The tourism market is constantly evolving and growing thanks to the technological innovation that allows us to create new opportunities. An increasing number of people use the internet to plan their trip, for pleasure or work necessities; as a result, the role of reviews and social media in decision-making processes is becoming increasingly important. In particular, It is estimated that 95% of travelers consult the reviews before booking a hotel, and they are strongly used to find a restaurant or an attraction. However, day after day both travelers and residents of a city, find themselves faced with the widespread problem of information overload that leads them to read a large number of reviews before they can find what they need, and often these opinions are in conflict with each other. This situation involves a considerable loss of time and energy. Multiple researches show that users ask for a simplification of content and the born of numerous travel start-ups (1,497 since 2005) confirms the broad possibilities of growth in this market. In particular the start-ups highlighted here allow the personalization of the experience including services such as tour booking and activities, creation of customized routes and integration with the weather; in addition they provide a summary of the topics. In B2B, services are offered to manage and monitor the online reputation of their business. This study extracts and examines, throughout text processing techniques, text summarization and sentiment analysis with R, the TripAdvisor reviews of restaurants, hotels and places of interest with the aim of integrating them into a platform that can improve the experience of tourists and residents; this could simplify the decision making-process by providing a clear and easy-to-read interface that contains primary information and both positive and negative data concerning a given activity. Specifically, the data collected is based on the sample of the top 10 restaurants, hotels and places of interest in the city of Toronto and the result obtained is an interesting starting point for future projects that, in addition to expanding the dataset, integrates other review platforms in the tourism and non-tourism sectors.
Il mercato turistico è in continua evoluzione e crescita grazie all’innovazione tecnologica che permette di creare nuove opportunità. Un numero sempre maggiore di persone, per piacere o per necessità lavorative, fa ricorso ad internet per pianificare il proprio viaggio; di conseguenza acquista sempre più importanza il ruolo delle recensioni e dei social media nei processi decisionali. In particolare si stima che il 95% dei viaggiatori consulta le recensioni prima di prenotare un hotel, ed esse sono fortemente utilizzate per la ricerca di un ristorante o di un’attrazione. Tuttavia, giorno dopo giorno i viaggiatori, ma anche i residenti in una data città, si ritrovano ad affrontare il dilagante problema dell’Information Overload che li porta a dover leggere un gran numero di recensioni prima di riuscire a trovare ciò di cui hanno bisogno e spesso queste opinioni sono in contrasto tra loro. Tutto ciò comporta un notevole dispendio di tempo ed energie. Molteplici ricerche mostrano come gli utenti richiedano una semplificazione dei contenuti e l’avvento di numerose Start-up in ambito travel (1.497 dal 2005) conferma le ampie possibilità di crescita in questo mercato. In particolare le Start-up qui evidenziate consentono, oltre a fornire una sintesi degli argomenti, la personalizzazione dell’esperienza includendo servizi quali prenotazione di tour e attività, creazione di percorsi personalizzati e integrazione con il meteo. In abito B2B vengono offerti invece servizi per gestire e monitorare la reputazione online della propria attività. Questo studio estrae ed esamina, tramite tecniche di text processing, text summarization e sentiment analysis con R, le recensioni di Tripadvisor di ristoranti, hotel e luoghi di interesse con lo scopo di integrarle in una piattaforma che possa migliorare l’esperienza di turisti e residenti, semplificando il processo decisionale, fornendo un’interfaccia chiara e di facile lettura che contenga informazioni primarie e dati positivi e negativi riguardanti una data attività. Specificamente i dati raccolti si basano sul campione dei primi 10 ristoranti, hotel e luoghi di interesse della città di Toronto Il risultato così ottenuto rappresenta un interessante spunto per futuri progetti che, oltre ad espandere il dataset, integrino altre piattaforme di recensioni in ambito turismo e non.
A study on implementation of a platform based on extraction of useful information from human-generated content on TripAdvisor
ROSOLIA, GIUSEPPE DARIO
2017/2018
Abstract
The tourism market is constantly evolving and growing thanks to the technological innovation that allows us to create new opportunities. An increasing number of people use the internet to plan their trip, for pleasure or work necessities; as a result, the role of reviews and social media in decision-making processes is becoming increasingly important. In particular, It is estimated that 95% of travelers consult the reviews before booking a hotel, and they are strongly used to find a restaurant or an attraction. However, day after day both travelers and residents of a city, find themselves faced with the widespread problem of information overload that leads them to read a large number of reviews before they can find what they need, and often these opinions are in conflict with each other. This situation involves a considerable loss of time and energy. Multiple researches show that users ask for a simplification of content and the born of numerous travel start-ups (1,497 since 2005) confirms the broad possibilities of growth in this market. In particular the start-ups highlighted here allow the personalization of the experience including services such as tour booking and activities, creation of customized routes and integration with the weather; in addition they provide a summary of the topics. In B2B, services are offered to manage and monitor the online reputation of their business. This study extracts and examines, throughout text processing techniques, text summarization and sentiment analysis with R, the TripAdvisor reviews of restaurants, hotels and places of interest with the aim of integrating them into a platform that can improve the experience of tourists and residents; this could simplify the decision making-process by providing a clear and easy-to-read interface that contains primary information and both positive and negative data concerning a given activity. Specifically, the data collected is based on the sample of the top 10 restaurants, hotels and places of interest in the city of Toronto and the result obtained is an interesting starting point for future projects that, in addition to expanding the dataset, integrates other review platforms in the tourism and non-tourism sectors.| File | Dimensione | Formato | |
|---|---|---|---|
|
2018_07_Rosolia.pdf
solo utenti autorizzati dal 13/07/2019
Descrizione: Testo della tesi
Dimensione
7.37 MB
Formato
Adobe PDF
|
7.37 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/142120