A Tool-Based Methodology to Analyze Social Network Interactions in Cultural Fields: The Use Case “MuseumWeek”

Analysing Cultural Events on Twitter

“…A period before the event and a period during the event. This distinction is an empirical consequence of our studies on Museum Week 2014 et 2015 [3].…”

“…An exception worth mentioning is the work presented in Lovejoy and Saxton [11] in which the authors (Twitter users) analyze the global behavior of nonprofit organizations on Twitter based on three communication classes: Information, Community and Action classes. Recently, several studies on tweet classification have been carried out in NLP [9,1] but to the best of our knowledge, only [3] has classified cultural institutional tweets in communication categories based on NLP techniques.…”

“…The features selected were semio-linguistic (mostly lexical, but also including punctuation marks, emoticons), tweet-specific features (for example, the presence / absence of hashtags in tweets) as well as metadata such as the identity of the account. In the second stage, a classifier was built based on a corpus of 1000 tweets annotated by hand by cultural experts according to the categories defined in the previous step [3]. The classifier is based on the Naive Bayes and SVM models, with unanimous vote.…”

The Museum Week Event: Analyzing Social Network Interactions in Cultural Fields

“…First, we would like to summarize our tool-based methodology [17] to analyze social network interaction. In the first step, based on a Tuser taxonomy built by experts in the studied field, all the tweet accounts are categorized.…”

An associative engines based approach supporting collaborative analytics in the Internet of cultural things

“…International Workshops and continuous research tracks have been devoted to Tweet Contextualization [10], and related geographic information received much attention: in [15] authors propose and evaluate a probabilis-135 tic framework for estimating a Twitter user's city-level location based purely on the content of the user's tweets, while in [16] geo-space relationships and proximity influence are associated to tweet contents, since people in close geographic proximity can provide real-time information and eyewitness updates for one another about events of local interest. In the domain of Cultural Her-140 itage, the introduction of #culturalheritage in the official Twitter hashtag set has provided evidence of the increasing interest of users about this topic, and an interesting study appeared about tweets usage for Museum activities and events promotions [17]. A recent study [5] reports on the association of indoor with outdoor activities, in order to both suggest possible venues and provide 145 associated content at opportune location and times in novel ways; in this paper we rather explore the associations of information coming from territorial visits and on-site events by focusing the analysis on CH-related tweets, based on their semantic, geographic and temporal attributes.…”