On stopwords, filtering and data sparsity for sentiment analysis of Twitter

Sentiment classification over Twitter is usually affected by the noisy nature (abbreviations, irregular forms) of tweets data. A popular procedure to reduce the noise of textual data is to remove stopwords by using pre-compiled stopword lists or more sophisticated methods for dynamic stopword identification. However, the effectiveness of removing stopwords in the context of Twitter sentiment classification has been debated in the last few years. In this paper we investigate whether removing stopwords helps or hampers the effectiveness of Twitter sentiment classification methods. To this end, we apply six different stopword identification methods to Twitter data from six different datasets and observe how removing stopwords affects two well-known supervised sentiment classification methods. We assess the impact of removing stopwords by observing fluctuations on the level of data sparsity, the size of the classifier's feature space and its classification performance. Our results show that using pre-compiled lists of stopwords negatively impacts the performance of Twitter sentiment classification approaches. On the other hand, the dynamic generation of stopword lists, by removing those infrequent terms appearing only once in the corpus, appears to be the optimal method to maintaining a high classification performance while reducing the data sparsity and substantially shrinking the feature space

Text mining na análise de sentimentos em contextos de big data

Dissertação de mestrado integrado em Engenharia e Gestão de Sistemas de Informação

Opinion Mining

In this thesis we study the field of opinion mining by giving a comprehensive review of the available research that has been done in this topic. Also using this available knowledge we present a case study of a multilevel opinion mining system for a student organization's sales management system. We describe the field of opinion mining by discussing its historical roots, its motivations and applications as well as the different scientific approaches that have been used to solve this challenging problem of mining opinions. To deal with this huge subfield of natural language processing, we first give an abstraction of the problem of opinion mining and describe the theoretical frameworks that are available for dealing with appraisal language. Then we discuss the relation between opinion mining and computational linguistics which is a crucial pre-processing step for the accuracy of the subsequent steps of opinion mining. The second part of our thesis deals with the semantics of opinions where we describe the different ways used to collect lists of opinion words as well as the methods and techniques available for extracting knowledge from opinions present in unstructured textual data. In the part about collecting lists of opinion words we describe manual, semi manual and automatic ways to do so and give a review of the available lists that are used as gold standards in opinion mining research. For the methods and techniques of opinion mining we divide the task into three levels that are the document, sentence and feature level. The techniques that are presented in the document and sentence level are divided into supervised and unsupervised approaches that are used to determine the subjectivity and polarity of texts and sentences at these levels of analysis. At the feature level we give a description of the techniques available for finding the opinion targets, the polarity of the opinions about these opinion targets and the opinion holders. Also at the feature level we discuss the various ways to summarize and visualize the results of this level of analysis. In the third part of our thesis we present a case study of a sales management system that uses free form text and that can benefit from an opinion mining system. Using the knowledge gathered in the review of this field we provide a theoretical multi level opinion mining system (MLOM) that can perform most of the tasks needed from an opinion mining system. Based on the previous research we give some hints that many of the laborious market research tasks that are done by the sales force, which uses this sales management system, can improve their insight about their partners and by that increase the quality of their sales services and their overall results.

Elaborazione del linguaggio naturale nell'IA e tecnologie moderne: Sentiment Analysis come caso di studio.

