Opinion mining is the task of extracting from a set of documents opinions expressed by a source on a specified target. This article presents a comparative study on the methods and resources that can be employed for mining opinions from quotations (reported speech) in newspaper articles. We show the difficulty of this task, motivated by the presence of different possible targets and the large variety of affect phenomena that quotes contain. We evaluate our approaches using annotated quotations extracted from news provided by the EMM news gathering engine. We conclude that a generic opinion mining system requires both the use of large lexicons, as well as specialised training and testing data.

Abstract — This paper addresses a relatively new text categorization problem: classifying a political blog as either ‘liberal ’ or ‘conservative’, based on its political leaning. Instead of simply using “Bag of Words ” features (BoW) as in previous work, we have explored subjectivity manifested in blogs and used subjectivity information thus found to help build political leaning classifiers. Specifically, our subjectivity based approach is two fold: 1) we identify subjective sentences that contain at least two strong subjective clues based on the General Inquirer dictionary; 2) from subjective sentences identified, we extract opinion expressions and BoW features to build political leaning classifiers. Experiments with a political blog corpus we built show that by using features from subjective sentences can significantly improve the classification performance. In addition, by extracting opinion expressions from subjective sentences, we are able to reveal opinions that are characteristic of a specific political orientation to some extent.

Automatic sentiment analysis in texts has attracted considerable attention in recent years. Most of the approaches developed to classify texts or sentences as positive or negative rest on a very specific kind of language resource: emotional lexicons. To build these resources, several automatic techniques have been proposed. Some of them are based on dictionaries while others use corpora. One of the main advantages of the corpora techniques is that they can build lexicons that are tailored for a specific application simply by using a specific corpus. Currently, only anecdotal observations and data from other areas of language processing plead in favour of the utility of specific corpora. This research aims to test this hypothesis. An experiment based on 702 sentences evaluated by judges shows that automatic techniques developed for estimating the valence from relatively small corpora are more efficient if the corpora used contain texts similar to the one that must be evaluated. 1.

Summary. This paper describes the part of a recommendation system designed for the recognition of film reviews (RRSS). Such a system allows the automatic collection, evaluation and rating of reviews and opinions of the movies. First the system searches and retrieves texts supposed to be movie reviews from the Internet. Subsequently the system carries out an evaluation and rating of the movie reviews. Finally, the system automatically associates a digital assessment with each review. The goal of the system is to give the score of reviews associated with the user who wrote them. All of this data is the input to the cognitive engine. Data from our base allows the making of correspondences, which are required for cognitive algorithms to improve, advanced recommending functionalities for e-business and e-purchase websites. In this paper we will describe the different methods on automatically identifying opinions using natural language knowledge and techniques of classification. 1 Introduction and

de opiniones con ontologías

ii Foreword Recent years have marked the beginning and expansion of the Social Web, in which people freely express and respond to opinion on a whole variety of topics. While the growing volume of subjective information available allows for better and more informed decisions of the users, the quantity of data to be analyzed imposed the development of specialized Natural Language Processing (NLP) systems that automatically detect subjectivity and sentiment in text and subsequently extract, classify and summarize the opinions available on different topics. Although the subjectivity and sentiment analysis research fields have been highly dynamic in the past years, dealing with subjectivity and sentiment in text has proven to be a complex, interdisciplinary problem that remains far from being solved. Its challenges include the need to address the issue from different perspectives and at different levels, depending on the characteristics of the textual genre, the language(s) treated and the final application for which the analysis is done. Inspired by the objectives we aimed at in the first edition of the Workshop on Computational Approaches to Subjectivity Analysis (WASSA 2010) and the final outcome, the purpose of the

This paper describes the preliminary results of a system for extracting sentiments opinioned with regard with named entities. It also combines rule-based classification, statistics and machine learning in a new method. The accuracy and speed of extraction and classification are crucial. The service oriented architecture permits the end-user to work with a flexible interface in order to produce applications that range from aggregating consumer feedback on commercial products to measuring public opinion on political issues from blog and forums. The experiment has two versions available for testing, one with concrete extraction results and sentiment calculus and the other with internal metrics validation results. 1

Treating sentiment analysis as a classification problem has proven extremely useful, but it misses the blended, continuous nature of sentiment expression in natural language. Using data from the Experience Project, we study texts as distributions over sentiment categories. Analysis of the document collection shows the texts contain blended sentiment information substantially different from a categorization view of sentiment. We introduce a statistical vector-space model that learns from distributions over emotive categories, in addition to capturing basic semantic information in an unsupervised fashion. Our model outperforms several baselines in predicting sentiment distributions given only the text of a document. 1

This paper focuses on the task of collocation polarity disambiguation. The collocation refers to a binary tuple of a polarity word and a target (such as ⟨long, battery life ⟩ or ⟨long, startup⟩), in which the sentiment orientation of the polarity word (“long”) changes along with different targets (“battery life ” or “startup”). To disambiguate a collocation’s polarity, previous work always turned to investigate the polarities of its surrounding contexts, and then assigned the majority polarity to the collocation. However, these contexts are limited, thus the resulting polarity is insufficient to be reliable. We therefore propose an unsupervised three-component framework to expand some pseudo contexts from web, to help disambiguate a collocation’s polarity.Without using any additional labeled data, experiments show that our method is effective. 1