In the data age, businesses are revolutionizing into smart businesses by storing a significant amount of their customer’s data and revealing customer behavioral patterns by systematically analyzing and scrutinizing those data. To secure maximum profit and user satisfaction, the need for mining relevant and accurate patterns is increasing rapidly. The Association Rule Mining or ARM technique has been used as a key method to distinguish association in the customer’s purchase pattern. From ARM, we receive Association Rules that describe the shrouded relationship between items in a transaction database. Over time, several ARM algorithms have been recommended like Apriori, FP-Growth, EFP-Growth, ECLAT, FUFP, FIUFP etc., but the demand for relevant rule mining was not met by them. Apriori, FP-Growth, EFP-Growth, and ECLAT algorithms were concentrated on mitigating the time and space complexity, while the FUFP and FIUFP algorithms worked on optimizing the execution time in terms of incremental transaction database. In this paper, we have proposed a new ARM technique, named Dominant-Tree or D-Tree, which focuses on amplifying the relevance of the mined Association Rules to meet the demand of authentic patterns. D-Tree introduces a unique method for handling transactions, called InitSet, along with an innovative approach that uses a Dominant Node to construct the dominant tree, and hence the name. It also leverages the Non-Repeated Values or NRV to prioritize node insertion within the tree. The performance of our proposed model has been assessed on a trust-evolution supported experimental environment along with the other two models, namely FP-Growth and EFP-Growth by applying a benchmark dataset. The experimental results demonstrate that the proposed D-Tree technique significantly outperforms its counterparts.