þdx ð6Þ
Step 7: The index of the magnitude of change in the pollutants concentration, iscalculated based on the amount of relative change of each parameters value.
RelativeChange ¼ FutureValue� InitialValueInitialValue
ð7Þ
4 Results and Discussion
After applying various CC scenarios (including various potential changes in temper-ature and precipitation) under the GFDL_CM2.0 model for the center of Athens, theforecasted Relative Changes (RC) in the concentration of pollutants were obtained.Finally, 36 scenarios were developed by the combination of temperature and therainfall. Table 5 presents the most important of the forecasted values.
Attempting a thorough presentation of the most important RC observed, it isobvious that a zero increase in temperature (0 °C), combined with a large reduction ofrainfall (−219.98 mm) contributes significantly to the growth of all primary pollutantswhile it does not affect the secondary pollutants such as O3 (ID1). On the other hand, avery slight decrease in rainfall (−108.48 mm) helps to increase the SO2, while helpingto reduce NO2, NO, O3 and CO (ID2). Increasing the temperature to the value of themedian (+1.43 °C) in combination with a minimum reduction of precipitation(−108.48 mm) allows the reduction of both primary and secondary pollutants (ID4),whereas absolutely the opposite (ID3) is observed when the rainfall is reduced
Table 5. Relative changes of air pollutants based of the climate change scenarios
ID AirTemp (AT) RF O3 SO2 CO NO NO2
1 high negative (+0 °C) high negative(−219.98 mm)
0 0.2 0.06 0.05 0.09
2 high negative (+0 °C) high positive(−108.48 mm)
−0.02 0.03 −0.02 −0.01 −0.33
3 low negative (+1.43 °C) high negative(−219.98 mm)
0.03 0.18 0.06 0.05 0.12
4 low negative (+1.43 °C) high positive(−108.48 mm)
−0.01 −0.06 −0.05 −0.02 −0.21
5 high positive (+2.86 °C) high negative(−219.98 mm)
0.19 0.14 0.05 0.03 0.14
6 high positive (+2.86 °C) high positive(−108.48 mm)
0.11 −0.38 −0.29 −0.28 −0.11
Fuzzy Cognitive Maps for Long-Term Prognosis 183
kdemertz@fmenr.duth.gr
significantly (−219.98 mm). A large increase in the temperature higher than the median(1.43 °C) and as high as the upper limit of the CC model (2.86 °C) combined with ahigh decrease in the millimeters of rainfall (−219.98 mm) contributes to the increase ofall pollutants, especially to the increase of O3 (ID5). Finally, a very small reduction inthe millimeters of rainfall (−108.48 mm) leads to a significant reduction of all pollu-tants concentration except for the values of O3 which rises significantly (ID6). An ATincrease from 0 °C to 2.86 regardless the rainfall reduction results to an increase of O3
and reduces the values of SO2 and NO, CO. The highest values for O3 and NO2 appearin the extreme AT scenario (ID5) whereas the lowest air pollutants values appear in theID2 (zero AT increase) and the highest values for SO2, CO, NO are observed in ID1and ID3 (stable or low increase of AT). It is important to mention that we observe thehighest values for all pollutants with the highest decrease of rainfall (−219,98 mm) inID1 and ID5 and their lowest in ID2 and ID6 scenario with the least rainfall reduction(−108, 48 mm). At Table 6 the average values of historical period of 2000–2012 usedas initial values for the estimation of forecasted values. The forecasted RelativeChanges (RC) determine the values of each pollutant. The last two decades of the 21stcentury (2080–2099) are very interesting, due to the extreme scenarios observed (ID5).According to this case the temperature will increase by 15 % (2.86 °C) which repre-sents the Linguistic High Positive change (+++) and the precipitation will be reducedby 49 % (−219.98 mm) corresponding to the Linguistic High Negative (−−−) change.This combination will cause significant increase of all primary and secondary pollu-tants (presented in the following Table 6).
