Visualizing museums through the visitors’ eye: An n-gram model-based text analysis approach

Abstract

All over the world Museums are developing a Customer Service mindset characteristic of large businesses. Part of this shift can be seen in the rapid adoption of museum visitor surveys that gather feedback about a visitor’s experience in relation to their demographic characteristics — such as likes, dislikes, etc. as per their age, gender and income. The concept of listening to the visitor through such census data collection is not new, it was there from the very beginning, but in certain aspects, it appears to be deficient. For example, it becomes arduous and difficult to identify if the visitors find the place “memorable”. Did they feel “welcome” from the beginning? Do they think “science” is given a due share in that science museum? Can we build an at-a-glance graphical representation of the museum’s image in the visitor’s mental map? Census type data collection is very manpower intensive and is often biased as the visitors feel righteous to answer the questionnaire more “correctly” than “frankly”. They are not generally considered “easy in-person survey tools”. In this paper, we will discuss a very new way of listening to the visitors and reconstructing their free and fair post-visit overall mental map of the museum. In this new effort, we have tried to construct an automated AI-based word-cloud image from visitors’ feedback offered voluntarily in social networks like, say, in Google review. This image is helpful in summarizing and analysing a large amount of feedback data on a single page by focusing on the keywords and phrases adopted by the visitor in describing her experience. This approach is especially useful for centralized analysis of feedback data on museums having a large number of branches. Hence, we have applied this method to process feedback on the National Council of Science Museums, the largest body of science museums in India managing twenty-five centres all over the country. For the very first time applying this method, we have been able to visualize our museums through the visitor’s eye rendering the analytic result in the form of a word-cloud that reflects what the visitors fancy most when they recall the museum experience after the visit.