L'informatica e le sue tecnologie nella società moderna si riassumono spesso in un assioma fuorviante: essa, infatti, è comunemente legata al concetto che ciò che le tecnologie ci offrono può essere accessibile da tutti e sfruttato, all'interno della propria quotidianità, in modi più o meno semplici. Anche se quello appena descritto è un obiettivo fondamentale del mondo high-tech, occorre chiarire subito una questione: l'informatica non è semplicemente tutto ciò che le tecnologie ci offrono, perchè questo pensiero sommario fa presagire ad un'informatica "generalizzante"; l'informatica invece si divide tra molteplici ambiti, toccando diversi mondi inter-disciplinari. L'importanza di queste tecnologie nella società moderna deve spingerci a porre domande, riflessioni sul perchè l'informatica, in tutte le sue sfaccettature, negli ultimi decenni, ha portato una vera e propria rivoluzione nelle nostre vite, nelle nostre abitudini, e non di meno importanza, nel nostro contesto lavorativo e aziendale, e non ha alcuna intenzione (per fortuna) di fermare le proprie possibilità di sviluppo. In questo trattato ci occuperemo di definire una particolare tecnica moderna relativa a una parte di quel mondo complesso che viene definito come "Intelligenza Artificiale". L'intelligenza Artificiale (IA) è una scienza che si è sviluppata proprio con il progresso tecnologico e dei suoi potenti strumenti, che non sono solo informatici, ma soprattutto teorico-matematici (probabilistici) e anche inerenti l'ambito Elettronico-TLC (basti pensare alla Robotica): ecco l'interdisciplinarità. Concetto che è fondamentale per poi affrontare il nocciolo del percorso presentato nel secondo capitolo del documento proposto: i due approcci possibili, semantico e probabilistico, verso l'elaborazione del linguaggio naturale(NLP), branca fondamentale di IA. Per quanto darò un buono spazio nella tesi a come le tecniche di NLP semantiche e statistiche si siano sviluppate nel tempo, verrà prestata attenzione soprattutto ai concetti fondamentali di questi ambiti, perché, come già detto sopra, anche se è fondamentale farsi delle basi e conoscere l'evoluzione di queste tecnologie nel tempo, l'obiettivo è quello a un certo punto di staccarsi e studiare il livello tecnologico moderno inerenti a questo mondo, con uno sguardo anche al domani: in questo caso, la Sentiment Analysis (capitolo 3). Sentiment Analysis (SA) è una tecnica di NLP che si sta definendo proprio ai giorni nostri, tecnica che si è sviluppata soprattutto in relazione all'esplosione del fenomeno Social Network, che viviamo e "tocchiamo" costantemente. L'approfondimento centrale della tesi verterà sulla presentazione di alcuni esempi moderni e modelli di SA che riguardano entrambi gli approcci (statistico e semantico), con particolare attenzione a modelli di SA che sono stati proposti per Twitter in questi ultimi anni, valutando quali sono gli scenari che propone questa tecnica moderna, e a quali conseguenze contestuali (e non) potrebbe portare questa particolare tecnica.

Self-training from labeled features for sentiment analysis

Sentiment analysis concerns about automatically identifying sentiment or opinion expressed in a given piece of text. Most prior work either use prior lexical knowledge defined as sentiment polarity of words or view the task as a text classification problem and rely on labeled corpora to train a sentiment classifier. While lexicon-based approaches do not adapt well to different domains, corpus-based approaches require expensive manual annotation effort. In this paper, we propose a novel framework where an initial classifier is learned by incorporating prior information extracted from an existing sentiment lexicon with preferences on expectations of sentiment labels of those lexicon words being expressed using generalized expectation criteria. Documents classified with high confidence are then used as pseudo-labeled examples for automatical domain-specific feature acquisition. The word-class distributions of such self-learned features are estimated from the pseudo-labeled examples and are used to train another classifier by constraining the model's predictions on unlabeled instances. Experiments on both the movie-review data and the multi-domain sentiment dataset show that our approach attains comparable or better performance than existing weakly-supervised sentiment classification methods despite using no labeled documents.

Text Analytics of Social Media: Sentiment Analysis, Event Detection and Summarization

In the last decade, large numbers of social media services have emerged and been widely used in people's daily life as important information sharing and acquisition tools. With a substantial amount of user-contributed text data on social media, it becomes a necessity to develop methods and tools for text analysis for this emerging data, in order to better utilize it to deliver meaningful information to users. Previous work on text analytics in last several decades is mainly focused on traditional types of text like emails, news and academic literatures, and several critical issues to text data on social media have not been well explored: 1) how to detect sentiment from text on social media; 2) how to make use of social media's real-time nature; 3) how to address information overload for flexible information needs. In this dissertation, we focus on these three problems. First, to detect sentiment of text on social media, we propose a non-negative matrix tri-factorization (tri-NMF) based dual active supervision method to minimize human labeling efforts for the new type of data. Second, to make use of social media's real-time nature, we propose approaches to detect events from text streams on social media. Third, to address information overload for flexible information needs, we propose two summarization framework, dominating set based summarization framework and learning-to-rank based summarization framework. The dominating set based summarization framework can be applied for different types of summarization problems, while the learning-to-rank based summarization framework helps utilize the existing training data to guild the new summarization tasks. In addition, we integrate these techneques in an application study of event summarization for sports games as an example of how to better utilize social media data.

Identification of research trends in the field of separation processes. Application of epidemiological model, citation analysis, text mining, and technical analysis of the financial markets

Choice of industrial development options and the relevant allocation of the research funds become more and more difficult because of the increasing R&D costs and pressure for shorter development period. Forecast of the research progress is based on the analysis of the publications activity in the field of interest as well as on the dynamics of its change. Moreover, allocation of funds is hindered by exponential growth in the number of publications and patents. Thematic clusters become more and more difficult to identify, and their evolution hard to follow. The existing approaches of research field structuring and identification of its development are very limited. They do not identify the thematic clusters with adequate precision while the identified trends are often ambiguous. Therefore, there is a clear need to develop methods and tools, which are able to identify developing fields of research. The main objective of this Thesis is to develop tools and methods helping in the identification of the promising research topics in the field of separation processes. Two structuring methods as well as three approaches for identification of the development trends have been proposed. The proposed methods have been applied to the analysis of the research on distillation and filtration. The results show that the developed methods are universal and could be used to study of the various fields of research. The identified thematic clusters and the forecasted trends of their development have been confirmed in almost all tested cases. It proves the universality of the proposed methods. The results allow for identification of the fast-growing scientific fields as well as the topics characterized by stagnant or diminishing research activity.