A potential verification of the extreme scenario will cause a significant increase inthe concentrations of O3 and NOx which will imply the increase of the cardiovascularand respiratory diseases and will be accompanied by frequent presence of thick summerphotochemical smog. Also the high values of SO2 and CO, will increase the frequencyof smog during the winter months, while serious problems may occur on the surfaces ofmonuments and historic buildings from acid rain. Since there are no actual similarfuture projections and forecasting efforts for Athens we cannot check the accuracy ofthe obtained results. However, both the methodology employed and the producedscenarios with the forecasted evolution of air pollution till 2099 are very useful as theywill motivate researchers towards more flexible modeling attempts beyond the typicalstatistical ones.
Table 6. Air pollutants forecasted values based on the extreme scenario
Period O3 SO2 CO NO NO2
2000–2012 21.64 20.53 2.47 115.79 87.112080–2099 25.75 23.40 2.59 119.26 99.30Absolute increase value +4.11 +2.87 +0.12 +3.47 +12.19
184 V.-D. Anezakis et al.
kdemertz@fmenr.duth.gr
5 Conclusions
In this paper we have proposed the use of an innovative method for analysis andmodeling of air quality. We have also developed and applied an air pollution andforecasting system which is based on CC models and uses Soft Computing techniques.More specifically, Linguistic representation of the correlations between atmosphericpollutants is performed by employing Fuzzy Cognitive Maps. Based on the aboveapproach we have obtained relative changes of air pollutants values for the center ofAthens for the period 2000–2012. Additionally, a projection in the future is performedregarding the evolution of the pollutants concentrations, based on the fluctuation oftemperature and precipitation till 2100, as reflected by the projections of climate modelGFDL_CM2.0. The workings of the model were tested in various scenarios and pre-sented important information about the hazards of air quality and the effects of airpollution. From this point of view, though the forecasted results cannot be checked foraccuracy, the fact that this paper introduces a flexible Soft Computing approach toproduce long term projection of air quality based on scenarios of CC opens innovativehorizons to researchers introducing an alternative algorithm. In the future we will try toimprove the model by employing optimization methods (e.g. genetic algorithms,swarm intelligence) or hybrid approaches.
References
1. Amer, M., Jetter, A.J., Daim, T.U.: Scenario planning for the national wind energy sectorthrough fuzzy cognitive maps. In: Technology Management in the IT-Driven Services(PICMET) Proceedings of PICMET 2013, pp. 2153–2162 (2013)
2. Bougoudis, I., Demertzis, K., Iliadis, L.: HISYCOL a hybrid computational intelligencesystem for combined machine learning: the case of air pollution modeling in Athens. NeuralComput. Appl. 27, 1191–1206 (2015). doi:10.1007/s00521-015-1927-7. Springer
3. Bougoudis, I., Demertzis, K., Iliadis, L.: Fast and low cost prediction of extreme airpollution values with hybrid unsupervised learning. In: Integrated Computer-AidedEngineering, Vol. Preprint. NO. Preprint, pp. 1–13. IOS Press (2015). doi:10.3233/ICA-150505
4. Bougoudis, I., Iliadis, L., Papaleonidas, A.: Fuzzy inference ANN ensembles for airpollutants modeling in a major urban area: the case of Athens. Eng. Appl. Neural Netw.Commun. Comput. Inf. Sci. 459, 1–14 (2014)
5. Fons, S., Achari, G., Ross, T.: A fuzzy cognitive mapping analysis of the impacts of aneco-industrial park. J. Intell. Fuzzy Syst. 15(2), 75–88 (2004)
6. Gordaliza, J.A., Florez, R.E.V.: Using fuzzy cognitive maps to support complexenvironmental issues learning. In: Proceedings of New Perspectives in Science EducationConference, 2nd edn. (2013)
7. Griffies, S.M.: Fundamentals of Ocean Models, p. 496. Princeton University Press, Princeton(2004)
