Abstract: A method of manufacturing an expanded granular material comprises: forming a mixture comprising a silicate material, an alkali compound and water; curing the mixture to form a solid precursor; crushing and/or milling the solid precursor to form an expandable granular material; and heating the granular material to form an expanded granular material.

Abstract: The present disclosure describes a system, method, and computer program for automatically predicting the emotion(s) to which a user is most likely to respond based on user characteristics and for tailoring a marketing message to a user based on the predicted emotion(s). A statistical model is created that predicts how a user with certain characteristics will respond to certain emotions. User characteristics are the input to the model, and, for each of a set of emotions, the output is the probability of a corresponding user responding to a message with the emotion. The statistical model is used to generate promotional messages that are tailored to each user based on the emotion to which the user is predicted to respond best, given the user's characteristics.

Abstract: The disclosure is directed to a system, method, and computer program product for detecting and responding to changes in the performance of an ad. In a first phase, an expected performance value for an ad is calculated. In a second phase, the performance of the ad is monitored in intervals, wherein for each interval, the response rate of the ad during the interval is calculated and compared to the expected performance value, an upper performance limit, and a lower performance limit. The ad is replaced or the expected performance value is recalculated, depending on how the response rate compares to the expected performance value, the upper performance limit, and the lower performance limit.