With the advent of the digital age and its entry into the business environment, marketing trends have undergone several changes, such as the expansion of data-driven marketing and the digitalization of the business environment that requires further investigation. As a result, the aim of this study is to present a novel approach in marketing research and identify value components among the large volume of customer feedback in virtual networks using machine learning, big data analysis, and a predictive marketing strategy. This research is applied in terms of purpose, qualitative-quantitative (mixed) in terms of method and descriptive-survey in terms of data collection that has used an inductive approach. For this purpose, the Iranian tourism industry and tourist areas of Tehran province was selected as a case study and 9325 comments from customers about hotels used in their travels to Tehran were collected from the Internet between the summer of 2020 to winter 2021 and using "data clustering" and "Association Rules Extraction" methods, the value components were extracted and RapidMiner software and Python programming language were used to perform data mining, text mining and big data analysis processes. In summary, the findings demonstrate that by employing big data analysis and machine learning, the process of "marketing research" can be performed with greater speed, accuracy, and extensiveness, as well as at a lower relative cost. The findings and suggestions of this study are used to inform marketing researchers, as well as to raise the awareness of managers, especially in hotel industry.