8. Griffies, S.M., Gnanadesikan, A., Pacanowski, R., Larichev, V., Dukowicz, J.K., Smith, R.D.: Isopycnal mixing in a z-coordinate ocean model. J. Phys. Oceanogr. 28, 805–830 (1998)
9. Griffies, S.M.: Gent–McWilliams skew flux. J. Phys. Oceanogr. 28, 831–841 (1998)
Fuzzy Cognitive Maps for Long-Term Prognosis 185
kdemertz@fmenr.duth.gr
10. Large, W., Danasbogulu, G., McWilliams, J., Gent, P., Bryan, F.O.: Equatorial circulation ofa global ocean climate model with anisotropic viscosity. J. Phys. Oceanogr. 31, 518–536(2001)
11. Lock, P., Brown, R., Bush, R., Martin, M., Smith, B.: A new boundary layer mixing scheme.Scheme description and single-column model tests. Mon. Weather Rev. 128, 3187–3199(2000)
12. Luiz, J., Muller, E.: Greenhouse gas emission reduction under the kyoto protocol: the SouthAfrican example. Int. Bus. Econ. Res. J. 7, 75–92 (2008)
13. Marco, F., Chalabi, Z., Foss, M.: Assessing framing assumptions in quantitative healthimpact assessments: a housing intervention example. Environ. Int. 59, 133–140 (2013)
14. Papageorgiou, E.I., Salmeron, J.L.: A review of fuzzy cognitive maps research during thelast decade. IEEE Trans. Fuzzy Syst. 21(1), 66–79 (2013)
15. Paschalidou, A.: University of Ioannina, Ph.d. thesis development of box model for the airpollution forecasting in medium size cities (2007). (in Greek)
16. Pathinathan, T., Ponnivalavan, K.: The study of hazards of plastic pollution using inducedfuzzy cognitive maps (IFCMS). J. Comput. Algorithm 3, 671–674 (2014)
17. Salmeron, J.L., Froelich, W.: Dynamic optimization of fuzzy cognitive maps for time seriesforecasting. Knowl. Based Syst. 105, 29–37 (2016). Forthcoming
18. Vidal, R., Salmeron, J.L., Mena, A., Chulvi, V.: Fuzzy cognitive map-based selection ofTRIZ trends for eco-innovation of ceramic industry products. J. Cleaner Prod. 107, 202–214(2015)
19. Winton, M.: A reformulated three-layer sea ice model. J. Atmos. Oceanic Technol. 17,525–531 (2000)
20. Zhang, H., Song, J., Su, C., He, M.: Human attitudes in environmental management: fuzzycognitive maps and policy option simulations analysis for a coal-mine ecosystem in China.J. Environ. Manag. 115, 227–234 (2013)
21. http://www.climatechange2013.org/images/report/WG1AR5_Chapter09_FINAL.pdf
186 V.-D. Anezakis et al.
kdemertz@fmenr.duth.gr
My Publications
Cyber Security informatics
1. Demertzis, K., Iliadis, L., 2018. A Computational Intelligence System Identifying Cyber-
Attacks on Smart Energy Grids, in: Daras, N.J., Rassias, T.M. (Eds.), Modern Discrete
Mathematics and Analysis: With Applications in Cryptography, Information Systems
and Modeling, Springer Optimization and Its Applications. Springer International
Publishing, Cham, pp. 97–116. https://doi.org/10.1007/978-3-319-74325-7_5
2. Demertzis, K., Iliadis, L., 2017. Computational intelligence anti-malware framework
for android OS. Vietnam J Comput Sci 4, 245–259. https://doi.org/10/gdp86x
3. Demertzis, K., Iliadis, L., 2016. Bio-inspired Hybrid Intelligent Method for Detecting
Android Malware, in: Kunifuji, S., Papadopoulos, G.A., Skulimowski, A.M.J., Kacprzyk,
J. (Eds.), Knowledge, Information and Creativity Support Systems, Advances in
Intelligent Systems and Computing. Springer International Publishing, pp. 289–304.
4. Demertzis, K., Iliadis, L., 2015. A Bio-Inspired Hybrid Artificial Intelligence Framework
for Cyber Security, in: Daras, N.J., Rassias, M.T. (Eds.), Computation, Cryptography,
and Network Security. Springer International Publishing, Cham, pp. 161–193.
https://doi.org/10.1007/978-3-319-18275-9_7
5. Demertzis, K., Iliadis, L., 2015. Evolving Smart URL Filter in a Zone-Based Policy Firewall
for Detecting Algorithmically Generated Malicious Domains, in: Gammerman, A.,
Vovk, V., Papadopoulos, H. (Eds.), Statistical Learning and Data Sciences, Lecture
Notes in Computer Science. Springer International Publishing, pp. 223–233.
