The focus of this research is on assessing the public perception of service quality with a multi-aspect sentiment analysis technique and visualizing the assessment results by means of a customer-oriented web application. Both the methodological and development issues are covered. The methodology includes two main steps: 1) creating digitalized linguistic resources and 2) developing a web application based on the pipelined precreated resources. The first step includes raw data acquisition and data specificity analysis, preprocessing to eliminate public induced linguistic mistakes, as well as creation of a gold standard annotated corpus based on a domain- and task-oriented lexical and ontological knowledge base and disambiguation rules. The second step describes specific methods for the automatic calculation of sentiment scores for multiple service aspects prioritized by general public. The calculations are performed by means of a web application that works on the pipelined gold standard annotated corpus and specific knowledge for template-based natural language generation (NLG). The methodology is tested as applied to measure public satisfaction with medical services in the city of Chelyabinsk (the capital of the South Ural). Practical results of the methodology application are presented. Both the methodogical issues and practical results of the current work are to be used for further research and may also deserve the attention of both computational linguists and sociologists.