Mining the change of event trends for decision support in environmental scanning

Extraction of Unexpected Rules from Twitter Hashtags and its Application to Sport Events

“…ARM is a data mining technique used for extracting remarkable associations between different arrays of items in transactional databases (Market Basket), relational databases (Personal Details), or any other information warehouses [12] in the form of rules. It analyses the frequent if/then (antecedent/consequent) patterns by using the support and confidence measures to discover most significant relationships [5].…”

“…We employ the method of Liu et al [12] and Song et al [20] for calculating similarities and differences between two rules at 2 different period. Given that the methods implemented in [12] and [20] were aimed for ARs detected from relational datasets (general association rules). Our method defines the similarity based on the principles of degree of similarity proposed in [12] and [20].…”

“…Given that the methods implemented in [12] and [20] were aimed for ARs detected from relational datasets (general association rules). Our method defines the similarity based on the principles of degree of similarity proposed in [12] and [20]. Details of the calculations and notation used are stated as follows:…”

“…Equations 1 and 2 were adopted from Liu et al [12]. They demonstrate the degree of similarity in the conditional p ij and consequent q ij parts of rule i and rule j at different time t and t + 1 in that order.…”

SoMEST

“…Event extraction and tracking techniques have been used, for example, to support environmental scanning [10]. Furthermore, ETA has been used in computer software that is to use timeline-based guidance of the analysis progress [7,8].…”