6. Demertzis, K., Iliadis, L., 2015. SAME: An Intelligent Anti-malware Extension for
Android ART Virtual Machine, in: Núñez, M., Nguyen, N.T., Camacho, D., Trawiński, B.
(Eds.), Computational Collective Intelligence, Lecture Notes in Computer Science.
Springer International Publishing, pp. 235–245.
7. Demertzis, K., Iliadis, L., 2014. A Hybrid Network Anomaly and Intrusion Detection
Approach Based on Evolving Spiking Neural Network Classification, in: Sideridis, A.B.,
Kardasiadou, Z., Yialouris, C.P., Zorkadis, V. (Eds.), E-Democracy, Security, Privacy and
Trust in a Digital World, Communications in Computer and Information Science.
Springer International Publishing, pp. 11–23.
8. Demertzis, K., Iliadis, L., 2014. Evolving Computational Intelligence System for
Malware Detection, in: Iliadis, L., Papazoglou, M., Pohl, K. (Eds.), Advanced
Information Systems Engineering Workshops, Lecture Notes in Business Information
Processing. Springer International Publishing, pp. 322–334.
9. Demertzis, K., Iliadis, L., Anezakis, V., 2018. MOLESTRA: A Multi-Task Learning
Approach for Real-Time Big Data Analytics, in: 2018 Innovations in Intelligent Systems
and Applications (INISTA). Presented at the 2018 Innovations in Intelligent Systems
and Applications (INISTA), pp. 1–8. https://doi.org/10.1109/INISTA.2018.8466306
10. Demertzis, Konstantinos, Iliadis, L., Anezakis, V.-D., 2018. A Dynamic Ensemble
Learning Framework for Data Stream Analysis and Real-Time Threat Detection, in:
Kůrková, V., Manolopoulos, Y., Hammer, B., Iliadis, L., Maglogiannis, I. (Eds.), Artificial
Neural Networks and Machine Learning – ICANN 2018, Lecture Notes in Computer
Science. Springer International Publishing, pp. 669–681.
11. Demertzis, Konstantinos, Iliadis, L., Spartalis, S., 2017. A Spiking One-Class Anomaly
Detection Framework for Cyber-Security on Industrial Control Systems, in: Boracchi,
G., Iliadis, L., Jayne, C., Likas, A. (Eds.), Engineering Applications of Neural Networks,
Communications in Computer and Information Science. Springer International
Publishing, pp. 122–134.
12. Demertzis, Konstantinos, Iliadis, L.S., Anezakis, V.-D., 2018. An innovative soft
computing system for smart energy grids cybersecurity. Advances in Building Energy
Research 12, 3–24. https://doi.org/10/gdp862
13. Demertzis, Konstantinos, Kikiras, P., Tziritas, N., Sanchez, S.L., Iliadis, L., 2018. The
Next Generation Cognitive Security Operations Center: Network Flow Forensics Using
Cybersecurity Intelligence. Big Data and Cognitive Computing 2, 35.
https://doi.org/10/gfkhpp
14. Rantos, K., Drosatos, G., Demertzis, K., Ilioudis, C., Papanikolaou, A., 2018. Blockchain-
based Consents Management for Personal Data Processing in the IoT Ecosystem.
Presented at the International Conference on Security and Cryptography, pp. 572–
577.
15. Demertzis, Konstantinos, Iliadis, L.S., 2018. Real-time Computational Intelligence
Protection Framework Against Advanced Persistent Threats. Book entitled "Cyber-
Security and Information Warfare", Series: Cybercrime and Cybersecurity Research,
NOVA science publishers, ISBN: 978-1-53614-385-0, Chapter 5.
16. Demertzis, Konstantinos, Iliadis, L.S., 2016. Ladon: A Cyber Threat Bio-Inspired
Intelligence Management System. Journal of Applied Mathematics & Bioinformatics,
vol.6, no.3, 2016, 45-64, ISSN: 1792-6602 (print), 1792-6939 (online), Scienpress Ltd,
2016.
