Analyzing the opinions and emotions of Internet customers using deep ensemble learning based on rbm

Background: the emotions and opinions of internet users are critical, as they directly influence the provision of proper services. the aim of this study was analyzing the opinions and emotions of internet customers using deep ensemble learning based on rbm. methods: method of this study was based on the deep ensemble learning technique which uses a deep ensemble neural network based on gaussian restricted boltzmann machine and cost- sensitive tree the opinions and emotions of internet customers were analyzed in terms of semantics and linguistics in virtual shops. to analyze behavior or emotions, existing algorithms were divided into groups of semantic approach, language approach and machine learning. the semantic, linguistic and group learning aspects (machine learning) were considered together. the opinions, feelings, and behaviors of internet customers were analyzed. the proposed method was implemented in matlab software. to evaluate this method, conventional criteria that were /applied in data mining applications have been used including accuracy, recall, and f score. results: based on the experiments performed and by evaluating this method against individual and ensemble methods plus the approaches presented in data mining so far, it was revealed that the proposed model outperforms other methods regarding data mining assessment criteria. conclusion: based on social engineering, the suggested model is provided to forecast consumer behavior. in addition to analyzing customers' behavior which examined their emotions and feelings based on their opinions. the results of this study can be used by planners in the field of competitive internet markets.