Seite - 332 - in Intelligent Environments 2019 - Workshop Proceedings of the 15th International Conference on Intelligent Environments

wasa tiebetween the frequencyof twocompetingemotions recorded; this is thecaseas shown infigure4(A) (B4,C5, andD1), and inFigure4(B)C3. Figure 3. Emotional frequencies in the geographical areawhere the experimentwas carried out. Text pro- cessed by IBM. Rowswith zeros have been omitted(A), and Classification of tweet within their respective quadrant(B). Table2. Emotional frequencies in thegeographicalareawhere theexperimentwascarriedout.Textevaluated by theuser.Rowswithzeroshavebeenomitted. Quadrant Anger Disgust Fear Joy Sadness total A2 0 0 0 1 0 1 B1 0 0 1 0 0 1 B3 0 0 0 1 0 1 B4 1 0 0 1 1 3 C2 0 0 2 2 11 15 C3 1 0 0 4 1 6 C4 0 0 1 0 0 1 C5 0 0 0 1 1 2 D1 0 0 1 1 1 3 D3 2 0 0 3 1 6 D4 2 1 3 10 3 19 D5 3 2 6 17 10 38 E4 0 0 0 2 0 2 Figure5(A), showsdata from0:00 to11:59,quadrantA2andB4,whichareassoci- atedwitharesidentialarea.HereweseeahigherfrequencyforJoy;onthecentralzoneof thecity,cellC3.Here thepredominantemotionwassadness,and thecoordinatesD4and D5,which covers the institute also showsadness and joy respectively. For Figure 5(B), showingdata from12:00to24:59,quadrantB1andB3showsadnessas thepredominant emotion. InC2,which is another residential zone, it showspredominance in Joy,unlike C4, andC5,which show anger and sadness respectively. ZoneD1, on the outskirts of the city, shows sadness as thepredominant feelingand in the same row,D3,D4andD5 J.J.ManriquezSantosetal. /ACrowdsourcingBasedComputationalMethodology332