17. Demertzis, K.; Tziritas, N.; Kikiras, P.; Sanchez, S.L.; Iliadis, L. The Next Generation
Cognitive Security Operations Center: Adaptive Analytic Lambda Architecture for
Efficient Defense against Adversarial Attacks. Big Data Cogn. Comput. 2019, 3, 6.
18. Rantos K., Drosatos G., Demertzis K., Ilioudis C., Papanikolaou A., Kritsas A. (2019)
ADvoCATE: A Consent Management Platform for Personal Data Processing in the IoT
Using Blockchain Technology. In: Lanet JL., Toma C. (eds) Innovative Security Solutions
for Information Technology and Communications. SECITC 2018. Lecture Notes in
Computer Science, vol 11359. Springer, Cham.
19. Demertzis, K.; Iliadis, L.. Cognitive Web Application Firewall to Critical Infrastructures
Protection from Phishing Attacks, Journal of Computations & Modelling, vol.9, no.2,
2019, 1-26, ISSN: 1792-7625 (print), 1792-8850 (online), Scienpress Ltd, 2019.
20. Demertzis K., Iliadis L., Kikiras P., Tziritas N. (2019) Cyber-Typhon: An Online Multi-
task Anomaly Detection Framework. In: MacIntyre J., Maglogiannis I., Iliadis L.,
Pimenidis E. (eds) Artificial Intelligence Applications and Innovations. AIAI 2019. IFIP
Advances in Information and Communication Technology, vol 559. Springer, Cham
21. Xing, L., Demertzis, K. & Yang, J. Neural Comput & Applic (2019).
https://doi.org/10.1007/s00521-019-04288-5.
Environmental informatics
22. Anezakis, V., Mallinis, G., Iliadis, L., Demertzis, K., 2018. Soft computing forecasting of
cardiovascular and respiratory incidents based on climate change scenarios, in: 2018
IEEE Conference on Evolving and Adaptive Intelligent Systems (EAIS). Presented at the
2018 IEEE Conference on Evolving and Adaptive Intelligent Systems (EAIS), pp. 1–8.
https://doi.org/10.1109/EAIS.2018.8397174
23. Anezakis, V.-D., Demertzis, K., Iliadis, L., 2018. Classifying with fuzzy chi-square test:
The case of invasive species. AIP Conference Proceedings 1978, 290003.
https://doi.org/10/gdtm5q
24. Anezakis, V.-D., Demertzis, K., Iliadis, L., Spartalis, S., 2018. Hybrid intelligent modeling
of wild fires risk. Evolving Systems 9, 267–283. https://doi.org/10/gdp863
25. Anezakis, V.-D., Demertzis, K., Iliadis, L., Spartalis, S., 2016. A Hybrid Soft Computing
Approach Producing Robust Forest Fire Risk Indices, in: Iliadis, L., Maglogiannis, I.
(Eds.), Artificial Intelligence Applications and Innovations, IFIP Advances in
Information and Communication Technology. Springer International Publishing, pp.
191–203.
26. Anezakis, V.-D., Dermetzis, K., Iliadis, L., Spartalis, S., 2016. Fuzzy Cognitive Maps for
Long-Term Prognosis of the Evolution of Atmospheric Pollution, Based on Climate
Change Scenarios: The Case of Athens, in: Nguyen, N.-T., Iliadis, L., Manolopoulos, Y.,
Trawiński, B. (Eds.), Computational Collective Intelligence, Lecture Notes in Computer
Science. Springer International Publishing, pp. 175–186.
27. Anezakis, V.-D., Iliadis, L., Demertzis, K., Mallinis, G., 2017. Hybrid Soft Computing
Analytics of Cardiorespiratory Morbidity and Mortality Risk Due to Air Pollution, in:
Dokas, I.M., Bellamine-Ben Saoud, N., Dugdale, J., Díaz, P. (Eds.), Information Systems
for Crisis Response and Management in Mediterranean Countries, Lecture Notes in
Business Information Processing. Springer International Publishing, pp. 87–105.
