About the project

Our project is fantastic!

The Reviewland project focuses on extracting, elaborating, and interpretating online users' reviews on top-ranked and top-visited e-advise websites, such as Booking.com and Tripadvisor. We will provide refined information regarding tourism, offered services and products, especially centering the attention on the territory of Lucca, a popular medieval town located in Tuscany, Italy.

Reviewland aims to extract - from online reviews - valuable information describing the perception of the services and the products available in Lucca, making then such information easily accessible by both consumers and service providers.

Reviewland handles the users' online reviews, providing the analysis of a big mole of data and the disclosure of hidden patterns, exploiting state-of-art data-mining techniques and natural language processing.

The project started in August, 2016, and it will last 18 months. It is partially funded by Fondazione Cassa di Risparmio di Lucca.

Analyses and results

In this section, we show a list of analyses, and analyses results, achieved during the first year of the project (September 2016 - August 2017)

First Year Report (in Italian)

Data Tables: In this section, we provide data tables in CSV format, that has been produced during the analysis

Evaluations for the province of Lucca, according to the following dimensions:

Tripadvisor dataset: Analysis on the number of reviews

Text analysis

Tripadvisor: Number of occurrences of stemmed words, reviews about Lucca:

Booking: Number of occurrences for stemmed words, reviews about Lucca:

Booking: Top 10 words, Lucca, Paris and New York

Tables for frequent sets of localities that are reviewed togheter:

Publications acknowledging ReviewLand:

Michela Fazzolari, Marinella Petrocchi, Angelo Spognardi: Predicting online review scores across reviewer categories. In submission: Knowledge and Information Systems, Springer

Michela Fazzolari, Massimiliano La Gala, Marinella Petrocchi: Enriching traveling information through online reviews analyses. In submission: Information Technology and People, Emerald Publishing

Michela Fazzolari, Marinella Petrocchi, Alessandro Tommasi, and Cesare Zavattari: Mining Worse and Better Opinions: Unsupervised and Agnostic Aggregation of Online Reviews (arXiv version) 17th International Conference on Web Engineering, Rome, June 2017.

Stefano Cresci, Roberto Di Pietro, Marinella Petrocchi, Angelo Spognardi, and Maurizio Tesconi: Exploiting digital DNA for the analysis of similarities in Twitter behaviours (Tech. Rep. version) 4th IEEE International Conference on Data Science and Advanced Analytics - DSAA 2017

Stefano Cresci, Roberto Di Pietro, Marinella Petrocchi, Angelo Spognardi, Maurizio Tesconi: Social fingerprinting: detection of spambot groups through DNA-inspired behavioral modeling (arXiv version) IEEE Transactions on Dependable and Secure Computing, In press (2017)

Stefano Cresci, Roberto Di Pietro, Marinella Petrocchi, Angelo Spognardi, and Maurizio Tesconi: The paradigm-shift of social spambots: Evidence, theories, and tools for the arms race (arXiv version) World Wide Web 2017