In the evolving landscape of retail, understanding customer perceptions toward data mining techniques is vital for impacting the purchase decisions. This analytical study discovers how various data mining methods, including Association Rule Mining (ARM), RFM Technique, Customer Segmentation, Market Basket Analysis (MBA), and Time Series Analysis (TSA), impacted the consumer behaviour in retail environments. After the inspecting these techniques, the research goal to discover patterns in customer preferences, such as recognizing frequently co-purchased items through ARM and MBA, classifying buyers via Customer Segmentation, evaluating loyalty with RFM, and predicting periodic trends using TSA. Data was collected from 100 retail customers through a Likert rating scale questionnaire, capturing responses on a 5-point Likert scale for perceptions of data mining techniques. The Structural Equation Modelling (SEM) was executed using AMOS software. The model fit was examine using indices such as RMSEA (0.06), CFI (0.94), and TLI (0.93), confirmatory a robust structure after minor modifications. The findings reveal significant positive influences of data mining techniques on customer perceptions and decisions.