28. Bougoudis, I., Demertzis, K., Iliadis, L., 2016. Fast and low cost prediction of extreme
air pollution values with hybrid unsupervised learning. Integrated Computer-Aided
Engineering 23, 115–127. https://doi.org/10/f8dt4t
29. Bougoudis, I., Demertzis, K., Iliadis, L., 2016. HISYCOL a hybrid computational
intelligence system for combined machine learning: the case of air pollution modeling
in Athens. Neural Comput & Applic 27, 1191–1206. https://doi.org/10/f8r7vf
30. Bougoudis, I., Demertzis, K., Iliadis, L., Anezakis, V.-D., Papaleonidas, A., 2018.
FuSSFFra, a fuzzy semi-supervised forecasting framework: the case of the air pollution
in Athens. Neural Comput & Applic 29, 375–388. https://doi.org/10/gc9bbf
31. Bougoudis, I., Demertzis, K., Iliadis, L., Anezakis, V.-D., Papaleonidas, A., 2016. Semi-
supervised Hybrid Modeling of Atmospheric Pollution in Urban Centers, in: Jayne, C.,
Iliadis, L. (Eds.), Engineering Applications of Neural Networks, Communications in
Computer and Information Science. Springer International Publishing, pp. 51–63.
32. Demertzis, Konstantinos, Anezakis, V.-D., Iliadis, L., Spartalis, S., 2018. Temporal
Modeling of Invasive Species’ Migration in Greece from Neighboring Countries Using
Fuzzy Cognitive Maps, in: Iliadis, L., Maglogiannis, I., Plagianakos, V. (Eds.), Artificial
Intelligence Applications and Innovations, IFIP Advances in Information and
Communication Technology. Springer International Publishing, pp. 592–605.
33. Demertzis, K., Iliadis, L., 2018. The Impact of Climate Change on Biodiversity: The
Ecological Consequences of Invasive Species in Greece, in: Leal Filho, W., Manolas, E.,
Azul, A.M., Azeiteiro, U.M., McGhie, H. (Eds.), Handbook of Climate Change
Communication: Vol. 1: Theory of Climate Change Communication, Climate Change
Management. Springer International Publishing, Cham, pp. 15–38.
https://doi.org/10.1007/978-3-319-69838-0_2
34. Demertzis, K., Iliadis, L., 2017. Adaptive Elitist Differential Evolution Extreme Learning
Machines on Big Data: Intelligent Recognition of Invasive Species, in: Angelov, P.,
Manolopoulos, Y., Iliadis, L., Roy, A., Vellasco, M. (Eds.), Advances in Big Data,
Advances in Intelligent Systems and Computing. Springer International Publishing, pp.
333–345.
35. Demertzis, K., Iliadis, L., 2015. Intelligent Bio-Inspired Detection of Food Borne
Pathogen by DNA Barcodes: The Case of Invasive Fish Species Lagocephalus
Sceleratus, in: Iliadis, L., Jayne, C. (Eds.), Engineering Applications of Neural Networks,
Communications in Computer and Information Science. Springer International
Publishing, pp. 89–99.
36. Demertzis, K., Iliadis, L., Anezakis, V., 2017. A deep spiking machine-hearing system
for the case of invasive fish species, in: 2017 IEEE International Conference on
INnovations in Intelligent SysTems and Applications (INISTA). Presented at the 2017
IEEE International Conference on INnovations in Intelligent SysTems and Applications
(INISTA), pp. 23–28. https://doi.org/10.1109/INISTA.2017.8001126
37. Demertzis, Konstantinos, Iliadis, L., Anezakis, V.-D., 2017. Commentary: Aedes
albopictus and Aedes japonicus—two invasive mosquito species with different
temperature niches in Europe. Front. Environ. Sci. 5. https://doi.org/10/gdp865
38. Demertzis, K., Iliadis, L., Avramidis, S., El-Kassaby, Y.A., 2017. Machine learning use in
predicting interior spruce wood density utilizing progeny test information. Neural
Comput & Applic 28, 505–519. https://doi.org/10/gdp86z
39. Demertzis, Konstantinos, Iliadis, L.S., Anezakis, V.-D., 2018. Extreme deep learning in
biosecurity: the case of machine hearing for marine species identification. Journal of
Information and Telecommunication 2, 492–510. https://doi.org/10/gdwszn
40. Dimou, V., Anezakis, V.-D., Demertzis, K., Iliadis, L., 2018. Comparative analysis of
exhaust emissions caused by chainsaws with soft computing and statistical
approaches. Int. J. Environ. Sci. Technol. 15, 1597–1608. https://doi.org/10/gdp864
41. Iliadis, L., Anezakis, V.-D., Demertzis, K., Mallinis, G., 2017. Hybrid Unsupervised
Modeling of Air Pollution Impact to Cardiovascular and Respiratory Diseases.