Entity extraction e text classification come strumenti per la sentiment analysis di testi giornalistici

In this thesis we are going to talk about technologies which allow us to approach sentiment analysis on newspapers articles. The final goal of this work is to help social scholars to do content analysis on big corpora of texts in a faster way thanks to the support of automatic text classification.

Linguistic linked data for sentiment analysis

In this paper we describe the specification of amodel for the semantically interoperable representation of language resources for sentiment analysis. The model integrates "lemon", an RDF-based model for the specification of ontology-lexica (Buitelaar et al. 2009), which is used increasinglyfor the representation of language resources asLinked Data, with Marl, an RDF-based model for the representation of sentiment annotations (West-erski et al., 2011; Sánchez-Rada et al., 2013)

EUROSENTIMENT: Linked Data Sentiment Analysis

Sentiment and Emotion Analysis strongly depend on quality language resources, especially sentiment dictionaries. These resources are usually scattered, heterogeneous and limited to specific domains of appli- cation by simple algorithms. The EUROSENTIMENT project addresses these issues by 1) developing a common language resource representation model for sentiment analysis, and APIs for sentiment analysis services based on established Linked Data formats (lemon, Marl, NIF and ONYX) 2) by creating a Language Resource Pool (a.k.a. LRP) that makes avail- able to the community existing scattered language resources and services for sentiment analysis in an interoperable way. In this paper we describe the available language resources and services in the LRP and some sam- ple applications that can be developed on top of the EUROSENTIMENT LRP.

Evaluating EmotiBlog robustness for sentiment analysis tasks

EmotiBlog is a corpus labelled with the homonymous annotation schema designed for detecting subjectivity in the new textual genres. Preliminary research demonstrated its relevance as a Machine Learning resource to detect opinionated data. In this paper we compare EmotiBlog with the JRC corpus in order to check the EmotiBlog robustness of annotation. For this research we concentrate on its coarse-grained labels. We carry out a deep ML experimentation also with the inclusion of lexical resources. The results obtained show a similarity with the ones obtained with the JRC demonstrating the EmotiBlog validity as a resource for the SA task.

Incorporating sentiment prior knowledge for weakly supervised sentiment analysis

This article presents two novel approaches for incorporating sentiment prior knowledge into the topic model for weakly supervised sentiment analysis where sentiment labels are considered as topics. One is by modifying the Dirichlet prior for topic-word distribution (LDA-DP), the other is by augmenting the model objective function through adding terms that express preferences on expectations of sentiment labels of the lexicon words using generalized expectation criteria (LDA-GE). We conducted extensive experiments on English movie review data and multi-domain sentiment dataset as well as Chinese product reviews about mobile phones, digital cameras, MP3 players, and monitors. The results show that while both LDA-DP and LDAGE perform comparably to existing weakly supervised sentiment classification algorithms, they are much simpler and computationally efficient, rendering themmore suitable for online and real-time sentiment classification on the Web. We observed that LDA-GE is more effective than LDA-DP, suggesting that it should be preferred when considering employing the topic model for sentiment analysis. Moreover, both models are able to extract highly domain-salient polarity words from text.

Learning higher-level features with convolutional restricted Boltzmann machines for sentiment analysis

In recent years, learning word vector representations has attracted much interest in Natural Language Processing. Word representations or embeddings learned using unsupervised methods help addressing the problem of traditional bag-of-word approaches which fail to capture contextual semantics. In this paper we go beyond the vector representations at the word level and propose a novel framework that learns higher-level feature representations of n-grams, phrases and sentences using a deep neural network built from stacked Convolutional Restricted Boltzmann Machines (CRBMs). These representations have been shown to map syntactically and semantically related n-grams to closeby locations in the hidden feature space. We have experimented to additionally incorporate these higher-level features into supervised classifier training for two sentiment analysis tasks: subjectivity classification and sentiment classification. Our results have demonstrated the success of our proposed framework with 4% improvement in accuracy observed for subjectivity classification and improved the results achieved for sentiment classification over models trained without our higher level features.