Opinion Mining Framework


Opinion Mining Framework


Franco Tuveri, valorisation@crs4.it


Nowadays, information and communications technologies strongly influence the way in which we use, produce and exchange words. This affects the behaviour of companies, their ability to make profit, their organization, the production of goods, services and innovation and the benefits offered to consumers. The growth of unstructured data sources, such as online reviews, posts and social media conversations, and structured data sources, such as stakeholders letters, strategic documents and patents, did not generate a better understanding of the reality in which they operate, but rather made more complex the work of analysts.

New automatic tools able to exploit the power of words are necessary to answer the following questions:

  • How the words used in conversations and strategic documents influence individual and organizational performances?
  • Through which strategies, the ability of companies to analyze and exploit the words may constitute a new source of competitive advantage?


The framework, still under development, is designed to support applications of semantic text analysis, by means of opinion mining, brand reputation, document classification, contents analysis and user profiling tools.
The most common usage contexts are:

  • Companies: Social Media Intelligence tools to monitor, create interest, analyze and extract contents from public conversations in social media, forums and notice boards.
  • Policy makers: tools for territory promotion by means of the tourism demand forecasting at local and regional level, based on the interpretation of social data related to various venues such as accommodations, typical restaurants, cultural heritage and other attractions.

Innovative features

  • opinion analysis: analyze opinions related to events or facts also under way, even when they are not related to specific topics;
  • features extraction: extract aspects and significant information contained in the opinions, related to different contexts not always well defined, starting from multiple sources of reviews;
  • domain specific features: contextualize the features through the use of tools for the semantic classification, the management of semantic networks, and the use of ad-hoc linguistic resources;
  • reporting (opinion summarization): aggregate and represent the processed results so that they become useful information in decision making.

Potential users

Developers, Business managers, Policy makers, Customers, Citizens.

Impact sectors

Tourism - Business and Trade - Industries and Market Research Organizations – Policy making

Other resources

  1. Manuela Angioni, Maria Laura Clemente, Franco Tuveri. Evaluating Potential Improvements of Collaborative Filtering with Opinion Mining. ICEIS 2015 - Proceeedings of the 17th International Conference on Enterprise Information Systems – 2015
  2. Manuela Angioni, Maria Laura Clemente, Franco Tuveri. Combining Opinion Mining with Collaborative Filtering. WEBIST 2015 - 11th International Conference on Web Information Systems and Technologies – 2015
  3. Manuela Angioni, Franco Tuveri. An Opinion Mining Model for Generic Domains. Springer Berlin Heidelberg, Volume 515, page 51--64 - 2014
  4. Manuela Angioni, Franco Tuveri. A Linguistic Approach to Opinion Mining. Springer Berlin Heidelberg, Volume 439, page 113--129 - 2012

Questo sito utilizza cookie tecnici e assimilati. Possono essere presenti anche cookie profilazione di terze parti. Se vuoi saperne di più o negare il consenso a tutti o ad alcuni cookie leggi l'informativa completa. Proseguendo nella navigazione (anche con il semplice scrolling) acconsenti all'uso dei cookie. This site uses technical and anonymized analytics cookies only. There may also be profiling third-party cookies. Please read the cookie information page to learn more about how we use cookies or blocking them. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.