Evaluation of the possibility of predicting social reactions using in-depth analysis of information published on web-portals

Abstrakt

Based on a literature analysis of the applicability of econometric and analytical models in the forecasting of social responses it was concluded that a dominating role is played by conditions and rules deriving from statistical physics. These models examine the straight behaviour and reactions of group members in a specific situation. There are no models where the behaviour of participants is related to phenomena occurring in their environment (closer and further) and dynamically changing states of this environment (in particular the information environment). A model for the study of the web information impact on social responses has been proposed to complement the identified research gap. The model has been evaluated on the basis of investment decisions made at the Warsaw Stock Exchange. The conducted analyzes show the usefulness of the model and the possibility of its further development in the wider context of social applications.