Article Preview
TopIntroduction
In the era of Big Data users-generated data are increasingly collected, stored and analyzed by modern organizations. These data typically describe different dimensions of the daily social life and are the heart of a knowledge society, where the understanding of complex social phenomena is sustained by the knowledge extracted from the miners of big data across the various social dimensions by using mining technologies.
Social media shatters the boundaries between the real world and the virtual world. Actually are available technologies and capabilities to integrate social theories with computational methods. The uniqueness of social media data calls for novel data mining techniques that can effectively handle user generated content with rich social relations. The study and development of these new techniques are under the purview of social media mining, an emerging discipline under the umbrella of data mining. Social Media Mining is the process of representing, analyzing, and extracting actionable patterns from social media data, encompassing the tools to formally represent, measure, model, and mine meaningful patterns from large-scale social media data.
It is well versed in social and computational theories, specialized to analyze recalcitrant social media data, and aims to fill in the gap from what we know (social and computational theories) to what we want to know about the vast social media world with computational tools. Social media mining is, anyway, an emerging field where there are more problems than ready solutions.
A challenging and useful direction may be the relatively new character of combining social media pulses, and more generally User Content Generation (UGC), with business intelligence, aimed to reveal the underlying parameters that determine the performance of an organization. In order to understand the influence of social media contents on an organization performance, key performance indicators (“KPIs”) are required and links between such KPIs and social media parameters and phenomena have to be taken in account, because KPIs measure the performance of an organization with respect to its strategy.
Cultural Heritage (CH) is among the sectors which could advantage from social business intelligence, driving cultural operators and organizations to design and implement novel strategies for engaging and attracting more visitors. The increasing interest into such a topic was empirically evidenced in the Social Network Twitter, where users introduced in 2015 the hashtag #culturalheritage to refer clearly cultural related messages.
The opportunities offered by linking both concepts are acknowledged in the recent scientific literature, but a relative underexposure of quantitative studies devoted to assess the potential and effectiveness of social pulses and activities for supporting organizations promoting Cultural Heritage was recorded. In this work a Big Data driven approach is proposed, exploiting the computing capabilities of advanced associative and in-memory technologies, such as the Qlik View Engine (QlikTech International, 2015) to perform Social Big Data mining efficiently. Our attention is focused on textual data produced by the micro blogging platform Twitter; so, a specific issue faced in this research is the text mining over Twitter messages, in order to extract meaningful information useful to infer, preliminary, the social users’ trends in their affection and sensitivity towards Cultural topics and Events. Cultural Sensitivity Mining is inferred by calculating a set of KPIs, early proposed in (Chianese, 2016), as the result of a text mining process, exploiting semantic and lexical resources (cultural domain specific and more general ontologies). The calculation of such KPIs provides a quantitative estimation of Cultural Heritage Sensitivity as expressed by social network users.
Furthermore, advanced features for text elements classification (e.g., the most popular hashtags in the CH context) and opinion and sentiment polarity estimation are performed.
Huge datasets of tweets, covering a time horizon lasting over the whole 2015 and geo-localized in Italy, are processed combining Natural Language Processing (NLP) techniques, semantic technologies, geo-referencing and temporal analysis. Examples of computed measures for characterizing people’s interest and sensitivity into CH subjects, are geographical density of CH resources and temporal proximity to CH-related events.