Abstract：Constructing a sentiment analysis model for LibQUALcomments PurposeThe purpose of this paper is to establish a data mining model for performing sentiment analysis on open-ended qualitative LibQUALcommentsproviding a further method for year-to-year comparison of user satisfactionboth of the library as a whole and individual topicsDesignmethodologyapproachA training set of 514 commentsselected at random from five LibQUALsurvey responseswas manually reviewed and labeled as having a positive or negative sentimentUsing the open-source RapidMiner data mining platformthose comments provided the framework for creating library-specific positive and negative word vectors to power the sentiment analysis modelA further process was created to help isolate individual topics within the larger commentsallowing for more nuanced sentiment analysisFindingsApplied to LibQUALcomments for a Canadian mid-sized academic research librarythe model suggested a fairly even distribution of positive and negative sentiment in overall commentsWhen filtering comments into affect of serviceinformation control and library as placethe three dimensionsrelative polarity mirrored the results of the quantitative LibQUALquestionswith highest scores for affect of service and lowest for library as placePractical implicationsThe sentiment analysis model provides a complementary tool to the LibQUALquantitative resultsallowing for simpletime-efficientyear-to-year analysis of open-ended commentsFurthermorethe process provides the means to isolate specific topics based on specified keywordsallowing individual institutions to tailor results for more in-depth analysisOriginalityvalueTo best account for library-specific terminology and phrasingthe sentiment model was created using LibQUALopen-ended comments as the foundation for the sentiment models classification processThe process also allows individual topicschosen to meet individual library needsto be isolated and independently analyzedproviding more precise examinationConsulting ServiceThrough emotional analysisthe library can better understand the needs and feedback of usersthereby providing better consulting servicesRetrieval and IndexingThrough emotional analysis and data explorationlibraries can more effectively extract and analyze the emotions in user comments to better retrieve and indexThis article explores how the modern academic library is facing changes with the advancement of technologyWhen the image of the traditional library has passedthe needs of modern students have changedand advanced digital services and global books are requiredDespite such violent changesmodern academic libraries are still prosperousAmong themthe Libqualsurvey is a major feedback mechanismbut each survey will generate a lot of commentsand the analysis of these comments may require a lot of human resourcesTo solve this problememotional analysis or opinion mining can be used to analyze these comments