Good location, terrible food: detecting feature sentiment in user-generated reviews

Abstract

A growing corpus of online informal reviews is generated every day by non-experts, on social networks and blogs, about an unlimited range of products and services. Users do not only express holistic opinions, but often focus on specific features of their interest. The automatic understanding of "what people think" at the feature level can greatly support decision making, both for consumers and producers. In this paper, we present an approach to feature-level sentiment detection that integrates natural language processing with statistical techniques, in order to extract users' opinions about specific features of products and services from usergenerated reviews. First, we extract domain features, and each review is modelled as a lexical dependency graph. Second, for each review, we estimate the polarity relative to the features by leveraging the syntactic dependencies between the terms. The approach is evaluated against a ground truth consisting of set of usergenerated reviews, manually annotated by 39 human subjects and available online, showing its human-like ability to capture feature-level opinions.

Geo‐text data and data‐driven geospatial semantics

“…Wang et al () performed text mining on geotagged tweets in the response to a wildfire and detected the attitudes of people, such as their appreciation to fire fighters. Looking into TripAdvisor hotel reviews, Cataldi et al () proposed an approach for detecting the sentiments of people toward different aspects of a hotel, such as its location convenience and food quality. Wang and Zhou () performed sentiment analysis on TripAdvisor hotel reviews within the same city and found that spatial dependence exists in the satisfaction of customers.…”

Advertising recommendation system based on dynamic data analysis on Turkish speaking Twitter users

“…Bosch et al have developed a system that conducts real-time filtering in order to acquire the contents that appeal to the customer out of the share clusters the user is exposed to [10]. Based on the shares that are made, Cataldi et al have tried to make inferences about the issues on agenda [11]. On the other hand, Kang et al sought to determine the connection between the events in addition to the event inferences [12].…”

Onward Ho! Studying Place Emotions, Language Style, and Social Tendencies in Travel Blogs

“…This strand of natural language processing aims at the classification of a textual document into a subjective category (e.g., good/bad) or along a single continuous dimension from negative to positive [12]. Domain-specific sentiment analysis techniques have been tailored to online reviews of hotels and travel destinations, adopting syntactic parsing [5], as well as supervised machine learning approaches [20]. Opinions can be explicit or implicit in a text, the latter being especially difficult to extract from simple methods based on syntactic matching.…”

“…The measurement of public sentiment about a given topic in such outlets has already tangible applications in politics, marketing, and business analytics [10]. However, existing techniques aim at quantifying sentiment on a negative-positive spectrum, ignoring more nuanced aspects of emotions [5]. While some research has targeted this corpus from a geographic perspective [14,1], no attempt has been made to explore places with respect to their emotional structure.…”