IJISCRAM 9, 13–35. https://doi.org/10/gfkhpm
42. Iliadis, L., Anezakis, V.-D., Demertzis, K., Spartalis, S., 2018. Hybrid Soft Computing for
Atmospheric Pollution-Climate Change Data Mining, in: Thanh Nguyen, N., Kowalczyk,
R. (Eds.), Transactions on Computational Collective Intelligence XXX, Lecture Notes in
Computer Science. Springer International Publishing, Cham, pp. 152–177.
https://doi.org/10.1007/978-3-319-99810-7_8
43. Demertzis, K., Iliadis, L., 2017. Detecting invasive species with a bio-inspired semi-
supervised neurocomputing approach: the case of Lagocephalus sceleratus. Neural
Comput & Applic 28, 1225–1234. https://doi.org/10/gbkgb7
44. Κωνσταντίνος Δεμερτζής, Λάζαρος Ηλιάδης, 2015, Γενετική Ταυτοποίηση
Χωροκατακτητικών Ειδών με Εξελιγμένες Μεθόδους Τεχνητής Νοημοσύνης: Η
Περίπτωση του Ασιατικού Κουνουπιού Τίγρης (Aedes Αlbopictus). Θέματα
Δασολογίας & Διαχείρισης Περιβάλλοντος & Φυσικών Πόρων, 7ος τόμος, Κλιματική
Αλλαγή: Διεπιστημονικές Προσεγγίσεις, ISSN: 1791-7824, ISBN: 978-960-9698-11-5,
Eκδοτικός Oίκος: Δημοκρίτειο Πανεπιστήμιο Θράκης
45. Βαρδής-Δημήτριος Ανεζάκης, Κωνσταντίνος Δεμερτζής, Λάζαρος Ηλιάδης. Πρόβλεψη
Χαλαζοπτώσεων Μέσω Μηχανικής Μάθησης. 3o Πανελλήνιο Συνέδριο Πολιτικής
Προστασίας «SafeEvros 2016: Οι νέες τεχνολογίες στην υπηρεσία της Πολιτικής
Προστασίας», Proceedings, ISBN : 978-960-89345-7-3, Ιούνιος 2017, Eκδοτικός Oίκος:
∆ημοκρίτειο Πανεπιστήμιο Θράκης.
46. Demertzis K., Iliadis L., Anezakis VD. (2019) A Machine Hearing Framework for Real-
Time Streaming Analytics Using Lambda Architecture. In: Macintyre J., Iliadis L.,
Maglogiannis I., Jayne C. (eds) Engineering Applications of Neural Networks. EANN
2019. Communications in Computer and Information Science, vol 1000. Springer,
Cham
Other
47. Κωνσταντίνος Δεμερτζής. Ενίσχυση της Διοικητικής Ικανότητας των Δήμων Μέσω της
Ηλεκτρονικής Διακυβέρνησης: Η Στρατηγική των «Έξυπνων Πόλεων» με Σκοπό την
Αειφόρο Ανάπτυξη. Θέματα Δασολογίας και Διαχείρισης Περιβάλλοντος και
Φυσικών Πόρων, 10ος Τόμος: Περιβαλλοντική Πολιτική: Καλές Πρακτικές,
Προβλήματα και Προοπτικές, σελ. 84 - 100, ISSN: 1791-7824, ISBN: 978-960-9698-14-
6, Νοέμβριος 2018, Eκδοτικός Oίκος: Δημοκρίτειο Πανεπιστήμιο Θράκης.
Recommended