Constructing a sentiment analysis model for LibQUAL+ comments

Abstract

Purpose The purpose of this paper is to establish a data mining model for performing sentiment analysis on open-ended qualitative LibQUAL+ comments, providing a further method for year-to-year comparison of user satisfaction, both of the library as a whole and individual topics. Design/methodology/approach A training set of 514 comments, selected at random from five LibQUAL+ survey responses, was manually reviewed and labeled as having a positive or negative sentiment. Using the open-source RapidMiner data mining platform, those comments provided the framework for creating library-specific positive and negative word vectors to power the sentiment analysis model. A further process was created to help isolate individual topics within the larger comments, allowing for more nuanced sentiment analysis. Findings Applied to LibQUAL+ comments for a Canadian mid-sized academic research library, the model suggested a fairly even distribution of positive and negative sentiment in overall comments. When filtering comments into affect of service, information control and library as place, the three dimensions’ relative polarity mirrored the results of the quantitative LibQUAL+ questions, with highest scores for affect of service and lowest for library as place. Practical implications The sentiment analysis model provides a complementary tool to the LibQUAL+ quantitative results, allowing for simple, time-efficient, year-to-year analysis of open-ended comments. Furthermore, the process provides the means to isolate specific topics based on specified keywords, allowing individual institutions to tailor results for more in-depth analysis. Originality/value To best account for library-specific terminology and phrasing, the sentiment model was created using LibQUAL+ open-ended comments as the foundation for the sentiment model’s classification process. The process also allows individual topics, chosen to meet individual library needs, to be isolated and independently analyzed, providing more precise examination.

Unveiling the research pattern and trends in library service quality studies: A meta-narrative review

“…Another qualitative study was conducted at a Canadian academic mid-sized academic library by using comments of LibQUAL+ surveys from the past 5 years, that is, 2008, 2010, 2013, 2015 and 2016. The data were analysed with the help of RapidMiner data mining software to perform sentiment analysis and findings have been categorized under three dimensions, that is, AS, IC, LP and the filtered results reflected highest score in AS, whereas lowest score in LP (Moore, 2017). There are few studies at different university libraries by using different sample, whose findings indicated that AS service dimension of LibQUAL+ tool considered as the top-rated and highly important dimension across all other dimensions of LibQUAL+ (Choshaly and Mirabolghasemi, 2018; Mallya and Payini, 2019; Ziaei and Korjan, 2018).…”

“…Regarding the sub-categories of user perception, there are various instances where the user perception was examined and explored on different samples from various libraries across the globe. Users' viewpoints/judgements/perceptions or their open-comments recorded through LibQUAL+ tool were explored to determine the service quality of libraries (Cabrerizo et al, 2017;De Jager, 2015;Guder, 2017;Kumar and Mahajan, 2019b;Luther, 2015;McCaffrey and Breen, 2016;Manuela La Fata and Lupo, 2017;Moore, 2017;Natesan et al, 2015;Sullo, 2019). Besides examining user perceptions through LibQUAL+ studies, researchers found evidence of ServQUAL studies too where the perceptions of library users were analysed by adopting TQM methodology and DEMATEL technique (Chen, 2016;Gathoni and Van der Walt, 2019).…”

Anatomy of Sentiment Analysis of Tweets Using Machine Learning Approach

“…For example, the sentence "I find this MP4 player really useful" expresses a sentiment about the entity i.e., mp3 player. The target of the entity could be a person, a product, an organisation, or an event, among others [21]. • In its simplest form, the sentiment polarity is the degree of expressing a sentiment that can be negative or positive.…”

Principles and practice in impact assessment for academic libraries

“…Qualitative data analysis can be time consuming and coding needs to be reliable, as an analysis of interactive chat transcripts, studying changes in user satisfaction with librarian behaviour indicates (Baumgart et al , 2014). Techniques such as sentiment analysis might identify trends in perception and value estimations, as long as there are suitable data sets containing free-text comments about library services (Moore, 2017 describes sentiment analysis of some LibQUAL+ data using RapidMiner). More work is probably necessary to develop specific algorithms for specialized topics (Thelwall and Buckley, 2013).…”