Automatic linguistic reporting of customer activity patterns in open malls

Abstract In this work, we present a complete system to produce an automatic linguistic reporting about the customer activity patterns inside open malls, mixed distribution of classical malls joined with shops on street. These reports can assist design marketing campaigns by means identifying best places catch attention customers. Activity are estimated process mining techniques and key information localization. Localization is obtained parallelized solution based WiFi fingerprint speed up solution. agreement practices for human evaluation natural language generation systems, quality generated report was evaluated 41 experts who filled in online questionnaire. Results encouraging, since average global score dimension 6.17 (0.76 standard deviation) 7-point Likert scale. This expresses high degree satisfaction validates adequacy textual as complementary tool model visualization.